==================================================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_006/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: other_op
Start group testing...

ops group_cases_0 with 162 cases start to running, all cases are below:
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.chunk')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.gather')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.cat')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.cumsum')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.index_select')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.index_select')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.masked_select')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.split')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.split')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nonzero')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.flatten')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.interpolate(mode="bilinear")')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.interpolate(mode="trilinear")')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.interpolate(mode="bicubic")')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.interpolate(mode="linear")')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.interpolate(mode="nearest")-1d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.interpolate(mode="nearest")-2d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.interpolate(mode="nearest")-3d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.repeat_interleave')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.repeat_interleave')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.repeat')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.arange')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.select')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.select')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.one_hot')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.reshape')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.matmul')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.broadcast_to')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.matmul')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.BatchNorm1d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.BatchNorm2d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.BatchNorm3d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.batch_norm')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.binary_cross_entropy')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.binary_cross_entropy_with_logits')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.l1_loss')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.log_softmax')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.mse_loss')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.softmax')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.unique')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.unique')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.clamp')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.clamp')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.stack')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.gather')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.expand_as')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.scatter')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.dropout')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.pad(mode="constant")')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.pad(mode="reflect")')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.pad(mode="replicate")')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.pad(mode="circular")')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.conv1d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.conv2d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.conv3d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.linear')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.Linear')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.Conv1d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.masked_scatter')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.masked_scatter_')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.add_')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.layer_norm')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.GroupNorm')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.LayerNorm')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.permute')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.transpose')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.unsqueeze')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.topk')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.normal')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.adaptive_avg_pool1d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.adaptive_avg_pool2d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.adaptive_avg_pool3d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.avg_pool2d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.functional.max_pool2d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.AdaptiveAvgPool2d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'mint.nn.AdaptiveAvgPool3d')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.transpose')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.unsqueeze')
case: (<function test_other_op_reference_forward at 0xffff979c13a0>, 'pynative', 'Tensor.topk')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.chunk')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.gather')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.cat')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.cumsum')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.index_select')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.index_select')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.masked_select')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.split')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.split')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nonzero')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.flatten')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.interpolate(mode="bilinear")')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.interpolate(mode="trilinear")')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.interpolate(mode="bicubic")')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.interpolate(mode="linear")')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.interpolate(mode="nearest")-1d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.interpolate(mode="nearest")-2d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.interpolate(mode="nearest")-3d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.repeat_interleave')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.repeat_interleave')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.repeat')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.arange')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.select')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.select')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.one_hot')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.reshape')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.matmul')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.broadcast_to')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.matmul')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.BatchNorm1d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.BatchNorm2d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.BatchNorm3d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.batch_norm')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.binary_cross_entropy')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.binary_cross_entropy_with_logits')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.l1_loss')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.log_softmax')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.mse_loss')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.softmax')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.unique')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.unique')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.clamp')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.clamp')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.stack')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.gather')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.expand_as')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.scatter')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.dropout')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.pad(mode="constant")')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.pad(mode="reflect")')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.pad(mode="replicate")')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.pad(mode="circular")')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.conv1d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.conv2d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.conv3d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.linear')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.Linear')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.Conv1d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.masked_scatter')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.masked_scatter_')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.add_')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.layer_norm')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.GroupNorm')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.LayerNorm')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.permute')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.transpose')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.unsqueeze')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.topk')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.normal')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.adaptive_avg_pool1d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.adaptive_avg_pool2d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.adaptive_avg_pool3d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.avg_pool2d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.functional.max_pool2d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.AdaptiveAvgPool2d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'mint.nn.AdaptiveAvgPool3d')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.transpose')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.unsqueeze')
case: (<function test_other_op_reference_backward at 0xffff979c1430>, 'pynative', 'Tensor.topk')
case: (<function test_other_op_reference_forward_kbk at 0xffff979c14c0>, 'kbk', 'mint.chunk')
case: (<function test_other_op_reference_forward_kbk at 0xffff979c14c0>, 'kbk', 'mint.gather')
case: (<function test_other_op_reference_backward_kbk at 0xffff979c1550>, 'kbk', 'mint.chunk')
case: (<function test_other_op_reference_backward_kbk at 0xffff979c1550>, 'kbk', 'mint.gather')
ops group_cases_0 total running memory: 2592M, memory threshold: 51200M

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

op_name: mint.split, mode:pynative, test_op_reference...

op_name: mint.index_select, mode:pynative, test_op_reference...

op_name: mint.cumsum, mode:pynative, test_op_reference...

op_name: mint.masked_select, mode:pynative, test_op_reference...

op_name: mint.gather, mode:pynative, test_op_reference...

op_name: mint.chunk, mode:pynative, test_op_reference...

op_name: mint.cat, mode:pynative, test_op_reference...

op_name: Tensor.split, mode:pynative, test_op_reference...
..[LOG_WARNING] can not open file, file: /home/jenkins/ascend/log/debug/plog/plog-169938_20260129173655645.log, possible reason: Permission denied..........[LOG_WARNING] can not open file, file: /home/jenkins/ascend/log/debug/plog/plog-169938_20260129173655645.log, possible reason: Permission denied................................................[LOG_WARNING] can not open file, file: /home/jenkins/ascend/log/debug/plog/plog-169938_20260129173655645.log, possible reason: Permission denied.........[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:01.492.208 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdd01b3a60> should be wrapped by jit only once.

TotalTime = 0.0239473, [30]
[bootstrap]: 0.0004718
[type_inference]: 0.0111556
[event_method]: 2.156e-05
[auto_monad]: 0.00013745
[graph_reusing]: 6.21e-06
[pre_auto_parallel]: 1.05e-05
[py_interpret_to_execute]: 4.321e-05
[rewriter_before_opt_a]: 0.00030872
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.00919996, [2]
    [Cycle 1]: 0.00190157, [27]
        [switch_simplify]: 0.00020801
        [loop_unroll]: 3.587e-05
        [a_1]: 0.00063572
        [with_stream_mark]: 1.788e-05
        [recompute_prepare]: 7.78999e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 3.9e-06
        [updatestate_loads_eliminate]: 3.11001e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.89999e-06
        [updatestate_useless_node_eliminater]: 5.94999e-06
        [accelerated_algorithm]: 6.24999e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 6.26e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.741e-05
        [j_node_and_user_rematch]: 1.003e-05
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 1.097e-05
        [inline_without_move]: 6.02999e-06
        [renormalize]: 0.00063813
        [add_forward_monad_depend]: 2.28e-05
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.513e-05
        [cse]: 3.312e-05
        [replace_applicator]: 1.2e-05
    [Cycle 2]: 0.00035164, [27]
        [switch_simplify]: 6.64999e-06
        [loop_unroll]: 5.67001e-06
        [a_1]: 0.00011466
        [with_stream_mark]: 8.2e-06
        [recompute_prepare]: 5.59e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 6.16e-06
        [accelerated_algorithm]: 5.66003e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 2.61999e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.321e-05
        [j_node_and_user_rematch]: 8.63001e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 8.82e-06
        [inline_without_move]: 5.48002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.79999e-06
        [cse]: 1.233e-05
        [replace_applicator]: 6.06e-06
[py_interpret_to_execute_after_opt_a]: 1.057e-05
[rewriter_after_opt_a]: 0.00015444
[convert_after_rewriter]: 9.322e-05
[order_py_execute_after_rewriter]: 6.79001e-06
[mutable_eliminate]: 0.00053402
[jit_opt_b]: 5.424e-05, [1]
    [Cycle 1]: 4.771e-05, [2]
        [frontend_op_eliminate]: 1.878e-05
        [inline_after_opt_a]: 1.746e-05
[cconv]: 2.343e-05
[loop_unroll]: 0.00041876
[jit_opt_after_cconv]: 0.00014996, [1]
    [Cycle 1]: 0.00014382, [11]
        [c_1]: 2.474e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.59e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [cse]: 2.12e-05
        [call_graph_tuple_transform]: 2.007e-05
        [tuple_list_get_item_eliminator]: 6.38003e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 6.23e-06
[remove_dup_value]: 1.494e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 2.322e-05
[add_recomputation]: 5.261e-05
[cse_after_recomputation]: 2.521e-05, [1]
    [Cycle 1]: 1.981e-05, [1]
        [cse]: 1.341e-05
[auto_monad_reorder]: 2.422e-05
[get_jit_bprop_graph]: 1.66e-06
[rewriter_after_jit_bprop_graph]: 0.00014864
[opt_after_jit_grad]: 0.00049008
[symbol_engine_optimizer]: 7.594e-05, [1]
    [Cycle 1]: 6.986e-05, [6]
        [build]: 3.67002e-06
        [elim_shapecalc]: 8.65001e-06
        [elim_not_effective]: 1.343e-05
        [opt_reshape]: 6.33998e-06
        [fold_const_symbol]: 9.09e-06
        [renormalize]: 4.19997e-07
[validate]: 8.536e-05
Sums
    bootstrap                            :     0.000472s :  2.88%
    type_inference                       :     0.011156s : 68.06%
    event_method                         :     0.000022s :  0.13%
    auto_monad                           :     0.000137s :  0.84%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000011s :  0.06%
    py_interpret_to_execute              :     0.000043s :  0.26%
    rewriter_before_opt_a                :     0.000309s :  1.88%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000215s :  1.31%
    jit_opt_a.loop_unroll                :     0.000042s :  0.25%
    jit_opt_a.a_1                        :     0.000750s :  4.58%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.16%
    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        :     0.000003s :  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 :  0.02%
    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  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    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        :     0.000012s :  0.07%
    jit_opt_a.renormalize                :     0.000638s :  3.89%
    jit_opt_a.add_forward_monad_depend   :     0.000024s :  0.15%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000045s :  0.28%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000154s :  0.94%
    convert_after_rewriter               :     0.000093s :  0.57%
    order_py_execute_after_rewriter      :     0.000007s :  0.04%
    mutable_eliminate                    :     0.000534s :  3.26%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.11%
    cconv                                :     0.000023s :  0.14%
    loop_unroll                          :     0.000419s :  2.55%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.15%
    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 :     0.000003s :  0.02%
    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 :     0.000006s :  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 :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000023s :  0.14%
    add_recomputation                    :     0.000053s :  0.32%
    cse_after_recomputation.cse          :     0.000013s :  0.08%
    auto_monad_reorder                   :     0.000024s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000149s :  0.91%
    opt_after_jit_grad                   :     0.000490s :  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 :     0.000013s :  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                             :     0.000085s :  0.52%


Time group info:
------[substitution.]   0.000227    28
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000006s :      4: substitution.graph_param_transform
82.86% :     0.000188s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000005s :      4: substitution.replace_old_param
 7.76% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011084     2
90.89% :     0.010074s :      1: type_inference.infer
 9.11% :     0.001010s :      1: type_inference.specialize
------[replace.]   0.000084     8
57.09% :     0.000048s :      4: replace.inline
42.91% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000201     8
92.17% :     0.000185s :      4: match.inline
 7.83% :     0.000016s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     13: predicate.addn_check_dump
 1.28% :     0.000002s :     13: predicate.addn_zero_filter
 1.85% :     0.000003s :     13: predicate.arithmetic_simplify
 1.41% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     13: predicate.compare_switch_simplify
 1.39% :     0.000002s :     13: predicate.depend_value_elim
 1.20% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      4: predicate.elim_not_effective
 0.60% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.46% :     0.000002s :     13: predicate.environ_get_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.98% :     0.000001s :      8: predicate.get_grad_eliminate
 0.37% :     0.000000s :      4: predicate.graph_param_transform
 5.57% :     0.000008s :     29: predicate.inline
 0.98% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      8: predicate.less_batch_normalization
 1.71% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.03% :     0.000003s :     21: predicate.load_eliminater
 1.27% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.06% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     13: predicate.merge_addn
 1.17% :     0.000002s :     13: predicate.minmaximum_grad
 1.47% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.50% :     0.000003s :     21: predicate.partial_eliminate
 1.23% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     13: predicate.reduce_eliminate
 1.67% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.91% :     0.000003s :     25: predicate.replace_applicator
 0.63% :     0.000001s :      8: predicate.replace_old_param
 0.31% :     0.000000s :      4: predicate.reset_defer_inline
 1.28% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.41% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     21: predicate.switch_defer_inline
 2.25% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.84% :     0.000011s :     67: predicate.switch_simplify
 1.24% :     0.000002s :     13: predicate.tile_eliminate
 1.19% :     0.000002s :     13: predicate.transpose_eliminate
 1.40% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.95% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000703    12
53.49% :     0.000376s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.56% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.95% :     0.000295s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025595    72
 0.22% :     0.000056s :      1: add_recomputation
 0.55% :     0.000142s :      1: auto_monad
 0.10% :     0.000027s :      1: auto_monad_reorder
 1.93% :     0.000495s :      1: bootstrap
 0.10% :     0.000026s :      1: cconv
 0.38% :     0.000096s :      1: convert_after_rewriter
 0.11% :     0.000027s :      1: cse_after_recomputation
 0.10% :     0.000026s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
35.96% :     0.009203s :      1: jit_opt_a
 0.60% :     0.000153s :      1: jit_opt_after_cconv
 0.22% :     0.000057s :      1: jit_opt_b
 1.67% :     0.000426s :      1: loop_unroll
 2.12% :     0.000542s :      1: mutable_eliminate
 3.93% :     0.001006s :     26: opt.transform.jit_opt_a
 0.21% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      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
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.95% :     0.000498s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.05% :     0.000012s :      1: pre_auto_parallel
 0.18% :     0.000046s :      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
 1.39% :     0.000355s :      1: renormalize.infer
 1.08% :     0.000276s :      1: renormalize.specialize
 0.60% :     0.000152s :      1: rewriter_after_jit_bprop_graph
 0.62% :     0.000158s :      1: rewriter_after_opt_a
 1.22% :     0.000312s :      1: rewriter_before_opt_a
 0.31% :     0.000079s :      1: symbol_engine_optimizer
43.64% :     0.011168s :      1: type_inference

[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:01.664.024 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafc79a60> should be wrapped by jit only once.

TotalTime = 0.600195, [33]
[bootstrap]: 0.0005825
[type_inference]: 0.0524145
[event_method]: 2.326e-05
[auto_monad]: 0.00033765
[graph_reusing]: 7.39002e-06
[pre_auto_parallel]: 2.107e-05
[py_interpret_to_execute]: 0.00015784
[rewriter_before_opt_a]: 0.00016346
[expand_dump_flag]: 3.23998e-06
[jit_opt_a]: 0.0120439, [2]
    [Cycle 1]: 0.00280449, [27]
        [switch_simplify]: 0.0001084
        [loop_unroll]: 3.672e-05
        [a_1]: 0.00081649
        [with_stream_mark]: 2.797e-05
        [recompute_prepare]: 1.235e-05
        [updatestate_depend_eliminate]: 6.23002e-06
        [updatestate_assign_eliminate]: 6.02999e-06
        [updatestate_loads_eliminate]: 4.63001e-06
        [parameter_eliminate]: 2.07001e-06
        [specialize_transform]: 9.58997e-06
        [updatestate_useless_node_eliminater]: 1.205e-05
        [accelerated_algorithm]: 8.84e-06
        [meta_shard_fg_expand]: 3.73001e-06
        [get_grad_eliminate_]: 7.85e-06
        [merge_forward]: 5.83002e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.55e-05
        [j_node_and_user_rematch]: 1.509e-05
        [meta_fg_expand]: 3.86999e-06
        [replace_old_param]: 1.363e-05
        [inline_without_move]: 8.43999e-06
        [renormalize]: 0.00133934
        [add_forward_monad_depend]: 2.017e-05
        [auto_monad_grad]: 3.10998e-06
        [auto_monad_eliminator]: 2.404e-05
        [cse]: 3.936e-05
        [replace_applicator]: 1.809e-05
    [Cycle 2]: 0.00046496, [27]
        [switch_simplify]: 8.84e-06
        [loop_unroll]: 7.6e-06
        [a_1]: 0.00016532
        [with_stream_mark]: 1.167e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 4.76002e-06
        [updatestate_loads_eliminate]: 4.47e-06
        [parameter_eliminate]: 1.93997e-06
        [specialize_transform]: 8e-06
        [updatestate_useless_node_eliminater]: 1.089e-05
        [accelerated_algorithm]: 8.15e-06
        [meta_shard_fg_expand]: 1.69e-06
        [get_grad_eliminate_]: 7.28e-06
        [merge_forward]: 4.71002e-06
        [cell_reuse_recompute_pass]: 1.77999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.673e-05
        [j_node_and_user_rematch]: 1.283e-05
        [meta_fg_expand]: 3.04001e-06
        [replace_old_param]: 1.044e-05
        [inline_without_move]: 7.95e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 1.006e-05
        [cse]: 2.044e-05
        [replace_applicator]: 7.83999e-06
[py_interpret_to_execute_after_opt_a]: 1.717e-05
[rewriter_after_opt_a]: 0.00034833
[convert_after_rewriter]: 1.284e-05
[order_py_execute_after_rewriter]: 7.31001e-06
[mutable_eliminate]: 0.00099065
[jit_opt_b]: 7.066e-05, [1]
    [Cycle 1]: 6.185e-05, [2]
        [frontend_op_eliminate]: 2.438e-05
        [inline_after_opt_a]: 2.462e-05
[cconv]: 3.322e-05
[loop_unroll]: 0.00046739
[jit_opt_after_cconv]: 0.00021012, [1]
    [Cycle 1]: 0.00020301, [11]
        [c_1]: 4.842e-05
        [parameter_eliminate]: 3.68e-06
        [updatestate_depend_eliminate]: 1.024e-05
        [updatestate_assign_eliminate]: 5.15001e-06
        [updatestate_loads_eliminate]: 4.60001e-06
        [cse]: 3.436e-05
        [call_graph_tuple_transform]: 2.264e-05
        [tuple_list_get_item_eliminator]: 8.35001e-06
        [none_parameter_eliminate]: 2.24001e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 9.59e-06
[remove_dup_value]: 2.005e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 4.151e-05
[add_recomputation]: 0.00010964
[cse_after_recomputation]: 3.459e-05, [1]
    [Cycle 1]: 2.748e-05, [1]
        [cse]: 2.004e-05
[auto_monad_reorder]: 3.483e-05
[get_jit_bprop_graph]: 2.22999e-06
[rewriter_after_jit_bprop_graph]: 3.25e-06
[opt_after_jit_grad]: 0.0005258
[symbol_engine_optimizer]: 0.00012206, [1]
    [Cycle 1]: 0.0001146, [6]
        [build]: 2.891e-05
        [elim_shapecalc]: 1.278e-05
        [elim_not_effective]: 1.832e-05
        [opt_reshape]: 9.14998e-06
        [fold_const_symbol]: 1.35e-05
        [renormalize]: 9.30013e-07
[validate]: 9.612e-05
[backend_pass]: 1.24998e-06
[task_emit]: 0.530993
[execute]: 9.34998e-06
Sums
    bootstrap                            :     0.000583s :  0.10%
    type_inference                       :     0.052414s :  8.87%
    event_method                         :     0.000023s :  0.00%
    auto_monad                           :     0.000338s :  0.06%
    graph_reusing                        :     0.000007s :  0.00%
    pre_auto_parallel                    :     0.000021s :  0.00%
    py_interpret_to_execute              :     0.000158s :  0.03%
    rewriter_before_opt_a                :     0.000163s :  0.03%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000117s :  0.02%
    jit_opt_a.loop_unroll                :     0.000044s :  0.01%
    jit_opt_a.a_1                        :     0.000982s :  0.17%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.01%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.00%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.00%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.00%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.00%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.00%
    jit_opt_a.specialize_transform       :     0.000018s :  0.00%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.00%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.00%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.00%
    jit_opt_a.merge_forward              :     0.000011s :  0.00%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000042s :  0.01%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.00%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.00%
    jit_opt_a.replace_old_param          :     0.000024s :  0.00%
    jit_opt_a.inline_without_move        :     0.000016s :  0.00%
    jit_opt_a.renormalize                :     0.001339s :  0.23%
    jit_opt_a.add_forward_monad_depend   :     0.000021s :  0.00%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000034s :  0.01%
    jit_opt_a.cse                        :     0.000060s :  0.01%
    jit_opt_a.replace_applicator         :     0.000026s :  0.00%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.00%
    rewriter_after_opt_a                 :     0.000348s :  0.06%
    convert_after_rewriter               :     0.000013s :  0.00%
    order_py_execute_after_rewriter      :     0.000007s :  0.00%
    mutable_eliminate                    :     0.000991s :  0.17%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.00%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.00%
    cconv                                :     0.000033s :  0.01%
    loop_unroll                          :     0.000467s :  0.08%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.00%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.00%
    jit_opt_after_cconv.cse              :     0.000034s :  0.01%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.00%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000010s :  0.00%
    remove_dup_value                     :     0.000020s :  0.00%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000042s :  0.01%
    add_recomputation                    :     0.000110s :  0.02%
    cse_after_recomputation.cse          :     0.000020s :  0.00%
    auto_monad_reorder                   :     0.000035s :  0.01%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000003s :  0.00%
    opt_after_jit_grad                   :     0.000526s :  0.09%
    symbol_engine_optimizer.build        :     0.000029s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.00%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.00%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.00%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000096s :  0.02%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.530993s : 89.90%
    execute                              :     0.000009s :  0.00%


Time group info:
------[substitution.]   0.000357    47
 2.92% :     0.000010s :      2: substitution.depend_value_elim
 0.69% :     0.000002s :      4: substitution.elim_not_effective
 0.64% :     0.000002s :      4: substitution.fold_const_symbol
 1.96% :     0.000007s :      5: substitution.graph_param_transform
78.71% :     0.000281s :      4: substitution.inline
 1.48% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.19% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.62% :     0.000006s :      2: substitution.replace_old_param
 3.86% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.06% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 2.87% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.052323     2
95.93% :     0.050192s :      1: type_inference.infer
 4.07% :     0.002131s :      1: type_inference.specialize
------[replace.]   0.000113     6
68.15% :     0.000077s :      4: replace.inline
31.85% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000291     6
95.62% :     0.000278s :      4: match.inline
 4.38% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 0.95% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     15: predicate.addn_check_dump
 1.27% :     0.000002s :     15: predicate.addn_zero_filter
 2.28% :     0.000004s :     15: predicate.arithmetic_simplify
 1.71% :     0.000003s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.10% :     0.000002s :     15: predicate.compare_switch_simplify
 1.57% :     0.000003s :     15: predicate.depend_value_elim
 1.13% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.95% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.57% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.01% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 5.50% :     0.000010s :     31: predicate.inline
 0.99% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :     10: predicate.less_batch_normalization
 1.39% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.75% :     0.000003s :     22: predicate.load_eliminater
 1.34% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.59% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.87% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.20% :     0.000002s :     15: predicate.minmaximum_grad
 2.26% :     0.000004s :      5: predicate.mutable_eliminate
 0.56% :     0.000001s :      5: predicate.opt_reshape
 2.33% :     0.000004s :     22: predicate.partial_eliminate
 1.23% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.65% :     0.000003s :     15: predicate.reduce_eliminate
 1.48% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.72% :     0.000003s :     27: predicate.replace_applicator
 0.71% :     0.000001s :     10: predicate.replace_old_param
 0.48% :     0.000001s :      5: predicate.reset_defer_inline
 1.22% :     0.000002s :     15: predicate.reshape_eliminate
 1.16% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.99% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.01% :     0.000002s :     10: predicate.special_op_eliminate
 1.09% :     0.000002s :     10: predicate.specialize_transform
 1.44% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.34% :     0.000004s :     21: predicate.switch_defer_inline
 1.84% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.69% :     0.000013s :     68: predicate.switch_simplify
 1.16% :     0.000002s :     15: predicate.tile_eliminate
 1.21% :     0.000002s :     15: predicate.transpose_eliminate
 1.68% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.33% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.78% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.32% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001278    13
45.07% :     0.000576s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.36% :     0.000069s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.57% :     0.000633s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.603035    76
 0.02% :     0.000113s :      1: add_recomputation
 0.06% :     0.000344s :      1: auto_monad
 0.01% :     0.000038s :      1: auto_monad_reorder
 0.00% :     0.000005s :      1: backend_pass
 0.10% :     0.000607s :      1: bootstrap
 0.01% :     0.000036s :      1: cconv
 0.00% :     0.000016s :      1: convert_after_rewriter
 0.01% :     0.000037s :      1: cse_after_recomputation
 0.01% :     0.000045s :      1: environ_conv
 0.00% :     0.000028s :      1: event_method
 0.00% :     0.000014s :      1: execute
 0.00% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.00% :     0.000010s :      1: graph_reusing
 2.00% :     0.012048s :      1: jit_opt_a
 0.04% :     0.000213s :      1: jit_opt_after_cconv
 0.01% :     0.000074s :      1: jit_opt_b
 0.08% :     0.000476s :      1: loop_unroll
 0.17% :     0.001001s :      1: mutable_eliminate
 0.22% :     0.001325s :     26: opt.transform.jit_opt_a
 0.01% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.01% :     0.000042s :      4: opt.transform.jit_opt_b
 0.00% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.00% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.01% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.01% :     0.000049s :      4: opt.transform.symbol_engine_opt
 0.09% :     0.000536s :      1: opt_after_jit_grad
 0.00% :     0.000010s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000024s :      1: pre_auto_parallel
 0.03% :     0.000163s :      1: py_interpret_to_execute
 0.00% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.00% :     0.000023s :      1: remove_dup_value
 0.14% :     0.000857s :      1: renormalize.infer
 0.08% :     0.000473s :      1: renormalize.specialize
 0.00% :     0.000005s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000354s :      1: rewriter_after_opt_a
 0.03% :     0.000168s :      1: rewriter_before_opt_a
 0.02% :     0.000125s :      1: symbol_engine_optimizer
88.06% :     0.531018s :      1: task_emit
 8.69% :     0.052433s :      1: type_inference
 0.02% :     0.000133s :      1: validate


TotalTime = 0.641051, [33]
[bootstrap]: 0.00091847
[type_inference]: 0.0833705
[event_method]: 0.00012964
[auto_monad]: 0.00036846
[graph_reusing]: 8.22998e-06
[pre_auto_parallel]: 1.206e-05
[py_interpret_to_execute]: 4.296e-05
[rewriter_before_opt_a]: 0.00016394
[expand_dump_flag]: 3.54002e-06
[jit_opt_a]: 0.0718477, [3]
    [Cycle 1]: 0.0609836, [27]
        [switch_simplify]: 0.00012934
        [loop_unroll]: 4.9e-05
        [a_1]: 0.00124408
        [with_stream_mark]: 4.149e-05
        [recompute_prepare]: 2.598e-05
        [updatestate_depend_eliminate]: 1.282e-05
        [updatestate_assign_eliminate]: 1.058e-05
        [updatestate_loads_eliminate]: 9.96e-06
        [parameter_eliminate]: 2.82002e-06
        [specialize_transform]: 1.933e-05
        [updatestate_useless_node_eliminater]: 2.183e-05
        [accelerated_algorithm]: 1.999e-05
        [meta_shard_fg_expand]: 4.65001e-06
        [get_grad_eliminate_]: 1.749e-05
        [merge_forward]: 1.106e-05
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 4.004e-05
        [j_node_and_user_rematch]: 4.321e-05
        [meta_fg_expand]: 0.0410949
        [replace_old_param]: 0.00011629
        [inline_without_move]: 0.00011314
        [renormalize]: 0.0171176
        [add_forward_monad_depend]: 3.638e-05
        [auto_monad_grad]: 9.62001e-06
        [auto_monad_eliminator]: 7.797e-05
        [cse]: 0.00024997
        [replace_applicator]: 0.00018148
    [Cycle 2]: 0.0028022, [27]
        [switch_simplify]: 5.856e-05
        [loop_unroll]: 5.561e-05
        [a_1]: 0.00100198
        [with_stream_mark]: 1.803e-05
        [recompute_prepare]: 1.192e-05
        [updatestate_depend_eliminate]: 2.461e-05
        [updatestate_assign_eliminate]: 6.94999e-06
        [updatestate_loads_eliminate]: 4.94e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 8.90001e-06
        [updatestate_useless_node_eliminater]: 1.139e-05
        [accelerated_algorithm]: 9.41e-06
        [meta_shard_fg_expand]: 2.59999e-06
        [get_grad_eliminate_]: 7.81001e-06
        [merge_forward]: 5.84999e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.904e-05
        [j_node_and_user_rematch]: 1.274e-05
        [meta_fg_expand]: 8.155e-05
        [replace_old_param]: 1.256e-05
        [inline_without_move]: 7.74002e-06
        [renormalize]: 0.00117177
        [add_forward_monad_depend]: 4.82998e-06
        [auto_monad_grad]: 2.00002e-06
        [auto_monad_eliminator]: 1.623e-05
        [cse]: 6.121e-05
        [replace_applicator]: 1.63e-05
    [Cycle 3]: 0.00045749, [27]
        [switch_simplify]: 8.68001e-06
        [loop_unroll]: 7.78999e-06
        [a_1]: 0.00015897
        [with_stream_mark]: 1.073e-05
        [recompute_prepare]: 7.87998e-06
        [updatestate_depend_eliminate]: 5.63002e-06
        [updatestate_assign_eliminate]: 4.47e-06
        [updatestate_loads_eliminate]: 3.94002e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 8.22e-06
        [updatestate_useless_node_eliminater]: 1.074e-05
        [accelerated_algorithm]: 7.98999e-06
        [meta_shard_fg_expand]: 1.67001e-06
        [get_grad_eliminate_]: 7.49002e-06
        [merge_forward]: 4.46002e-06
        [cell_reuse_recompute_pass]: 1.68997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.565e-05
        [j_node_and_user_rematch]: 1.239e-05
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 1.054e-05
        [inline_without_move]: 7.31001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 9.92999e-06
        [cse]: 2.39e-05
        [replace_applicator]: 8e-06
[py_interpret_to_execute_after_opt_a]: 1.512e-05
[rewriter_after_opt_a]: 6.292e-05
[convert_after_rewriter]: 9.20999e-06
[order_py_execute_after_rewriter]: 6.37001e-06
[mutable_eliminate]: 0.00074405
[jit_opt_b]: 7.62e-05, [1]
    [Cycle 1]: 6.83e-05, [2]
        [frontend_op_eliminate]: 3.038e-05
        [inline_after_opt_a]: 2.528e-05
[cconv]: 2.486e-05
[loop_unroll]: 0.00048924
[jit_opt_after_cconv]: 0.00019881, [1]
    [Cycle 1]: 0.00019142, [11]
        [c_1]: 4.424e-05
        [parameter_eliminate]: 3.46001e-06
        [updatestate_depend_eliminate]: 7.78001e-06
        [updatestate_assign_eliminate]: 4.55001e-06
        [updatestate_loads_eliminate]: 3.98999e-06
        [cse]: 3.184e-05
        [call_graph_tuple_transform]: 2.285e-05
        [tuple_list_get_item_eliminator]: 8.27e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 8.37e-06
[remove_dup_value]: 2.301e-05
[partial_unused_args_eliminate]: 2.31998e-06
[environ_conv]: 1.625e-05
[add_recomputation]: 8.769e-05
[cse_after_recomputation]: 3.311e-05, [1]
    [Cycle 1]: 2.683e-05, [1]
        [cse]: 2.047e-05
[auto_monad_reorder]: 3.178e-05
[get_jit_bprop_graph]: 2.05002e-06
[rewriter_after_jit_bprop_graph]: 0.00022577
[opt_after_jit_grad]: 0.00048279
[symbol_engine_optimizer]: 9.743e-05, [1]
    [Cycle 1]: 9.082e-05, [6]
        [build]: 1.084e-05
        [elim_shapecalc]: 1.088e-05
        [elim_not_effective]: 1.772e-05
        [opt_reshape]: 8.77e-06
        [fold_const_symbol]: 1.358e-05
        [renormalize]: 4.50003e-07
[validate]: 9.613e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.481188
[execute]: 7.53999e-06
Sums
    bootstrap                            :     0.000918s :  0.15%
    type_inference                       :     0.083370s : 13.18%
    event_method                         :     0.000130s :  0.02%
    auto_monad                           :     0.000368s :  0.06%
    graph_reusing                        :     0.000008s :  0.00%
    pre_auto_parallel                    :     0.000012s :  0.00%
    py_interpret_to_execute              :     0.000043s :  0.01%
    rewriter_before_opt_a                :     0.000164s :  0.03%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000197s :  0.03%
    jit_opt_a.loop_unroll                :     0.000112s :  0.02%
    jit_opt_a.a_1                        :     0.002405s :  0.38%
    jit_opt_a.with_stream_mark           :     0.000070s :  0.01%
    jit_opt_a.recompute_prepare          :     0.000046s :  0.01%
    jit_opt_a.updatestate_depend_eliminate :     0.000043s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000022s :  0.00%
    jit_opt_a.updatestate_loads_eliminate :     0.000019s :  0.00%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.00%
    jit_opt_a.specialize_transform       :     0.000036s :  0.01%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000044s :  0.01%
    jit_opt_a.accelerated_algorithm      :     0.000037s :  0.01%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000033s :  0.01%
    jit_opt_a.merge_forward              :     0.000021s :  0.00%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000075s :  0.01%
    jit_opt_a.j_node_and_user_rematch    :     0.000068s :  0.01%
    jit_opt_a.meta_fg_expand             :     0.041179s :  6.51%
    jit_opt_a.replace_old_param          :     0.000139s :  0.02%
    jit_opt_a.inline_without_move        :     0.000128s :  0.02%
    jit_opt_a.renormalize                :     0.018289s :  2.89%
    jit_opt_a.add_forward_monad_depend   :     0.000043s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000013s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000104s :  0.02%
    jit_opt_a.cse                        :     0.000335s :  0.05%
    jit_opt_a.replace_applicator         :     0.000206s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.00%
    rewriter_after_opt_a                 :     0.000063s :  0.01%
    convert_after_rewriter               :     0.000009s :  0.00%
    order_py_execute_after_rewriter      :     0.000006s :  0.00%
    mutable_eliminate                    :     0.000744s :  0.12%
    jit_opt_b.frontend_op_eliminate      :     0.000030s :  0.00%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.00%
    cconv                                :     0.000025s :  0.00%
    loop_unroll                          :     0.000489s :  0.08%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.00%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.cse              :     0.000032s :  0.01%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.00%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000008s :  0.00%
    remove_dup_value                     :     0.000023s :  0.00%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000016s :  0.00%
    add_recomputation                    :     0.000088s :  0.01%
    cse_after_recomputation.cse          :     0.000020s :  0.00%
    auto_monad_reorder                   :     0.000032s :  0.01%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000226s :  0.04%
    opt_after_jit_grad                   :     0.000483s :  0.08%
    symbol_engine_optimizer.build        :     0.000011s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.00%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.00%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.00%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000096s :  0.02%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.481188s : 76.08%
    execute                              :     0.000008s :  0.00%


Time group info:
------[substitution.]   0.001749   159
 1.19% :     0.000021s :      8: substitution.depend_value_elim
 0.15% :     0.000003s :      4: substitution.elim_not_effective
 0.13% :     0.000002s :      4: substitution.fold_const_symbol
59.08% :     0.001033s :      4: substitution.getattr_setattr_resolve
 0.39% :     0.000007s :      5: substitution.graph_param_transform
23.81% :     0.000416s :     10: substitution.inline
 1.97% :     0.000034s :      4: substitution.inline_without_move
 1.25% :     0.000022s :     20: substitution.j_node_and_user_rematch
 0.54% :     0.000010s :      5: substitution.minmaximum_grad
 0.74% :     0.000013s :      3: substitution.partial_eliminate
 0.84% :     0.000015s :     20: substitution.remove_not_recompute_node
 2.61% :     0.000046s :     13: substitution.replace_applicator
 0.80% :     0.000014s :     17: substitution.replace_old_param
 0.16% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.04% :     0.000018s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.70% :     0.000012s :      5: substitution.tuple_list_get_item_depend_reorder
 2.11% :     0.000037s :     10: substitution.tuple_list_get_item_eliminator
 0.89% :     0.000016s :      8: substitution.updatestate_pure_node_eliminater
 1.60% :     0.000028s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.083165     2
97.43% :     0.081030s :      1: type_inference.infer
 2.57% :     0.002135s :      1: type_inference.specialize
------[replace.]   0.000362    20
16.96% :     0.000061s :      3: replace.getattr_setattr_resolve
41.20% :     0.000149s :     10: replace.inline
11.95% :     0.000043s :      1: replace.replace_applicator
23.97% :     0.000087s :      5: replace.tuple_list_get_item_eliminator
 5.91% :     0.000021s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001415    20
68.32% :     0.000967s :      3: match.getattr_setattr_resolve
28.99% :     0.000410s :     10: match.inline
 0.68% :     0.000010s :      1: match.replace_applicator
 1.31% :     0.000019s :      5: match.tuple_list_get_item_eliminator
 0.70% :     0.000010s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000437  2975
 1.38% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.43% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.33% :     0.000006s :     46: predicate.addn_check_dump
 1.47% :     0.000006s :     46: predicate.addn_zero_filter
 1.87% :     0.000008s :     46: predicate.arithmetic_simplify
 1.57% :     0.000007s :     46: predicate.cast_eliminate
 0.16% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.28% :     0.000006s :     46: predicate.compare_switch_simplify
 1.57% :     0.000007s :     46: predicate.depend_value_elim
 1.30% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.50% :     0.000007s :     46: predicate.dict_get_item_eliminator
 1.39% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.35% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.12% :     0.000001s :      5: predicate.elim_not_effective
 0.27% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.29% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.37% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.43% :     0.000006s :     46: predicate.environ_get_eliminate
 1.33% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.13% :     0.000001s :      5: predicate.fold_const_symbol
 1.05% :     0.000005s :     27: predicate.get_grad_eliminate
 1.50% :     0.000007s :     20: predicate.getattr_setattr_resolve
 0.16% :     0.000001s :      5: predicate.graph_param_transform
 3.95% :     0.000017s :     72: predicate.inline
 3.14% :     0.000014s :     89: predicate.inline_without_move
 0.43% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.40% :     0.000006s :     27: predicate.less_batch_normalization
 1.55% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.84% :     0.000008s :     56: predicate.load_eliminater
 0.50% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.83% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.71% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000006s :     46: predicate.merge_addn
 1.45% :     0.000006s :     46: predicate.minmaximum_grad
 0.57% :     0.000002s :      5: predicate.mutable_eliminate
 0.21% :     0.000001s :      5: predicate.opt_reshape
 2.06% :     0.000009s :     56: predicate.partial_eliminate
 1.47% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.76% :     0.000008s :     46: predicate.reduce_eliminate
 1.59% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000002s :     27: predicate.remove_not_recompute_node
 2.94% :     0.000013s :    126: predicate.replace_applicator
 1.56% :     0.000007s :     89: predicate.replace_old_param
 0.13% :     0.000001s :      5: predicate.reset_defer_inline
 1.54% :     0.000007s :     46: predicate.reshape_eliminate
 1.51% :     0.000007s :     46: predicate.row_tensor_add_zeros_like
 0.26% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.49% :     0.000007s :     46: predicate.same_eliminate
 0.56% :     0.000002s :     29: predicate.set_cell_output_no_recompute
 0.46% :     0.000002s :     10: predicate.special_op_eliminate
 0.98% :     0.000004s :     27: predicate.specialize_transform
 1.65% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.24% :     0.000010s :     62: predicate.switch_defer_inline
 2.10% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.58% :     0.000029s :    188: predicate.switch_simplify
 1.32% :     0.000006s :     46: predicate.tile_eliminate
 1.35% :     0.000006s :     46: predicate.transpose_eliminate
 1.76% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 3.07% :     0.000013s :     61: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000008s :     46: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.88% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000013s :     84: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000007s :     46: predicate.value_based_eliminate
 0.17% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.30% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003602    48
65.56% :     0.002361s :     27: func_graph_cloner_run.FuncGraphClonerGraph
34.44% :     0.001241s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.664112    93
 0.01% :     0.000091s :      1: add_recomputation
 0.06% :     0.000377s :      1: auto_monad
 0.01% :     0.000035s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.14% :     0.000942s :      1: bootstrap
 0.00% :     0.000028s :      1: cconv
 0.00% :     0.000012s :      1: convert_after_rewriter
 0.01% :     0.000035s :      1: cse_after_recomputation
 0.00% :     0.000019s :      1: environ_conv
 0.02% :     0.000137s :      1: event_method
 0.00% :     0.000012s :      1: execute
 0.00% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.00% :     0.000012s :      1: graph_reusing
10.82% :     0.071852s :      1: jit_opt_a
 0.03% :     0.000202s :      1: jit_opt_after_cconv
 0.01% :     0.000079s :      1: jit_opt_b
 0.07% :     0.000498s :      1: loop_unroll
 0.11% :     0.000753s :      1: mutable_eliminate
 0.52% :     0.003460s :     39: opt.transform.jit_opt_a
 0.01% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.01% :     0.000048s :      4: opt.transform.jit_opt_b
 0.00% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.00% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.00% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.001156s :      2: opt.transform.opt_resolve
 0.01% :     0.000048s :      4: opt.transform.symbol_engine_opt
 0.07% :     0.000491s :      1: opt_after_jit_grad
 0.00% :     0.000009s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000015s :      1: pre_auto_parallel
 0.01% :     0.000046s :      1: py_interpret_to_execute
 0.00% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.00% :     0.000026s :      1: remove_dup_value
 2.26% :     0.015008s :      2: renormalize.infer
 0.49% :     0.003261s :      2: renormalize.specialize
 0.03% :     0.000229s :      1: rewriter_after_jit_bprop_graph
 0.01% :     0.000066s :      1: rewriter_after_opt_a
 0.03% :     0.000167s :      1: rewriter_before_opt_a
 0.02% :     0.000100s :      1: symbol_engine_optimizer
72.46% :     0.481208s :      1: task_emit
12.56% :     0.083390s :      1: type_inference
 0.02% :     0.000128s :      1: validate

[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.167.203 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdc8475d30> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.168.289 [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 = 0.0444718, [33]
[bootstrap]: 0.00034121
[type_inference]: 0.0227079
[event_method]: 2.145e-05
[auto_monad]: 0.00015098
[graph_reusing]: 6.43e-06
[pre_auto_parallel]: 3.03998e-06
[py_interpret_to_execute]: 0.0001164
[rewriter_before_opt_a]: 9.931e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00633221, [2]
    [Cycle 1]: 0.00220142, [27]
        [switch_simplify]: 5.95e-05
        [loop_unroll]: 3.692e-05
        [a_1]: 0.00069212
        [with_stream_mark]: 2.128e-05
        [recompute_prepare]: 9.97001e-06
        [updatestate_depend_eliminate]: 5.98002e-06
        [updatestate_assign_eliminate]: 6.03002e-06
        [updatestate_loads_eliminate]: 4.84003e-06
        [parameter_eliminate]: 2.21e-06
        [specialize_transform]: 9.25999e-06
        [updatestate_useless_node_eliminater]: 1.147e-05
        [accelerated_algorithm]: 8.48001e-06
        [meta_shard_fg_expand]: 2.99999e-06
        [get_grad_eliminate_]: 8.15e-06
        [merge_forward]: 5.17999e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.896e-05
        [j_node_and_user_rematch]: 1.325e-05
        [meta_fg_expand]: 3.5e-06
        [replace_old_param]: 1.08e-05
        [inline_without_move]: 8.02e-06
        [renormalize]: 0.00098792
        [add_forward_monad_depend]: 1.156e-05
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 2.049e-05
        [cse]: 3.913e-05
        [replace_applicator]: 1.602e-05
    [Cycle 2]: 0.00050053, [27]
        [switch_simplify]: 8.95001e-06
        [loop_unroll]: 7.74002e-06
        [a_1]: 0.00016615
        [with_stream_mark]: 1.019e-05
        [recompute_prepare]: 8.19002e-06
        [updatestate_depend_eliminate]: 1.671e-05
        [updatestate_assign_eliminate]: 4.79e-06
        [updatestate_loads_eliminate]: 3.80998e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 9.00999e-06
        [updatestate_useless_node_eliminater]: 1.196e-05
        [accelerated_algorithm]: 1.71e-05
        [meta_shard_fg_expand]: 1.82999e-06
        [get_grad_eliminate_]: 7.88001e-06
        [merge_forward]: 4.28001e-06
        [cell_reuse_recompute_pass]: 2.09999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.597e-05
        [j_node_and_user_rematch]: 1.261e-05
        [meta_fg_expand]: 2.77002e-06
        [replace_old_param]: 1.007e-05
        [inline_without_move]: 7.43999e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 1.086e-05
        [cse]: 2.946e-05
        [replace_applicator]: 9.20001e-06
[py_interpret_to_execute_after_opt_a]: 1.21e-05
[rewriter_after_opt_a]: 0.00030088
[convert_after_rewriter]: 1.216e-05
[order_py_execute_after_rewriter]: 7.42002e-06
[mutable_eliminate]: 0.00057899
[jit_opt_b]: 7.188e-05, [1]
    [Cycle 1]: 6.41e-05, [2]
        [frontend_op_eliminate]: 2.553e-05
        [inline_after_opt_a]: 2.59e-05
[cconv]: 2.625e-05
[loop_unroll]: 0.00046757
[jit_opt_after_cconv]: 0.00020134, [1]
    [Cycle 1]: 0.00019425, [11]
        [c_1]: 4.747e-05
        [parameter_eliminate]: 2.76e-06
        [updatestate_depend_eliminate]: 8.28999e-06
        [updatestate_assign_eliminate]: 4.62e-06
        [updatestate_loads_eliminate]: 4.43001e-06
        [cse]: 2.828e-05
        [call_graph_tuple_transform]: 2.425e-05
        [tuple_list_get_item_eliminator]: 9.36e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 9.04e-06
[remove_dup_value]: 2.133e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 1.165e-05
[add_recomputation]: 6.311e-05
[cse_after_recomputation]: 3.18e-05, [1]
    [Cycle 1]: 2.581e-05, [1]
        [cse]: 1.966e-05
[auto_monad_reorder]: 2.586e-05
[get_jit_bprop_graph]: 2.00002e-06
[rewriter_after_jit_bprop_graph]: 6.01998e-06
[opt_after_jit_grad]: 0.00052164
[symbol_engine_optimizer]: 0.0001072, [1]
    [Cycle 1]: 0.00010044, [6]
        [build]: 1.188e-05
        [elim_shapecalc]: 1.243e-05
        [elim_not_effective]: 2.037e-05
        [opt_reshape]: 9.32999e-06
        [fold_const_symbol]: 1.501e-05
        [renormalize]: 4.80009e-07
[validate]: 5.806e-05
[backend_pass]: 1.09998e-06
[task_emit]: 0.0119238
[execute]: 6.34999e-06
Sums
    bootstrap                            :     0.000341s :  0.85%
    type_inference                       :     0.022708s : 56.51%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000151s :  0.38%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000116s :  0.29%
    rewriter_before_opt_a                :     0.000099s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000858s :  2.14%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000023s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    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 :     0.000035s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.000988s :  2.46%
    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      :     0.000031s :  0.08%
    jit_opt_a.cse                        :     0.000069s :  0.17%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000301s :  0.75%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000579s :  1.44%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000468s :  1.16%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.12%
    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              :     0.000028s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000063s :  0.16%
    cse_after_recomputation.cse          :     0.000020s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000522s :  1.30%
    symbol_engine_optimizer.build        :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    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                             :     0.000058s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011924s : 29.68%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000267    47
 3.03% :     0.000008s :      2: substitution.depend_value_elim
 1.07% :     0.000003s :      4: substitution.elim_not_effective
 0.79% :     0.000002s :      4: substitution.fold_const_symbol
 2.59% :     0.000007s :      5: substitution.graph_param_transform
76.06% :     0.000203s :      4: substitution.inline
 1.52% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.57% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.32% :     0.000004s :      2: substitution.replace_old_param
 4.64% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.11% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.32% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022641     2
94.06% :     0.021296s :      1: type_inference.infer
 5.94% :     0.001345s :      1: type_inference.specialize
------[replace.]   0.000088     6
66.03% :     0.000058s :      4: replace.inline
33.97% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000211     6
94.67% :     0.000200s :      4: match.inline
 5.33% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.26% :     0.000002s :     15: predicate.addn_zero_filter
 1.85% :     0.000003s :     15: predicate.arithmetic_simplify
 1.29% :     0.000002s :     15: predicate.cast_eliminate
 0.48% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     15: predicate.compare_switch_simplify
 1.41% :     0.000002s :     15: predicate.depend_value_elim
 1.19% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.02% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.67% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.30% :     0.000000s :      5: predicate.fold_const_symbol
 1.06% :     0.000002s :     10: predicate.get_grad_eliminate
 0.29% :     0.000000s :      5: predicate.graph_param_transform
 5.62% :     0.000010s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :     10: predicate.less_batch_normalization
 1.32% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.99% :     0.000003s :     22: predicate.load_eliminater
 1.25% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.45% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.09% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.25% :     0.000002s :     15: predicate.merge_addn
 1.19% :     0.000002s :     15: predicate.minmaximum_grad
 1.43% :     0.000002s :      5: predicate.mutable_eliminate
 0.58% :     0.000001s :      5: predicate.opt_reshape
 2.35% :     0.000004s :     22: predicate.partial_eliminate
 1.21% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.71% :     0.000003s :     15: predicate.reduce_eliminate
 1.58% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.68% :     0.000003s :     27: predicate.replace_applicator
 0.70% :     0.000001s :     10: predicate.replace_old_param
 0.53% :     0.000001s :      5: predicate.reset_defer_inline
 1.19% :     0.000002s :     15: predicate.reshape_eliminate
 1.29% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     15: predicate.same_eliminate
 0.61% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.03% :     0.000002s :     10: predicate.special_op_eliminate
 1.20% :     0.000002s :     10: predicate.specialize_transform
 1.60% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.18% :     0.000004s :     21: predicate.switch_defer_inline
 2.04% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.60% :     0.000011s :     68: predicate.switch_simplify
 1.21% :     0.000002s :     15: predicate.tile_eliminate
 1.44% :     0.000002s :     15: predicate.transpose_eliminate
 1.54% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.39% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000789    13
45.30% :     0.000357s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.99% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.70% :     0.000384s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046825    76
 0.14% :     0.000066s :      1: add_recomputation
 0.33% :     0.000157s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.76% :     0.000357s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.06% :     0.000026s :      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
13.53% :     0.006336s :      1: jit_opt_a
 0.44% :     0.000205s :      1: jit_opt_after_cconv
 0.16% :     0.000075s :      1: jit_opt_b
 1.02% :     0.000476s :      1: loop_unroll
 1.26% :     0.000588s :      1: mutable_eliminate
 2.49% :     0.001166s :     26: opt.transform.jit_opt_a
 0.18% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000044s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      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% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.13% :     0.000531s :      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
 0.26% :     0.000121s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000024s :      1: remove_dup_value
 1.27% :     0.000594s :      1: renormalize.infer
 0.83% :     0.000387s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.65% :     0.000305s :      1: rewriter_after_opt_a
 0.22% :     0.000104s :      1: rewriter_before_opt_a
 0.23% :     0.000110s :      1: symbol_engine_optimizer
25.49% :     0.011937s :      1: task_emit
48.53% :     0.022725s :      1: type_inference
 0.19% :     0.000088s :      1: validate


TotalTime = 0.0130876, [30]
[bootstrap]: 0.00033942
[type_inference]: 0.00473213
[event_method]: 2.216e-05
[auto_monad]: 6.386e-05
[graph_reusing]: 6.19999e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.623e-05
[rewriter_before_opt_a]: 8.934e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.00566769, [2]
    [Cycle 1]: 0.0016503, [27]
        [switch_simplify]: 5.658e-05
        [loop_unroll]: 3.62e-05
        [a_1]: 0.00062071
        [with_stream_mark]: 1.642e-05
        [recompute_prepare]: 7.92e-06
        [updatestate_depend_eliminate]: 3.96001e-06
        [updatestate_assign_eliminate]: 3.66001e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 6.68998e-06
        [updatestate_useless_node_eliminater]: 6.26e-06
        [accelerated_algorithm]: 6.41998e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.92999e-06
        [merge_forward]: 3.69002e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.54e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 2.41998e-06
        [replace_old_param]: 1.064e-05
        [inline_without_move]: 5.82001e-06
        [renormalize]: 0.0005916
        [add_forward_monad_depend]: 9.71e-06
        [auto_monad_grad]: 2.24999e-06
        [auto_monad_eliminator]: 1.442e-05
        [cse]: 3.092e-05
        [replace_applicator]: 1.248e-05
    [Cycle 2]: 0.00035382, [27]
        [switch_simplify]: 6.71e-06
        [loop_unroll]: 5.69e-06
        [a_1]: 0.00011391
        [with_stream_mark]: 8.98002e-06
        [recompute_prepare]: 5.55001e-06
        [updatestate_depend_eliminate]: 3.12002e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [parameter_eliminate]: 1.29003e-06
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 5.74e-06
        [accelerated_algorithm]: 5.85002e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 5.62999e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.335e-05
        [j_node_and_user_rematch]: 8.43999e-06
        [meta_fg_expand]: 2.00002e-06
        [replace_old_param]: 8.73001e-06
        [inline_without_move]: 5.56998e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.16997e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 5.92001e-06
        [cse]: 1.269e-05
        [replace_applicator]: 6.21e-06
[py_interpret_to_execute_after_opt_a]: 1.136e-05
[rewriter_after_opt_a]: 3.476e-05
[convert_after_rewriter]: 7.26999e-06
[order_py_execute_after_rewriter]: 5.40001e-06
[mutable_eliminate]: 0.00055669
[jit_opt_b]: 5.507e-05, [1]
    [Cycle 1]: 4.831e-05, [2]
        [frontend_op_eliminate]: 1.922e-05
        [inline_after_opt_a]: 1.752e-05
[cconv]: 2.631e-05
[loop_unroll]: 0.0004189
[jit_opt_after_cconv]: 0.00015297, [1]
    [Cycle 1]: 0.00014652, [11]
        [c_1]: 2.582e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 5.91998e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.40997e-06
        [cse]: 2.05e-05
        [call_graph_tuple_transform]: 2.073e-05
        [tuple_list_get_item_eliminator]: 6.60997e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 6.11e-06
[remove_dup_value]: 1.73e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 5.64998e-06
[add_recomputation]: 4.869e-05
[cse_after_recomputation]: 2.424e-05, [1]
    [Cycle 1]: 1.876e-05, [1]
        [cse]: 1.266e-05
[auto_monad_reorder]: 1.769e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 4.35e-06
[opt_after_jit_grad]: 0.00046051
[symbol_engine_optimizer]: 7.604e-05, [1]
    [Cycle 1]: 6.928e-05, [6]
        [build]: 3.17002e-06
        [elim_shapecalc]: 8.13001e-06
        [elim_not_effective]: 1.409e-05
        [opt_reshape]: 6.26e-06
        [fold_const_symbol]: 9.46e-06
        [renormalize]: 4.09986e-07
[validate]: 3.626e-05
Sums
    bootstrap                            :     0.000339s :  3.83%
    type_inference                       :     0.004732s : 53.36%
    event_method                         :     0.000022s :  0.25%
    auto_monad                           :     0.000064s :  0.72%
    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                :     0.000089s :  1.01%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.71%
    jit_opt_a.loop_unroll                :     0.000042s :  0.47%
    jit_opt_a.a_1                        :     0.000735s :  8.28%
    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 :     0.000007s :  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       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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_        :     0.000012s :  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 :     0.000029s :  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          :     0.000019s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000592s :  6.67%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.23%
    jit_opt_a.cse                        :     0.000044s :  0.49%
    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 :  0.06%
    mutable_eliminate                    :     0.000557s :  6.28%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000026s :  0.30%
    loop_unroll                          :     0.000419s :  4.72%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.29%
    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              :     0.000021s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  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  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000017s :  0.20%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000049s :  0.55%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000461s :  5.19%
    symbol_engine_optimizer.build        :     0.000003s :  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  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.41%


Time group info:
------[substitution.]   0.000210    28
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000006s :      4: substitution.graph_param_transform
82.26% :     0.000173s :      4: substitution.inline
 1.27% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000005s :      4: substitution.replace_old_param
 7.75% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004676     2
80.69% :     0.003773s :      1: type_inference.infer
19.31% :     0.000903s :      1: type_inference.specialize
------[replace.]   0.000084     8
57.18% :     0.000048s :      4: replace.inline
42.82% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     8
92.28% :     0.000170s :      4: match.inline
 7.72% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.26% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     13: predicate.addn_check_dump
 1.20% :     0.000002s :     13: predicate.addn_zero_filter
 1.91% :     0.000003s :     13: predicate.arithmetic_simplify
 1.42% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     13: predicate.compare_switch_simplify
 1.12% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.52% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     13: predicate.environ_get_eliminate
 1.07% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.94% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.39% :     0.000007s :     29: predicate.inline
 0.89% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      8: predicate.less_batch_normalization
 1.67% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.02% :     0.000003s :     21: predicate.load_eliminater
 1.29% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.25% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     13: predicate.merge_addn
 1.07% :     0.000001s :     13: predicate.minmaximum_grad
 1.65% :     0.000002s :      4: predicate.mutable_eliminate
 0.48% :     0.000001s :      4: predicate.opt_reshape
 2.56% :     0.000004s :     21: predicate.partial_eliminate
 1.19% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     13: predicate.reduce_eliminate
 1.71% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.63% :     0.000001s :      8: predicate.replace_old_param
 0.39% :     0.000001s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.15% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      8: predicate.special_op_eliminate
 0.97% :     0.000001s :      8: predicate.specialize_transform
 1.30% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     21: predicate.switch_defer_inline
 2.24% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.67% :     0.000011s :     67: predicate.switch_simplify
 1.18% :     0.000002s :     13: predicate.tile_eliminate
 1.20% :     0.000002s :     13: predicate.transpose_eliminate
 1.64% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.27% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     13: predicate.value_based_eliminate
 0.42% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000582    12
48.79% :     0.000284s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.66% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.55% :     0.000265s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014730    72
 0.35% :     0.000052s :      1: add_recomputation
 0.46% :     0.000067s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 2.41% :     0.000355s :      1: bootstrap
 0.20% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.18% :     0.000026s :      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
38.50% :     0.005671s :      1: jit_opt_a
 1.06% :     0.000156s :      1: jit_opt_after_cconv
 0.39% :     0.000058s :      1: jit_opt_b
 2.89% :     0.000426s :      1: loop_unroll
 3.84% :     0.000566s :      1: mutable_eliminate
 6.58% :     0.000970s :     26: opt.transform.jit_opt_a
 0.38% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000031s :      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
 0.16% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.18% :     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
 0.27% :     0.000039s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000020s :      1: remove_dup_value
 2.15% :     0.000317s :      1: renormalize.infer
 1.81% :     0.000267s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000038s :      1: rewriter_after_opt_a
 0.63% :     0.000092s :      1: rewriter_before_opt_a
 0.53% :     0.000079s :      1: symbol_engine_optimizer
32.20% :     0.004743s :      1: type_inference


TotalTime = 0.0721615, [33]
[bootstrap]: 0.00031947
[type_inference]: 0.0322127
[event_method]: 0.00012418
[auto_monad]: 0.00022887
[graph_reusing]: 8.55999e-06
[pre_auto_parallel]: 3.36001e-06
[py_interpret_to_execute]: 4.622e-05
[rewriter_before_opt_a]: 0.00012764
[expand_dump_flag]: 4.06001e-06
[jit_opt_a]: 0.0279931, [3]
    [Cycle 1]: 0.0208739, [27]
        [switch_simplify]: 8.993e-05
        [loop_unroll]: 4.888e-05
        [a_1]: 0.00107392
        [with_stream_mark]: 2.545e-05
        [recompute_prepare]: 2.492e-05
        [updatestate_depend_eliminate]: 1.179e-05
        [updatestate_assign_eliminate]: 9.85002e-06
        [updatestate_loads_eliminate]: 8.93002e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 2.033e-05
        [updatestate_useless_node_eliminater]: 2.274e-05
        [accelerated_algorithm]: 1.929e-05
        [meta_shard_fg_expand]: 3.9e-06
        [get_grad_eliminate_]: 1.859e-05
        [merge_forward]: 1.035e-05
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 3.443e-05
        [j_node_and_user_rematch]: 3.337e-05
        [meta_fg_expand]: 0.00545102
        [replace_old_param]: 0.00010924
        [inline_without_move]: 0.00010831
        [renormalize]: 0.0129274
        [add_forward_monad_depend]: 2.536e-05
        [auto_monad_grad]: 9.46998e-06
        [auto_monad_eliminator]: 7.967e-05
        [cse]: 0.00027736
        [replace_applicator]: 0.00018754
    [Cycle 2]: 0.00297809, [27]
        [switch_simplify]: 5.982e-05
        [loop_unroll]: 5.793e-05
        [a_1]: 0.00098058
        [with_stream_mark]: 1.979e-05
        [recompute_prepare]: 1.333e-05
        [updatestate_depend_eliminate]: 2.688e-05
        [updatestate_assign_eliminate]: 6.39999e-06
        [updatestate_loads_eliminate]: 4.48001e-06
        [parameter_eliminate]: 2.29999e-06
        [specialize_transform]: 1.049e-05
        [updatestate_useless_node_eliminater]: 1.172e-05
        [accelerated_algorithm]: 9.14998e-06
        [meta_shard_fg_expand]: 2.81e-06
        [get_grad_eliminate_]: 8.03999e-06
        [merge_forward]: 6.63003e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.977e-05
        [j_node_and_user_rematch]: 1.421e-05
        [meta_fg_expand]: 8.966e-05
        [replace_old_param]: 1.319e-05
        [inline_without_move]: 8.23001e-06
        [renormalize]: 0.0013279
        [add_forward_monad_depend]: 7.37997e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.764e-05
        [cse]: 6.888e-05
        [replace_applicator]: 1.778e-05
    [Cycle 3]: 0.00049692, [27]
        [switch_simplify]: 9.02999e-06
        [loop_unroll]: 8.35001e-06
        [a_1]: 0.00018201
        [with_stream_mark]: 1.39e-05
        [recompute_prepare]: 8.50001e-06
        [updatestate_depend_eliminate]: 6.14001e-06
        [updatestate_assign_eliminate]: 4.96002e-06
        [updatestate_loads_eliminate]: 4.07e-06
        [parameter_eliminate]: 1.29998e-06
        [specialize_transform]: 8.13001e-06
        [updatestate_useless_node_eliminater]: 1.145e-05
        [accelerated_algorithm]: 7.81001e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 7.53e-06
        [merge_forward]: 4.89e-06
        [cell_reuse_recompute_pass]: 3.29001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.755e-05
        [j_node_and_user_rematch]: 1.244e-05
        [meta_fg_expand]: 3.46001e-06
        [replace_old_param]: 1.071e-05
        [inline_without_move]: 7.56001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.81e-06
        [auto_monad_grad]: 1.84998e-06
        [auto_monad_eliminator]: 1.113e-05
        [cse]: 2.246e-05
        [replace_applicator]: 7.73001e-06
[py_interpret_to_execute_after_opt_a]: 1.634e-05
[rewriter_after_opt_a]: 6.002e-05
[convert_after_rewriter]: 9.19e-06
[order_py_execute_after_rewriter]: 6.34001e-06
[mutable_eliminate]: 0.00079626
[jit_opt_b]: 7.047e-05, [1]
    [Cycle 1]: 6.244e-05, [2]
        [frontend_op_eliminate]: 2.588e-05
        [inline_after_opt_a]: 2.503e-05
[cconv]: 2.647e-05
[loop_unroll]: 0.00044849
[jit_opt_after_cconv]: 0.00020323, [1]
    [Cycle 1]: 0.00019608, [11]
        [c_1]: 4.454e-05
        [parameter_eliminate]: 2.70002e-06
        [updatestate_depend_eliminate]: 7.90998e-06
        [updatestate_assign_eliminate]: 4.88001e-06
        [updatestate_loads_eliminate]: 4.41002e-06
        [cse]: 3.353e-05
        [call_graph_tuple_transform]: 2.294e-05
        [tuple_list_get_item_eliminator]: 9.22999e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 8.97e-06
[remove_dup_value]: 2.357e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 1.073e-05
[add_recomputation]: 7.118e-05
[cse_after_recomputation]: 3.719e-05, [1]
    [Cycle 1]: 3.147e-05, [1]
        [cse]: 2.504e-05
[auto_monad_reorder]: 2.499e-05
[get_jit_bprop_graph]: 1.52999e-06
[rewriter_after_jit_bprop_graph]: 5.20001e-06
[opt_after_jit_grad]: 0.00051938
[symbol_engine_optimizer]: 0.00010184, [1]
    [Cycle 1]: 9.486e-05, [6]
        [build]: 9.59999e-06
        [elim_shapecalc]: 1.274e-05
        [elim_not_effective]: 2.002e-05
        [opt_reshape]: 9.00001e-06
        [fold_const_symbol]: 1.395e-05
        [renormalize]: 4.50003e-07
[validate]: 5.054e-05
[backend_pass]: 1.27e-06
[task_emit]: 0.00836956
[execute]: 8.1e-06
Sums
    bootstrap                            :     0.000319s :  0.47%
    type_inference                       :     0.032213s : 47.63%
    event_method                         :     0.000124s :  0.18%
    auto_monad                           :     0.000229s :  0.34%
    graph_reusing                        :     0.000009s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000046s :  0.07%
    rewriter_before_opt_a                :     0.000128s :  0.19%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000159s :  0.23%
    jit_opt_a.loop_unroll                :     0.000115s :  0.17%
    jit_opt_a.a_1                        :     0.002237s :  3.31%
    jit_opt_a.with_stream_mark           :     0.000059s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000047s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000045s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000021s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000017s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.01%
    jit_opt_a.specialize_transform       :     0.000039s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000046s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000036s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000034s :  0.05%
    jit_opt_a.merge_forward              :     0.000022s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000072s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000060s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.005544s :  8.20%
    jit_opt_a.replace_old_param          :     0.000133s :  0.20%
    jit_opt_a.inline_without_move        :     0.000124s :  0.18%
    jit_opt_a.renormalize                :     0.014255s : 21.08%
    jit_opt_a.add_forward_monad_depend   :     0.000035s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000014s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000108s :  0.16%
    jit_opt_a.cse                        :     0.000369s :  0.55%
    jit_opt_a.replace_applicator         :     0.000213s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.02%
    rewriter_after_opt_a                 :     0.000060s :  0.09%
    convert_after_rewriter               :     0.000009s :  0.01%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000796s :  1.18%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.04%
    cconv                                :     0.000026s :  0.04%
    loop_unroll                          :     0.000448s :  0.66%
    jit_opt_after_cconv.c_1              :     0.000045s :  0.07%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.01%
    remove_dup_value                     :     0.000024s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000011s :  0.02%
    add_recomputation                    :     0.000071s :  0.11%
    cse_after_recomputation.cse          :     0.000025s :  0.04%
    auto_monad_reorder                   :     0.000025s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000519s :  0.77%
    symbol_engine_optimizer.build        :     0.000010s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000051s :  0.07%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008370s : 12.37%
    execute                              :     0.000008s :  0.01%


Time group info:
------[substitution.]   0.001433   160
 1.54% :     0.000022s :      8: substitution.depend_value_elim
 0.20% :     0.000003s :      4: substitution.elim_not_effective
 0.14% :     0.000002s :      4: substitution.fold_const_symbol
57.07% :     0.000818s :      4: substitution.getattr_setattr_resolve
 0.43% :     0.000006s :      5: substitution.graph_param_transform
23.80% :     0.000341s :     10: substitution.inline
 1.89% :     0.000027s :      4: substitution.inline_without_move
 0.84% :     0.000012s :     20: substitution.j_node_and_user_rematch
 0.67% :     0.000010s :      5: substitution.minmaximum_grad
 0.26% :     0.000004s :      3: substitution.partial_eliminate
 1.00% :     0.000014s :     20: substitution.remove_not_recompute_node
 3.47% :     0.000050s :     13: substitution.replace_applicator
 0.95% :     0.000014s :     18: substitution.replace_old_param
 0.18% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.37% :     0.000020s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.89% :     0.000013s :      5: substitution.tuple_list_get_item_depend_reorder
 2.36% :     0.000034s :     10: substitution.tuple_list_get_item_eliminator
 1.00% :     0.000014s :      8: substitution.updatestate_pure_node_eliminater
 1.93% :     0.000028s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.032133     2
94.03% :     0.030214s :      1: type_inference.infer
 5.97% :     0.001919s :      1: type_inference.specialize
------[replace.]   0.000285    20
16.91% :     0.000048s :      3: replace.getattr_setattr_resolve
30.23% :     0.000086s :     10: replace.inline
14.11% :     0.000040s :      1: replace.replace_applicator
30.71% :     0.000088s :      5: replace.tuple_list_get_item_eliminator
 8.05% :     0.000023s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001132    20
67.34% :     0.000762s :      3: match.getattr_setattr_resolve
29.55% :     0.000334s :     10: match.inline
 1.04% :     0.000012s :      1: match.replace_applicator
 1.25% :     0.000014s :      5: match.tuple_list_get_item_eliminator
 0.82% :     0.000009s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000442  2979
 1.42% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.55% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.27% :     0.000006s :     46: predicate.addn_check_dump
 1.49% :     0.000007s :     46: predicate.addn_zero_filter
 2.07% :     0.000009s :     46: predicate.arithmetic_simplify
 1.55% :     0.000007s :     46: predicate.cast_eliminate
 0.20% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.32% :     0.000006s :     46: predicate.compare_switch_simplify
 1.45% :     0.000006s :     46: predicate.depend_value_elim
 1.30% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.53% :     0.000007s :     46: predicate.dict_get_item_eliminator
 1.44% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.41% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      5: predicate.elim_not_effective
 0.22% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.32% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.29% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.46% :     0.000006s :     46: predicate.environ_get_eliminate
 1.28% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      5: predicate.fold_const_symbol
 1.00% :     0.000004s :     27: predicate.get_grad_eliminate
 1.44% :     0.000006s :     20: predicate.getattr_setattr_resolve
 0.12% :     0.000001s :      5: predicate.graph_param_transform
 3.95% :     0.000017s :     72: predicate.inline
 3.25% :     0.000014s :     91: predicate.inline_without_move
 0.43% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.32% :     0.000006s :     27: predicate.less_batch_normalization
 1.64% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.81% :     0.000008s :     56: predicate.load_eliminater
 0.49% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.82% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.75% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000006s :     46: predicate.merge_addn
 1.34% :     0.000006s :     46: predicate.minmaximum_grad
 0.72% :     0.000003s :      5: predicate.mutable_eliminate
 0.20% :     0.000001s :      5: predicate.opt_reshape
 2.15% :     0.000010s :     56: predicate.partial_eliminate
 1.39% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.82% :     0.000008s :     46: predicate.reduce_eliminate
 1.55% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000003s :     27: predicate.remove_not_recompute_node
 2.90% :     0.000013s :    126: predicate.replace_applicator
 1.70% :     0.000008s :     91: predicate.replace_old_param
 0.12% :     0.000001s :      5: predicate.reset_defer_inline
 1.42% :     0.000006s :     46: predicate.reshape_eliminate
 1.43% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.36% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.53% :     0.000007s :     46: predicate.same_eliminate
 0.61% :     0.000003s :     29: predicate.set_cell_output_no_recompute
 0.44% :     0.000002s :     10: predicate.special_op_eliminate
 1.07% :     0.000005s :     27: predicate.specialize_transform
 1.60% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000007s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.21% :     0.000010s :     62: predicate.switch_defer_inline
 2.04% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.29% :     0.000028s :    188: predicate.switch_simplify
 1.36% :     0.000006s :     46: predicate.tile_eliminate
 1.42% :     0.000006s :     46: predicate.transpose_eliminate
 1.74% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.58% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 2.98% :     0.000013s :     61: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000008s :     46: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.72% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000013s :     84: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000008s :     46: predicate.value_based_eliminate
 0.18% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.26% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003250    48
63.73% :     0.002071s :     27: func_graph_cloner_run.FuncGraphClonerGraph
36.27% :     0.001179s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.090775    93
 0.08% :     0.000075s :      1: add_recomputation
 0.26% :     0.000238s :      1: auto_monad
 0.03% :     0.000028s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.37% :     0.000334s :      1: bootstrap
 0.03% :     0.000029s :      1: cconv
 0.01% :     0.000012s :      1: convert_after_rewriter
 0.04% :     0.000039s :      1: cse_after_recomputation
 0.01% :     0.000013s :      1: environ_conv
 0.14% :     0.000131s :      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% :     0.000012s :      1: graph_reusing
30.84% :     0.027997s :      1: jit_opt_a
 0.23% :     0.000206s :      1: jit_opt_after_cconv
 0.08% :     0.000073s :      1: jit_opt_b
 0.50% :     0.000456s :      1: loop_unroll
 0.89% :     0.000804s :      1: mutable_eliminate
 3.60% :     0.003266s :     39: opt.transform.jit_opt_a
 0.09% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000044s :      4: opt.transform.jit_opt_b
 0.02% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 1.01% :     0.000915s :      2: opt.transform.opt_resolve
 0.06% :     0.000052s :      4: opt.transform.symbol_engine_opt
 0.58% :     0.000528s :      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% :     0.000006s :      1: pre_auto_parallel
 0.05% :     0.000049s :      1: py_interpret_to_execute
 0.02% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000026s :      1: remove_dup_value
11.95% :     0.010850s :      2: renormalize.infer
 3.73% :     0.003386s :      2: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000063s :      1: rewriter_after_opt_a
 0.14% :     0.000131s :      1: rewriter_before_opt_a
 0.12% :     0.000104s :      1: symbol_engine_optimizer
 9.24% :     0.008384s :      1: task_emit
35.50% :     0.032226s :      1: type_inference
 0.09% :     0.000084s :      1: validate

[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.250.689 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdc82cfc10> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.251.385 [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 = 0.0435289, [33]
[bootstrap]: 0.00027208
[type_inference]: 0.0215288
[event_method]: 2.018e-05
[auto_monad]: 0.00015012
[graph_reusing]: 5.97999e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 0.000131
[rewriter_before_opt_a]: 9.469e-05
[expand_dump_flag]: 4.18001e-06
[jit_opt_a]: 0.00606899, [2]
    [Cycle 1]: 0.00223565, [27]
        [switch_simplify]: 5.647e-05
        [loop_unroll]: 3.63e-05
        [a_1]: 0.00069351
        [with_stream_mark]: 1.763e-05
        [recompute_prepare]: 1.116e-05
        [updatestate_depend_eliminate]: 5.40999e-06
        [updatestate_assign_eliminate]: 5.08002e-06
        [updatestate_loads_eliminate]: 4.2e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 9.10001e-06
        [updatestate_useless_node_eliminater]: 1.199e-05
        [accelerated_algorithm]: 8.61002e-06
        [meta_shard_fg_expand]: 2.96001e-06
        [get_grad_eliminate_]: 8.17e-06
        [merge_forward]: 5.19e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.698e-05
        [j_node_and_user_rematch]: 1.321e-05
        [meta_fg_expand]: 3.59002e-06
        [replace_old_param]: 1.126e-05
        [inline_without_move]: 8e-06
        [renormalize]: 0.00105575
        [add_forward_monad_depend]: 5.77999e-06
        [auto_monad_grad]: 2.87002e-06
        [auto_monad_eliminator]: 2.064e-05
        [cse]: 3.445e-05
        [replace_applicator]: 1.546e-05
    [Cycle 2]: 0.00045847, [27]
        [switch_simplify]: 8.63001e-06
        [loop_unroll]: 7.96001e-06
        [a_1]: 0.00016505
        [with_stream_mark]: 1.172e-05
        [recompute_prepare]: 8.17e-06
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 4.53999e-06
        [updatestate_loads_eliminate]: 4.03001e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 7.92003e-06
        [updatestate_useless_node_eliminater]: 1.051e-05
        [accelerated_algorithm]: 8.28999e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 8.02998e-06
        [merge_forward]: 4.33001e-06
        [cell_reuse_recompute_pass]: 2.15002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.54e-05
        [j_node_and_user_rematch]: 1.242e-05
        [meta_fg_expand]: 2.79999e-06
        [replace_old_param]: 1.002e-05
        [inline_without_move]: 7.35e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.29003e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 1.028e-05
        [cse]: 1.97e-05
        [replace_applicator]: 7.65e-06
[py_interpret_to_execute_after_opt_a]: 1.316e-05
[rewriter_after_opt_a]: 0.00028029
[convert_after_rewriter]: 1.036e-05
[order_py_execute_after_rewriter]: 7.36999e-06
[mutable_eliminate]: 0.00061579
[jit_opt_b]: 6.864e-05, [1]
    [Cycle 1]: 6.043e-05, [2]
        [frontend_op_eliminate]: 2.435e-05
        [inline_after_opt_a]: 2.35e-05
[cconv]: 2.495e-05
[loop_unroll]: 0.00047873
[jit_opt_after_cconv]: 0.00019772, [1]
    [Cycle 1]: 0.0001903, [11]
        [c_1]: 4.485e-05
        [parameter_eliminate]: 3.04999e-06
        [updatestate_depend_eliminate]: 7.87998e-06
        [updatestate_assign_eliminate]: 4.78001e-06
        [updatestate_loads_eliminate]: 4.55001e-06
        [cse]: 3.125e-05
        [call_graph_tuple_transform]: 2.193e-05
        [tuple_list_get_item_eliminator]: 8.55999e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 3.50003e-07
        [switch_simplify]: 8.29002e-06
[remove_dup_value]: 2.091e-05
[partial_unused_args_eliminate]: 2.61e-06
[environ_conv]: 1.185e-05
[add_recomputation]: 6.086e-05
[cse_after_recomputation]: 2.961e-05, [1]
    [Cycle 1]: 2.414e-05, [1]
        [cse]: 1.798e-05
[auto_monad_reorder]: 2.432e-05
[get_jit_bprop_graph]: 2.39999e-06
[rewriter_after_jit_bprop_graph]: 4.75999e-06
[opt_after_jit_grad]: 0.00055372
[symbol_engine_optimizer]: 0.00010597, [1]
    [Cycle 1]: 9.922e-05, [6]
        [build]: 1.318e-05
        [elim_shapecalc]: 1.373e-05
        [elim_not_effective]: 1.919e-05
        [opt_reshape]: 9.10001e-06
        [fold_const_symbol]: 1.355e-05
        [renormalize]: 4.39992e-07
[validate]: 5.67e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.0124567
[execute]: 8.12998e-06
Sums
    bootstrap                            :     0.000272s :  0.69%
    type_inference                       :     0.021529s : 54.46%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000150s :  0.38%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000131s :  0.33%
    rewriter_before_opt_a                :     0.000095s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.16%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000859s :  2.17%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  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 :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001056s :  2.67%
    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      :     0.000031s :  0.08%
    jit_opt_a.cse                        :     0.000054s :  0.14%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000280s :  0.71%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000616s :  1.56%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000025s :  0.06%
    loop_unroll                          :     0.000479s :  1.21%
    jit_opt_after_cconv.c_1              :     0.000045s :  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 :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000061s :  0.15%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000024s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000554s :  1.40%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000057s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012457s : 31.51%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000261    47
 3.15% :     0.000008s :      2: substitution.depend_value_elim
 1.13% :     0.000003s :      4: substitution.elim_not_effective
 0.83% :     0.000002s :      4: substitution.fold_const_symbol
 2.54% :     0.000007s :      5: substitution.graph_param_transform
75.51% :     0.000197s :      4: substitution.inline
 1.64% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.48% :     0.000006s :      8: substitution.remove_not_recompute_node
 1.47% :     0.000004s :      2: substitution.replace_old_param
 4.63% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.34% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.27% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021464     2
92.97% :     0.019955s :      1: type_inference.infer
 7.03% :     0.001509s :      1: type_inference.specialize
------[replace.]   0.000095     6
66.43% :     0.000063s :      4: replace.inline
33.57% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     6
94.61% :     0.000194s :      4: match.inline
 5.39% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1021
 1.30% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     15: predicate.addn_check_dump
 1.38% :     0.000002s :     15: predicate.addn_zero_filter
 2.00% :     0.000003s :     15: predicate.arithmetic_simplify
 1.30% :     0.000002s :     15: predicate.cast_eliminate
 0.49% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     15: predicate.compare_switch_simplify
 1.36% :     0.000002s :     15: predicate.depend_value_elim
 1.15% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.90% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.53% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.31% :     0.000001s :      5: predicate.graph_param_transform
 5.11% :     0.000009s :     31: predicate.inline
 1.10% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.58% :     0.000003s :     10: predicate.less_batch_normalization
 1.44% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.90% :     0.000003s :     22: predicate.load_eliminater
 1.36% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.50% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.21% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     15: predicate.merge_addn
 1.14% :     0.000002s :     15: predicate.minmaximum_grad
 1.48% :     0.000003s :      5: predicate.mutable_eliminate
 0.52% :     0.000001s :      5: predicate.opt_reshape
 2.33% :     0.000004s :     22: predicate.partial_eliminate
 1.22% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     15: predicate.reduce_eliminate
 1.51% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.77% :     0.000003s :     27: predicate.replace_applicator
 0.57% :     0.000001s :     10: predicate.replace_old_param
 0.37% :     0.000001s :      5: predicate.reset_defer_inline
 1.26% :     0.000002s :     15: predicate.reshape_eliminate
 1.26% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     15: predicate.same_eliminate
 0.61% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.05% :     0.000002s :     10: predicate.special_op_eliminate
 1.08% :     0.000002s :     10: predicate.specialize_transform
 1.49% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.17% :     0.000004s :     21: predicate.switch_defer_inline
 1.95% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.47% :     0.000011s :     68: predicate.switch_simplify
 1.19% :     0.000002s :     15: predicate.tile_eliminate
 1.42% :     0.000002s :     15: predicate.transpose_eliminate
 1.60% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.75% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.91% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.35% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.93% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000771    13
42.85% :     0.000330s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.26% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.89% :     0.000400s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045921    76
 0.14% :     0.000064s :      1: add_recomputation
 0.34% :     0.000155s :      1: auto_monad
 0.06% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.62% :     0.000283s :      1: bootstrap
 0.06% :     0.000028s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000024s :      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% :     0.000009s :      1: graph_reusing
13.22% :     0.006073s :      1: jit_opt_a
 0.44% :     0.000201s :      1: jit_opt_after_cconv
 0.16% :     0.000071s :      1: jit_opt_b
 1.06% :     0.000488s :      1: loop_unroll
 1.36% :     0.000625s :      1: mutable_eliminate
 2.51% :     0.001151s :     26: opt.transform.jit_opt_a
 0.17% :     0.000080s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.23% :     0.000564s :      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
 0.30% :     0.000136s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000023s :      1: remove_dup_value
 1.33% :     0.000610s :      1: renormalize.infer
 0.95% :     0.000438s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.62% :     0.000285s :      1: rewriter_after_opt_a
 0.21% :     0.000099s :      1: rewriter_before_opt_a
 0.24% :     0.000109s :      1: symbol_engine_optimizer
27.16% :     0.012473s :      1: task_emit
46.91% :     0.021541s :      1: type_inference
 0.17% :     0.000080s :      1: validate


TotalTime = 0.0122591, [30]
[bootstrap]: 0.00027303
[type_inference]: 0.00462171
[event_method]: 2.088e-05
[auto_monad]: 6.206e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 3.564e-05
[rewriter_before_opt_a]: 9.059e-05
[expand_dump_flag]: 3.01999e-06
[jit_opt_a]: 0.00513266, [2]
    [Cycle 1]: 0.00166601, [27]
        [switch_simplify]: 8.766e-05
        [loop_unroll]: 3.762e-05
        [a_1]: 0.0006133
        [with_stream_mark]: 1.57e-05
        [recompute_prepare]: 8.2e-06
        [updatestate_depend_eliminate]: 3.83001e-06
        [updatestate_assign_eliminate]: 3.45e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [parameter_eliminate]: 2.42001e-06
        [specialize_transform]: 6.73e-06
        [updatestate_useless_node_eliminater]: 6.12001e-06
        [accelerated_algorithm]: 6.21998e-06
        [meta_shard_fg_expand]: 2.11998e-06
        [get_grad_eliminate_]: 5.86e-06
        [merge_forward]: 4.17998e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.564e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 1.04e-05
        [inline_without_move]: 6.06e-06
        [renormalize]: 0.00058546
        [add_forward_monad_depend]: 6.19001e-06
        [auto_monad_grad]: 2.01e-06
        [auto_monad_eliminator]: 1.411e-05
        [cse]: 2.974e-05
        [replace_applicator]: 1.25e-05
    [Cycle 2]: 0.00036049, [27]
        [switch_simplify]: 6.54001e-06
        [loop_unroll]: 6.07999e-06
        [a_1]: 0.00011212
        [with_stream_mark]: 8.14997e-06
        [recompute_prepare]: 1.182e-05
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 7.41001e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.51002e-06
        [get_grad_eliminate_]: 5.77001e-06
        [merge_forward]: 2.69999e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.395e-05
        [j_node_and_user_rematch]: 8.83001e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 5.51002e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 6.11e-06
        [cse]: 1.289e-05
        [replace_applicator]: 5.98002e-06
[py_interpret_to_execute_after_opt_a]: 8.85999e-06
[rewriter_after_opt_a]: 3.389e-05
[convert_after_rewriter]: 7.45e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00047437
[jit_opt_b]: 5.38e-05, [1]
    [Cycle 1]: 4.729e-05, [2]
        [frontend_op_eliminate]: 1.845e-05
        [inline_after_opt_a]: 1.719e-05
[cconv]: 2.522e-05
[loop_unroll]: 0.00041683
[jit_opt_after_cconv]: 0.00014962, [1]
    [Cycle 1]: 0.00014322, [11]
        [c_1]: 2.524e-05
        [parameter_eliminate]: 2.34001e-06
        [updatestate_depend_eliminate]: 5.97001e-06
        [updatestate_assign_eliminate]: 2.76999e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [cse]: 2.004e-05
        [call_graph_tuple_transform]: 1.994e-05
        [tuple_list_get_item_eliminator]: 6.38e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 6.11e-06
[remove_dup_value]: 1.339e-05
[partial_unused_args_eliminate]: 2.30002e-06
[environ_conv]: 5.92001e-06
[add_recomputation]: 4.569e-05
[cse_after_recomputation]: 2.356e-05, [1]
    [Cycle 1]: 1.812e-05, [1]
        [cse]: 1.227e-05
[auto_monad_reorder]: 1.697e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 4.84998e-06
[opt_after_jit_grad]: 0.00045177
[symbol_engine_optimizer]: 7.586e-05, [1]
    [Cycle 1]: 6.991e-05, [6]
        [build]: 3.12002e-06
        [elim_shapecalc]: 8.57e-06
        [elim_not_effective]: 1.358e-05
        [opt_reshape]: 6.56e-06
        [fold_const_symbol]: 9.66e-06
        [renormalize]: 4.40021e-07
[validate]: 3.548e-05
Sums
    bootstrap                            :     0.000273s :  3.18%
    type_inference                       :     0.004622s : 53.75%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000062s :  0.72%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.41%
    rewriter_before_opt_a                :     0.000091s :  1.05%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000094s :  1.10%
    jit_opt_a.loop_unroll                :     0.000044s :  0.51%
    jit_opt_a.a_1                        :     0.000725s :  8.44%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.23%
    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 :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000014s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000012s :  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 :     0.000030s :  0.34%
    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          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000586s :  6.81%
    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                        :     0.000043s :  0.50%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000474s :  5.52%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000025s :  0.29%
    loop_unroll                          :     0.000417s :  4.85%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.29%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.23%
    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      :     0.000000s :  0.00%
    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                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000046s :  0.53%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000017s :  0.20%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000452s :  5.25%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000035s :  0.41%


Time group info:
------[substitution.]   0.000205    28
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.72% :     0.000006s :      4: substitution.graph_param_transform
81.66% :     0.000167s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.45% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000005s :      4: substitution.replace_old_param
 7.67% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004561     2
79.98% :     0.003648s :      1: type_inference.infer
20.02% :     0.000913s :      1: type_inference.specialize
------[replace.]   0.000083     8
57.46% :     0.000048s :      4: replace.inline
42.54% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     8
92.28% :     0.000164s :      4: match.inline
 7.72% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   912
 1.21% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.10% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     13: predicate.addn_check_dump
 1.26% :     0.000002s :     13: predicate.addn_zero_filter
 1.85% :     0.000003s :     13: predicate.arithmetic_simplify
 1.27% :     0.000002s :     13: predicate.cast_eliminate
 0.48% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.19% :     0.000002s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.14% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      4: predicate.elim_not_effective
 0.56% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.22% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     13: predicate.environ_get_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 0.97% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.43% :     0.000007s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.13% :     0.000002s :      8: predicate.less_batch_normalization
 1.76% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     21: predicate.load_eliminater
 1.34% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.45% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 1.31% :     0.000002s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.55% :     0.000004s :     21: predicate.partial_eliminate
 1.13% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     13: predicate.reduce_eliminate
 1.55% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.92% :     0.000003s :     25: predicate.replace_applicator
 0.72% :     0.000001s :      8: predicate.replace_old_param
 0.39% :     0.000001s :      4: predicate.reset_defer_inline
 1.24% :     0.000002s :     13: predicate.reshape_eliminate
 1.14% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      8: predicate.special_op_eliminate
 0.94% :     0.000001s :      8: predicate.specialize_transform
 1.30% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     21: predicate.switch_defer_inline
 2.27% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.78% :     0.000011s :     67: predicate.switch_simplify
 1.18% :     0.000002s :     13: predicate.tile_eliminate
 1.30% :     0.000002s :     13: predicate.transpose_eliminate
 1.46% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000599    12
48.15% :     0.000288s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.81% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.03% :     0.000282s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013926    72
 0.35% :     0.000049s :      1: add_recomputation
 0.47% :     0.000065s :      1: auto_monad
 0.14% :     0.000019s :      1: auto_monad_reorder
 2.07% :     0.000289s :      1: bootstrap
 0.20% :     0.000028s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.18% :     0.000025s :      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
36.88% :     0.005135s :      1: jit_opt_a
 1.09% :     0.000152s :      1: jit_opt_after_cconv
 0.40% :     0.000056s :      1: jit_opt_b
 3.05% :     0.000425s :      1: loop_unroll
 3.46% :     0.000482s :      1: mutable_eliminate
 7.21% :     0.001004s :     26: opt.transform.jit_opt_a
 0.39% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000030s :      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
 0.16% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.30% :     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
 0.28% :     0.000038s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 2.18% :     0.000304s :      1: renormalize.infer
 1.97% :     0.000274s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000037s :      1: rewriter_after_opt_a
 0.67% :     0.000093s :      1: rewriter_before_opt_a
 0.56% :     0.000078s :      1: symbol_engine_optimizer
33.27% :     0.004633s :      1: type_inference


TotalTime = 0.0131222, [30]
[bootstrap]: 0.00027239
[type_inference]: 0.00479349
[event_method]: 2.273e-05
[auto_monad]: 6.578e-05
[graph_reusing]: 5.54e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 3.923e-05
[rewriter_before_opt_a]: 9.006e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.0055696, [2]
    [Cycle 1]: 0.00178899, [27]
        [switch_simplify]: 6.112e-05
        [loop_unroll]: 3.567e-05
        [a_1]: 0.00064019
        [with_stream_mark]: 1.667e-05
        [recompute_prepare]: 7.97998e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 3.45e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 7.05e-06
        [updatestate_useless_node_eliminater]: 6.17999e-06
        [accelerated_algorithm]: 6.53998e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 6.26e-06
        [merge_forward]: 4.37e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.551e-05
        [j_node_and_user_rematch]: 9.66e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 1.102e-05
        [inline_without_move]: 6.12999e-06
        [renormalize]: 0.00070365
        [add_forward_monad_depend]: 5.75001e-06
        [auto_monad_grad]: 2.55002e-06
        [auto_monad_eliminator]: 1.364e-05
        [cse]: 3.251e-05
        [replace_applicator]: 1.263e-05
    [Cycle 2]: 0.00035594, [27]
        [switch_simplify]: 6.96001e-06
        [loop_unroll]: 5.76e-06
        [a_1]: 0.00011659
        [with_stream_mark]: 8.89e-06
        [recompute_prepare]: 5.84e-06
        [updatestate_depend_eliminate]: 3.19001e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.79e-06
        [updatestate_useless_node_eliminater]: 5.62999e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 1.36998e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.329e-05
        [j_node_and_user_rematch]: 8.88002e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 9.22999e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 5.98998e-06
        [cse]: 1.285e-05
        [replace_applicator]: 5.52999e-06
[py_interpret_to_execute_after_opt_a]: 1.123e-05
[rewriter_after_opt_a]: 3.222e-05
[convert_after_rewriter]: 7.24001e-06
[order_py_execute_after_rewriter]: 5.40999e-06
[mutable_eliminate]: 0.00056823
[jit_opt_b]: 5.562e-05, [1]
    [Cycle 1]: 4.85e-05, [2]
        [frontend_op_eliminate]: 1.944e-05
        [inline_after_opt_a]: 1.776e-05
[cconv]: 2.604e-05
[loop_unroll]: 0.00044117
[jit_opt_after_cconv]: 0.00015158, [1]
    [Cycle 1]: 0.00014548, [11]
        [c_1]: 2.528e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 5.57999e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.77002e-06
        [cse]: 2.153e-05
        [call_graph_tuple_transform]: 1.966e-05
        [tuple_list_get_item_eliminator]: 6.08998e-06
        [none_parameter_eliminate]: 1.96e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 6.04001e-06
[remove_dup_value]: 1.486e-05
[partial_unused_args_eliminate]: 2.43002e-06
[environ_conv]: 5.91e-06
[add_recomputation]: 4.79e-05
[cse_after_recomputation]: 2.345e-05, [1]
    [Cycle 1]: 1.799e-05, [1]
        [cse]: 1.196e-05
[auto_monad_reorder]: 7.561e-05
[get_jit_bprop_graph]: 2.47001e-06
[rewriter_after_jit_bprop_graph]: 5.74e-06
[opt_after_jit_grad]: 0.00049942
[symbol_engine_optimizer]: 7.541e-05, [1]
    [Cycle 1]: 6.938e-05, [6]
        [build]: 3.06001e-06
        [elim_shapecalc]: 8.52e-06
        [elim_not_effective]: 1.306e-05
        [opt_reshape]: 6.15002e-06
        [fold_const_symbol]: 9.82999e-06
        [renormalize]: 4.99975e-07
[validate]: 3.88e-05
Sums
    bootstrap                            :     0.000272s :  2.98%
    type_inference                       :     0.004793s : 52.49%
    event_method                         :     0.000023s :  0.25%
    auto_monad                           :     0.000066s :  0.72%
    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                :     0.000090s :  0.99%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000068s :  0.75%
    jit_opt_a.loop_unroll                :     0.000041s :  0.45%
    jit_opt_a.a_1                        :     0.000757s :  8.29%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000014s :  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       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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_        :     0.000012s :  0.13%
    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 :     0.000029s :  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          :     0.000020s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000704s :  7.71%
    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 :  0.21%
    jit_opt_a.cse                        :     0.000045s :  0.50%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000032s :  0.35%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000568s :  6.22%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.19%
    cconv                                :     0.000026s :  0.29%
    loop_unroll                          :     0.000441s :  4.83%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.28%
    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 :     0.000003s :  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  :     0.000006s :  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                    :     0.000048s :  0.52%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000076s :  0.83%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000499s :  5.47%
    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 :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000039s :  0.42%


Time group info:
------[substitution.]   0.000226    28
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000006s :      4: substitution.graph_param_transform
82.86% :     0.000187s :      4: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000005s :      4: substitution.replace_old_param
 7.51% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004729     2
79.59% :     0.003764s :      1: type_inference.infer
20.41% :     0.000965s :      1: type_inference.specialize
------[replace.]   0.000088     8
56.65% :     0.000050s :      4: replace.inline
43.35% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     8
92.35% :     0.000184s :      4: match.inline
 7.65% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   912
 1.20% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.09% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     13: predicate.addn_check_dump
 1.39% :     0.000002s :     13: predicate.addn_zero_filter
 1.82% :     0.000003s :     13: predicate.arithmetic_simplify
 1.45% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.18% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      4: predicate.elim_not_effective
 0.52% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.22% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      4: predicate.fold_const_symbol
 0.99% :     0.000001s :      8: predicate.get_grad_eliminate
 0.29% :     0.000000s :      4: predicate.graph_param_transform
 5.45% :     0.000008s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      8: predicate.less_batch_normalization
 1.61% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     21: predicate.load_eliminater
 1.27% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.12% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     13: predicate.merge_addn
 1.10% :     0.000002s :     13: predicate.minmaximum_grad
 1.67% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.50% :     0.000003s :     21: predicate.partial_eliminate
 1.10% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     13: predicate.reduce_eliminate
 1.53% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.97% :     0.000003s :     25: predicate.replace_applicator
 0.81% :     0.000001s :      8: predicate.replace_old_param
 0.45% :     0.000001s :      4: predicate.reset_defer_inline
 1.21% :     0.000002s :     13: predicate.reshape_eliminate
 1.14% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     13: predicate.same_eliminate
 0.55% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.11% :     0.000002s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.40% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     21: predicate.switch_defer_inline
 2.15% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.61% :     0.000011s :     67: predicate.switch_simplify
 1.23% :     0.000002s :     13: predicate.tile_eliminate
 1.24% :     0.000002s :     13: predicate.transpose_eliminate
 1.43% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.15% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     13: predicate.value_based_eliminate
 0.35% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000647    12
50.60% :     0.000327s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.86% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.54% :     0.000288s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014898    72
 0.34% :     0.000051s :      1: add_recomputation
 0.47% :     0.000069s :      1: auto_monad
 0.53% :     0.000079s :      1: auto_monad_reorder
 1.93% :     0.000288s :      1: bootstrap
 0.19% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.18% :     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
37.41% :     0.005572s :      1: jit_opt_a
 1.04% :     0.000155s :      1: jit_opt_after_cconv
 0.39% :     0.000058s :      1: jit_opt_b
 3.01% :     0.000449s :      1: loop_unroll
 3.86% :     0.000575s :      1: mutable_eliminate
 6.71% :     0.000999s :     26: opt.transform.jit_opt_a
 0.36% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000031s :      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% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.41% :     0.000508s :      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% :     0.000042s :      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
 2.61% :     0.000389s :      1: renormalize.infer
 2.05% :     0.000306s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000035s :      1: rewriter_after_opt_a
 0.63% :     0.000093s :      1: rewriter_before_opt_a
 0.52% :     0.000078s :      1: symbol_engine_optimizer
32.27% :     0.004808s :      1: type_inference


TotalTime = 0.0127977, [30]
[bootstrap]: 0.0002937
[type_inference]: 0.00488341
[event_method]: 2.194e-05
[auto_monad]: 6.564e-05
[graph_reusing]: 5.92999e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.601e-05
[rewriter_before_opt_a]: 8.782e-05
[expand_dump_flag]: 2.94001e-06
[jit_opt_a]: 0.00534433, [2]
    [Cycle 1]: 0.00174614, [27]
        [switch_simplify]: 5.78e-05
        [loop_unroll]: 3.464e-05
        [a_1]: 0.00064601
        [with_stream_mark]: 1.837e-05
        [recompute_prepare]: 7.97e-06
        [updatestate_depend_eliminate]: 3.89002e-06
        [updatestate_assign_eliminate]: 3.92998e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 6.78998e-06
        [updatestate_useless_node_eliminater]: 6.01e-06
        [accelerated_algorithm]: 6.44999e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 6.66999e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.565e-05
        [j_node_and_user_rematch]: 1.029e-05
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 1.098e-05
        [inline_without_move]: 5.77001e-06
        [renormalize]: 0.00064719
        [add_forward_monad_depend]: 5.30001e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 1.431e-05
        [cse]: 3.626e-05
        [replace_applicator]: 1.328e-05
    [Cycle 2]: 0.00036291, [27]
        [switch_simplify]: 6.64999e-06
        [loop_unroll]: 5.44998e-06
        [a_1]: 0.00011316
        [with_stream_mark]: 9.34e-06
        [recompute_prepare]: 5.79e-06
        [updatestate_depend_eliminate]: 2.95998e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 5.70001e-06
        [updatestate_useless_node_eliminater]: 7.86001e-06
        [accelerated_algorithm]: 5.47999e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 5.17999e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.384e-05
        [j_node_and_user_rematch]: 9.56e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 9.69e-06
        [inline_without_move]: 5.51e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 6.64001e-06
        [cse]: 1.301e-05
        [replace_applicator]: 5.84e-06
[py_interpret_to_execute_after_opt_a]: 1.035e-05
[rewriter_after_opt_a]: 3.495e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00049102
[jit_opt_b]: 5.349e-05, [1]
    [Cycle 1]: 4.686e-05, [2]
        [frontend_op_eliminate]: 1.834e-05
        [inline_after_opt_a]: 1.716e-05
[cconv]: 2.572e-05
[loop_unroll]: 0.00042592
[jit_opt_after_cconv]: 0.00014855, [1]
    [Cycle 1]: 0.00014202, [11]
        [c_1]: 2.432e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 5.91e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [cse]: 2.037e-05
        [call_graph_tuple_transform]: 1.94e-05
        [tuple_list_get_item_eliminator]: 6.44001e-06
        [none_parameter_eliminate]: 1.53002e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 6.19001e-06
[remove_dup_value]: 1.628e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 5.89999e-06
[add_recomputation]: 4.915e-05
[cse_after_recomputation]: 2.273e-05, [1]
    [Cycle 1]: 1.725e-05, [1]
        [cse]: 1.198e-05
[auto_monad_reorder]: 1.711e-05
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 4.84003e-06
[opt_after_jit_grad]: 0.00045084
[symbol_engine_optimizer]: 7.755e-05, [1]
    [Cycle 1]: 7.081e-05, [6]
        [build]: 4.23999e-06
        [elim_shapecalc]: 8.79e-06
        [elim_not_effective]: 1.423e-05
        [opt_reshape]: 6.06998e-06
        [fold_const_symbol]: 9.41998e-06
        [renormalize]: 3.59985e-07
[validate]: 3.562e-05
Sums
    bootstrap                            :     0.000294s :  3.27%
    type_inference                       :     0.004883s : 54.33%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000066s :  0.73%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.40%
    rewriter_before_opt_a                :     0.000088s :  0.98%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000064s :  0.72%
    jit_opt_a.loop_unroll                :     0.000040s :  0.45%
    jit_opt_a.a_1                        :     0.000759s :  8.45%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.31%
    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 :     0.000006s :  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 :     0.000014s :  0.15%
    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_        :     0.000012s :  0.13%
    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 :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000021s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000647s :  7.20%
    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                        :     0.000049s :  0.55%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000035s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000491s :  5.46%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000026s :  0.29%
    loop_unroll                          :     0.000426s :  4.74%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.27%
    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              :     0.000020s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  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                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000049s :  0.55%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000017s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000451s :  5.02%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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 :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.40%


Time group info:
------[substitution.]   0.000229    28
 1.06% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000006s :      4: substitution.graph_param_transform
82.52% :     0.000189s :      4: substitution.inline
 1.45% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000006s :      4: substitution.replace_old_param
 7.42% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004815     2
80.81% :     0.003891s :      1: type_inference.infer
19.19% :     0.000924s :      1: type_inference.specialize
------[replace.]   0.000082     8
57.08% :     0.000047s :      4: replace.inline
42.92% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000201     8
92.61% :     0.000186s :      4: match.inline
 7.39% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   912
 1.30% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     13: predicate.addn_check_dump
 1.22% :     0.000002s :     13: predicate.addn_zero_filter
 1.75% :     0.000002s :     13: predicate.arithmetic_simplify
 1.34% :     0.000002s :     13: predicate.cast_eliminate
 0.41% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.24% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.50% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     13: predicate.environ_get_eliminate
 1.21% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 1.07% :     0.000001s :      8: predicate.get_grad_eliminate
 0.40% :     0.000001s :      4: predicate.graph_param_transform
 5.48% :     0.000007s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.13% :     0.000002s :      8: predicate.less_batch_normalization
 1.65% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.08% :     0.000003s :     21: predicate.load_eliminater
 1.15% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.09% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 1.31% :     0.000002s :      4: predicate.mutable_eliminate
 0.40% :     0.000001s :      4: predicate.opt_reshape
 2.52% :     0.000003s :     21: predicate.partial_eliminate
 1.24% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     13: predicate.reduce_eliminate
 1.64% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.97% :     0.000003s :     25: predicate.replace_applicator
 0.77% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000000s :      4: predicate.reset_defer_inline
 1.33% :     0.000002s :     13: predicate.reshape_eliminate
 1.17% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      8: predicate.special_op_eliminate
 0.95% :     0.000001s :      8: predicate.specialize_transform
 1.38% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     21: predicate.switch_defer_inline
 2.23% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.65% :     0.000010s :     67: predicate.switch_simplify
 1.22% :     0.000002s :     13: predicate.tile_eliminate
 1.24% :     0.000002s :     13: predicate.transpose_eliminate
 1.42% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.87% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000629    12
48.80% :     0.000307s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.23% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.97% :     0.000289s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014519    72
 0.36% :     0.000052s :      1: add_recomputation
 0.48% :     0.000070s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 2.15% :     0.000313s :      1: bootstrap
 0.20% :     0.000029s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.18% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
36.83% :     0.005347s :      1: jit_opt_a
 1.04% :     0.000151s :      1: jit_opt_after_cconv
 0.39% :     0.000056s :      1: jit_opt_b
 2.99% :     0.000434s :      1: loop_unroll
 3.44% :     0.000500s :      1: mutable_eliminate
 6.88% :     0.001000s :     26: opt.transform.jit_opt_a
 0.37% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000030s :      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% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.16% :     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
 0.27% :     0.000039s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000019s :      1: remove_dup_value
 2.38% :     0.000346s :      1: renormalize.infer
 2.02% :     0.000293s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000038s :      1: rewriter_after_opt_a
 0.62% :     0.000091s :      1: rewriter_before_opt_a
 0.55% :     0.000080s :      1: symbol_engine_optimizer
33.72% :     0.004896s :      1: type_inference


TotalTime = 0.0781091, [33]
[bootstrap]: 0.00022245
[type_inference]: 0.0349886
[event_method]: 0.00014483
[auto_monad]: 0.00022297
[graph_reusing]: 8.33001e-06
[pre_auto_parallel]: 3.23e-06
[py_interpret_to_execute]: 4.1e-05
[rewriter_before_opt_a]: 0.00013239
[expand_dump_flag]: 3.54002e-06
[jit_opt_a]: 0.0302678, [3]
    [Cycle 1]: 0.022002, [27]
        [switch_simplify]: 7.991e-05
        [loop_unroll]: 4.909e-05
        [a_1]: 0.00114232
        [with_stream_mark]: 3.404e-05
        [recompute_prepare]: 2.602e-05
        [updatestate_depend_eliminate]: 1.303e-05
        [updatestate_assign_eliminate]: 1.136e-05
        [updatestate_loads_eliminate]: 9.52999e-06
        [parameter_eliminate]: 3.08e-06
        [specialize_transform]: 2.032e-05
        [updatestate_useless_node_eliminater]: 2.269e-05
        [accelerated_algorithm]: 1.862e-05
        [meta_shard_fg_expand]: 5.36002e-06
        [get_grad_eliminate_]: 1.872e-05
        [merge_forward]: 1.133e-05
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.862e-05
        [j_node_and_user_rematch]: 9.575e-05
        [meta_fg_expand]: 0.00559563
        [replace_old_param]: 0.00011817
        [inline_without_move]: 0.00011418
        [renormalize]: 0.0137024
        [add_forward_monad_depend]: 1.844e-05
        [auto_monad_grad]: 1.032e-05
        [auto_monad_eliminator]: 8.347e-05
        [cse]: 0.00030606
        [replace_applicator]: 0.00019097
    [Cycle 2]: 0.00331337, [27]
        [switch_simplify]: 6.096e-05
        [loop_unroll]: 5.83e-05
        [a_1]: 0.00103291
        [with_stream_mark]: 2.278e-05
        [recompute_prepare]: 1.33e-05
        [updatestate_depend_eliminate]: 2.711e-05
        [updatestate_assign_eliminate]: 6.16e-06
        [updatestate_loads_eliminate]: 4.80001e-06
        [parameter_eliminate]: 2.21998e-06
        [specialize_transform]: 1.061e-05
        [updatestate_useless_node_eliminater]: 1.142e-05
        [accelerated_algorithm]: 9.56998e-06
        [meta_shard_fg_expand]: 3.4e-06
        [get_grad_eliminate_]: 8.80999e-06
        [merge_forward]: 6.28998e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.099e-05
        [j_node_and_user_rematch]: 1.498e-05
        [meta_fg_expand]: 9.406e-05
        [replace_old_param]: 1.452e-05
        [inline_without_move]: 8.87999e-06
        [renormalize]: 0.0015594
        [add_forward_monad_depend]: 8.25e-06
        [auto_monad_grad]: 3.38e-06
        [auto_monad_eliminator]: 2.244e-05
        [cse]: 8.307e-05
        [replace_applicator]: 2.082e-05
    [Cycle 3]: 0.00068538, [27]
        [switch_simplify]: 1.069e-05
        [loop_unroll]: 8.65999e-06
        [a_1]: 0.00018113
        [with_stream_mark]: 1.512e-05
        [recompute_prepare]: 8.75001e-06
        [updatestate_depend_eliminate]: 7.4e-06
        [updatestate_assign_eliminate]: 5.26002e-06
        [updatestate_loads_eliminate]: 4.15999e-06
        [parameter_eliminate]: 1.60001e-06
        [specialize_transform]: 1.03e-05
        [updatestate_useless_node_eliminater]: 8.769e-05
        [accelerated_algorithm]: 1.196e-05
        [meta_shard_fg_expand]: 3.19001e-06
        [get_grad_eliminate_]: 9.77999e-06
        [merge_forward]: 7.68999e-06
        [cell_reuse_recompute_pass]: 3.91001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.201e-05
        [j_node_and_user_rematch]: 4.18e-05
        [meta_fg_expand]: 4.01001e-06
        [replace_old_param]: 1.441e-05
        [inline_without_move]: 8.35001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 3.32002e-06
        [auto_monad_grad]: 1.56998e-06
        [auto_monad_eliminator]: 1.849e-05
        [cse]: 3.237e-05
        [replace_applicator]: 1.111e-05
[py_interpret_to_execute_after_opt_a]: 1.899e-05
[rewriter_after_opt_a]: 6.753e-05
[convert_after_rewriter]: 9.81e-06
[order_py_execute_after_rewriter]: 6.64001e-06
[mutable_eliminate]: 0.00080155
[jit_opt_b]: 7.007e-05, [1]
    [Cycle 1]: 6.201e-05, [2]
        [frontend_op_eliminate]: 2.385e-05
        [inline_after_opt_a]: 2.502e-05
[cconv]: 3.025e-05
[loop_unroll]: 0.00055056
[jit_opt_after_cconv]: 0.00021978, [1]
    [Cycle 1]: 0.00021216, [11]
        [c_1]: 4.747e-05
        [parameter_eliminate]: 2.98998e-06
        [updatestate_depend_eliminate]: 8.28999e-06
        [updatestate_assign_eliminate]: 4.53001e-06
        [updatestate_loads_eliminate]: 5.20001e-06
        [cse]: 4.096e-05
        [call_graph_tuple_transform]: 2.704e-05
        [tuple_list_get_item_eliminator]: 8.02003e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 8.48001e-06
[remove_dup_value]: 2.606e-05
[partial_unused_args_eliminate]: 2.61999e-06
[environ_conv]: 1.198e-05
[add_recomputation]: 7.967e-05
[cse_after_recomputation]: 4.207e-05, [1]
    [Cycle 1]: 3.468e-05, [1]
        [cse]: 2.669e-05
[auto_monad_reorder]: 3.189e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 7.13998e-06
[opt_after_jit_grad]: 0.00059715
[symbol_engine_optimizer]: 0.00010704, [1]
    [Cycle 1]: 0.00010048, [6]
        [build]: 1.392e-05
        [elim_shapecalc]: 1.227e-05
        [elim_not_effective]: 1.929e-05
        [opt_reshape]: 9.27001e-06
        [fold_const_symbol]: 1.508e-05
        [renormalize]: 3.39991e-07
[validate]: 5.548e-05
[backend_pass]: 1.10999e-06
[task_emit]: 0.00908461
[execute]: 8.37e-06
Sums
    bootstrap                            :     0.000222s :  0.31%
    type_inference                       :     0.034989s : 48.02%
    event_method                         :     0.000145s :  0.20%
    auto_monad                           :     0.000223s :  0.31%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000041s :  0.06%
    rewriter_before_opt_a                :     0.000132s :  0.18%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000152s :  0.21%
    jit_opt_a.loop_unroll                :     0.000116s :  0.16%
    jit_opt_a.a_1                        :     0.002356s :  3.23%
    jit_opt_a.with_stream_mark           :     0.000072s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000048s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000048s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000023s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000018s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000007s :  0.01%
    jit_opt_a.specialize_transform       :     0.000041s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000122s :  0.17%
    jit_opt_a.accelerated_algorithm      :     0.000040s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000012s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000037s :  0.05%
    jit_opt_a.merge_forward              :     0.000025s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000082s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000153s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.005694s :  7.81%
    jit_opt_a.replace_old_param          :     0.000147s :  0.20%
    jit_opt_a.inline_without_move        :     0.000131s :  0.18%
    jit_opt_a.renormalize                :     0.015262s : 20.94%
    jit_opt_a.add_forward_monad_depend   :     0.000030s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000015s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000124s :  0.17%
    jit_opt_a.cse                        :     0.000422s :  0.58%
    jit_opt_a.replace_applicator         :     0.000223s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.03%
    rewriter_after_opt_a                 :     0.000068s :  0.09%
    convert_after_rewriter               :     0.000010s :  0.01%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000802s :  1.10%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.03%
    cconv                                :     0.000030s :  0.04%
    loop_unroll                          :     0.000551s :  0.76%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.07%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000041s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.04%
    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      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.01%
    remove_dup_value                     :     0.000026s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000012s :  0.02%
    add_recomputation                    :     0.000080s :  0.11%
    cse_after_recomputation.cse          :     0.000027s :  0.04%
    auto_monad_reorder                   :     0.000032s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000597s :  0.82%
    symbol_engine_optimizer.build        :     0.000014s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000055s :  0.08%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009085s : 12.47%
    execute                              :     0.000008s :  0.01%


Time group info:
------[substitution.]   0.001479   160
 1.41% :     0.000021s :      8: substitution.depend_value_elim
 0.18% :     0.000003s :      4: substitution.elim_not_effective
 0.18% :     0.000003s :      4: substitution.fold_const_symbol
53.34% :     0.000789s :      4: substitution.getattr_setattr_resolve
 0.54% :     0.000008s :      5: substitution.graph_param_transform
26.52% :     0.000392s :     10: substitution.inline
 2.15% :     0.000032s :      4: substitution.inline_without_move
 0.89% :     0.000013s :     20: substitution.j_node_and_user_rematch
 0.66% :     0.000010s :      5: substitution.minmaximum_grad
 0.26% :     0.000004s :      3: substitution.partial_eliminate
 1.03% :     0.000015s :     20: substitution.remove_not_recompute_node
 3.37% :     0.000050s :     13: substitution.replace_applicator
 1.04% :     0.000015s :     18: substitution.replace_old_param
 0.22% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.39% :     0.000021s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.90% :     0.000013s :      5: substitution.tuple_list_get_item_depend_reorder
 2.60% :     0.000038s :     10: substitution.tuple_list_get_item_eliminator
 1.17% :     0.000017s :      8: substitution.updatestate_pure_node_eliminater
 2.16% :     0.000032s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.034893     2
93.97% :     0.032790s :      1: type_inference.infer
 6.03% :     0.002102s :      1: type_inference.specialize
------[replace.]   0.000317    20
15.70% :     0.000050s :      3: replace.getattr_setattr_resolve
30.70% :     0.000097s :     10: replace.inline
14.30% :     0.000045s :      1: replace.replace_applicator
32.09% :     0.000102s :      5: replace.tuple_list_get_item_eliminator
 7.21% :     0.000023s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001168    20
63.63% :     0.000743s :      3: match.getattr_setattr_resolve
33.04% :     0.000386s :     10: match.inline
 0.89% :     0.000010s :      1: match.replace_applicator
 1.66% :     0.000019s :      5: match.tuple_list_get_item_eliminator
 0.78% :     0.000009s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000458  2979
 1.39% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.49% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.24% :     0.000006s :     46: predicate.addn_check_dump
 1.51% :     0.000007s :     46: predicate.addn_zero_filter
 2.16% :     0.000010s :     46: predicate.arithmetic_simplify
 1.74% :     0.000008s :     46: predicate.cast_eliminate
 0.26% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.30% :     0.000006s :     46: predicate.compare_switch_simplify
 1.43% :     0.000007s :     46: predicate.depend_value_elim
 1.34% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.54% :     0.000007s :     46: predicate.dict_get_item_eliminator
 1.42% :     0.000007s :     46: predicate.dict_set_item_eliminator
 0.28% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.15% :     0.000001s :      5: predicate.elim_not_effective
 0.24% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000007s :     46: predicate.environ_add_const_eliminate
 1.34% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.28% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.42% :     0.000007s :     46: predicate.environ_get_eliminate
 1.23% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      5: predicate.fold_const_symbol
 1.16% :     0.000005s :     27: predicate.get_grad_eliminate
 1.31% :     0.000006s :     20: predicate.getattr_setattr_resolve
 0.10% :     0.000000s :      5: predicate.graph_param_transform
 3.72% :     0.000017s :     72: predicate.inline
 3.39% :     0.000016s :     91: predicate.inline_without_move
 0.44% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.50% :     0.000007s :     27: predicate.less_batch_normalization
 1.70% :     0.000008s :     51: predicate.list_to_tuple_eliminator_
 1.67% :     0.000008s :     56: predicate.load_eliminater
 0.55% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.74% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.71% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.31% :     0.000006s :     46: predicate.merge_addn
 1.33% :     0.000006s :     46: predicate.minmaximum_grad
 0.58% :     0.000003s :      5: predicate.mutable_eliminate
 0.23% :     0.000001s :      5: predicate.opt_reshape
 2.10% :     0.000010s :     56: predicate.partial_eliminate
 1.37% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.75% :     0.000008s :     46: predicate.reduce_eliminate
 1.68% :     0.000008s :     51: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000003s :     27: predicate.remove_not_recompute_node
 2.82% :     0.000013s :    126: predicate.replace_applicator
 1.75% :     0.000008s :     91: predicate.replace_old_param
 0.11% :     0.000001s :      5: predicate.reset_defer_inline
 1.45% :     0.000007s :     46: predicate.reshape_eliminate
 1.46% :     0.000007s :     46: predicate.row_tensor_add_zeros_like
 0.24% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.42% :     0.000007s :     46: predicate.same_eliminate
 0.57% :     0.000003s :     29: predicate.set_cell_output_no_recompute
 0.41% :     0.000002s :     10: predicate.special_op_eliminate
 1.01% :     0.000005s :     27: predicate.specialize_transform
 1.56% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.22% :     0.000010s :     62: predicate.switch_defer_inline
 2.06% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.31% :     0.000029s :    188: predicate.switch_simplify
 1.38% :     0.000006s :     46: predicate.tile_eliminate
 1.40% :     0.000006s :     46: predicate.transpose_eliminate
 1.71% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.57% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 3.20% :     0.000015s :     61: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000008s :     46: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.67% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000014s :     84: predicate.updatestate_useless_node_eliminater
 2.00% :     0.000009s :     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.]   0.003578    48
64.65% :     0.002313s :     27: func_graph_cloner_run.FuncGraphClonerGraph
35.35% :     0.001265s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.098017    93
 0.09% :     0.000084s :      1: add_recomputation
 0.24% :     0.000231s :      1: auto_monad
 0.04% :     0.000036s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.24% :     0.000234s :      1: bootstrap
 0.03% :     0.000033s :      1: cconv
 0.01% :     0.000013s :      1: convert_after_rewriter
 0.05% :     0.000044s :      1: cse_after_recomputation
 0.02% :     0.000015s :      1: environ_conv
 0.15% :     0.000152s :      1: event_method
 0.01% :     0.000014s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000012s :      1: graph_reusing
30.88% :     0.030272s :      1: jit_opt_a
 0.23% :     0.000223s :      1: jit_opt_after_cconv
 0.07% :     0.000072s :      1: jit_opt_b
 0.57% :     0.000559s :      1: loop_unroll
 0.82% :     0.000808s :      1: mutable_eliminate
 3.66% :     0.003587s :     39: opt.transform.jit_opt_a
 0.09% :     0.000087s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000041s :      4: opt.transform.jit_opt_b
 0.02% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.91% :     0.000890s :      2: opt.transform.opt_resolve
 0.05% :     0.000051s :      4: opt.transform.symbol_engine_opt
 0.62% :     0.000605s :      1: opt_after_jit_grad
 0.01% :     0.000009s :      1: order_py_execute_after_rewriter
 0.00% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.04% :     0.000044s :      1: py_interpret_to_execute
 0.02% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000029s :      1: remove_dup_value
12.04% :     0.011797s :      2: renormalize.infer
 3.51% :     0.003440s :      2: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000070s :      1: rewriter_after_opt_a
 0.14% :     0.000135s :      1: rewriter_before_opt_a
 0.11% :     0.000110s :      1: symbol_engine_optimizer
 9.29% :     0.009103s :      1: task_emit
35.72% :     0.035008s :      1: type_inference
 0.08% :     0.000083s :      1: validate


TotalTime = 0.0439221, [33]
[bootstrap]: 0.00029753
[type_inference]: 0.02224
[event_method]: 2.161e-05
[auto_monad]: 0.00015164
[graph_reusing]: 6.49001e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 0.00012272
[rewriter_before_opt_a]: 9.558e-05
[expand_dump_flag]: 4.03001e-06
[jit_opt_a]: 0.00673181, [2]
    [Cycle 1]: 0.00249486, [27]
        [switch_simplify]: 5.574e-05
        [loop_unroll]: 3.608e-05
        [a_1]: 0.00076357
        [with_stream_mark]: 1.898e-05
        [recompute_prepare]: 1.074e-05
        [updatestate_depend_eliminate]: 6.38e-06
        [updatestate_assign_eliminate]: 4.78001e-06
        [updatestate_loads_eliminate]: 4.46002e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 9.84999e-06
        [updatestate_useless_node_eliminater]: 1.217e-05
        [accelerated_algorithm]: 8.57e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 8.00999e-06
        [merge_forward]: 4.89e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.693e-05
        [j_node_and_user_rematch]: 1.356e-05
        [meta_fg_expand]: 3.71999e-06
        [replace_old_param]: 1.223e-05
        [inline_without_move]: 7.97003e-06
        [renormalize]: 0.0012377
        [add_forward_monad_depend]: 7.77e-06
        [auto_monad_grad]: 2.36998e-06
        [auto_monad_eliminator]: 2.05e-05
        [cse]: 2.696e-05
        [replace_applicator]: 1.751e-05
    [Cycle 2]: 0.00047563, [27]
        [switch_simplify]: 8.99998e-06
        [loop_unroll]: 7.84002e-06
        [a_1]: 0.0001705
        [with_stream_mark]: 1.246e-05
        [recompute_prepare]: 7.82e-06
        [updatestate_depend_eliminate]: 4.80999e-06
        [updatestate_assign_eliminate]: 5.22999e-06
        [updatestate_loads_eliminate]: 3.98999e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 8.17998e-06
        [updatestate_useless_node_eliminater]: 1.042e-05
        [accelerated_algorithm]: 8.22e-06
        [meta_shard_fg_expand]: 2.18002e-06
        [get_grad_eliminate_]: 8.3e-06
        [merge_forward]: 4.42e-06
        [cell_reuse_recompute_pass]: 2.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.632e-05
        [j_node_and_user_rematch]: 1.241e-05
        [meta_fg_expand]: 3.13e-06
        [replace_old_param]: 1.044e-05
        [inline_without_move]: 7.45003e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 1.09998e-06
        [auto_monad_eliminator]: 1.03e-05
        [cse]: 2.097e-05
        [replace_applicator]: 8.57e-06
[py_interpret_to_execute_after_opt_a]: 1.441e-05
[rewriter_after_opt_a]: 0.00032084
[convert_after_rewriter]: 1.103e-05
[order_py_execute_after_rewriter]: 6.99001e-06
[mutable_eliminate]: 0.00072047
[jit_opt_b]: 7.548e-05, [1]
    [Cycle 1]: 6.655e-05, [2]
        [frontend_op_eliminate]: 2.594e-05
        [inline_after_opt_a]: 2.682e-05
[cconv]: 2.567e-05
[loop_unroll]: 0.00054105
[jit_opt_after_cconv]: 0.00021093, [1]
    [Cycle 1]: 0.0002023, [11]
        [c_1]: 4.755e-05
        [parameter_eliminate]: 3.67998e-06
        [updatestate_depend_eliminate]: 8.87999e-06
        [updatestate_assign_eliminate]: 5.05001e-06
        [updatestate_loads_eliminate]: 4.52998e-06
        [cse]: 3.376e-05
        [call_graph_tuple_transform]: 2.301e-05
        [tuple_list_get_item_eliminator]: 8.97999e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 2.9002e-07
        [switch_simplify]: 8.74e-06
[remove_dup_value]: 1.958e-05
[partial_unused_args_eliminate]: 2.65002e-06
[environ_conv]: 1.288e-05
[add_recomputation]: 6.185e-05
[cse_after_recomputation]: 3.519e-05, [1]
    [Cycle 1]: 2.816e-05, [1]
        [cse]: 2.012e-05
[auto_monad_reorder]: 2.299e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 5.87001e-06
[opt_after_jit_grad]: 0.00059431
[symbol_engine_optimizer]: 0.00010126, [1]
    [Cycle 1]: 9.425e-05, [6]
        [build]: 1.057e-05
        [elim_shapecalc]: 1.173e-05
        [elim_not_effective]: 1.822e-05
        [opt_reshape]: 8.94998e-06
        [fold_const_symbol]: 1.493e-05
        [renormalize]: 6.89994e-07
[validate]: 5.705e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.0111741
[execute]: 7.91001e-06
Sums
    bootstrap                            :     0.000298s :  0.75%
    type_inference                       :     0.022240s : 56.30%
    event_method                         :     0.000022s :  0.05%
    auto_monad                           :     0.000152s :  0.38%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000123s :  0.31%
    rewriter_before_opt_a                :     0.000096s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.16%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000934s :  2.36%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001238s :  3.13%
    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      :     0.000031s :  0.08%
    jit_opt_a.cse                        :     0.000048s :  0.12%
    jit_opt_a.replace_applicator         :     0.000026s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.04%
    rewriter_after_opt_a                 :     0.000321s :  0.81%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000720s :  1.82%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.07%
    cconv                                :     0.000026s :  0.06%
    loop_unroll                          :     0.000541s :  1.37%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000062s :  0.16%
    cse_after_recomputation.cse          :     0.000020s :  0.05%
    auto_monad_reorder                   :     0.000023s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000594s :  1.50%
    symbol_engine_optimizer.build        :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000057s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011174s : 28.29%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000306    47
 2.89% :     0.000009s :      2: substitution.depend_value_elim
 0.81% :     0.000002s :      4: substitution.elim_not_effective
 0.79% :     0.000002s :      4: substitution.fold_const_symbol
 2.18% :     0.000007s :      5: substitution.graph_param_transform
77.51% :     0.000237s :      4: substitution.inline
 1.35% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.47% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.66% :     0.000005s :      2: substitution.replace_old_param
 4.40% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 2.75% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.19% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022169     2
93.62% :     0.020753s :      1: type_inference.infer
 6.38% :     0.001415s :      1: type_inference.specialize
------[replace.]   0.000108     6
64.32% :     0.000070s :      4: replace.inline
35.68% :     0.000039s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     6
95.01% :     0.000234s :      4: match.inline
 4.99% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000176  1021
 1.32% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     15: predicate.addn_check_dump
 1.27% :     0.000002s :     15: predicate.addn_zero_filter
 2.66% :     0.000005s :     15: predicate.arithmetic_simplify
 1.23% :     0.000002s :     15: predicate.cast_eliminate
 0.49% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     15: predicate.compare_switch_simplify
 1.33% :     0.000002s :     15: predicate.depend_value_elim
 1.12% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.12% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      5: predicate.elim_not_effective
 0.53% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.57% :     0.000003s :     15: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     15: predicate.environ_get_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.12% :     0.000002s :     10: predicate.get_grad_eliminate
 0.35% :     0.000001s :      5: predicate.graph_param_transform
 4.84% :     0.000008s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.40% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.49% :     0.000003s :     10: predicate.less_batch_normalization
 2.10% :     0.000004s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     22: predicate.load_eliminater
 1.05% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.42% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.12% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     15: predicate.merge_addn
 1.16% :     0.000002s :     15: predicate.minmaximum_grad
 1.65% :     0.000003s :      5: predicate.mutable_eliminate
 0.60% :     0.000001s :      5: predicate.opt_reshape
 2.26% :     0.000004s :     22: predicate.partial_eliminate
 1.25% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.60% :     0.000003s :     15: predicate.reduce_eliminate
 1.45% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.66% :     0.000003s :     27: predicate.replace_applicator
 0.56% :     0.000001s :     10: predicate.replace_old_param
 0.36% :     0.000001s :      5: predicate.reset_defer_inline
 1.28% :     0.000002s :     15: predicate.reshape_eliminate
 1.32% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.96% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.03% :     0.000002s :     10: predicate.special_op_eliminate
 1.16% :     0.000002s :     10: predicate.specialize_transform
 1.41% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.46% :     0.000004s :     21: predicate.switch_defer_inline
 1.92% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.49% :     0.000011s :     68: predicate.switch_simplify
 1.22% :     0.000002s :     15: predicate.tile_eliminate
 1.47% :     0.000003s :     15: predicate.transpose_eliminate
 1.40% :     0.000002s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 2.07% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.34% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.49% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000799    13
44.62% :     0.000356s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.29% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.08% :     0.000392s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046587    76
 0.14% :     0.000065s :      1: add_recomputation
 0.33% :     0.000156s :      1: auto_monad
 0.06% :     0.000026s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000313s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.08% :     0.000037s :      1: cse_after_recomputation
 0.03% :     0.000016s :      1: environ_conv
 0.05% :     0.000025s :      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
14.46% :     0.006735s :      1: jit_opt_a
 0.46% :     0.000214s :      1: jit_opt_after_cconv
 0.17% :     0.000078s :      1: jit_opt_b
 1.18% :     0.000550s :      1: loop_unroll
 1.57% :     0.000732s :      1: mutable_eliminate
 2.65% :     0.001232s :     26: opt.transform.jit_opt_a
 0.18% :     0.000084s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000044s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.29% :     0.000602s :      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
 0.27% :     0.000126s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.64% :     0.000764s :      1: renormalize.infer
 1.00% :     0.000465s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.70% :     0.000327s :      1: rewriter_after_opt_a
 0.21% :     0.000100s :      1: rewriter_before_opt_a
 0.22% :     0.000104s :      1: symbol_engine_optimizer
24.03% :     0.011193s :      1: task_emit
47.76% :     0.022249s :      1: type_inference
 0.17% :     0.000080s :      1: validate

[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.343.226 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffe2caf60d0> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.343.951 [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 = 0.0122128, [30]
[bootstrap]: 0.00027672
[type_inference]: 0.00459933
[event_method]: 2.3e-05
[auto_monad]: 6.43e-05
[graph_reusing]: 5.93002e-06
[pre_auto_parallel]: 2.42001e-06
[py_interpret_to_execute]: 3.603e-05
[rewriter_before_opt_a]: 8.8e-05
[expand_dump_flag]: 3.56999e-06
[jit_opt_a]: 0.00507233, [2]
    [Cycle 1]: 0.00164154, [27]
        [switch_simplify]: 5.596e-05
        [loop_unroll]: 3.482e-05
        [a_1]: 0.00061328
        [with_stream_mark]: 1.59e-05
        [recompute_prepare]: 8.17003e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 3.47002e-06
        [updatestate_loads_eliminate]: 3.13e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.68e-06
        [updatestate_useless_node_eliminater]: 6.04999e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.513e-05
        [j_node_and_user_rematch]: 1.008e-05
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 1.091e-05
        [inline_without_move]: 5.89e-06
        [renormalize]: 0.00059181
        [add_forward_monad_depend]: 5.28002e-06
        [auto_monad_grad]: 2.14e-06
        [auto_monad_eliminator]: 1.455e-05
        [cse]: 3.301e-05
        [replace_applicator]: 1.298e-05
    [Cycle 2]: 0.00035876, [27]
        [switch_simplify]: 6.86001e-06
        [loop_unroll]: 5.57999e-06
        [a_1]: 0.00011272
        [with_stream_mark]: 8.25e-06
        [recompute_prepare]: 5.56e-06
        [updatestate_depend_eliminate]: 2.87002e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.61e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.334e-05
        [j_node_and_user_rematch]: 9.13002e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 9.57001e-06
        [inline_without_move]: 5.94e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 6.14999e-06
        [cse]: 1.354e-05
        [replace_applicator]: 5.92001e-06
[py_interpret_to_execute_after_opt_a]: 9.49e-06
[rewriter_after_opt_a]: 3.281e-05
[convert_after_rewriter]: 6.96001e-06
[order_py_execute_after_rewriter]: 5.31998e-06
[mutable_eliminate]: 0.00046645
[jit_opt_b]: 5.361e-05, [1]
    [Cycle 1]: 4.685e-05, [2]
        [frontend_op_eliminate]: 1.796e-05
        [inline_after_opt_a]: 1.72e-05
[cconv]: 2.355e-05
[loop_unroll]: 0.00040408
[jit_opt_after_cconv]: 0.0001972, [1]
    [Cycle 1]: 0.00019091, [11]
        [c_1]: 2.398e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.49e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 1.967e-05
        [call_graph_tuple_transform]: 2.025e-05
        [tuple_list_get_item_eliminator]: 6.11e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 6.59999e-06
[remove_dup_value]: 1.584e-05
[partial_unused_args_eliminate]: 2.70002e-06
[environ_conv]: 5.52999e-06
[add_recomputation]: 4.874e-05
[cse_after_recomputation]: 2.4e-05, [1]
    [Cycle 1]: 1.847e-05, [1]
        [cse]: 1.287e-05
[auto_monad_reorder]: 1.841e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.55001e-06
[opt_after_jit_grad]: 0.00044874
[symbol_engine_optimizer]: 7.554e-05, [1]
    [Cycle 1]: 6.941e-05, [6]
        [build]: 3.37997e-06
        [elim_shapecalc]: 8.42e-06
        [elim_not_effective]: 1.361e-05
        [opt_reshape]: 6.34999e-06
        [fold_const_symbol]: 9.11002e-06
        [renormalize]: 5.59987e-07
[validate]: 3.431e-05
Sums
    bootstrap                            :     0.000277s :  3.25%
    type_inference                       :     0.004599s : 53.94%
    event_method                         :     0.000023s :  0.27%
    auto_monad                           :     0.000064s :  0.75%
    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                :     0.000088s :  1.03%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.74%
    jit_opt_a.loop_unroll                :     0.000040s :  0.47%
    jit_opt_a.a_1                        :     0.000726s :  8.51%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000014s :  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 :     0.000005s :  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      :     0.000012s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    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  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000020s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000592s :  6.94%
    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      :     0.000021s :  0.24%
    jit_opt_a.cse                        :     0.000047s :  0.55%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000033s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000466s :  5.47%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000024s :  0.28%
    loop_unroll                          :     0.000404s :  4.74%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.28%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.23%
    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 :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.08%
    remove_dup_value                     :     0.000016s :  0.19%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000049s :  0.57%
    cse_after_recomputation.cse          :     0.000013s :  0.15%
    auto_monad_reorder                   :     0.000018s :  0.22%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000449s :  5.26%
    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 :  0.16%
    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 :  0.40%


Time group info:
------[substitution.]   0.000210    28
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000006s :      4: substitution.graph_param_transform
81.76% :     0.000172s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.20% :     0.000005s :      4: substitution.replace_old_param
 7.87% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004538     2
80.34% :     0.003646s :      1: type_inference.infer
19.66% :     0.000892s :      1: type_inference.specialize
------[replace.]   0.000080     8
56.20% :     0.000045s :      4: replace.inline
43.80% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     8
92.00% :     0.000169s :      4: match.inline
 8.00% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   912
 1.31% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.96% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     13: predicate.addn_check_dump
 1.20% :     0.000002s :     13: predicate.addn_zero_filter
 1.96% :     0.000003s :     13: predicate.arithmetic_simplify
 1.38% :     0.000002s :     13: predicate.cast_eliminate
 0.42% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     13: predicate.compare_switch_simplify
 1.17% :     0.000002s :     13: predicate.depend_value_elim
 1.17% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      4: predicate.elim_not_effective
 0.55% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     13: predicate.environ_get_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      4: predicate.fold_const_symbol
 0.93% :     0.000001s :      8: predicate.get_grad_eliminate
 0.37% :     0.000001s :      4: predicate.graph_param_transform
 5.44% :     0.000007s :     29: predicate.inline
 0.88% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      8: predicate.less_batch_normalization
 1.65% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.94% :     0.000003s :     21: predicate.load_eliminater
 1.17% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.14% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.02% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     13: predicate.merge_addn
 1.11% :     0.000001s :     13: predicate.minmaximum_grad
 1.44% :     0.000002s :      4: predicate.mutable_eliminate
 0.49% :     0.000001s :      4: predicate.opt_reshape
 2.55% :     0.000003s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     13: predicate.reduce_eliminate
 1.54% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.90% :     0.000003s :     25: predicate.replace_applicator
 0.83% :     0.000001s :      8: predicate.replace_old_param
 0.38% :     0.000001s :      4: predicate.reset_defer_inline
 1.17% :     0.000002s :     13: predicate.reshape_eliminate
 1.30% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     21: predicate.switch_defer_inline
 2.16% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.92% :     0.000011s :     67: predicate.switch_simplify
 1.17% :     0.000002s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.46% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.86% :     0.000002s :     21: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000594    12
50.59% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.11% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.30% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013843    72
 0.37% :     0.000052s :      1: add_recomputation
 0.49% :     0.000068s :      1: auto_monad
 0.15% :     0.000021s :      1: auto_monad_reorder
 2.12% :     0.000294s :      1: bootstrap
 0.19% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.19% :     0.000026s :      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
36.66% :     0.005075s :      1: jit_opt_a
 1.45% :     0.000200s :      1: jit_opt_after_cconv
 0.41% :     0.000056s :      1: jit_opt_b
 2.98% :     0.000412s :      1: loop_unroll
 3.43% :     0.000475s :      1: mutable_eliminate
 6.95% :     0.000961s :     26: opt.transform.jit_opt_a
 0.39% :     0.000054s :      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% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.30% :     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% :     0.000039s :      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
 2.25% :     0.000312s :      1: renormalize.infer
 1.97% :     0.000272s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000036s :      1: rewriter_after_opt_a
 0.66% :     0.000091s :      1: rewriter_before_opt_a
 0.56% :     0.000078s :      1: symbol_engine_optimizer
33.31% :     0.004612s :      1: type_inference


TotalTime = 0.0118159, [30]
[bootstrap]: 0.00024914
[type_inference]: 0.0043013
[event_method]: 2.152e-05
[auto_monad]: 6.394e-05
[graph_reusing]: 6.58e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 3.678e-05
[rewriter_before_opt_a]: 8.959e-05
[expand_dump_flag]: 2.76999e-06
[jit_opt_a]: 0.00500336, [2]
    [Cycle 1]: 0.00157426, [27]
        [switch_simplify]: 5.5e-05
        [loop_unroll]: 3.46e-05
        [a_1]: 0.00059974
        [with_stream_mark]: 1.617e-05
        [recompute_prepare]: 8.43999e-06
        [updatestate_depend_eliminate]: 4.00998e-06
        [updatestate_assign_eliminate]: 3.20998e-06
        [updatestate_loads_eliminate]: 3.3e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 6.98e-06
        [updatestate_useless_node_eliminater]: 6.04001e-06
        [accelerated_algorithm]: 6.05002e-06
        [meta_shard_fg_expand]: 2.25002e-06
        [get_grad_eliminate_]: 5.91e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.564e-05
        [j_node_and_user_rematch]: 9.22001e-06
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 1.068e-05
        [inline_without_move]: 5.87001e-06
        [renormalize]: 0.00054832
        [add_forward_monad_depend]: 4.90001e-06
        [auto_monad_grad]: 2.16e-06
        [auto_monad_eliminator]: 1.346e-05
        [cse]: 2.95e-05
        [replace_applicator]: 1.227e-05
    [Cycle 2]: 0.00034705, [27]
        [switch_simplify]: 6.43e-06
        [loop_unroll]: 5.86e-06
        [a_1]: 0.00011218
        [with_stream_mark]: 8.22e-06
        [recompute_prepare]: 5.99e-06
        [updatestate_depend_eliminate]: 3.04001e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.28998e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 5.69e-06
        [accelerated_algorithm]: 5.81998e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.269e-05
        [j_node_and_user_rematch]: 8.53001e-06
        [meta_fg_expand]: 1.83997e-06
        [replace_old_param]: 8.72998e-06
        [inline_without_move]: 5.80002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 1.02998e-06
        [auto_monad_eliminator]: 6.04001e-06
        [cse]: 1.242e-05
        [replace_applicator]: 5.84e-06
[py_interpret_to_execute_after_opt_a]: 9.82999e-06
[rewriter_after_opt_a]: 3.313e-05
[convert_after_rewriter]: 6.63e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00046465
[jit_opt_b]: 5.386e-05, [1]
    [Cycle 1]: 4.712e-05, [2]
        [frontend_op_eliminate]: 1.858e-05
        [inline_after_opt_a]: 1.683e-05
[cconv]: 2.439e-05
[loop_unroll]: 0.00047777
[jit_opt_after_cconv]: 0.00014847, [1]
    [Cycle 1]: 0.00014217, [11]
        [c_1]: 2.509e-05
        [parameter_eliminate]: 2.23998e-06
        [updatestate_depend_eliminate]: 5.87999e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [cse]: 2.002e-05
        [call_graph_tuple_transform]: 1.978e-05
        [tuple_list_get_item_eliminator]: 6.09999e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 3.50003e-07
        [switch_simplify]: 6.06998e-06
[remove_dup_value]: 1.51e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.27999e-06
[add_recomputation]: 4.609e-05
[cse_after_recomputation]: 2.276e-05, [1]
    [Cycle 1]: 1.763e-05, [1]
        [cse]: 1.207e-05
[auto_monad_reorder]: 1.681e-05
[get_jit_bprop_graph]: 1.35999e-06
[rewriter_after_jit_bprop_graph]: 4.68999e-06
[opt_after_jit_grad]: 0.00043641
[symbol_engine_optimizer]: 7.449e-05, [1]
    [Cycle 1]: 6.831e-05, [6]
        [build]: 2.86999e-06
        [elim_shapecalc]: 8.66997e-06
        [elim_not_effective]: 1.343e-05
        [opt_reshape]: 6.43e-06
        [fold_const_symbol]: 9.37999e-06
        [renormalize]: 4.30009e-07
[validate]: 3.332e-05
Sums
    bootstrap                            :     0.000249s :  3.04%
    type_inference                       :     0.004301s : 52.52%
    event_method                         :     0.000022s :  0.26%
    auto_monad                           :     0.000064s :  0.78%
    graph_reusing                        :     0.000007s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.45%
    rewriter_before_opt_a                :     0.000090s :  1.09%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.75%
    jit_opt_a.loop_unroll                :     0.000040s :  0.49%
    jit_opt_a.a_1                        :     0.000712s :  8.69%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.09%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.07%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000011s :  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 :  0.35%
    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          :     0.000019s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000548s :  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      :     0.000020s :  0.24%
    jit_opt_a.cse                        :     0.000042s :  0.51%
    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                    :     0.000465s :  5.67%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000024s :  0.30%
    loop_unroll                          :     0.000478s :  5.83%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.31%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.24%
    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 :     0.000002s :  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                    :     0.000046s :  0.56%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000017s :  0.21%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000436s :  5.33%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    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.]   0.000198    28
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000005s :      4: substitution.graph_param_transform
81.19% :     0.000161s :      4: substitution.inline
 1.48% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000004s :      4: substitution.replace_old_param
 8.40% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004243     2
78.95% :     0.003350s :      1: type_inference.infer
21.05% :     0.000893s :      1: type_inference.specialize
------[replace.]   0.000080     8
56.99% :     0.000045s :      4: replace.inline
43.01% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     8
91.55% :     0.000158s :      4: match.inline
 8.45% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.30% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.97% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     13: predicate.addn_check_dump
 1.25% :     0.000002s :     13: predicate.addn_zero_filter
 1.96% :     0.000003s :     13: predicate.arithmetic_simplify
 1.32% :     0.000002s :     13: predicate.cast_eliminate
 0.51% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000001s :     13: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      4: predicate.elim_not_effective
 0.50% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.18% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      4: predicate.fold_const_symbol
 0.98% :     0.000001s :      8: predicate.get_grad_eliminate
 0.39% :     0.000001s :      4: predicate.graph_param_transform
 5.33% :     0.000007s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      8: predicate.less_batch_normalization
 1.76% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.91% :     0.000003s :     21: predicate.load_eliminater
 1.31% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.16% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     13: predicate.merge_addn
 1.21% :     0.000002s :     13: predicate.minmaximum_grad
 1.72% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.55% :     0.000003s :     21: predicate.partial_eliminate
 1.24% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     13: predicate.reduce_eliminate
 1.58% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.92% :     0.000003s :     25: predicate.replace_applicator
 0.78% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000000s :      4: predicate.reset_defer_inline
 1.26% :     0.000002s :     13: predicate.reshape_eliminate
 1.24% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.57% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      8: predicate.special_op_eliminate
 1.02% :     0.000001s :      8: predicate.specialize_transform
 1.27% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     21: predicate.switch_defer_inline
 2.14% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.67% :     0.000010s :     67: predicate.switch_simplify
 1.22% :     0.000002s :     13: predicate.tile_eliminate
 1.25% :     0.000002s :     13: predicate.transpose_eliminate
 1.45% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.21% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.95% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000543    12
46.64% :     0.000253s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.11% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.25% :     0.000262s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013389    72
 0.37% :     0.000049s :      1: add_recomputation
 0.50% :     0.000068s :      1: auto_monad
 0.14% :     0.000019s :      1: auto_monad_reorder
 1.97% :     0.000263s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.19% :     0.000025s :      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
37.39% :     0.005007s :      1: jit_opt_a
 1.13% :     0.000151s :      1: jit_opt_after_cconv
 0.42% :     0.000056s :      1: jit_opt_b
 3.63% :     0.000486s :      1: loop_unroll
 3.53% :     0.000472s :      1: mutable_eliminate
 7.05% :     0.000944s :     26: opt.transform.jit_opt_a
 0.40% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000030s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.32% :     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% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000040s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 2.20% :     0.000295s :      1: renormalize.infer
 1.84% :     0.000246s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000036s :      1: rewriter_after_opt_a
 0.69% :     0.000092s :      1: rewriter_before_opt_a
 0.58% :     0.000077s :      1: symbol_engine_optimizer
32.21% :     0.004313s :      1: type_inference


TotalTime = 0.0129366, [30]
[bootstrap]: 0.00028411
[type_inference]: 0.00494207
[event_method]: 2.221e-05
[auto_monad]: 6.4e-05
[graph_reusing]: 5.73997e-06
[pre_auto_parallel]: 2.71e-06
[py_interpret_to_execute]: 3.603e-05
[rewriter_before_opt_a]: 9.04e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00540377, [2]
    [Cycle 1]: 0.00177644, [27]
        [switch_simplify]: 6.15e-05
        [loop_unroll]: 3.489e-05
        [a_1]: 0.00062135
        [with_stream_mark]: 1.976e-05
        [recompute_prepare]: 7.85998e-06
        [updatestate_depend_eliminate]: 4.18001e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 3.29001e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 7.03998e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 6.14999e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 5.81998e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.506e-05
        [j_node_and_user_rematch]: 9.69e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 1.085e-05
        [inline_without_move]: 6.22001e-06
        [renormalize]: 0.00069985
        [add_forward_monad_depend]: 5.41002e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.5e-05
        [cse]: 3.212e-05
        [replace_applicator]: 1.296e-05
    [Cycle 2]: 0.00036506, [27]
        [switch_simplify]: 7.28e-06
        [loop_unroll]: 5.79e-06
        [a_1]: 0.00011412
        [with_stream_mark]: 9.36e-06
        [recompute_prepare]: 6.01998e-06
        [updatestate_depend_eliminate]: 3.32002e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 5.71003e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.435e-05
        [j_node_and_user_rematch]: 8.93002e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 1.029e-05
        [inline_without_move]: 6.01e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.16997e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 6.49001e-06
        [cse]: 1.39e-05
        [replace_applicator]: 6.18998e-06
[py_interpret_to_execute_after_opt_a]: 1.269e-05
[rewriter_after_opt_a]: 3.604e-05
[convert_after_rewriter]: 7.63001e-06
[order_py_execute_after_rewriter]: 5.42001e-06
[mutable_eliminate]: 0.00051272
[jit_opt_b]: 5.546e-05, [1]
    [Cycle 1]: 4.823e-05, [2]
        [frontend_op_eliminate]: 1.85e-05
        [inline_after_opt_a]: 1.758e-05
[cconv]: 2.5e-05
[loop_unroll]: 0.00040604
[jit_opt_after_cconv]: 0.00015104, [1]
    [Cycle 1]: 0.00014434, [11]
        [c_1]: 2.456e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 5.71e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [cse]: 2.088e-05
        [call_graph_tuple_transform]: 2.019e-05
        [tuple_list_get_item_eliminator]: 6.15002e-06
        [none_parameter_eliminate]: 1.79e-06
        [renormalize]: 7.30011e-07
        [switch_simplify]: 6.12999e-06
[remove_dup_value]: 1.606e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 5.76e-06
[add_recomputation]: 4.991e-05
[cse_after_recomputation]: 2.343e-05, [1]
    [Cycle 1]: 1.78e-05, [1]
        [cse]: 1.178e-05
[auto_monad_reorder]: 1.83e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 4.79e-06
[opt_after_jit_grad]: 0.00045507
[symbol_engine_optimizer]: 7.495e-05, [1]
    [Cycle 1]: 6.851e-05, [6]
        [build]: 3.7e-06
        [elim_shapecalc]: 8.48001e-06
        [elim_not_effective]: 1.329e-05
        [opt_reshape]: 6.23002e-06
        [fold_const_symbol]: 9.56998e-06
        [renormalize]: 3.59985e-07
[validate]: 3.836e-05
Sums
    bootstrap                            :     0.000284s :  3.13%
    type_inference                       :     0.004942s : 54.40%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000064s :  0.70%
    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                :     0.000090s :  1.00%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000069s :  0.76%
    jit_opt_a.loop_unroll                :     0.000041s :  0.45%
    jit_opt_a.a_1                        :     0.000735s :  8.10%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.32%
    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 :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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              :     0.000007s :  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    :     0.000019s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000021s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000700s :  7.70%
    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      :     0.000021s :  0.24%
    jit_opt_a.cse                        :     0.000046s :  0.51%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    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                    :     0.000513s :  5.64%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.19%
    cconv                                :     0.000025s :  0.28%
    loop_unroll                          :     0.000406s :  4.47%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.27%
    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 :  0.03%
    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 :  0.07%
    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                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000050s :  0.55%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000455s :  5.01%
    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 :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.42%


Time group info:
------[substitution.]   0.000214    28
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000002s :      2: substitution.fold_const_symbol
 2.57% :     0.000005s :      4: substitution.graph_param_transform
81.66% :     0.000174s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000005s :      4: substitution.replace_old_param
 8.11% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004870     2
80.93% :     0.003941s :      1: type_inference.infer
19.07% :     0.000929s :      1: type_inference.specialize
------[replace.]   0.000082     8
56.86% :     0.000047s :      4: replace.inline
43.14% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000187     8
91.90% :     0.000172s :      4: match.inline
 8.10% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.26% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.96% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     13: predicate.addn_check_dump
 1.38% :     0.000002s :     13: predicate.addn_zero_filter
 1.74% :     0.000002s :     13: predicate.arithmetic_simplify
 1.29% :     0.000002s :     13: predicate.cast_eliminate
 0.53% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     13: predicate.compare_switch_simplify
 1.23% :     0.000002s :     13: predicate.depend_value_elim
 1.20% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      4: predicate.elim_not_effective
 0.52% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     13: predicate.environ_get_eliminate
 1.30% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 0.92% :     0.000001s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 5.85% :     0.000008s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      8: predicate.less_batch_normalization
 1.63% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.96% :     0.000003s :     21: predicate.load_eliminater
 1.15% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     13: predicate.merge_addn
 1.09% :     0.000001s :     13: predicate.minmaximum_grad
 1.66% :     0.000002s :      4: predicate.mutable_eliminate
 0.41% :     0.000001s :      4: predicate.opt_reshape
 2.53% :     0.000003s :     21: predicate.partial_eliminate
 1.15% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.05% :     0.000003s :     25: predicate.replace_applicator
 0.64% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.14% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      8: predicate.special_op_eliminate
 1.02% :     0.000001s :      8: predicate.specialize_transform
 1.27% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     21: predicate.switch_defer_inline
 2.20% :     0.000003s :     21: predicate.switch_layer_defer_inline
 8.05% :     0.000011s :     67: predicate.switch_simplify
 1.20% :     0.000002s :     13: predicate.tile_eliminate
 1.27% :     0.000002s :     13: predicate.transpose_eliminate
 1.41% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.23% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     13: predicate.value_based_eliminate
 0.36% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000618    12
50.91% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.65% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.43% :     0.000275s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014684    72
 0.36% :     0.000053s :      1: add_recomputation
 0.46% :     0.000068s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 2.06% :     0.000303s :      1: bootstrap
 0.19% :     0.000028s :      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
 0.18% :     0.000027s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
36.82% :     0.005407s :      1: jit_opt_a
 1.05% :     0.000154s :      1: jit_opt_after_cconv
 0.40% :     0.000058s :      1: jit_opt_b
 2.82% :     0.000415s :      1: loop_unroll
 3.55% :     0.000522s :      1: mutable_eliminate
 6.67% :     0.000979s :     26: opt.transform.jit_opt_a
 0.37% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000030s :      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
 0.16% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.16% :     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
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000039s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000018s :      1: remove_dup_value
 2.61% :     0.000383s :      1: renormalize.infer
 2.11% :     0.000309s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000039s :      1: rewriter_after_opt_a
 0.64% :     0.000093s :      1: rewriter_before_opt_a
 0.53% :     0.000078s :      1: symbol_engine_optimizer
33.77% :     0.004959s :      1: type_inference


TotalTime = 0.0782063, [33]
[bootstrap]: 0.00026421
[type_inference]: 0.0350941
[event_method]: 0.00017327
[auto_monad]: 0.00023088
[graph_reusing]: 8.54998e-06
[pre_auto_parallel]: 4.31002e-06
[py_interpret_to_execute]: 3.927e-05
[rewriter_before_opt_a]: 0.00012889
[expand_dump_flag]: 4.08999e-06
[jit_opt_a]: 0.0308177, [3]
    [Cycle 1]: 0.0226328, [27]
        [switch_simplify]: 9.363e-05
        [loop_unroll]: 5.353e-05
        [a_1]: 0.0013028
        [with_stream_mark]: 3.471e-05
        [recompute_prepare]: 3.239e-05
        [updatestate_depend_eliminate]: 1.369e-05
        [updatestate_assign_eliminate]: 1.124e-05
        [updatestate_loads_eliminate]: 9.57999e-06
        [parameter_eliminate]: 3.11999e-06
        [specialize_transform]: 2.322e-05
        [updatestate_useless_node_eliminater]: 2.512e-05
        [accelerated_algorithm]: 2.042e-05
        [meta_shard_fg_expand]: 5.82001e-06
        [get_grad_eliminate_]: 2.056e-05
        [merge_forward]: 1.171e-05
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 4.421e-05
        [j_node_and_user_rematch]: 3.745e-05
        [meta_fg_expand]: 0.00653905
        [replace_old_param]: 0.00013023
        [inline_without_move]: 0.00017171
        [renormalize]: 0.0131671
        [add_forward_monad_depend]: 1.7e-05
        [auto_monad_grad]: 1.048e-05
        [auto_monad_eliminator]: 8.222e-05
        [cse]: 0.00028878
        [replace_applicator]: 0.00018409
    [Cycle 2]: 0.00331761, [27]
        [switch_simplify]: 6.05e-05
        [loop_unroll]: 5.799e-05
        [a_1]: 0.00101936
        [with_stream_mark]: 2.412e-05
        [recompute_prepare]: 1.457e-05
        [updatestate_depend_eliminate]: 2.932e-05
        [updatestate_assign_eliminate]: 6.83e-06
        [updatestate_loads_eliminate]: 4.99e-06
        [parameter_eliminate]: 2.07001e-06
        [specialize_transform]: 1.028e-05
        [updatestate_useless_node_eliminater]: 1.303e-05
        [accelerated_algorithm]: 9.02e-06
        [meta_shard_fg_expand]: 2.89999e-06
        [get_grad_eliminate_]: 8.30999e-06
        [merge_forward]: 6.83e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.954e-05
        [j_node_and_user_rematch]: 1.422e-05
        [meta_fg_expand]: 0.00010021
        [replace_old_param]: 1.456e-05
        [inline_without_move]: 8.32e-06
        [renormalize]: 0.00155285
        [add_forward_monad_depend]: 8.18999e-06
        [auto_monad_grad]: 2.89999e-06
        [auto_monad_eliminator]: 2.234e-05
        [cse]: 8.92e-05
        [replace_applicator]: 2.395e-05
    [Cycle 3]: 0.00051937, [27]
        [switch_simplify]: 9.25999e-06
        [loop_unroll]: 8.79998e-06
        [a_1]: 0.00018149
        [with_stream_mark]: 1.553e-05
        [recompute_prepare]: 8.55001e-06
        [updatestate_depend_eliminate]: 7.46001e-06
        [updatestate_assign_eliminate]: 5.04e-06
        [updatestate_loads_eliminate]: 4.68999e-06
        [parameter_eliminate]: 2.12001e-06
        [specialize_transform]: 8.82e-06
        [updatestate_useless_node_eliminater]: 1.205e-05
        [accelerated_algorithm]: 9.12001e-06
        [meta_shard_fg_expand]: 2.73e-06
        [get_grad_eliminate_]: 8.47e-06
        [merge_forward]: 5.98002e-06
        [cell_reuse_recompute_pass]: 3.85998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.881e-05
        [j_node_and_user_rematch]: 1.347e-05
        [meta_fg_expand]: 3.26999e-06
        [replace_old_param]: 1.229e-05
        [inline_without_move]: 7.87e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 1.31002e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 1.146e-05
        [cse]: 2.555e-05
        [replace_applicator]: 8.12e-06
[py_interpret_to_execute_after_opt_a]: 2.1e-05
[rewriter_after_opt_a]: 6.524e-05
[convert_after_rewriter]: 9.64e-06
[order_py_execute_after_rewriter]: 6.59001e-06
[mutable_eliminate]: 0.00076576
[jit_opt_b]: 7.315e-05, [1]
    [Cycle 1]: 6.404e-05, [2]
        [frontend_op_eliminate]: 2.651e-05
        [inline_after_opt_a]: 2.532e-05
[cconv]: 2.892e-05
[loop_unroll]: 0.00049134
[jit_opt_after_cconv]: 0.00021253, [1]
    [Cycle 1]: 0.00020575, [11]
        [c_1]: 4.667e-05
        [parameter_eliminate]: 3.11001e-06
        [updatestate_depend_eliminate]: 7.92e-06
        [updatestate_assign_eliminate]: 4.58999e-06
        [updatestate_loads_eliminate]: 4.36002e-06
        [cse]: 3.92e-05
        [call_graph_tuple_transform]: 2.66e-05
        [tuple_list_get_item_eliminator]: 8.17e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 8.24002e-06
[remove_dup_value]: 2.412e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 1.203e-05
[add_recomputation]: 6.847e-05
[cse_after_recomputation]: 3.591e-05, [1]
    [Cycle 1]: 2.876e-05, [1]
        [cse]: 2.223e-05
[auto_monad_reorder]: 2.498e-05
[get_jit_bprop_graph]: 2.73e-06
[rewriter_after_jit_bprop_graph]: 7.87998e-06
[opt_after_jit_grad]: 0.00060846
[symbol_engine_optimizer]: 0.00010873, [1]
    [Cycle 1]: 0.00010224, [6]
        [build]: 1.118e-05
        [elim_shapecalc]: 1.197e-05
        [elim_not_effective]: 2.104e-05
        [opt_reshape]: 1.012e-05
        [fold_const_symbol]: 1.526e-05
        [renormalize]: 4.60015e-07
[validate]: 5.534e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.0085481
[execute]: 8.71002e-06
Sums
    bootstrap                            :     0.000264s :  0.36%
    type_inference                       :     0.035094s : 48.17%
    event_method                         :     0.000173s :  0.24%
    auto_monad                           :     0.000231s :  0.32%
    graph_reusing                        :     0.000009s :  0.01%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.05%
    rewriter_before_opt_a                :     0.000129s :  0.18%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000163s :  0.22%
    jit_opt_a.loop_unroll                :     0.000120s :  0.17%
    jit_opt_a.a_1                        :     0.002504s :  3.44%
    jit_opt_a.with_stream_mark           :     0.000074s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000056s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000050s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000023s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000019s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000007s :  0.01%
    jit_opt_a.specialize_transform       :     0.000042s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000050s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000039s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000011s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000037s :  0.05%
    jit_opt_a.merge_forward              :     0.000025s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000083s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000065s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.006643s :  9.12%
    jit_opt_a.replace_old_param          :     0.000157s :  0.22%
    jit_opt_a.inline_without_move        :     0.000188s :  0.26%
    jit_opt_a.renormalize                :     0.014720s : 20.20%
    jit_opt_a.add_forward_monad_depend   :     0.000027s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000015s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000116s :  0.16%
    jit_opt_a.cse                        :     0.000404s :  0.55%
    jit_opt_a.replace_applicator         :     0.000216s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.03%
    rewriter_after_opt_a                 :     0.000065s :  0.09%
    convert_after_rewriter               :     0.000010s :  0.01%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000766s :  1.05%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.03%
    cconv                                :     0.000029s :  0.04%
    loop_unroll                          :     0.000491s :  0.67%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.06%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000039s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.04%
    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      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.01%
    remove_dup_value                     :     0.000024s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000012s :  0.02%
    add_recomputation                    :     0.000068s :  0.09%
    cse_after_recomputation.cse          :     0.000022s :  0.03%
    auto_monad_reorder                   :     0.000025s :  0.03%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.01%
    opt_after_jit_grad                   :     0.000608s :  0.84%
    symbol_engine_optimizer.build        :     0.000011s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000055s :  0.08%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008548s : 11.73%
    execute                              :     0.000009s :  0.01%


Time group info:
------[substitution.]   0.001697   159
 1.39% :     0.000024s :      8: substitution.depend_value_elim
 0.18% :     0.000003s :      4: substitution.elim_not_effective
 0.17% :     0.000003s :      4: substitution.fold_const_symbol
52.71% :     0.000895s :      4: substitution.getattr_setattr_resolve
 0.41% :     0.000007s :      5: substitution.graph_param_transform
26.95% :     0.000457s :     10: substitution.inline
 3.98% :     0.000067s :      4: substitution.inline_without_move
 0.77% :     0.000013s :     20: substitution.j_node_and_user_rematch
 0.59% :     0.000010s :      5: substitution.minmaximum_grad
 0.29% :     0.000005s :      3: substitution.partial_eliminate
 0.88% :     0.000015s :     20: substitution.remove_not_recompute_node
 3.20% :     0.000054s :     13: substitution.replace_applicator
 0.91% :     0.000015s :     17: substitution.replace_old_param
 0.20% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.31% :     0.000022s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.80% :     0.000014s :      5: substitution.tuple_list_get_item_depend_reorder
 2.37% :     0.000040s :     10: substitution.tuple_list_get_item_eliminator
 1.07% :     0.000018s :      8: substitution.updatestate_pure_node_eliminater
 1.81% :     0.000031s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.034991     2
93.70% :     0.032786s :      1: type_inference.infer
 6.30% :     0.002205s :      1: type_inference.specialize
------[replace.]   0.000327    20
18.91% :     0.000062s :      3: replace.getattr_setattr_resolve
31.33% :     0.000102s :     10: replace.inline
12.26% :     0.000040s :      1: replace.replace_applicator
30.36% :     0.000099s :      5: replace.tuple_list_get_item_eliminator
 7.14% :     0.000023s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001338    20
63.22% :     0.000846s :      3: match.getattr_setattr_resolve
33.54% :     0.000449s :     10: match.inline
 0.92% :     0.000012s :      1: match.replace_applicator
 1.54% :     0.000021s :      5: match.tuple_list_get_item_eliminator
 0.77% :     0.000010s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000479  2975
 1.40% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.41% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.28% :     0.000006s :     46: predicate.addn_check_dump
 1.59% :     0.000008s :     46: predicate.addn_zero_filter
 2.50% :     0.000012s :     46: predicate.arithmetic_simplify
 1.79% :     0.000009s :     46: predicate.cast_eliminate
 0.27% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.27% :     0.000006s :     46: predicate.compare_switch_simplify
 1.45% :     0.000007s :     46: predicate.depend_value_elim
 1.21% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.66% :     0.000008s :     46: predicate.dict_get_item_eliminator
 1.57% :     0.000007s :     46: predicate.dict_set_item_eliminator
 0.28% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      5: predicate.elim_not_effective
 0.26% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000007s :     46: predicate.environ_add_const_eliminate
 1.25% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.24% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.39% :     0.000007s :     46: predicate.environ_get_eliminate
 1.24% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000005s :     27: predicate.get_grad_eliminate
 1.56% :     0.000007s :     20: predicate.getattr_setattr_resolve
 0.10% :     0.000000s :      5: predicate.graph_param_transform
 3.72% :     0.000018s :     72: predicate.inline
 3.60% :     0.000017s :     89: predicate.inline_without_move
 0.45% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.35% :     0.000006s :     27: predicate.less_batch_normalization
 1.55% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.81% :     0.000009s :     56: predicate.load_eliminater
 0.53% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.73% :     0.000018s :    121: predicate.loop_unroll_before_grad
 1.63% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000006s :     46: predicate.merge_addn
 1.26% :     0.000006s :     46: predicate.minmaximum_grad
 0.41% :     0.000002s :      5: predicate.mutable_eliminate
 0.22% :     0.000001s :      5: predicate.opt_reshape
 2.05% :     0.000010s :     56: predicate.partial_eliminate
 1.38% :     0.000007s :     46: predicate.print_const_string_wrapper
 1.81% :     0.000009s :     46: predicate.reduce_eliminate
 1.51% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000003s :     27: predicate.remove_not_recompute_node
 2.69% :     0.000013s :    126: predicate.replace_applicator
 1.61% :     0.000008s :     89: predicate.replace_old_param
 0.10% :     0.000000s :      5: predicate.reset_defer_inline
 1.38% :     0.000007s :     46: predicate.reshape_eliminate
 1.41% :     0.000007s :     46: predicate.row_tensor_add_zeros_like
 0.33% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.64% :     0.000008s :     46: predicate.same_eliminate
 0.58% :     0.000003s :     29: predicate.set_cell_output_no_recompute
 0.51% :     0.000002s :     10: predicate.special_op_eliminate
 1.04% :     0.000005s :     27: predicate.specialize_transform
 1.78% :     0.000008s :     46: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000007s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.44% :     0.000012s :     62: predicate.switch_defer_inline
 1.98% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.25% :     0.000030s :    188: predicate.switch_simplify
 1.32% :     0.000006s :     46: predicate.tile_eliminate
 1.31% :     0.000006s :     46: predicate.transpose_eliminate
 1.60% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.48% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 3.13% :     0.000015s :     61: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000009s :     46: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000008s :     51: predicate.tuple_to_list_eliminator_
 1.82% :     0.000009s :     56: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000014s :     84: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000009s :     46: predicate.value_based_eliminate
 0.17% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.20% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003620    48
62.86% :     0.002276s :     27: func_graph_cloner_run.FuncGraphClonerGraph
37.14% :     0.001344s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.097777    93
 0.07% :     0.000072s :      1: add_recomputation
 0.25% :     0.000242s :      1: auto_monad
 0.03% :     0.000028s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.29% :     0.000285s :      1: bootstrap
 0.03% :     0.000032s :      1: cconv
 0.01% :     0.000012s :      1: convert_after_rewriter
 0.04% :     0.000038s :      1: cse_after_recomputation
 0.01% :     0.000014s :      1: environ_conv
 0.19% :     0.000182s :      1: event_method
 0.01% :     0.000014s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000012s :      1: graph_reusing
31.52% :     0.030822s :      1: jit_opt_a
 0.22% :     0.000215s :      1: jit_opt_after_cconv
 0.08% :     0.000076s :      1: jit_opt_b
 0.51% :     0.000499s :      1: loop_unroll
 0.79% :     0.000774s :      1: mutable_eliminate
 3.74% :     0.003658s :     39: opt.transform.jit_opt_a
 0.09% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000044s :      4: opt.transform.jit_opt_b
 0.02% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 1.04% :     0.001022s :      2: opt.transform.opt_resolve
 0.06% :     0.000054s :      4: opt.transform.symbol_engine_opt
 0.63% :     0.000616s :      1: opt_after_jit_grad
 0.01% :     0.000009s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000007s :      1: pre_auto_parallel
 0.04% :     0.000042s :      1: py_interpret_to_execute
 0.02% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000027s :      1: remove_dup_value
11.41% :     0.011155s :      2: renormalize.infer
 3.62% :     0.003541s :      2: renormalize.specialize
 0.01% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000068s :      1: rewriter_after_opt_a
 0.14% :     0.000133s :      1: rewriter_before_opt_a
 0.11% :     0.000111s :      1: symbol_engine_optimizer
 8.76% :     0.008566s :      1: task_emit
35.91% :     0.035115s :      1: type_inference
 0.08% :     0.000082s :      1: validate


TotalTime = 0.0462288, [33]
[bootstrap]: 0.00030825
[type_inference]: 0.0243176
[event_method]: 2.305e-05
[auto_monad]: 0.00018413
[graph_reusing]: 6.81999e-06
[pre_auto_parallel]: 2.94001e-06
[py_interpret_to_execute]: 0.00014874
[rewriter_before_opt_a]: 0.00010537
[expand_dump_flag]: 3.33998e-06
[jit_opt_a]: 0.00685562, [2]
    [Cycle 1]: 0.00241963, [27]
        [switch_simplify]: 5.782e-05
        [loop_unroll]: 3.577e-05
        [a_1]: 0.00074375
        [with_stream_mark]: 1.773e-05
        [recompute_prepare]: 1.059e-05
        [updatestate_depend_eliminate]: 5.62001e-06
        [updatestate_assign_eliminate]: 4.86002e-06
        [updatestate_loads_eliminate]: 4.1e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 9.61e-06
        [updatestate_useless_node_eliminater]: 1.178e-05
        [accelerated_algorithm]: 8.77999e-06
        [meta_shard_fg_expand]: 2.04999e-06
        [get_grad_eliminate_]: 8.37e-06
        [merge_forward]: 4.52e-06
        [cell_reuse_recompute_pass]: 1.75001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.677e-05
        [j_node_and_user_rematch]: 1.314e-05
        [meta_fg_expand]: 3.39001e-06
        [replace_old_param]: 1.195e-05
        [inline_without_move]: 8.22e-06
        [renormalize]: 0.00116936
        [add_forward_monad_depend]: 6.02999e-06
        [auto_monad_grad]: 2.94001e-06
        [auto_monad_eliminator]: 2.246e-05
        [cse]: 2.586e-05
        [replace_applicator]: 1.7e-05
    [Cycle 2]: 0.00048399, [27]
        [switch_simplify]: 8.58001e-06
        [loop_unroll]: 8.25999e-06
        [a_1]: 0.00017626
        [with_stream_mark]: 1.368e-05
        [recompute_prepare]: 8.05e-06
        [updatestate_depend_eliminate]: 4.73001e-06
        [updatestate_assign_eliminate]: 5.52001e-06
        [updatestate_loads_eliminate]: 4.38001e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 8.38999e-06
        [updatestate_useless_node_eliminater]: 1.106e-05
        [accelerated_algorithm]: 8.35999e-06
        [meta_shard_fg_expand]: 2.28998e-06
        [get_grad_eliminate_]: 7.60998e-06
        [merge_forward]: 4.67998e-06
        [cell_reuse_recompute_pass]: 2.39999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.633e-05
        [j_node_and_user_rematch]: 1.292e-05
        [meta_fg_expand]: 2.88998e-06
        [replace_old_param]: 1.108e-05
        [inline_without_move]: 7.60998e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 1.16002e-06
        [auto_monad_eliminator]: 1.077e-05
        [cse]: 2.002e-05
        [replace_applicator]: 7.97003e-06
[py_interpret_to_execute_after_opt_a]: 1.494e-05
[rewriter_after_opt_a]: 0.00030983
[convert_after_rewriter]: 1.201e-05
[order_py_execute_after_rewriter]: 7.11999e-06
[mutable_eliminate]: 0.00067898
[jit_opt_b]: 7.055e-05, [1]
    [Cycle 1]: 6.331e-05, [2]
        [frontend_op_eliminate]: 2.49e-05
        [inline_after_opt_a]: 2.68e-05
[cconv]: 2.581e-05
[loop_unroll]: 0.00047664
[jit_opt_after_cconv]: 0.00020423, [1]
    [Cycle 1]: 0.00019734, [11]
        [c_1]: 4.519e-05
        [parameter_eliminate]: 4.01001e-06
        [updatestate_depend_eliminate]: 8.88002e-06
        [updatestate_assign_eliminate]: 5.47001e-06
        [updatestate_loads_eliminate]: 5.00001e-06
        [cse]: 3.354e-05
        [call_graph_tuple_transform]: 2.287e-05
        [tuple_list_get_item_eliminator]: 8.60999e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 3.9002e-07
        [switch_simplify]: 8.58001e-06
[remove_dup_value]: 1.811e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 1.194e-05
[add_recomputation]: 5.988e-05
[cse_after_recomputation]: 2.935e-05, [1]
    [Cycle 1]: 2.39e-05, [1]
        [cse]: 1.738e-05
[auto_monad_reorder]: 2.21e-05
[get_jit_bprop_graph]: 1.71e-06
[rewriter_after_jit_bprop_graph]: 5.49e-06
[opt_after_jit_grad]: 0.00055448
[symbol_engine_optimizer]: 0.00010487, [1]
    [Cycle 1]: 9.813e-05, [6]
        [build]: 1.247e-05
        [elim_shapecalc]: 1.174e-05
        [elim_not_effective]: 1.972e-05
        [opt_reshape]: 8.53001e-06
        [fold_const_symbol]: 1.401e-05
        [renormalize]: 4.39992e-07
[validate]: 5.772e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.0113719
[execute]: 8.95999e-06
Sums
    bootstrap                            :     0.000308s :  0.74%
    type_inference                       :     0.024318s : 58.44%
    event_method                         :     0.000023s :  0.06%
    auto_monad                           :     0.000184s :  0.44%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000149s :  0.36%
    rewriter_before_opt_a                :     0.000105s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.16%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000920s :  2.21%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    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 :     0.000033s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001169s :  2.81%
    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      :     0.000033s :  0.08%
    jit_opt_a.cse                        :     0.000046s :  0.11%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000310s :  0.74%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000679s :  1.63%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.06%
    cconv                                :     0.000026s :  0.06%
    loop_unroll                          :     0.000477s :  1.15%
    jit_opt_after_cconv.c_1              :     0.000045s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000018s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000060s :  0.14%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000022s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000554s :  1.33%
    symbol_engine_optimizer.build        :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000058s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011372s : 27.33%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000295    47
 2.64% :     0.000008s :      2: substitution.depend_value_elim
 1.00% :     0.000003s :      4: substitution.elim_not_effective
 0.66% :     0.000002s :      4: substitution.fold_const_symbol
 2.31% :     0.000007s :      5: substitution.graph_param_transform
77.92% :     0.000230s :      4: substitution.inline
 1.44% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.34% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.78% :     0.000005s :      2: substitution.replace_old_param
 4.35% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 2.73% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 2.82% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024240     2
93.61% :     0.022692s :      1: type_inference.infer
 6.39% :     0.001548s :      1: type_inference.specialize
------[replace.]   0.000101     6
67.44% :     0.000068s :      4: replace.inline
32.56% :     0.000033s :      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.]   0.000176  1021
 1.33% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     15: predicate.addn_check_dump
 1.67% :     0.000003s :     15: predicate.addn_zero_filter
 2.51% :     0.000004s :     15: predicate.arithmetic_simplify
 1.37% :     0.000002s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     15: predicate.compare_switch_simplify
 1.27% :     0.000002s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.67% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      5: predicate.elim_not_effective
 0.60% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.41% :     0.000002s :     15: predicate.environ_get_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.19% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 4.52% :     0.000008s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.40% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.56% :     0.000003s :     10: predicate.less_batch_normalization
 1.50% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.09% :     0.000004s :     22: predicate.load_eliminater
 1.12% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.46% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.43% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.34% :     0.000002s :     15: predicate.minmaximum_grad
 1.42% :     0.000002s :      5: predicate.mutable_eliminate
 0.53% :     0.000001s :      5: predicate.opt_reshape
 2.23% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 2.13% :     0.000004s :     15: predicate.reduce_eliminate
 1.45% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.69% :     0.000003s :     27: predicate.replace_applicator
 0.57% :     0.000001s :     10: predicate.replace_old_param
 0.35% :     0.000001s :      5: predicate.reset_defer_inline
 1.25% :     0.000002s :     15: predicate.reshape_eliminate
 1.56% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 1.00% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     15: predicate.same_eliminate
 0.54% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.02% :     0.000002s :     10: predicate.special_op_eliminate
 1.13% :     0.000002s :     10: predicate.specialize_transform
 1.41% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.26% :     0.000004s :     21: predicate.switch_defer_inline
 2.02% :     0.000004s :     21: predicate.switch_layer_defer_inline
 5.91% :     0.000010s :     68: predicate.switch_simplify
 1.33% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.65% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.86% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.67% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000889    13
43.96% :     0.000391s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.79% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.25% :     0.000447s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048806    76
 0.13% :     0.000063s :      1: add_recomputation
 0.39% :     0.000189s :      1: auto_monad
 0.05% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.67% :     0.000325s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.06% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000014s :      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% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
14.05% :     0.006859s :      1: jit_opt_a
 0.42% :     0.000207s :      1: jit_opt_after_cconv
 0.15% :     0.000074s :      1: jit_opt_b
 0.99% :     0.000484s :      1: loop_unroll
 1.41% :     0.000688s :      1: mutable_eliminate
 2.50% :     0.001220s :     26: opt.transform.jit_opt_a
 0.17% :     0.000081s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000044s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.16% :     0.000565s :      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
 0.31% :     0.000154s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.41% :     0.000689s :      1: renormalize.infer
 0.96% :     0.000471s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.64% :     0.000315s :      1: rewriter_after_opt_a
 0.22% :     0.000109s :      1: rewriter_before_opt_a
 0.22% :     0.000108s :      1: symbol_engine_optimizer
23.34% :     0.011390s :      1: task_emit
49.85% :     0.024328s :      1: type_inference
 0.17% :     0.000081s :      1: validate

[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.433.411 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffe2caf65e0> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.434.148 [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 = 0.787065, [33]
[bootstrap]: 0.00135339
[type_inference]: 0.0509831
[event_method]: 4.657e-05
[auto_monad]: 0.000195
[graph_reusing]: 6.81001e-06
[pre_auto_parallel]: 1.104e-05
[py_interpret_to_execute]: 4.179e-05
[rewriter_before_opt_a]: 0.00015904
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.0353306, [3]
    [Cycle 1]: 0.0237007, [27]
        [switch_simplify]: 0.00011653
        [loop_unroll]: 4.677e-05
        [a_1]: 0.00101013
        [with_stream_mark]: 3.756e-05
        [recompute_prepare]: 2.192e-05
        [updatestate_depend_eliminate]: 1.016e-05
        [updatestate_assign_eliminate]: 7.23e-06
        [updatestate_loads_eliminate]: 7.4e-06
        [parameter_eliminate]: 2.82002e-06
        [specialize_transform]: 1.483e-05
        [updatestate_useless_node_eliminater]: 1.352e-05
        [accelerated_algorithm]: 1.365e-05
        [meta_shard_fg_expand]: 3.55e-06
        [get_grad_eliminate_]: 1.401e-05
        [merge_forward]: 8.70999e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.232e-05
        [j_node_and_user_rematch]: 3.551e-05
        [meta_fg_expand]: 0.00237879
        [replace_old_param]: 7.595e-05
        [inline_without_move]: 6.307e-05
        [renormalize]: 0.0190577
        [add_forward_monad_depend]: 5.451e-05
        [auto_monad_grad]: 7.45e-06
        [auto_monad_eliminator]: 6.223e-05
        [cse]: 0.00021335
        [replace_applicator]: 8.879e-05
    [Cycle 2]: 0.00227407, [27]
        [switch_simplify]: 4.131e-05
        [loop_unroll]: 4.017e-05
        [a_1]: 0.00112055
        [with_stream_mark]: 1.596e-05
        [recompute_prepare]: 9.07001e-06
        [updatestate_depend_eliminate]: 4.58999e-06
        [updatestate_assign_eliminate]: 3.42002e-06
        [updatestate_loads_eliminate]: 3.09999e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 7e-06
        [updatestate_useless_node_eliminater]: 6.42001e-06
        [accelerated_algorithm]: 6.95002e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 6.31e-06
        [merge_forward]: 4.13001e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.566e-05
        [j_node_and_user_rematch]: 9.44998e-06
        [meta_fg_expand]: 4.2e-05
        [replace_old_param]: 1.269e-05
        [inline_without_move]: 6.14001e-06
        [renormalize]: 0.00070451
        [add_forward_monad_depend]: 4.89e-06
        [auto_monad_grad]: 1.86e-06
        [auto_monad_eliminator]: 1.271e-05
        [cse]: 2.281e-05
        [replace_applicator]: 1.322e-05
    [Cycle 3]: 0.00039866, [27]
        [switch_simplify]: 6.96001e-06
        [loop_unroll]: 5.92999e-06
        [a_1]: 0.00012056
        [with_stream_mark]: 9.48002e-06
        [recompute_prepare]: 5.91e-06
        [updatestate_depend_eliminate]: 3.33998e-06
        [updatestate_assign_eliminate]: 2.60002e-06
        [updatestate_loads_eliminate]: 2.28002e-06
        [parameter_eliminate]: 1.43002e-06
        [specialize_transform]: 5.87001e-06
        [updatestate_useless_node_eliminater]: 5.92999e-06
        [accelerated_algorithm]: 6.35002e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 7.13e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.475e-05
        [j_node_and_user_rematch]: 1.002e-05
        [meta_fg_expand]: 2.48998e-06
        [replace_old_param]: 1.062e-05
        [inline_without_move]: 6.04001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.35001e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 6.94001e-06
        [cse]: 1.619e-05
        [replace_applicator]: 6.57002e-06
[py_interpret_to_execute_after_opt_a]: 1.34e-05
[rewriter_after_opt_a]: 4.624e-05
[convert_after_rewriter]: 7.7e-06
[order_py_execute_after_rewriter]: 5.50001e-06
[mutable_eliminate]: 0.00077291
[jit_opt_b]: 6.622e-05, [1]
    [Cycle 1]: 5.777e-05, [2]
        [frontend_op_eliminate]: 2.65e-05
        [inline_after_opt_a]: 1.905e-05
[cconv]: 2.623e-05
[loop_unroll]: 0.00043693
[jit_opt_after_cconv]: 0.00016015, [1]
    [Cycle 1]: 0.00015351, [11]
        [c_1]: 2.508e-05
        [parameter_eliminate]: 2.89001e-06
        [updatestate_depend_eliminate]: 6.56e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [cse]: 2.514e-05
        [call_graph_tuple_transform]: 2.238e-05
        [tuple_list_get_item_eliminator]: 6.46999e-06
        [none_parameter_eliminate]: 1.76003e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 6.49001e-06
[remove_dup_value]: 1.843e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 1.138e-05
[add_recomputation]: 4.899e-05
[cse_after_recomputation]: 2.643e-05, [1]
    [Cycle 1]: 2.095e-05, [1]
        [cse]: 1.515e-05
[auto_monad_reorder]: 2.437e-05
[get_jit_bprop_graph]: 2.12001e-06
[rewriter_after_jit_bprop_graph]: 0.00021908
[opt_after_jit_grad]: 0.00048403
[symbol_engine_optimizer]: 7.843e-05, [1]
    [Cycle 1]: 7.22e-05, [6]
        [build]: 3.73001e-06
        [elim_shapecalc]: 9.32001e-06
        [elim_not_effective]: 1.429e-05
        [opt_reshape]: 6.71999e-06
        [fold_const_symbol]: 9.87999e-06
        [renormalize]: 8.29983e-07
[validate]: 0.00010077
[backend_pass]: 1.17999e-06
[task_emit]: 0.696081
[execute]: 7e-06
Sums
    bootstrap                            :     0.001353s :  0.17%
    type_inference                       :     0.050983s :  6.56%
    event_method                         :     0.000047s :  0.01%
    auto_monad                           :     0.000195s :  0.03%
    graph_reusing                        :     0.000007s :  0.00%
    pre_auto_parallel                    :     0.000011s :  0.00%
    py_interpret_to_execute              :     0.000042s :  0.01%
    rewriter_before_opt_a                :     0.000159s :  0.02%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000165s :  0.02%
    jit_opt_a.loop_unroll                :     0.000093s :  0.01%
    jit_opt_a.a_1                        :     0.002251s :  0.29%
    jit_opt_a.with_stream_mark           :     0.000063s :  0.01%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.00%
    jit_opt_a.updatestate_depend_eliminate :     0.000018s :  0.00%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.00%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.00%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.00%
    jit_opt_a.specialize_transform       :     0.000028s :  0.00%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.00%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.00%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000027s :  0.00%
    jit_opt_a.merge_forward              :     0.000017s :  0.00%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000063s :  0.01%
    jit_opt_a.j_node_and_user_rematch    :     0.000055s :  0.01%
    jit_opt_a.meta_fg_expand             :     0.002423s :  0.31%
    jit_opt_a.replace_old_param          :     0.000099s :  0.01%
    jit_opt_a.inline_without_move        :     0.000075s :  0.01%
    jit_opt_a.renormalize                :     0.019762s :  2.54%
    jit_opt_a.add_forward_monad_depend   :     0.000061s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000082s :  0.01%
    jit_opt_a.cse                        :     0.000252s :  0.03%
    jit_opt_a.replace_applicator         :     0.000109s :  0.01%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.00%
    rewriter_after_opt_a                 :     0.000046s :  0.01%
    convert_after_rewriter               :     0.000008s :  0.00%
    order_py_execute_after_rewriter      :     0.000006s :  0.00%
    mutable_eliminate                    :     0.000773s :  0.10%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.00%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.00%
    cconv                                :     0.000026s :  0.00%
    loop_unroll                          :     0.000437s :  0.06%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.00%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  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 :     0.000003s :  0.00%
    jit_opt_after_cconv.cse              :     0.000025s :  0.00%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.00%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  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  :     0.000006s :  0.00%
    remove_dup_value                     :     0.000018s :  0.00%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000011s :  0.00%
    add_recomputation                    :     0.000049s :  0.01%
    cse_after_recomputation.cse          :     0.000015s :  0.00%
    auto_monad_reorder                   :     0.000024s :  0.00%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000219s :  0.03%
    opt_after_jit_grad                   :     0.000484s :  0.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.00%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.00%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.00%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000101s :  0.01%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.696081s : 89.57%
    execute                              :     0.000007s :  0.00%


Time group info:
------[substitution.]   0.000633   113
 0.35% :     0.000002s :      2: substitution.elim_not_effective
 0.27% :     0.000002s :      2: substitution.fold_const_symbol
 0.97% :     0.000006s :      4: substitution.graph_param_transform
63.95% :     0.000405s :     12: substitution.inline
 3.49% :     0.000022s :      2: substitution.inline_without_move
 2.95% :     0.000019s :     12: substitution.j_node_and_user_rematch
 1.94% :     0.000012s :      7: substitution.minmaximum_grad
 2.75% :     0.000017s :      5: substitution.partial_eliminate
 1.66% :     0.000011s :     12: substitution.remove_not_recompute_node
 5.13% :     0.000032s :     10: substitution.replace_applicator
 2.01% :     0.000013s :     14: substitution.replace_old_param
 0.51% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.69% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.54% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 7.79% :     0.000049s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.050836     2
97.16% :     0.049392s :      1: type_inference.infer
 2.84% :     0.001444s :      1: type_inference.specialize
------[replace.]   0.000184    21
60.10% :     0.000111s :     12: replace.inline
39.90% :     0.000073s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000424    21
93.97% :     0.000398s :     12: match.inline
 6.03% :     0.000026s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000385  2828
 1.46% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.40% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.40% :     0.000005s :     47: predicate.addn_check_dump
 1.58% :     0.000006s :     47: predicate.addn_zero_filter
 2.00% :     0.000008s :     47: predicate.arithmetic_simplify
 1.71% :     0.000007s :     47: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.39% :     0.000005s :     47: predicate.compare_switch_simplify
 1.51% :     0.000006s :     47: predicate.depend_value_elim
 1.44% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.53% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.47% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.40% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.58% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.46% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.42% :     0.000005s :     47: predicate.environ_get_depend_swap
 1.55% :     0.000006s :     47: predicate.environ_get_eliminate
 1.41% :     0.000005s :     47: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.94% :     0.000004s :     21: predicate.get_grad_eliminate
 0.14% :     0.000001s :      4: predicate.graph_param_transform
 4.58% :     0.000018s :     76: predicate.inline
 1.98% :     0.000008s :     47: predicate.inline_without_move
 0.40% :     0.000002s :     21: predicate.j_node_and_user_rematch
 0.94% :     0.000004s :     21: predicate.less_batch_normalization
 1.89% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 1.92% :     0.000007s :     60: predicate.load_eliminater
 0.43% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.50% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.70% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.39% :     0.000005s :     47: predicate.merge_addn
 1.46% :     0.000006s :     47: predicate.minmaximum_grad
 0.53% :     0.000002s :      4: predicate.mutable_eliminate
 0.17% :     0.000001s :      4: predicate.opt_reshape
 2.38% :     0.000009s :     60: predicate.partial_eliminate
 1.41% :     0.000005s :     47: predicate.print_const_string_wrapper
 1.86% :     0.000007s :     47: predicate.reduce_eliminate
 1.81% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.73% :     0.000011s :    108: predicate.replace_applicator
 1.18% :     0.000005s :     47: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.46% :     0.000006s :     47: predicate.reshape_eliminate
 1.45% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.25% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.46% :     0.000006s :     47: predicate.same_eliminate
 0.55% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.89% :     0.000003s :     21: predicate.specialize_transform
 1.59% :     0.000006s :     47: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.77% :     0.000011s :     68: predicate.switch_defer_inline
 2.69% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.42% :     0.000025s :    169: predicate.switch_simplify
 1.47% :     0.000006s :     47: predicate.tile_eliminate
 1.43% :     0.000006s :     47: predicate.transpose_eliminate
 1.78% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.65% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000014s :     64: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.89% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000006s :     47: 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.]   0.002212    31
66.40% :     0.001469s :     15: func_graph_cloner_run.FuncGraphClonerGraph
33.60% :     0.000743s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.809916    91
 0.01% :     0.000052s :      1: add_recomputation
 0.02% :     0.000200s :      1: auto_monad
 0.00% :     0.000027s :      1: auto_monad_reorder
 0.00% :     0.000005s :      1: backend_pass
 0.17% :     0.001378s :      1: bootstrap
 0.00% :     0.000029s :      1: cconv
 0.00% :     0.000010s :      1: convert_after_rewriter
 0.00% :     0.000029s :      1: cse_after_recomputation
 0.00% :     0.000014s :      1: environ_conv
 0.01% :     0.000052s :      1: event_method
 0.00% :     0.000012s :      1: execute
 0.00% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.00% :     0.000010s :      1: graph_reusing
 4.36% :     0.035335s :      1: jit_opt_a
 0.02% :     0.000163s :      1: jit_opt_after_cconv
 0.01% :     0.000069s :      1: jit_opt_b
 0.05% :     0.000445s :      1: loop_unroll
 0.10% :     0.000782s :      1: mutable_eliminate
 0.37% :     0.002983s :     39: opt.transform.jit_opt_a
 0.01% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.00% :     0.000039s :      4: opt.transform.jit_opt_b
 0.00% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.00% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.00% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.00% :     0.000037s :      4: opt.transform.symbol_engine_opt
 0.06% :     0.000492s :      1: opt_after_jit_grad
 0.00% :     0.000008s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000014s :      1: pre_auto_parallel
 0.01% :     0.000044s :      1: py_interpret_to_execute
 0.00% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.00% :     0.000021s :      1: remove_dup_value
 2.21% :     0.017912s :      2: renormalize.infer
 0.23% :     0.001831s :      2: renormalize.specialize
 0.03% :     0.000223s :      1: rewriter_after_jit_bprop_graph
 0.01% :     0.000050s :      1: rewriter_after_opt_a
 0.02% :     0.000162s :      1: rewriter_before_opt_a
 0.01% :     0.000081s :      1: symbol_engine_optimizer
85.95% :     0.696104s :      1: task_emit
 6.30% :     0.051004s :      1: type_inference
 0.02% :     0.000159s :      1: validate

[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.487.421 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe8f670> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.488.630 [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 = 0.0407833, [33]
[bootstrap]: 0.00027446
[type_inference]: 0.0215763
[event_method]: 2.062e-05
[auto_monad]: 0.00014773
[graph_reusing]: 6.68e-06
[pre_auto_parallel]: 2.46998e-06
[py_interpret_to_execute]: 0.00013253
[rewriter_before_opt_a]: 9.388e-05
[expand_dump_flag]: 3.46999e-06
[jit_opt_a]: 0.0059142, [2]
    [Cycle 1]: 0.00215159, [27]
        [switch_simplify]: 5.64e-05
        [loop_unroll]: 3.729e-05
        [a_1]: 0.00068955
        [with_stream_mark]: 1.643e-05
        [recompute_prepare]: 1.072e-05
        [updatestate_depend_eliminate]: 5.69e-06
        [updatestate_assign_eliminate]: 4.87998e-06
        [updatestate_loads_eliminate]: 4.38001e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 9.58002e-06
        [updatestate_useless_node_eliminater]: 1.125e-05
        [accelerated_algorithm]: 8.84e-06
        [meta_shard_fg_expand]: 2.25002e-06
        [get_grad_eliminate_]: 8.40001e-06
        [merge_forward]: 4.51002e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.687e-05
        [j_node_and_user_rematch]: 1.337e-05
        [meta_fg_expand]: 3.35e-06
        [replace_old_param]: 1.1e-05
        [inline_without_move]: 8.57e-06
        [renormalize]: 0.00098167
        [add_forward_monad_depend]: 4.95001e-06
        [auto_monad_grad]: 2.12999e-06
        [auto_monad_eliminator]: 1.772e-05
        [cse]: 3.121e-05
        [replace_applicator]: 1.542e-05
    [Cycle 2]: 0.00045297, [27]
        [switch_simplify]: 9.12001e-06
        [loop_unroll]: 7.76001e-06
        [a_1]: 0.00016485
        [with_stream_mark]: 1.035e-05
        [recompute_prepare]: 7.66999e-06
        [updatestate_depend_eliminate]: 4.80999e-06
        [updatestate_assign_eliminate]: 4.35e-06
        [updatestate_loads_eliminate]: 3.68999e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 7.68001e-06
        [updatestate_useless_node_eliminater]: 9.99999e-06
        [accelerated_algorithm]: 7.54002e-06
        [meta_shard_fg_expand]: 1.82999e-06
        [get_grad_eliminate_]: 7.20998e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.79998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.547e-05
        [j_node_and_user_rematch]: 1.22e-05
        [meta_fg_expand]: 2.84001e-06
        [replace_old_param]: 9.92001e-06
        [inline_without_move]: 7.57998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.29998e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 9.94001e-06
        [cse]: 1.906e-05
        [replace_applicator]: 7.84002e-06
[py_interpret_to_execute_after_opt_a]: 1.078e-05
[rewriter_after_opt_a]: 0.00027661
[convert_after_rewriter]: 1.014e-05
[order_py_execute_after_rewriter]: 7.26001e-06
[mutable_eliminate]: 0.00055967
[jit_opt_b]: 6.721e-05, [1]
    [Cycle 1]: 6.047e-05, [2]
        [frontend_op_eliminate]: 2.451e-05
        [inline_after_opt_a]: 2.43e-05
[cconv]: 1.991e-05
[loop_unroll]: 0.00045926
[jit_opt_after_cconv]: 0.00019508, [1]
    [Cycle 1]: 0.00018801, [11]
        [c_1]: 4.559e-05
        [parameter_eliminate]: 2.94999e-06
        [updatestate_depend_eliminate]: 7.84002e-06
        [updatestate_assign_eliminate]: 4.78001e-06
        [updatestate_loads_eliminate]: 3.78999e-06
        [cse]: 2.749e-05
        [call_graph_tuple_transform]: 2.296e-05
        [tuple_list_get_item_eliminator]: 8.89998e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 9.15999e-06
[remove_dup_value]: 1.672e-05
[partial_unused_args_eliminate]: 2.18998e-06
[environ_conv]: 9.69e-06
[add_recomputation]: 5.507e-05
[cse_after_recomputation]: 3.061e-05, [1]
    [Cycle 1]: 2.465e-05, [1]
        [cse]: 1.841e-05
[auto_monad_reorder]: 2.206e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 5.19e-06
[opt_after_jit_grad]: 0.00061788
[symbol_engine_optimizer]: 0.00010223, [1]
    [Cycle 1]: 9.569e-05, [6]
        [build]: 1.034e-05
        [elim_shapecalc]: 1.222e-05
        [elim_not_effective]: 2e-05
        [opt_reshape]: 8.70001e-06
        [fold_const_symbol]: 1.403e-05
        [renormalize]: 4.69998e-07
[validate]: 5.072e-05
[backend_pass]: 1.14998e-06
[task_emit]: 0.00987314
[execute]: 6.86001e-06
Sums
    bootstrap                            :     0.000274s :  0.74%
    type_inference                       :     0.021576s : 58.55%
    event_method                         :     0.000021s :  0.06%
    auto_monad                           :     0.000148s :  0.40%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000133s :  0.36%
    rewriter_before_opt_a                :     0.000094s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.18%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000854s :  2.32%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000021s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  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 :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000982s :  2.66%
    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      :     0.000028s :  0.08%
    jit_opt_a.cse                        :     0.000050s :  0.14%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000277s :  0.75%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000560s :  1.52%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000020s :  0.05%
    loop_unroll                          :     0.000459s :  1.25%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.12%
    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              :     0.000027s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000055s :  0.15%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000022s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000618s :  1.68%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000051s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009873s : 26.79%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000244    47
 3.02% :     0.000007s :      2: substitution.depend_value_elim
 1.27% :     0.000003s :      4: substitution.elim_not_effective
 0.81% :     0.000002s :      4: substitution.fold_const_symbol
 2.60% :     0.000006s :      5: substitution.graph_param_transform
75.77% :     0.000185s :      4: substitution.inline
 1.74% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.72% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.47% :     0.000004s :      2: substitution.replace_old_param
 3.88% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
 3.33% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.38% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021514     2
93.48% :     0.020111s :      1: type_inference.infer
 6.52% :     0.001403s :      1: type_inference.specialize
------[replace.]   0.000091     6
65.25% :     0.000059s :      4: replace.inline
34.75% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
95.65% :     0.000182s :      4: match.inline
 4.35% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000172  1021
 1.57% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.67% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     15: predicate.addn_check_dump
 1.28% :     0.000002s :     15: predicate.addn_zero_filter
 1.99% :     0.000003s :     15: predicate.arithmetic_simplify
 1.29% :     0.000002s :     15: predicate.cast_eliminate
 0.49% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     15: predicate.compare_switch_simplify
 1.38% :     0.000002s :     15: predicate.depend_value_elim
 1.13% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      5: predicate.fold_const_symbol
 1.19% :     0.000002s :     10: predicate.get_grad_eliminate
 0.30% :     0.000001s :      5: predicate.graph_param_transform
 5.03% :     0.000009s :     31: predicate.inline
 1.08% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :     10: predicate.less_batch_normalization
 1.57% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.08% :     0.000004s :     22: predicate.load_eliminater
 1.62% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.46% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.18% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     15: predicate.merge_addn
 1.11% :     0.000002s :     15: predicate.minmaximum_grad
 1.47% :     0.000003s :      5: predicate.mutable_eliminate
 0.47% :     0.000001s :      5: predicate.opt_reshape
 2.33% :     0.000004s :     22: predicate.partial_eliminate
 1.18% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.67% :     0.000003s :     27: predicate.replace_applicator
 0.68% :     0.000001s :     10: predicate.replace_old_param
 0.38% :     0.000001s :      5: predicate.reset_defer_inline
 1.25% :     0.000002s :     15: predicate.reshape_eliminate
 1.27% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.92% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     15: predicate.same_eliminate
 0.60% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.08% :     0.000002s :     10: predicate.special_op_eliminate
 1.26% :     0.000002s :     10: predicate.specialize_transform
 1.38% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.56% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.30% :     0.000004s :     21: predicate.switch_defer_inline
 1.91% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.88% :     0.000012s :     68: predicate.switch_simplify
 1.21% :     0.000002s :     15: predicate.tile_eliminate
 1.23% :     0.000002s :     15: predicate.transpose_eliminate
 1.58% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.76% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000003s :     15: predicate.value_based_eliminate
 0.45% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000800    13
46.22% :     0.000370s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.90% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.88% :     0.000391s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043108    76
 0.13% :     0.000058s :      1: add_recomputation
 0.35% :     0.000153s :      1: auto_monad
 0.06% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000290s :      1: bootstrap
 0.05% :     0.000023s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.08% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.06% :     0.000025s :      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% :     0.000009s :      1: graph_reusing
13.73% :     0.005917s :      1: jit_opt_a
 0.46% :     0.000199s :      1: jit_opt_after_cconv
 0.16% :     0.000070s :      1: jit_opt_b
 1.09% :     0.000468s :      1: loop_unroll
 1.32% :     0.000569s :      1: mutable_eliminate
 2.66% :     0.001146s :     26: opt.transform.jit_opt_a
 0.19% :     0.000083s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.46% :     0.000628s :      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
 0.32% :     0.000137s :      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
 1.29% :     0.000556s :      1: renormalize.infer
 0.97% :     0.000418s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.65% :     0.000281s :      1: rewriter_after_opt_a
 0.23% :     0.000098s :      1: rewriter_before_opt_a
 0.24% :     0.000105s :      1: symbol_engine_optimizer
22.93% :     0.009884s :      1: task_emit
50.08% :     0.021589s :      1: type_inference
 0.17% :     0.000074s :      1: validate


TotalTime = 0.0685193, [33]
[bootstrap]: 0.00025513
[type_inference]: 0.0307786
[event_method]: 0.00014184
[auto_monad]: 0.00018974
[graph_reusing]: 6.43e-06
[pre_auto_parallel]: 2.30002e-06
[py_interpret_to_execute]: 3.765e-05
[rewriter_before_opt_a]: 0.00012228
[expand_dump_flag]: 2.72001e-06
[jit_opt_a]: 0.0270523, [3]
    [Cycle 1]: 0.0203065, [27]
        [switch_simplify]: 7.631e-05
        [loop_unroll]: 4.976e-05
        [a_1]: 0.00111495
        [with_stream_mark]: 3.122e-05
        [recompute_prepare]: 2.583e-05
        [updatestate_depend_eliminate]: 1.323e-05
        [updatestate_assign_eliminate]: 1.099e-05
        [updatestate_loads_eliminate]: 9.97999e-06
        [parameter_eliminate]: 2.57001e-06
        [specialize_transform]: 2.052e-05
        [updatestate_useless_node_eliminater]: 2.287e-05
        [accelerated_algorithm]: 1.901e-05
        [meta_shard_fg_expand]: 5.27001e-06
        [get_grad_eliminate_]: 1.848e-05
        [merge_forward]: 1.137e-05
        [cell_reuse_recompute_pass]: 1.04003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.73e-05
        [j_node_and_user_rematch]: 3.28e-05
        [meta_fg_expand]: 0.0053394
        [replace_old_param]: 0.00011255
        [inline_without_move]: 0.00011138
        [renormalize]: 0.012452
        [add_forward_monad_depend]: 1.353e-05
        [auto_monad_grad]: 8.65001e-06
        [auto_monad_eliminator]: 0.00012108
        [cse]: 0.00025193
        [replace_applicator]: 0.00016031
    [Cycle 2]: 0.00267454, [27]
        [switch_simplify]: 6.091e-05
        [loop_unroll]: 5.787e-05
        [a_1]: 0.0009521
        [with_stream_mark]: 1.709e-05
        [recompute_prepare]: 1.319e-05
        [updatestate_depend_eliminate]: 2.4e-05
        [updatestate_assign_eliminate]: 5.55001e-06
        [updatestate_loads_eliminate]: 5.00999e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 9.71e-06
        [updatestate_useless_node_eliminater]: 1.085e-05
        [accelerated_algorithm]: 8.50001e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 8.11002e-06
        [merge_forward]: 5.13002e-06
        [cell_reuse_recompute_pass]: 1.06002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.67e-05
        [j_node_and_user_rematch]: 1.291e-05
        [meta_fg_expand]: 7.275e-05
        [replace_old_param]: 1.2e-05
        [inline_without_move]: 8.22998e-06
        [renormalize]: 0.00109618
        [add_forward_monad_depend]: 4.78001e-06
        [auto_monad_grad]: 1.92999e-06
        [auto_monad_eliminator]: 1.625e-05
        [cse]: 6.937e-05
        [replace_applicator]: 1.728e-05
    [Cycle 3]: 0.00047211, [27]
        [switch_simplify]: 9.91998e-06
        [loop_unroll]: 8.34998e-06
        [a_1]: 0.00016772
        [with_stream_mark]: 1.14e-05
        [recompute_prepare]: 8.1e-06
        [updatestate_depend_eliminate]: 5.64998e-06
        [updatestate_assign_eliminate]: 4.25999e-06
        [updatestate_loads_eliminate]: 3.9e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 8.17e-06
        [updatestate_useless_node_eliminater]: 1.125e-05
        [accelerated_algorithm]: 7.87e-06
        [meta_shard_fg_expand]: 1.87999e-06
        [get_grad_eliminate_]: 7.34002e-06
        [merge_forward]: 4.04002e-06
        [cell_reuse_recompute_pass]: 1.59998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.718e-05
        [j_node_and_user_rematch]: 1.299e-05
        [meta_fg_expand]: 2.78e-06
        [replace_old_param]: 1.025e-05
        [inline_without_move]: 7.63999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.58002e-06
        [auto_monad_grad]: 9.90025e-07
        [auto_monad_eliminator]: 1.006e-05
        [cse]: 2.19e-05
        [replace_applicator]: 8e-06
[py_interpret_to_execute_after_opt_a]: 1.266e-05
[rewriter_after_opt_a]: 5.367e-05
[convert_after_rewriter]: 8.35999e-06
[order_py_execute_after_rewriter]: 6.28e-06
[mutable_eliminate]: 0.00068416
[jit_opt_b]: 6.824e-05, [1]
    [Cycle 1]: 6.132e-05, [2]
        [frontend_op_eliminate]: 2.466e-05
        [inline_after_opt_a]: 2.465e-05
[cconv]: 2.445e-05
[loop_unroll]: 0.00044578
[jit_opt_after_cconv]: 0.00020517, [1]
    [Cycle 1]: 0.00019871, [11]
        [c_1]: 4.659e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 8.67e-06
        [updatestate_assign_eliminate]: 4.63999e-06
        [updatestate_loads_eliminate]: 4.18999e-06
        [cse]: 3.233e-05
        [call_graph_tuple_transform]: 2.45e-05
        [tuple_list_get_item_eliminator]: 9.09e-06
        [none_parameter_eliminate]: 1.41002e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 8.17e-06
[remove_dup_value]: 2.134e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 1.043e-05
[add_recomputation]: 5.966e-05
[cse_after_recomputation]: 3.169e-05, [1]
    [Cycle 1]: 2.601e-05, [1]
        [cse]: 1.999e-05
[auto_monad_reorder]: 2.307e-05
[get_jit_bprop_graph]: 1.42e-06
[rewriter_after_jit_bprop_graph]: 5.42001e-06
[opt_after_jit_grad]: 0.00050988
[symbol_engine_optimizer]: 9.938e-05, [1]
    [Cycle 1]: 9.321e-05, [6]
        [build]: 9.05999e-06
        [elim_shapecalc]: 1.157e-05
        [elim_not_effective]: 1.938e-05
        [opt_reshape]: 9.42001e-06
        [fold_const_symbol]: 1.431e-05
        [renormalize]: 4.80009e-07
[validate]: 4.544e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.0073944
[execute]: 6.09001e-06
Sums
    bootstrap                            :     0.000255s :  0.40%
    type_inference                       :     0.030779s : 48.04%
    event_method                         :     0.000142s :  0.22%
    auto_monad                           :     0.000190s :  0.30%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000038s :  0.06%
    rewriter_before_opt_a                :     0.000122s :  0.19%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000147s :  0.23%
    jit_opt_a.loop_unroll                :     0.000116s :  0.18%
    jit_opt_a.a_1                        :     0.002235s :  3.49%
    jit_opt_a.with_stream_mark           :     0.000060s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000047s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000043s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000021s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000019s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000038s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000045s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000035s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000034s :  0.05%
    jit_opt_a.merge_forward              :     0.000021s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000071s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000059s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.005415s :  8.45%
    jit_opt_a.replace_old_param          :     0.000135s :  0.21%
    jit_opt_a.inline_without_move        :     0.000127s :  0.20%
    jit_opt_a.renormalize                :     0.013548s : 21.15%
    jit_opt_a.add_forward_monad_depend   :     0.000020s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000012s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000147s :  0.23%
    jit_opt_a.cse                        :     0.000343s :  0.54%
    jit_opt_a.replace_applicator         :     0.000186s :  0.29%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.02%
    rewriter_after_opt_a                 :     0.000054s :  0.08%
    convert_after_rewriter               :     0.000008s :  0.01%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000684s :  1.07%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.04%
    cconv                                :     0.000024s :  0.04%
    loop_unroll                          :     0.000446s :  0.70%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    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              :     0.000032s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  0.01%
    remove_dup_value                     :     0.000021s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000010s :  0.02%
    add_recomputation                    :     0.000060s :  0.09%
    cse_after_recomputation.cse          :     0.000020s :  0.03%
    auto_monad_reorder                   :     0.000023s :  0.04%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000510s :  0.80%
    symbol_engine_optimizer.build        :     0.000009s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.07%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007394s : 11.54%
    execute                              :     0.000006s :  0.01%


Time group info:
------[substitution.]   0.001388   160
 1.41% :     0.000019s :      8: substitution.depend_value_elim
 0.23% :     0.000003s :      4: substitution.elim_not_effective
 0.16% :     0.000002s :      4: substitution.fold_const_symbol
56.25% :     0.000780s :      4: substitution.getattr_setattr_resolve
 0.49% :     0.000007s :      5: substitution.graph_param_transform
24.75% :     0.000343s :     10: substitution.inline
 2.04% :     0.000028s :      4: substitution.inline_without_move
 0.80% :     0.000011s :     20: substitution.j_node_and_user_rematch
 0.63% :     0.000009s :      5: substitution.minmaximum_grad
 0.28% :     0.000004s :      3: substitution.partial_eliminate
 1.05% :     0.000015s :     20: substitution.remove_not_recompute_node
 3.10% :     0.000043s :     13: substitution.replace_applicator
 0.92% :     0.000013s :     18: substitution.replace_old_param
 0.22% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.38% :     0.000019s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.89% :     0.000012s :      5: substitution.tuple_list_get_item_depend_reorder
 2.61% :     0.000036s :     10: substitution.tuple_list_get_item_eliminator
 0.99% :     0.000014s :      8: substitution.updatestate_pure_node_eliminater
 1.81% :     0.000025s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.030639     2
94.22% :     0.028866s :      1: type_inference.infer
 5.78% :     0.001772s :      1: type_inference.specialize
------[replace.]   0.000275    20
17.27% :     0.000048s :      3: replace.getattr_setattr_resolve
31.83% :     0.000088s :     10: replace.inline
11.25% :     0.000031s :      1: replace.replace_applicator
31.64% :     0.000087s :      5: replace.tuple_list_get_item_eliminator
 8.00% :     0.000022s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001107    20
66.30% :     0.000734s :      3: match.getattr_setattr_resolve
30.41% :     0.000337s :     10: match.inline
 0.83% :     0.000009s :      1: match.replace_applicator
 1.75% :     0.000019s :      5: match.tuple_list_get_item_eliminator
 0.71% :     0.000008s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000444  2979
 1.38% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.46% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.31% :     0.000006s :     46: predicate.addn_check_dump
 1.52% :     0.000007s :     46: predicate.addn_zero_filter
 2.06% :     0.000009s :     46: predicate.arithmetic_simplify
 1.58% :     0.000007s :     46: predicate.cast_eliminate
 0.28% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.39% :     0.000006s :     46: predicate.compare_switch_simplify
 1.46% :     0.000006s :     46: predicate.depend_value_elim
 1.31% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.50% :     0.000007s :     46: predicate.dict_get_item_eliminator
 1.46% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.38% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.16% :     0.000001s :      5: predicate.elim_not_effective
 0.27% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.29% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.28% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.41% :     0.000006s :     46: predicate.environ_get_eliminate
 1.26% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      5: predicate.fold_const_symbol
 1.00% :     0.000004s :     27: predicate.get_grad_eliminate
 1.39% :     0.000006s :     20: predicate.getattr_setattr_resolve
 0.11% :     0.000001s :      5: predicate.graph_param_transform
 3.91% :     0.000017s :     72: predicate.inline
 3.34% :     0.000015s :     91: predicate.inline_without_move
 0.44% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.22% :     0.000005s :     27: predicate.less_batch_normalization
 1.66% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.77% :     0.000008s :     56: predicate.load_eliminater
 0.55% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.80% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.77% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000006s :     46: predicate.merge_addn
 1.32% :     0.000006s :     46: predicate.minmaximum_grad
 0.59% :     0.000003s :      5: predicate.mutable_eliminate
 0.21% :     0.000001s :      5: predicate.opt_reshape
 2.24% :     0.000010s :     56: predicate.partial_eliminate
 1.56% :     0.000007s :     46: predicate.print_const_string_wrapper
 1.89% :     0.000008s :     46: predicate.reduce_eliminate
 1.63% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000002s :     27: predicate.remove_not_recompute_node
 2.84% :     0.000013s :    126: predicate.replace_applicator
 1.61% :     0.000007s :     91: predicate.replace_old_param
 0.12% :     0.000001s :      5: predicate.reset_defer_inline
 1.32% :     0.000006s :     46: predicate.reshape_eliminate
 1.49% :     0.000007s :     46: predicate.row_tensor_add_zeros_like
 0.38% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.43% :     0.000006s :     46: predicate.same_eliminate
 0.61% :     0.000003s :     29: predicate.set_cell_output_no_recompute
 0.41% :     0.000002s :     10: predicate.special_op_eliminate
 1.08% :     0.000005s :     27: predicate.specialize_transform
 1.60% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000007s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.33% :     0.000010s :     62: predicate.switch_defer_inline
 2.01% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.37% :     0.000028s :    188: predicate.switch_simplify
 1.34% :     0.000006s :     46: predicate.tile_eliminate
 1.38% :     0.000006s :     46: predicate.transpose_eliminate
 1.72% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.54% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 2.94% :     0.000013s :     61: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000008s :     46: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.72% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000013s :     84: predicate.updatestate_useless_node_eliminater
 1.96% :     0.000009s :     46: predicate.value_based_eliminate
 0.18% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.20% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003227    48
67.70% :     0.002184s :     27: func_graph_cloner_run.FuncGraphClonerGraph
32.30% :     0.001042s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.086356    93
 0.07% :     0.000062s :      1: add_recomputation
 0.23% :     0.000198s :      1: auto_monad
 0.03% :     0.000026s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.32% :     0.000273s :      1: bootstrap
 0.03% :     0.000027s :      1: cconv
 0.01% :     0.000011s :      1: convert_after_rewriter
 0.04% :     0.000034s :      1: cse_after_recomputation
 0.01% :     0.000013s :      1: environ_conv
 0.17% :     0.000147s :      1: event_method
 0.01% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
31.33% :     0.027056s :      1: jit_opt_a
 0.24% :     0.000208s :      1: jit_opt_after_cconv
 0.08% :     0.000071s :      1: jit_opt_b
 0.53% :     0.000454s :      1: loop_unroll
 0.80% :     0.000692s :      1: mutable_eliminate
 3.74% :     0.003227s :     39: opt.transform.jit_opt_a
 0.10% :     0.000084s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000042s :      4: opt.transform.jit_opt_b
 0.02% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 1.02% :     0.000878s :      2: opt.transform.opt_resolve
 0.06% :     0.000052s :      4: opt.transform.symbol_engine_opt
 0.60% :     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
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.05% :     0.000041s :      1: py_interpret_to_execute
 0.02% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000024s :      1: remove_dup_value
12.27% :     0.010592s :      2: renormalize.infer
 3.41% :     0.002940s :      2: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000057s :      1: rewriter_after_opt_a
 0.15% :     0.000126s :      1: rewriter_before_opt_a
 0.12% :     0.000102s :      1: symbol_engine_optimizer
 8.57% :     0.007404s :      1: task_emit
35.66% :     0.030793s :      1: type_inference
 0.08% :     0.000067s :      1: validate

[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.512.725 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffe2caf6a60> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.513.390 [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 = 0.0406505, [33]
[bootstrap]: 0.00030513
[type_inference]: 0.0118086
[event_method]: 4.46e-05
[auto_monad]: 9.266e-05
[graph_reusing]: 6.73e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 2.715e-05
[rewriter_before_opt_a]: 0.00010594
[expand_dump_flag]: 2.66999e-06
[jit_opt_a]: 0.0176402, [3]
    [Cycle 1]: 0.0109255, [27]
        [switch_simplify]: 0.00010164
        [loop_unroll]: 4.61e-05
        [a_1]: 0.00093455
        [with_stream_mark]: 2.367e-05
        [recompute_prepare]: 1.95e-05
        [updatestate_depend_eliminate]: 8.64e-06
        [updatestate_assign_eliminate]: 6.83e-06
        [updatestate_loads_eliminate]: 6.38998e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 1.487e-05
        [updatestate_useless_node_eliminater]: 1.386e-05
        [accelerated_algorithm]: 1.409e-05
        [meta_shard_fg_expand]: 3.11999e-06
        [get_grad_eliminate_]: 1.378e-05
        [merge_forward]: 7.95e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.565e-05
        [j_node_and_user_rematch]: 2.472e-05
        [meta_fg_expand]: 0.00162786
        [replace_old_param]: 6.312e-05
        [inline_without_move]: 5.765e-05
        [renormalize]: 0.00733452
        [add_forward_monad_depend]: 1.7e-05
        [auto_monad_grad]: 7.04001e-06
        [auto_monad_eliminator]: 5.766e-05
        [cse]: 0.00020099
        [replace_applicator]: 8.271e-05
    [Cycle 2]: 0.00233642, [27]
        [switch_simplify]: 4.224e-05
        [loop_unroll]: 4.055e-05
        [a_1]: 0.00120637
        [with_stream_mark]: 1.778e-05
        [recompute_prepare]: 9.49e-06
        [updatestate_depend_eliminate]: 4.03001e-06
        [updatestate_assign_eliminate]: 3.7e-06
        [updatestate_loads_eliminate]: 2.95998e-06
        [parameter_eliminate]: 1.90001e-06
        [specialize_transform]: 6.78e-06
        [updatestate_useless_node_eliminater]: 6.38e-06
        [accelerated_algorithm]: 6.44999e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 7.21999e-06
        [merge_forward]: 4.97e-06
        [cell_reuse_recompute_pass]: 1.39998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.514e-05
        [j_node_and_user_rematch]: 1.015e-05
        [meta_fg_expand]: 4.612e-05
        [replace_old_param]: 1.217e-05
        [inline_without_move]: 6.26e-06
        [renormalize]: 0.00067045
        [add_forward_monad_depend]: 4.85999e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.15e-05
        [cse]: 2.519e-05
        [replace_applicator]: 1.297e-05
    [Cycle 3]: 0.0003912, [27]
        [switch_simplify]: 7.38999e-06
        [loop_unroll]: 6.78e-06
        [a_1]: 0.0001378
        [with_stream_mark]: 9.42001e-06
        [recompute_prepare]: 6.26e-06
        [updatestate_depend_eliminate]: 3.44001e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 6.30002e-06
        [updatestate_useless_node_eliminater]: 5.87999e-06
        [accelerated_algorithm]: 5.97999e-06
        [meta_shard_fg_expand]: 1.31998e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 2.09999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.32e-05
        [j_node_and_user_rematch]: 9.02999e-06
        [meta_fg_expand]: 2.00002e-06
        [replace_old_param]: 1.001e-05
        [inline_without_move]: 5.66e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 7.03998e-06
        [cse]: 1.613e-05
        [replace_applicator]: 6.05002e-06
[py_interpret_to_execute_after_opt_a]: 1.261e-05
[rewriter_after_opt_a]: 3.817e-05
[convert_after_rewriter]: 7.49002e-06
[order_py_execute_after_rewriter]: 5.13002e-06
[mutable_eliminate]: 0.00064111
[jit_opt_b]: 5.643e-05, [1]
    [Cycle 1]: 4.896e-05, [2]
        [frontend_op_eliminate]: 1.862e-05
        [inline_after_opt_a]: 1.827e-05
[cconv]: 2.405e-05
[loop_unroll]: 0.00043252
[jit_opt_after_cconv]: 0.00016532, [1]
    [Cycle 1]: 0.00015862, [11]
        [c_1]: 2.559e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 6.32001e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [cse]: 2.383e-05
        [call_graph_tuple_transform]: 2.084e-05
        [tuple_list_get_item_eliminator]: 6.51999e-06
        [none_parameter_eliminate]: 1.53002e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 1.679e-05
[remove_dup_value]: 1.732e-05
[partial_unused_args_eliminate]: 2.13002e-06
[environ_conv]: 5.87999e-06
[add_recomputation]: 5.226e-05
[cse_after_recomputation]: 2.723e-05, [1]
    [Cycle 1]: 2.178e-05, [1]
        [cse]: 1.563e-05
[auto_monad_reorder]: 1.831e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 4.92999e-06
[opt_after_jit_grad]: 0.0005175
[symbol_engine_optimizer]: 8.145e-05, [1]
    [Cycle 1]: 7.456e-05, [6]
        [build]: 3.71001e-06
        [elim_shapecalc]: 1.013e-05
        [elim_not_effective]: 1.455e-05
        [opt_reshape]: 7.05998e-06
        [fold_const_symbol]: 1.015e-05
        [renormalize]: 5.10016e-07
[validate]: 3.838e-05
[backend_pass]: 1.10001e-06
[task_emit]: 0.00825715
[execute]: 8.13999e-06
Sums
    bootstrap                            :     0.000305s :  0.85%
    type_inference                       :     0.011809s : 32.92%
    event_method                         :     0.000045s :  0.12%
    auto_monad                           :     0.000093s :  0.26%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.08%
    rewriter_before_opt_a                :     0.000106s :  0.30%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000151s :  0.42%
    jit_opt_a.loop_unroll                :     0.000093s :  0.26%
    jit_opt_a.a_1                        :     0.002279s :  6.35%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.04%
    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 :  0.01%
    jit_opt_a.specialize_transform       :     0.000028s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  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_        :     0.000027s :  0.07%
    jit_opt_a.merge_forward              :     0.000016s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.001676s :  4.67%
    jit_opt_a.replace_old_param          :     0.000085s :  0.24%
    jit_opt_a.inline_without_move        :     0.000070s :  0.19%
    jit_opt_a.renormalize                :     0.008005s : 22.32%
    jit_opt_a.add_forward_monad_depend   :     0.000023s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000076s :  0.21%
    jit_opt_a.cse                        :     0.000242s :  0.68%
    jit_opt_a.replace_applicator         :     0.000102s :  0.28%
    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                    :     0.000641s :  1.79%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.05%
    cconv                                :     0.000024s :  0.07%
    loop_unroll                          :     0.000433s :  1.21%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.07%
    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 :  0.07%
    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 :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000017s :  0.05%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000052s :  0.15%
    cse_after_recomputation.cse          :     0.000016s :  0.04%
    auto_monad_reorder                   :     0.000018s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000518s :  1.44%
    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 :  0.04%
    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                             :     0.000038s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008257s : 23.02%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000575   113
 0.39% :     0.000002s :      2: substitution.elim_not_effective
 0.22% :     0.000001s :      2: substitution.fold_const_symbol
 1.04% :     0.000006s :      4: substitution.graph_param_transform
66.54% :     0.000383s :     12: substitution.inline
 2.92% :     0.000017s :      2: substitution.inline_without_move
 1.47% :     0.000008s :     12: substitution.j_node_and_user_rematch
 1.87% :     0.000011s :      7: substitution.minmaximum_grad
 1.25% :     0.000007s :      5: substitution.partial_eliminate
 1.79% :     0.000010s :     12: substitution.remove_not_recompute_node
 5.27% :     0.000030s :     10: substitution.replace_applicator
 1.94% :     0.000011s :     14: substitution.replace_old_param
 0.43% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.12% :     0.000024s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.79% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 7.97% :     0.000046s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011736     2
90.18% :     0.010583s :      1: type_inference.infer
 9.82% :     0.001153s :      1: type_inference.specialize
------[replace.]   0.000177    21
58.42% :     0.000104s :     12: replace.inline
41.58% :     0.000074s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000399    21
94.11% :     0.000376s :     12: match.inline
 5.89% :     0.000024s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000408  2828
 1.52% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.47% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.39% :     0.000006s :     47: predicate.addn_check_dump
 1.52% :     0.000006s :     47: predicate.addn_zero_filter
 1.82% :     0.000007s :     47: predicate.arithmetic_simplify
 1.45% :     0.000006s :     47: predicate.cast_eliminate
 0.14% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.30% :     0.000005s :     47: predicate.compare_switch_simplify
 1.37% :     0.000006s :     47: predicate.depend_value_elim
 1.34% :     0.000005s :     47: predicate.dict_get_item_const_eliminator
 1.42% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.39% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.35% :     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
 1.39% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.35% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.37% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.43% :     0.000006s :     47: predicate.environ_get_eliminate
 1.34% :     0.000005s :     47: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.88% :     0.000004s :     21: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.17% :     0.000017s :     76: predicate.inline
 1.78% :     0.000007s :     47: predicate.inline_without_move
 0.38% :     0.000002s :     21: predicate.j_node_and_user_rematch
 0.91% :     0.000004s :     21: predicate.less_batch_normalization
 1.81% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 1.82% :     0.000007s :     60: predicate.load_eliminater
 0.46% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.39% :     0.000014s :     97: predicate.loop_unroll_before_grad
 1.65% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.35% :     0.000005s :     47: predicate.merge_addn
 1.39% :     0.000006s :     47: predicate.minmaximum_grad
 0.45% :     0.000002s :      4: predicate.mutable_eliminate
 0.17% :     0.000001s :      4: predicate.opt_reshape
 2.34% :     0.000010s :     60: predicate.partial_eliminate
 1.44% :     0.000006s :     47: predicate.print_const_string_wrapper
 1.85% :     0.000008s :     47: predicate.reduce_eliminate
 1.73% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.66% :     0.000011s :    108: predicate.replace_applicator
 1.08% :     0.000004s :     47: predicate.replace_old_param
 0.12% :     0.000001s :      4: predicate.reset_defer_inline
 1.39% :     0.000006s :     47: predicate.reshape_eliminate
 1.39% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.22% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.52% :     0.000006s :     47: predicate.same_eliminate
 0.47% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.32% :     0.000001s :      8: predicate.special_op_eliminate
 0.79% :     0.000003s :     21: predicate.specialize_transform
 1.60% :     0.000007s :     47: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.60% :     0.000011s :     68: predicate.switch_defer_inline
 2.41% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.02% :     0.000025s :    169: predicate.switch_simplify
 1.44% :     0.000006s :     47: predicate.tile_eliminate
 1.41% :     0.000006s :     47: predicate.transpose_eliminate
 1.81% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.62% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 3.26% :     0.000013s :     64: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 6.56% :     0.000027s :     56: predicate.tuple_to_list_eliminator_
 1.80% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000007s :     47: 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.]   0.001715    31
60.56% :     0.001038s :     15: func_graph_cloner_run.FuncGraphClonerGraph
39.44% :     0.000676s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.051770    91
 0.11% :     0.000055s :      1: add_recomputation
 0.19% :     0.000097s :      1: auto_monad
 0.04% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.61% :     0.000318s :      1: bootstrap
 0.05% :     0.000027s :      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% :     0.000049s :      1: event_method
 0.02% :     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
34.08% :     0.017644s :      1: jit_opt_a
 0.32% :     0.000168s :      1: jit_opt_after_cconv
 0.11% :     0.000059s :      1: jit_opt_b
 0.85% :     0.000440s :      1: loop_unroll
 1.25% :     0.000650s :      1: mutable_eliminate
 5.75% :     0.002976s :     39: opt.transform.jit_opt_a
 0.13% :     0.000066s :      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% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.02% :     0.000526s :      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
 0.06% :     0.000030s :      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
12.17% :     0.006303s :      2: renormalize.infer
 3.26% :     0.001686s :      2: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000041s :      1: rewriter_after_opt_a
 0.21% :     0.000109s :      1: rewriter_before_opt_a
 0.16% :     0.000084s :      1: symbol_engine_optimizer
15.97% :     0.008270s :      1: task_emit
22.83% :     0.011821s :      1: type_inference
 0.13% :     0.000065s :      1: validate

[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.554.321 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe8faf0> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.555.043 [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 = 0.0426161, [33]
[bootstrap]: 0.00035027
[type_inference]: 0.0219895
[event_method]: 2.122e-05
[auto_monad]: 0.00014848
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 2.83998e-06
[py_interpret_to_execute]: 0.00011555
[rewriter_before_opt_a]: 9.834e-05
[expand_dump_flag]: 3.66001e-06
[jit_opt_a]: 0.00642422, [2]
    [Cycle 1]: 0.00254779, [27]
        [switch_simplify]: 6.244e-05
        [loop_unroll]: 3.738e-05
        [a_1]: 0.00069635
        [with_stream_mark]: 1.847e-05
        [recompute_prepare]: 1.09e-05
        [updatestate_depend_eliminate]: 6.81999e-06
        [updatestate_assign_eliminate]: 5.77999e-06
        [updatestate_loads_eliminate]: 4.63001e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 9.17999e-06
        [updatestate_useless_node_eliminater]: 1.201e-05
        [accelerated_algorithm]: 8.11002e-06
        [meta_shard_fg_expand]: 3.13998e-06
        [get_grad_eliminate_]: 8.02e-06
        [merge_forward]: 6.14001e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.99e-05
        [j_node_and_user_rematch]: 1.299e-05
        [meta_fg_expand]: 3.41999e-06
        [replace_old_param]: 1.23e-05
        [inline_without_move]: 8.27e-06
        [renormalize]: 0.00132574
        [add_forward_monad_depend]: 6.93998e-06
        [auto_monad_grad]: 2.62001e-06
        [auto_monad_eliminator]: 2.245e-05
        [cse]: 3.937e-05
        [replace_applicator]: 1.672e-05
    [Cycle 2]: 0.00047559, [27]
        [switch_simplify]: 8.72998e-06
        [loop_unroll]: 7.97e-06
        [a_1]: 0.00017481
        [with_stream_mark]: 1.153e-05
        [recompute_prepare]: 8.21002e-06
        [updatestate_depend_eliminate]: 4.88001e-06
        [updatestate_assign_eliminate]: 4.73001e-06
        [updatestate_loads_eliminate]: 3.95e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 8.03001e-06
        [updatestate_useless_node_eliminater]: 1.127e-05
        [accelerated_algorithm]: 7.70998e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 7.31999e-06
        [merge_forward]: 4.51002e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.617e-05
        [j_node_and_user_rematch]: 1.269e-05
        [meta_fg_expand]: 2.76e-06
        [replace_old_param]: 1.047e-05
        [inline_without_move]: 7.51999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 1.004e-05
        [cse]: 2.194e-05
        [replace_applicator]: 8.07e-06
[py_interpret_to_execute_after_opt_a]: 1.378e-05
[rewriter_after_opt_a]: 0.00029297
[convert_after_rewriter]: 1.246e-05
[order_py_execute_after_rewriter]: 7.38e-06
[mutable_eliminate]: 0.00062644
[jit_opt_b]: 7.01e-05, [1]
    [Cycle 1]: 6.194e-05, [2]
        [frontend_op_eliminate]: 2.363e-05
        [inline_after_opt_a]: 2.492e-05
[cconv]: 2.68e-05
[loop_unroll]: 0.00047645
[jit_opt_after_cconv]: 0.00019483, [1]
    [Cycle 1]: 0.00018781, [11]
        [c_1]: 4.408e-05
        [parameter_eliminate]: 2.57001e-06
        [updatestate_depend_eliminate]: 8.97e-06
        [updatestate_assign_eliminate]: 4.58001e-06
        [updatestate_loads_eliminate]: 4e-06
        [cse]: 3.098e-05
        [call_graph_tuple_transform]: 2.126e-05
        [tuple_list_get_item_eliminator]: 7.97e-06
        [none_parameter_eliminate]: 2.02001e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 8.31002e-06
[remove_dup_value]: 1.859e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 1.048e-05
[add_recomputation]: 6.693e-05
[cse_after_recomputation]: 3.044e-05, [1]
    [Cycle 1]: 2.434e-05, [1]
        [cse]: 1.838e-05
[auto_monad_reorder]: 2.534e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 5.15999e-06
[opt_after_jit_grad]: 0.00054386
[symbol_engine_optimizer]: 0.00010334, [1]
    [Cycle 1]: 9.546e-05, [6]
        [build]: 1.152e-05
        [elim_shapecalc]: 1.214e-05
        [elim_not_effective]: 1.891e-05
        [opt_reshape]: 8.96002e-06
        [fold_const_symbol]: 1.387e-05
        [renormalize]: 9.89996e-07
[validate]: 5.512e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.0106444
[execute]: 7.28e-06
Sums
    bootstrap                            :     0.000350s :  0.91%
    type_inference                       :     0.021989s : 57.02%
    event_method                         :     0.000021s :  0.06%
    auto_monad                           :     0.000148s :  0.38%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000116s :  0.30%
    rewriter_before_opt_a                :     0.000098s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.18%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000871s :  2.26%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  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 :     0.000036s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001326s :  3.44%
    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      :     0.000032s :  0.08%
    jit_opt_a.cse                        :     0.000061s :  0.16%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.04%
    rewriter_after_opt_a                 :     0.000293s :  0.76%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000626s :  1.62%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000027s :  0.07%
    loop_unroll                          :     0.000476s :  1.24%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000031s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000067s :  0.17%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000025s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000544s :  1.41%
    symbol_engine_optimizer.build        :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000055s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010644s : 27.60%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000269    47
 3.43% :     0.000009s :      2: substitution.depend_value_elim
 0.96% :     0.000003s :      4: substitution.elim_not_effective
 0.82% :     0.000002s :      4: substitution.fold_const_symbol
 2.45% :     0.000007s :      5: substitution.graph_param_transform
74.66% :     0.000201s :      4: substitution.inline
 1.66% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.57% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.74% :     0.000005s :      2: substitution.replace_old_param
 4.95% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.27% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.50% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021917     2
93.71% :     0.020538s :      1: type_inference.infer
 6.29% :     0.001379s :      1: type_inference.specialize
------[replace.]   0.000091     6
66.13% :     0.000060s :      4: replace.inline
33.87% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
94.23% :     0.000198s :      4: match.inline
 5.77% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000172  1021
 1.31% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     15: predicate.addn_check_dump
 1.29% :     0.000002s :     15: predicate.addn_zero_filter
 2.03% :     0.000003s :     15: predicate.arithmetic_simplify
 1.34% :     0.000002s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.26% :     0.000002s :     15: predicate.depend_value_elim
 1.12% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.93% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      5: predicate.elim_not_effective
 0.69% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     15: predicate.environ_get_eliminate
 1.10% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.35% :     0.000001s :      5: predicate.graph_param_transform
 5.31% :     0.000009s :     31: predicate.inline
 1.05% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :     10: predicate.less_batch_normalization
 1.40% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     22: predicate.load_eliminater
 1.62% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.53% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     15: predicate.merge_addn
 1.10% :     0.000002s :     15: predicate.minmaximum_grad
 1.44% :     0.000002s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.44% :     0.000004s :     22: predicate.partial_eliminate
 1.18% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.92% :     0.000003s :     15: predicate.reduce_eliminate
 1.45% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.78% :     0.000003s :     27: predicate.replace_applicator
 0.82% :     0.000001s :     10: predicate.replace_old_param
 0.39% :     0.000001s :      5: predicate.reset_defer_inline
 1.56% :     0.000003s :     15: predicate.reshape_eliminate
 1.31% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.00% :     0.000002s :     10: predicate.special_op_eliminate
 1.14% :     0.000002s :     10: predicate.specialize_transform
 1.36% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.58% :     0.000004s :     21: predicate.switch_defer_inline
 1.85% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.77% :     0.000012s :     68: predicate.switch_simplify
 1.29% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.50% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.07% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.31% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     15: predicate.value_based_eliminate
 0.47% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000773    13
43.43% :     0.000336s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.51% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.06% :     0.000395s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045302    76
 0.15% :     0.000070s :      1: add_recomputation
 0.34% :     0.000154s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.80% :     0.000364s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.06% :     0.000026s :      1: event_method
 0.02% :     0.000011s :      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
14.19% :     0.006427s :      1: jit_opt_a
 0.44% :     0.000198s :      1: jit_opt_after_cconv
 0.16% :     0.000073s :      1: jit_opt_b
 1.07% :     0.000486s :      1: loop_unroll
 1.41% :     0.000637s :      1: mutable_eliminate
 2.59% :     0.001173s :     26: opt.transform.jit_opt_a
 0.17% :     0.000078s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.22% :     0.000553s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000120s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.27% :     0.000574s :      1: renormalize.infer
 1.64% :     0.000742s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.66% :     0.000298s :      1: rewriter_after_opt_a
 0.23% :     0.000103s :      1: rewriter_before_opt_a
 0.23% :     0.000106s :      1: symbol_engine_optimizer
23.52% :     0.010656s :      1: task_emit
48.58% :     0.022006s :      1: type_inference
 0.18% :     0.000080s :      1: validate


TotalTime = 0.0750987, [33]
[bootstrap]: 0.00020748
[type_inference]: 0.0310441
[event_method]: 0.00015103
[auto_monad]: 0.00020042
[graph_reusing]: 7.43999e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.952e-05
[rewriter_before_opt_a]: 0.00013531
[expand_dump_flag]: 3.5e-06
[jit_opt_a]: 0.0323909, [3]
    [Cycle 1]: 0.02507, [27]
        [switch_simplify]: 8.056e-05
        [loop_unroll]: 4.958e-05
        [a_1]: 0.00121087
        [with_stream_mark]: 3.639e-05
        [recompute_prepare]: 2.847e-05
        [updatestate_depend_eliminate]: 1.227e-05
        [updatestate_assign_eliminate]: 1.088e-05
        [updatestate_loads_eliminate]: 1.054e-05
        [parameter_eliminate]: 2.54999e-06
        [specialize_transform]: 2.111e-05
        [updatestate_useless_node_eliminater]: 2.377e-05
        [accelerated_algorithm]: 1.88e-05
        [meta_shard_fg_expand]: 5.30001e-06
        [get_grad_eliminate_]: 1.927e-05
        [merge_forward]: 1.142e-05
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.919e-05
        [j_node_and_user_rematch]: 3.331e-05
        [meta_fg_expand]: 0.0062133
        [replace_old_param]: 0.00011766
        [inline_without_move]: 0.00011571
        [renormalize]: 0.0162221
        [add_forward_monad_depend]: 1.417e-05
        [auto_monad_grad]: 9.74e-06
        [auto_monad_eliminator]: 7.716e-05
        [cse]: 0.00027429
        [replace_applicator]: 0.00015718
    [Cycle 2]: 0.00282566, [27]
        [switch_simplify]: 6.096e-05
        [loop_unroll]: 5.807e-05
        [a_1]: 0.00097002
        [with_stream_mark]: 1.658e-05
        [recompute_prepare]: 1.385e-05
        [updatestate_depend_eliminate]: 2.589e-05
        [updatestate_assign_eliminate]: 5.98002e-06
        [updatestate_loads_eliminate]: 4.41002e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 9.70002e-06
        [updatestate_useless_node_eliminater]: 1.201e-05
        [accelerated_algorithm]: 9.40001e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 8.40999e-06
        [merge_forward]: 5.00999e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.79e-05
        [j_node_and_user_rematch]: 1.421e-05
        [meta_fg_expand]: 7.999e-05
        [replace_old_param]: 1.406e-05
        [inline_without_move]: 8.57998e-06
        [renormalize]: 0.0011904
        [add_forward_monad_depend]: 5.64e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.824e-05
        [cse]: 7.856e-05
        [replace_applicator]: 1.841e-05
    [Cycle 3]: 0.00048231, [27]
        [switch_simplify]: 1.021e-05
        [loop_unroll]: 8.55999e-06
        [a_1]: 0.0001733
        [with_stream_mark]: 1.284e-05
        [recompute_prepare]: 8.59e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 4.42e-06
        [updatestate_loads_eliminate]: 4.18999e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 8.42998e-06
        [updatestate_useless_node_eliminater]: 1.114e-05
        [accelerated_algorithm]: 7.7e-06
        [meta_shard_fg_expand]: 1.68002e-06
        [get_grad_eliminate_]: 7.48e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 1.95001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.648e-05
        [j_node_and_user_rematch]: 1.253e-05
        [meta_fg_expand]: 2.73998e-06
        [replace_old_param]: 1.103e-05
        [inline_without_move]: 7.5e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.44998e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 1.015e-05
        [cse]: 2.248e-05
        [replace_applicator]: 8.02e-06
[py_interpret_to_execute_after_opt_a]: 1.467e-05
[rewriter_after_opt_a]: 5.816e-05
[convert_after_rewriter]: 8.42998e-06
[order_py_execute_after_rewriter]: 6.63e-06
[mutable_eliminate]: 0.00074103
[jit_opt_b]: 6.891e-05, [1]
    [Cycle 1]: 6.123e-05, [2]
        [frontend_op_eliminate]: 2.505e-05
        [inline_after_opt_a]: 2.459e-05
[cconv]: 2.619e-05
[loop_unroll]: 0.00044858
[jit_opt_after_cconv]: 0.0002062, [1]
    [Cycle 1]: 0.0001992, [11]
        [c_1]: 4.777e-05
        [parameter_eliminate]: 3.35003e-06
        [updatestate_depend_eliminate]: 8.70001e-06
        [updatestate_assign_eliminate]: 4.45999e-06
        [updatestate_loads_eliminate]: 3.96001e-06
        [cse]: 3.55e-05
        [call_graph_tuple_transform]: 2.247e-05
        [tuple_list_get_item_eliminator]: 8.82999e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 7.88999e-06
[remove_dup_value]: 2.193e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 1.063e-05
[add_recomputation]: 6.771e-05
[cse_after_recomputation]: 3.299e-05, [1]
    [Cycle 1]: 2.696e-05, [1]
        [cse]: 2.039e-05
[auto_monad_reorder]: 2.546e-05
[get_jit_bprop_graph]: 1.86998e-06
[rewriter_after_jit_bprop_graph]: 5.19e-06
[opt_after_jit_grad]: 0.00052874
[symbol_engine_optimizer]: 0.00010353, [1]
    [Cycle 1]: 9.697e-05, [6]
        [build]: 1.013e-05
        [elim_shapecalc]: 1.216e-05
        [elim_not_effective]: 1.882e-05
        [opt_reshape]: 1.011e-05
        [fold_const_symbol]: 1.54e-05
        [renormalize]: 3.59985e-07
[validate]: 4.617e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00806952
[execute]: 7.3e-06
Sums
    bootstrap                            :     0.000207s :  0.30%
    type_inference                       :     0.031044s : 44.35%
    event_method                         :     0.000151s :  0.22%
    auto_monad                           :     0.000200s :  0.29%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000040s :  0.06%
    rewriter_before_opt_a                :     0.000135s :  0.19%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000152s :  0.22%
    jit_opt_a.loop_unroll                :     0.000116s :  0.17%
    jit_opt_a.a_1                        :     0.002354s :  3.36%
    jit_opt_a.with_stream_mark           :     0.000066s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000051s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000044s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000021s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000019s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.01%
    jit_opt_a.specialize_transform       :     0.000039s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000047s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000036s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000035s :  0.05%
    jit_opt_a.merge_forward              :     0.000020s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000074s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000060s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.006296s :  8.99%
    jit_opt_a.replace_old_param          :     0.000143s :  0.20%
    jit_opt_a.inline_without_move        :     0.000132s :  0.19%
    jit_opt_a.renormalize                :     0.017413s : 24.88%
    jit_opt_a.add_forward_monad_depend   :     0.000021s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000013s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000106s :  0.15%
    jit_opt_a.cse                        :     0.000375s :  0.54%
    jit_opt_a.replace_applicator         :     0.000184s :  0.26%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.02%
    rewriter_after_opt_a                 :     0.000058s :  0.08%
    convert_after_rewriter               :     0.000008s :  0.01%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000741s :  1.06%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.04%
    cconv                                :     0.000026s :  0.04%
    loop_unroll                          :     0.000449s :  0.64%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    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              :     0.000036s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  0.01%
    remove_dup_value                     :     0.000022s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000011s :  0.02%
    add_recomputation                    :     0.000068s :  0.10%
    cse_after_recomputation.cse          :     0.000020s :  0.03%
    auto_monad_reorder                   :     0.000025s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000529s :  0.76%
    symbol_engine_optimizer.build        :     0.000010s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000046s :  0.07%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008070s : 11.53%
    execute                              :     0.000007s :  0.01%


Time group info:
------[substitution.]   0.001551   160
 1.56% :     0.000024s :      8: substitution.depend_value_elim
 0.19% :     0.000003s :      4: substitution.elim_not_effective
 0.14% :     0.000002s :      4: substitution.fold_const_symbol
57.17% :     0.000887s :      4: substitution.getattr_setattr_resolve
 0.44% :     0.000007s :      5: substitution.graph_param_transform
24.57% :     0.000381s :     10: substitution.inline
 2.02% :     0.000031s :      4: substitution.inline_without_move
 0.71% :     0.000011s :     20: substitution.j_node_and_user_rematch
 0.53% :     0.000008s :      5: substitution.minmaximum_grad
 0.25% :     0.000004s :      3: substitution.partial_eliminate
 1.04% :     0.000016s :     20: substitution.remove_not_recompute_node
 2.87% :     0.000044s :     13: substitution.replace_applicator
 0.93% :     0.000014s :     18: substitution.replace_old_param
 0.23% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.27% :     0.000020s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.80% :     0.000012s :      5: substitution.tuple_list_get_item_depend_reorder
 2.37% :     0.000037s :     10: substitution.tuple_list_get_item_eliminator
 1.13% :     0.000018s :      8: substitution.updatestate_pure_node_eliminater
 1.76% :     0.000027s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.030956     2
93.98% :     0.029094s :      1: type_inference.infer
 6.02% :     0.001862s :      1: type_inference.specialize
------[replace.]   0.000290    20
17.63% :     0.000051s :      3: replace.getattr_setattr_resolve
33.75% :     0.000098s :     10: replace.inline
 9.57% :     0.000028s :      1: replace.replace_applicator
31.76% :     0.000092s :      5: replace.tuple_list_get_item_eliminator
 7.28% :     0.000021s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001247    20
67.19% :     0.000838s :      3: match.getattr_setattr_resolve
29.96% :     0.000374s :     10: match.inline
 0.75% :     0.000009s :      1: match.replace_applicator
 1.45% :     0.000018s :      5: match.tuple_list_get_item_eliminator
 0.64% :     0.000008s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000465  2979
 1.41% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.27% :     0.000006s :     46: predicate.addn_check_dump
 1.46% :     0.000007s :     46: predicate.addn_zero_filter
 2.58% :     0.000012s :     46: predicate.arithmetic_simplify
 1.70% :     0.000008s :     46: predicate.cast_eliminate
 0.20% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.23% :     0.000006s :     46: predicate.compare_switch_simplify
 1.54% :     0.000007s :     46: predicate.depend_value_elim
 1.30% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.40% :     0.000007s :     46: predicate.dict_get_item_eliminator
 1.34% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.12% :     0.000001s :      5: predicate.elim_not_effective
 0.24% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.26% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.21% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.35% :     0.000006s :     46: predicate.environ_get_eliminate
 1.23% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      5: predicate.fold_const_symbol
 1.02% :     0.000005s :     27: predicate.get_grad_eliminate
 1.45% :     0.000007s :     20: predicate.getattr_setattr_resolve
 0.14% :     0.000001s :      5: predicate.graph_param_transform
 3.95% :     0.000018s :     72: predicate.inline
 3.45% :     0.000016s :     91: predicate.inline_without_move
 0.42% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.18% :     0.000005s :     27: predicate.less_batch_normalization
 1.57% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.66% :     0.000008s :     56: predicate.load_eliminater
 0.43% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.70% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.72% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.56% :     0.000007s :     46: predicate.merge_addn
 1.49% :     0.000007s :     46: predicate.minmaximum_grad
 0.54% :     0.000003s :      5: predicate.mutable_eliminate
 0.22% :     0.000001s :      5: predicate.opt_reshape
 1.96% :     0.000009s :     56: predicate.partial_eliminate
 1.36% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.79% :     0.000008s :     46: predicate.reduce_eliminate
 1.57% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000002s :     27: predicate.remove_not_recompute_node
 2.74% :     0.000013s :    126: predicate.replace_applicator
 1.59% :     0.000007s :     91: predicate.replace_old_param
 0.11% :     0.000000s :      5: predicate.reset_defer_inline
 1.40% :     0.000007s :     46: predicate.reshape_eliminate
 1.37% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.36% :     0.000006s :     46: predicate.same_eliminate
 0.57% :     0.000003s :     29: predicate.set_cell_output_no_recompute
 0.43% :     0.000002s :     10: predicate.special_op_eliminate
 1.04% :     0.000005s :     27: predicate.specialize_transform
 1.55% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.75% :     0.000008s :     46: predicate.stack_unstack_eliminate
 0.22% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.28% :     0.000011s :     62: predicate.switch_defer_inline
 1.96% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.12% :     0.000028s :    188: predicate.switch_simplify
 1.38% :     0.000006s :     46: predicate.tile_eliminate
 1.36% :     0.000006s :     46: predicate.transpose_eliminate
 2.00% :     0.000009s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.62% :     0.000008s :     46: predicate.tuple_list_get_item_depend_reorder
 2.93% :     0.000014s :     61: predicate.tuple_list_get_item_eliminator
 2.14% :     0.000010s :     46: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000008s :     51: predicate.tuple_to_list_eliminator_
 1.76% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 3.11% :     0.000014s :     84: predicate.updatestate_useless_node_eliminater
 2.14% :     0.000010s :     46: predicate.value_based_eliminate
 0.16% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003276    48
66.40% :     0.002175s :     27: func_graph_cloner_run.FuncGraphClonerGraph
33.60% :     0.001101s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.097066    93
 0.07% :     0.000071s :      1: add_recomputation
 0.21% :     0.000209s :      1: auto_monad
 0.03% :     0.000028s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.23% :     0.000220s :      1: bootstrap
 0.03% :     0.000029s :      1: cconv
 0.01% :     0.000011s :      1: convert_after_rewriter
 0.04% :     0.000035s :      1: cse_after_recomputation
 0.01% :     0.000013s :      1: environ_conv
 0.16% :     0.000158s :      1: event_method
 0.01% :     0.000011s :      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
33.37% :     0.032395s :      1: jit_opt_a
 0.22% :     0.000209s :      1: jit_opt_after_cconv
 0.07% :     0.000071s :      1: jit_opt_b
 0.47% :     0.000457s :      1: loop_unroll
 0.77% :     0.000751s :      1: mutable_eliminate
 3.47% :     0.003371s :     39: opt.transform.jit_opt_a
 0.09% :     0.000083s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000043s :      4: opt.transform.jit_opt_b
 0.02% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000042s :      1: opt.transform.opt_after_jit_grad
 1.02% :     0.000994s :      2: opt.transform.opt_resolve
 0.05% :     0.000053s :      4: opt.transform.symbol_engine_opt
 0.55% :     0.000537s :      1: opt_after_jit_grad
 0.01% :     0.000009s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000238s :      1: py_interpret_to_execute
 0.02% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000025s :      1: remove_dup_value
14.56% :     0.014128s :      2: renormalize.infer
 3.36% :     0.003265s :      2: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000061s :      1: rewriter_after_opt_a
 0.14% :     0.000139s :      1: rewriter_before_opt_a
 0.11% :     0.000106s :      1: symbol_engine_optimizer
 8.33% :     0.008081s :      1: task_emit
32.00% :     0.031059s :      1: type_inference
 0.07% :     0.000070s :      1: validate


TotalTime = 0.0389111, [33]
[bootstrap]: 0.00023302
[type_inference]: 0.0120327
[event_method]: 4.556e-05
[auto_monad]: 9.577e-05
[graph_reusing]: 6.96999e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 2.807e-05
[rewriter_before_opt_a]: 0.00011233
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.0169822, [3]
    [Cycle 1]: 0.0108209, [27]
        [switch_simplify]: 7.644e-05
        [loop_unroll]: 4.614e-05
        [a_1]: 0.00095494
        [with_stream_mark]: 2.741e-05
        [recompute_prepare]: 2.141e-05
        [updatestate_depend_eliminate]: 9.41e-06
        [updatestate_assign_eliminate]: 7.13998e-06
        [updatestate_loads_eliminate]: 6.57002e-06
        [parameter_eliminate]: 2.96001e-06
        [specialize_transform]: 1.62e-05
        [updatestate_useless_node_eliminater]: 1.457e-05
        [accelerated_algorithm]: 1.5e-05
        [meta_shard_fg_expand]: 4.99e-06
        [get_grad_eliminate_]: 1.433e-05
        [merge_forward]: 8.94998e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.053e-05
        [j_node_and_user_rematch]: 2.403e-05
        [meta_fg_expand]: 0.00170882
        [replace_old_param]: 6.479e-05
        [inline_without_move]: 5.951e-05
        [renormalize]: 0.00712649
        [add_forward_monad_depend]: 1.646e-05
        [auto_monad_grad]: 6.06e-06
        [auto_monad_eliminator]: 5.557e-05
        [cse]: 0.00020113
        [replace_applicator]: 7.77e-05
    [Cycle 2]: 0.00224287, [27]
        [switch_simplify]: 4.281e-05
        [loop_unroll]: 3.993e-05
        [a_1]: 0.00111422
        [with_stream_mark]: 1.403e-05
        [recompute_prepare]: 8.15999e-06
        [updatestate_depend_eliminate]: 4.10998e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 3.08998e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 6.62002e-06
        [updatestate_useless_node_eliminater]: 6.60997e-06
        [accelerated_algorithm]: 6.39999e-06
        [meta_shard_fg_expand]: 1.68002e-06
        [get_grad_eliminate_]: 6.23002e-06
        [merge_forward]: 3.37002e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.396e-05
        [j_node_and_user_rematch]: 9.25001e-06
        [meta_fg_expand]: 4.326e-05
        [replace_old_param]: 1.208e-05
        [inline_without_move]: 6.33002e-06
        [renormalize]: 0.00068471
        [add_forward_monad_depend]: 5.00999e-06
        [auto_monad_grad]: 1.94e-06
        [auto_monad_eliminator]: 1.248e-05
        [cse]: 2.673e-05
        [replace_applicator]: 1.419e-05
    [Cycle 3]: 0.00036622, [27]
        [switch_simplify]: 7.53e-06
        [loop_unroll]: 6.63e-06
        [a_1]: 0.00011669
        [with_stream_mark]: 1.056e-05
        [recompute_prepare]: 6.64001e-06
        [updatestate_depend_eliminate]: 3.4e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 5.87999e-06
        [accelerated_algorithm]: 5.96998e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 5.74999e-06
        [merge_forward]: 2.84001e-06
        [cell_reuse_recompute_pass]: 1.77999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.491e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 9.14e-06
        [inline_without_move]: 5.74999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 6.96999e-06
        [cse]: 1.387e-05
        [replace_applicator]: 5.82001e-06
[py_interpret_to_execute_after_opt_a]: 1.195e-05
[rewriter_after_opt_a]: 3.56e-05
[convert_after_rewriter]: 6.39001e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00076957
[jit_opt_b]: 5.692e-05, [1]
    [Cycle 1]: 5.028e-05, [2]
        [frontend_op_eliminate]: 1.983e-05
        [inline_after_opt_a]: 1.912e-05
[cconv]: 2.422e-05
[loop_unroll]: 0.00044389
[jit_opt_after_cconv]: 0.00016162, [1]
    [Cycle 1]: 0.00015519, [11]
        [c_1]: 2.631e-05
        [parameter_eliminate]: 3.19001e-06
        [updatestate_depend_eliminate]: 6.68e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [cse]: 2.704e-05
        [call_graph_tuple_transform]: 2.297e-05
        [tuple_list_get_item_eliminator]: 6.31998e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 2.60014e-07
        [switch_simplify]: 6.73998e-06
[remove_dup_value]: 1.554e-05
[partial_unused_args_eliminate]: 1.96e-06
[environ_conv]: 5.84e-06
[add_recomputation]: 5.239e-05
[cse_after_recomputation]: 2.61e-05, [1]
    [Cycle 1]: 2.037e-05, [1]
        [cse]: 1.461e-05
[auto_monad_reorder]: 1.669e-05
[get_jit_bprop_graph]: 1.74998e-06
[rewriter_after_jit_bprop_graph]: 5.34e-06
[opt_after_jit_grad]: 0.00048218
[symbol_engine_optimizer]: 7.538e-05, [1]
    [Cycle 1]: 6.916e-05, [6]
        [build]: 2.96001e-06
        [elim_shapecalc]: 9.04e-06
        [elim_not_effective]: 1.413e-05
        [opt_reshape]: 6.77002e-06
        [fold_const_symbol]: 9.46e-06
        [renormalize]: 3.7998e-07
[validate]: 3.624e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00691466
[execute]: 7.31001e-06
Sums
    bootstrap                            :     0.000233s :  0.67%
    type_inference                       :     0.012033s : 34.83%
    event_method                         :     0.000046s :  0.13%
    auto_monad                           :     0.000096s :  0.28%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.08%
    rewriter_before_opt_a                :     0.000112s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000127s :  0.37%
    jit_opt_a.loop_unroll                :     0.000093s :  0.27%
    jit_opt_a.a_1                        :     0.002186s :  6.33%
    jit_opt_a.with_stream_mark           :     0.000052s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000036s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  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       :     0.000029s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.08%
    jit_opt_a.merge_forward              :     0.000015s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000059s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.001754s :  5.08%
    jit_opt_a.replace_old_param          :     0.000086s :  0.25%
    jit_opt_a.inline_without_move        :     0.000072s :  0.21%
    jit_opt_a.renormalize                :     0.007811s : 22.61%
    jit_opt_a.add_forward_monad_depend   :     0.000023s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000075s :  0.22%
    jit_opt_a.cse                        :     0.000242s :  0.70%
    jit_opt_a.replace_applicator         :     0.000098s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000036s :  0.10%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000770s :  2.23%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.06%
    cconv                                :     0.000024s :  0.07%
    loop_unroll                          :     0.000444s :  1.28%
    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 :     0.000007s :  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              :     0.000027s :  0.08%
    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      :     0.000000s :  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                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000052s :  0.15%
    cse_after_recomputation.cse          :     0.000015s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000482s :  1.40%
    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 :  0.04%
    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                             :     0.000036s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006915s : 20.01%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000558   113
 0.33% :     0.000002s :      2: substitution.elim_not_effective
 0.23% :     0.000001s :      2: substitution.fold_const_symbol
 1.07% :     0.000006s :      4: substitution.graph_param_transform
65.34% :     0.000364s :     12: substitution.inline
 3.31% :     0.000018s :      2: substitution.inline_without_move
 1.44% :     0.000008s :     12: substitution.j_node_and_user_rematch
 1.93% :     0.000011s :      7: substitution.minmaximum_grad
 1.35% :     0.000008s :      5: substitution.partial_eliminate
 1.99% :     0.000011s :     12: substitution.remove_not_recompute_node
 5.29% :     0.000029s :     10: substitution.replace_applicator
 1.97% :     0.000011s :     14: substitution.replace_old_param
 0.53% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.04% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.74% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.44% :     0.000047s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011951     2
89.83% :     0.010736s :      1: type_inference.infer
10.17% :     0.001215s :      1: type_inference.specialize
------[replace.]   0.000178    21
59.40% :     0.000106s :     12: replace.inline
40.60% :     0.000072s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000383    21
93.35% :     0.000358s :     12: match.inline
 6.65% :     0.000025s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000391  2828
 1.48% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.42% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.40% :     0.000005s :     47: predicate.addn_check_dump
 1.67% :     0.000007s :     47: predicate.addn_zero_filter
 1.89% :     0.000007s :     47: predicate.arithmetic_simplify
 1.72% :     0.000007s :     47: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.38% :     0.000005s :     47: predicate.compare_switch_simplify
 1.44% :     0.000006s :     47: 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
 1.46% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.35% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.18% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.52% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.39% :     0.000005s :     47: predicate.environ_get_add_eliminate
 1.42% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.58% :     0.000006s :     47: predicate.environ_get_eliminate
 1.39% :     0.000005s :     47: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 0.85% :     0.000003s :     21: predicate.get_grad_eliminate
 0.15% :     0.000001s :      4: predicate.graph_param_transform
 4.34% :     0.000017s :     76: predicate.inline
 1.97% :     0.000008s :     47: predicate.inline_without_move
 0.37% :     0.000001s :     21: predicate.j_node_and_user_rematch
 0.95% :     0.000004s :     21: predicate.less_batch_normalization
 1.76% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 2.00% :     0.000008s :     60: predicate.load_eliminater
 0.39% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.51% :     0.000014s :     97: predicate.loop_unroll_before_grad
 1.93% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.43% :     0.000006s :     47: predicate.merge_addn
 1.49% :     0.000006s :     47: predicate.minmaximum_grad
 0.58% :     0.000002s :      4: predicate.mutable_eliminate
 0.16% :     0.000001s :      4: predicate.opt_reshape
 2.48% :     0.000010s :     60: predicate.partial_eliminate
 1.50% :     0.000006s :     47: predicate.print_const_string_wrapper
 1.82% :     0.000007s :     47: predicate.reduce_eliminate
 1.89% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.75% :     0.000011s :    108: predicate.replace_applicator
 1.03% :     0.000004s :     47: predicate.replace_old_param
 0.14% :     0.000001s :      4: predicate.reset_defer_inline
 1.45% :     0.000006s :     47: predicate.reshape_eliminate
 1.51% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.57% :     0.000006s :     47: predicate.same_eliminate
 0.53% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.46% :     0.000002s :      8: predicate.special_op_eliminate
 0.83% :     0.000003s :     21: predicate.specialize_transform
 1.65% :     0.000006s :     47: predicate.split_environ_get_set_with_tuple_value
 1.56% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.25% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.65% :     0.000010s :     68: predicate.switch_defer_inline
 2.47% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.51% :     0.000026s :    169: predicate.switch_simplify
 1.48% :     0.000006s :     47: predicate.tile_eliminate
 1.41% :     0.000006s :     47: predicate.transpose_eliminate
 1.90% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.59% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000013s :     64: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.87% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000007s :     47: 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.]   0.001829    31
62.34% :     0.001140s :     15: func_graph_cloner_run.FuncGraphClonerGraph
37.66% :     0.000689s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049714    91
 0.11% :     0.000055s :      1: add_recomputation
 0.20% :     0.000100s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.51% :     0.000253s :      1: bootstrap
 0.05% :     0.000027s :      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
 0.10% :     0.000050s :      1: event_method
 0.02% :     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
34.17% :     0.016985s :      1: jit_opt_a
 0.33% :     0.000164s :      1: jit_opt_after_cconv
 0.12% :     0.000060s :      1: jit_opt_b
 0.91% :     0.000452s :      1: loop_unroll
 1.57% :     0.000781s :      1: mutable_eliminate
 5.76% :     0.002862s :     39: opt.transform.jit_opt_a
 0.12% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000032s :      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
 0.05% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000036s :      4: opt.transform.symbol_engine_opt
 0.99% :     0.000490s :      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
 0.06% :     0.000030s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
12.26% :     0.006096s :      2: renormalize.infer
 3.42% :     0.001700s :      2: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000039s :      1: rewriter_after_opt_a
 0.23% :     0.000115s :      1: rewriter_before_opt_a
 0.16% :     0.000078s :      1: symbol_engine_optimizer
13.94% :     0.006931s :      1: task_emit
24.24% :     0.012051s :      1: type_inference
 0.11% :     0.000054s :      1: validate

[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.597.491 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdc82cfb80> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.598.167 [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. 
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.620.644 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe940d0> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.621.308 [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 = 0.041671, [33]
[bootstrap]: 0.00028026
[type_inference]: 0.0223721
[event_method]: 2.055e-05
[auto_monad]: 0.00015432
[graph_reusing]: 7.27002e-06
[pre_auto_parallel]: 3.03e-06
[py_interpret_to_execute]: 0.00011131
[rewriter_before_opt_a]: 9.647e-05
[expand_dump_flag]: 3.10998e-06
[jit_opt_a]: 0.00621814, [2]
    [Cycle 1]: 0.00225504, [27]
        [switch_simplify]: 6.293e-05
        [loop_unroll]: 3.636e-05
        [a_1]: 0.00071752
        [with_stream_mark]: 2.133e-05
        [recompute_prepare]: 1.181e-05
        [updatestate_depend_eliminate]: 6.36e-06
        [updatestate_assign_eliminate]: 6.13998e-06
        [updatestate_loads_eliminate]: 4.92999e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 9.15999e-06
        [updatestate_useless_node_eliminater]: 1.222e-05
        [accelerated_algorithm]: 8.85001e-06
        [meta_shard_fg_expand]: 3.03e-06
        [get_grad_eliminate_]: 7.99002e-06
        [merge_forward]: 5.37999e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.988e-05
        [j_node_and_user_rematch]: 1.336e-05
        [meta_fg_expand]: 3.75998e-06
        [replace_old_param]: 1.081e-05
        [inline_without_move]: 7.87003e-06
        [renormalize]: 0.00102004
        [add_forward_monad_depend]: 6.28e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.999e-05
        [cse]: 3.695e-05
        [replace_applicator]: 1.832e-05
    [Cycle 2]: 0.00052609, [27]
        [switch_simplify]: 9.27001e-06
        [loop_unroll]: 7.8e-06
        [a_1]: 0.00020844
        [with_stream_mark]: 1.496e-05
        [recompute_prepare]: 9.68002e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 4.85001e-06
        [updatestate_loads_eliminate]: 3.82002e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 8.22998e-06
        [updatestate_useless_node_eliminater]: 1.083e-05
        [accelerated_algorithm]: 7.43e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 7.65e-06
        [merge_forward]: 4.57003e-06
        [cell_reuse_recompute_pass]: 1.96e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.676e-05
        [j_node_and_user_rematch]: 1.256e-05
        [meta_fg_expand]: 2.99001e-06
        [replace_old_param]: 9.89999e-06
        [inline_without_move]: 7.48e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.60999e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 1.066e-05
        [cse]: 2.384e-05
        [replace_applicator]: 8.36002e-06
[py_interpret_to_execute_after_opt_a]: 1.362e-05
[rewriter_after_opt_a]: 0.0002946
[convert_after_rewriter]: 1.225e-05
[order_py_execute_after_rewriter]: 7.71999e-06
[mutable_eliminate]: 0.00059852
[jit_opt_b]: 6.754e-05, [1]
    [Cycle 1]: 6.035e-05, [2]
        [frontend_op_eliminate]: 2.329e-05
        [inline_after_opt_a]: 2.528e-05
[cconv]: 2.609e-05
[loop_unroll]: 0.00044456
[jit_opt_after_cconv]: 0.00018847, [1]
    [Cycle 1]: 0.00018185, [11]
        [c_1]: 4.356e-05
        [parameter_eliminate]: 2.46e-06
        [updatestate_depend_eliminate]: 8.02e-06
        [updatestate_assign_eliminate]: 4.49002e-06
        [updatestate_loads_eliminate]: 3.83001e-06
        [cse]: 2.853e-05
        [call_graph_tuple_transform]: 2.195e-05
        [tuple_list_get_item_eliminator]: 8.38999e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 8.60001e-06
[remove_dup_value]: 1.903e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 1.041e-05
[add_recomputation]: 6.195e-05
[cse_after_recomputation]: 3.081e-05, [1]
    [Cycle 1]: 2.46e-05, [1]
        [cse]: 1.844e-05
[auto_monad_reorder]: 2.583e-05
[get_jit_bprop_graph]: 2.17999e-06
[rewriter_after_jit_bprop_graph]: 4.58001e-06
[opt_after_jit_grad]: 0.00047861
[symbol_engine_optimizer]: 9.86e-05, [1]
    [Cycle 1]: 9.222e-05, [6]
        [build]: 1.02e-05
        [elim_shapecalc]: 1.156e-05
        [elim_not_effective]: 1.897e-05
        [opt_reshape]: 9.04e-06
        [fold_const_symbol]: 1.356e-05
        [renormalize]: 4.19997e-07
[validate]: 5.038e-05
[backend_pass]: 1.04003e-06
[task_emit]: 0.00975115
[execute]: 6.28998e-06
Sums
    bootstrap                            :     0.000280s :  0.75%
    type_inference                       :     0.022372s : 59.50%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000154s :  0.41%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000111s :  0.30%
    rewriter_before_opt_a                :     0.000096s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.19%
    jit_opt_a.loop_unroll                :     0.000044s :  0.12%
    jit_opt_a.a_1                        :     0.000926s :  2.46%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001020s :  2.71%
    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      :     0.000031s :  0.08%
    jit_opt_a.cse                        :     0.000061s :  0.16%
    jit_opt_a.replace_applicator         :     0.000027s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.04%
    rewriter_after_opt_a                 :     0.000295s :  0.78%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000599s :  1.59%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.07%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000445s :  1.18%
    jit_opt_after_cconv.c_1              :     0.000044s :  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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000062s :  0.16%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000479s :  1.27%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000050s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009751s : 25.93%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000271    47
 3.21% :     0.000009s :      2: substitution.depend_value_elim
 1.09% :     0.000003s :      4: substitution.elim_not_effective
 0.75% :     0.000002s :      4: substitution.fold_const_symbol
 2.37% :     0.000006s :      5: substitution.graph_param_transform
75.09% :     0.000203s :      4: substitution.inline
 1.77% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.62% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.46% :     0.000004s :      2: substitution.replace_old_param
 4.85% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.10% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.71% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022301     2
93.60% :     0.020873s :      1: type_inference.infer
 6.40% :     0.001428s :      1: type_inference.specialize
------[replace.]   0.000089     6
65.97% :     0.000059s :      4: replace.inline
34.03% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.35% :     0.000200s :      4: match.inline
 5.65% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000187  1021
 1.29% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.06% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.03% :     0.000002s :     15: predicate.addn_check_dump
 1.29% :     0.000002s :     15: predicate.addn_zero_filter
 1.91% :     0.000004s :     15: predicate.arithmetic_simplify
 1.21% :     0.000002s :     15: predicate.cast_eliminate
 0.43% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     15: predicate.compare_switch_simplify
 1.24% :     0.000002s :     15: predicate.depend_value_elim
 1.10% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.07% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.31% :     0.000001s :      5: predicate.elim_not_effective
 0.52% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.01% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.13% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      5: predicate.fold_const_symbol
 1.15% :     0.000002s :     10: predicate.get_grad_eliminate
 0.36% :     0.000001s :      5: predicate.graph_param_transform
 5.03% :     0.000009s :     31: predicate.inline
 0.94% :     0.000002s :     10: predicate.inline_without_move
 0.39% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :     10: predicate.less_batch_normalization
 1.22% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.73% :     0.000003s :     22: predicate.load_eliminater
 1.07% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.17% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.92% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000002s :     15: predicate.merge_addn
 1.08% :     0.000002s :     15: predicate.minmaximum_grad
 1.22% :     0.000002s :      5: predicate.mutable_eliminate
 0.48% :     0.000001s :      5: predicate.opt_reshape
 2.39% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.65% :     0.000003s :     15: predicate.reduce_eliminate
 9.21% :     0.000017s :     17: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.69% :     0.000003s :     27: predicate.replace_applicator
 0.55% :     0.000001s :     10: predicate.replace_old_param
 0.28% :     0.000001s :      5: predicate.reset_defer_inline
 1.24% :     0.000002s :     15: predicate.reshape_eliminate
 1.25% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.16% :     0.000002s :     15: predicate.same_eliminate
 0.52% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.92% :     0.000002s :     10: predicate.special_op_eliminate
 1.03% :     0.000002s :     10: predicate.specialize_transform
 1.45% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.56% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.99% :     0.000004s :     21: predicate.switch_defer_inline
 1.89% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.04% :     0.000011s :     68: predicate.switch_simplify
 1.15% :     0.000002s :     15: predicate.tile_eliminate
 1.18% :     0.000002s :     15: predicate.transpose_eliminate
 2.09% :     0.000004s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.25% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.67% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000003s :     15: predicate.value_based_eliminate
 0.39% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000785    13
44.67% :     0.000351s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.12% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.21% :     0.000394s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044109    76
 0.15% :     0.000065s :      1: add_recomputation
 0.36% :     0.000161s :      1: auto_monad
 0.07% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000293s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.08% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.06% :     0.000025s :      1: event_method
 0.02% :     0.000010s :      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
14.10% :     0.006222s :      1: jit_opt_a
 0.43% :     0.000191s :      1: jit_opt_after_cconv
 0.16% :     0.000070s :      1: jit_opt_b
 1.03% :     0.000452s :      1: loop_unroll
 1.38% :     0.000609s :      1: mutable_eliminate
 2.79% :     0.001231s :     26: opt.transform.jit_opt_a
 0.18% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.10% :     0.000487s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000115s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.37% :     0.000605s :      1: renormalize.infer
 0.92% :     0.000406s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.68% :     0.000300s :      1: rewriter_after_opt_a
 0.23% :     0.000101s :      1: rewriter_before_opt_a
 0.23% :     0.000101s :      1: symbol_engine_optimizer
22.13% :     0.009761s :      1: task_emit
50.76% :     0.022389s :      1: type_inference
 0.16% :     0.000073s :      1: validate


TotalTime = 0.0363304, [33]
[bootstrap]: 0.00019954
[type_inference]: 0.0100131
[event_method]: 4.284e-05
[auto_monad]: 7.759e-05
[graph_reusing]: 5.63002e-06
[pre_auto_parallel]: 2.27999e-06
[py_interpret_to_execute]: 2.443e-05
[rewriter_before_opt_a]: 0.00010479
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.016535, [3]
    [Cycle 1]: 0.010727, [27]
        [switch_simplify]: 6.776e-05
        [loop_unroll]: 4.738e-05
        [a_1]: 0.00092427
        [with_stream_mark]: 2.588e-05
        [recompute_prepare]: 2.101e-05
        [updatestate_depend_eliminate]: 8.61002e-06
        [updatestate_assign_eliminate]: 6.34001e-06
        [updatestate_loads_eliminate]: 5.72999e-06
        [parameter_eliminate]: 2.59001e-06
        [specialize_transform]: 1.648e-05
        [updatestate_useless_node_eliminater]: 1.443e-05
        [accelerated_algorithm]: 1.491e-05
        [meta_shard_fg_expand]: 3.41999e-06
        [get_grad_eliminate_]: 1.367e-05
        [merge_forward]: 8.08999e-06
        [cell_reuse_recompute_pass]: 9.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.724e-05
        [j_node_and_user_rematch]: 2.376e-05
        [meta_fg_expand]: 0.00164095
        [replace_old_param]: 6.71e-05
        [inline_without_move]: 5.92e-05
        [renormalize]: 0.00715634
        [add_forward_monad_depend]: 1.205e-05
        [auto_monad_grad]: 6.24999e-06
        [auto_monad_eliminator]: 5.416e-05
        [cse]: 0.00020354
        [replace_applicator]: 7.802e-05
    [Cycle 2]: 0.00221377, [27]
        [switch_simplify]: 4.148e-05
        [loop_unroll]: 4.013e-05
        [a_1]: 0.00111889
        [with_stream_mark]: 1.438e-05
        [recompute_prepare]: 8.35999e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 2.65002e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 7.05e-06
        [updatestate_useless_node_eliminater]: 6.19001e-06
        [accelerated_algorithm]: 6.83998e-06
        [meta_shard_fg_expand]: 1.79e-06
        [get_grad_eliminate_]: 6.64999e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.326e-05
        [j_node_and_user_rematch]: 9.60001e-06
        [meta_fg_expand]: 4.663e-05
        [replace_old_param]: 1.133e-05
        [inline_without_move]: 6.17999e-06
        [renormalize]: 0.00064873
        [add_forward_monad_depend]: 5.47001e-06
        [auto_monad_grad]: 1.77999e-06
        [auto_monad_eliminator]: 1.212e-05
        [cse]: 2.62e-05
        [replace_applicator]: 1.391e-05
    [Cycle 3]: 0.00035937, [27]
        [switch_simplify]: 6.51999e-06
        [loop_unroll]: 6.06e-06
        [a_1]: 0.0001146
        [with_stream_mark]: 1.074e-05
        [recompute_prepare]: 5.74e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 5.76998e-06
        [accelerated_algorithm]: 6.02001e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 5.62999e-06
        [merge_forward]: 2.84999e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.528e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 9.44e-06
        [inline_without_move]: 5.70001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.96e-06
        [cse]: 1.377e-05
        [replace_applicator]: 5.91003e-06
[py_interpret_to_execute_after_opt_a]: 1.052e-05
[rewriter_after_opt_a]: 3.469e-05
[convert_after_rewriter]: 6.51e-06
[order_py_execute_after_rewriter]: 5.07999e-06
[mutable_eliminate]: 0.00063665
[jit_opt_b]: 0.00013508, [1]
    [Cycle 1]: 0.00012864, [2]
        [frontend_op_eliminate]: 1.821e-05
        [inline_after_opt_a]: 2.027e-05
[cconv]: 1.974e-05
[loop_unroll]: 0.00045589
[jit_opt_after_cconv]: 0.00016747, [1]
    [Cycle 1]: 0.00016134, [11]
        [c_1]: 2.752e-05
        [parameter_eliminate]: 3.33e-06
        [updatestate_depend_eliminate]: 6.59999e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.38998e-06
        [cse]: 2.907e-05
        [call_graph_tuple_transform]: 2.411e-05
        [tuple_list_get_item_eliminator]: 6.68998e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 6.49999e-06
[remove_dup_value]: 1.416e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.55001e-06
[add_recomputation]: 4.113e-05
[cse_after_recomputation]: 2.487e-05, [1]
    [Cycle 1]: 1.973e-05, [1]
        [cse]: 1.399e-05
[auto_monad_reorder]: 1.473e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 6.73e-06
[opt_after_jit_grad]: 0.00049122
[symbol_engine_optimizer]: 7.806e-05, [1]
    [Cycle 1]: 7.155e-05, [6]
        [build]: 3.48999e-06
        [elim_shapecalc]: 9.25999e-06
        [elim_not_effective]: 1.465e-05
        [opt_reshape]: 6.82002e-06
        [fold_const_symbol]: 9.64999e-06
        [renormalize]: 4.39992e-07
[validate]: 3.727e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00692684
[execute]: 6.34999e-06
Sums
    bootstrap                            :     0.000200s :  0.62%
    type_inference                       :     0.010013s : 31.08%
    event_method                         :     0.000043s :  0.13%
    auto_monad                           :     0.000078s :  0.24%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.08%
    rewriter_before_opt_a                :     0.000105s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000116s :  0.36%
    jit_opt_a.loop_unroll                :     0.000094s :  0.29%
    jit_opt_a.a_1                        :     0.002158s :  6.70%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000035s :  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 :     0.000010s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000029s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.08%
    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_        :     0.000026s :  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 :     0.000056s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.001689s :  5.24%
    jit_opt_a.replace_old_param          :     0.000088s :  0.27%
    jit_opt_a.inline_without_move        :     0.000071s :  0.22%
    jit_opt_a.renormalize                :     0.007805s : 24.22%
    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      :     0.000072s :  0.22%
    jit_opt_a.cse                        :     0.000244s :  0.76%
    jit_opt_a.replace_applicator         :     0.000098s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000035s :  0.11%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000637s :  1.98%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.06%
    cconv                                :     0.000020s :  0.06%
    loop_unroll                          :     0.000456s :  1.41%
    jit_opt_after_cconv.c_1              :     0.000028s :  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 :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.09%
    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 :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000041s :  0.13%
    cse_after_recomputation.cse          :     0.000014s :  0.04%
    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                   :     0.000491s :  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 :     0.000015s :  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  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006927s : 21.50%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000532   113
 0.32% :     0.000002s :      2: substitution.elim_not_effective
 0.24% :     0.000001s :      2: substitution.fold_const_symbol
 1.13% :     0.000006s :      4: substitution.graph_param_transform
65.17% :     0.000347s :     12: substitution.inline
 3.56% :     0.000019s :      2: substitution.inline_without_move
 1.52% :     0.000008s :     12: substitution.j_node_and_user_rematch
 2.15% :     0.000011s :      7: substitution.minmaximum_grad
 1.16% :     0.000006s :      5: substitution.partial_eliminate
 1.98% :     0.000011s :     12: substitution.remove_not_recompute_node
 5.44% :     0.000029s :     10: substitution.replace_applicator
 1.97% :     0.000010s :     14: substitution.replace_old_param
 0.48% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.99% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.87% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.01% :     0.000043s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009946     2
88.74% :     0.008826s :      1: type_inference.infer
11.26% :     0.001120s :      1: type_inference.specialize
------[replace.]   0.000177    21
58.57% :     0.000104s :     12: replace.inline
41.43% :     0.000073s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000361    21
94.09% :     0.000340s :     12: match.inline
 5.91% :     0.000021s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000388  2828
 1.67% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.38% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.38% :     0.000005s :     47: predicate.addn_check_dump
 1.60% :     0.000006s :     47: predicate.addn_zero_filter
 1.94% :     0.000008s :     47: predicate.arithmetic_simplify
 1.57% :     0.000006s :     47: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.37% :     0.000005s :     47: predicate.compare_switch_simplify
 1.42% :     0.000005s :     47: predicate.depend_value_elim
 1.41% :     0.000005s :     47: predicate.dict_get_item_const_eliminator
 1.45% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.56% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.32% :     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
 1.47% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.41% :     0.000005s :     47: predicate.environ_get_add_eliminate
 1.38% :     0.000005s :     47: predicate.environ_get_depend_swap
 1.44% :     0.000006s :     47: predicate.environ_get_eliminate
 1.39% :     0.000005s :     47: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000003s :     21: predicate.get_grad_eliminate
 0.18% :     0.000001s :      4: predicate.graph_param_transform
 4.36% :     0.000017s :     76: predicate.inline
 1.94% :     0.000008s :     47: predicate.inline_without_move
 0.39% :     0.000002s :     21: predicate.j_node_and_user_rematch
 1.09% :     0.000004s :     21: predicate.less_batch_normalization
 1.86% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 1.95% :     0.000008s :     60: predicate.load_eliminater
 0.48% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.39% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.80% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.38% :     0.000005s :     47: predicate.merge_addn
 1.48% :     0.000006s :     47: predicate.minmaximum_grad
 0.57% :     0.000002s :      4: predicate.mutable_eliminate
 0.19% :     0.000001s :      4: predicate.opt_reshape
 2.47% :     0.000010s :     60: predicate.partial_eliminate
 1.40% :     0.000005s :     47: predicate.print_const_string_wrapper
 1.84% :     0.000007s :     47: predicate.reduce_eliminate
 1.81% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.76% :     0.000011s :    108: predicate.replace_applicator
 1.04% :     0.000004s :     47: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.64% :     0.000006s :     47: predicate.reshape_eliminate
 1.56% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.26% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.52% :     0.000006s :     47: predicate.same_eliminate
 0.49% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.87% :     0.000003s :     21: predicate.specialize_transform
 1.63% :     0.000006s :     47: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.24% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.78% :     0.000011s :     68: predicate.switch_defer_inline
 2.47% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.31% :     0.000024s :    169: predicate.switch_simplify
 1.44% :     0.000006s :     47: predicate.tile_eliminate
 1.47% :     0.000006s :     47: predicate.transpose_eliminate
 2.05% :     0.000008s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.96% :     0.000008s :     47: predicate.tuple_list_get_item_depend_reorder
 3.22% :     0.000013s :     64: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.92% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000007s :     47: 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.]   0.001716    31
61.89% :     0.001062s :     15: func_graph_cloner_run.FuncGraphClonerGraph
38.11% :     0.000654s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047085    91
 0.09% :     0.000045s :      1: add_recomputation
 0.17% :     0.000082s :      1: auto_monad
 0.04% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.45% :     0.000214s :      1: bootstrap
 0.05% :     0.000023s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.10% :     0.000047s :      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% :     0.000008s :      1: graph_reusing
35.12% :     0.016538s :      1: jit_opt_a
 0.36% :     0.000170s :      1: jit_opt_after_cconv
 0.29% :     0.000138s :      1: jit_opt_b
 0.99% :     0.000464s :      1: loop_unroll
 1.37% :     0.000645s :      1: mutable_eliminate
 6.00% :     0.002823s :     39: opt.transform.jit_opt_a
 0.13% :     0.000061s :      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% :     0.000017s :      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
 1.06% :     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
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.06% :     0.000027s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
12.95% :     0.006095s :      2: renormalize.infer
 3.60% :     0.001693s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000037s :      1: rewriter_after_opt_a
 0.23% :     0.000108s :      1: rewriter_before_opt_a
 0.17% :     0.000081s :      1: symbol_engine_optimizer
14.74% :     0.006939s :      1: task_emit
21.30% :     0.010028s :      1: type_inference
 0.12% :     0.000056s :      1: validate

[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.667.048 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe944c0> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.667.668 [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 = 0.0716224, [33]
[bootstrap]: 0.00021928
[type_inference]: 0.0301571
[event_method]: 0.00014211
[auto_monad]: 0.00020165
[graph_reusing]: 7.93999e-06
[pre_auto_parallel]: 3.25e-06
[py_interpret_to_execute]: 3.411e-05
[rewriter_before_opt_a]: 0.00014001
[expand_dump_flag]: 3.98999e-06
[jit_opt_a]: 0.0295499, [3]
    [Cycle 1]: 0.0219475, [27]
        [switch_simplify]: 7.613e-05
        [loop_unroll]: 4.997e-05
        [a_1]: 0.00114822
        [with_stream_mark]: 3.337e-05
        [recompute_prepare]: 2.713e-05
        [updatestate_depend_eliminate]: 1.302e-05
        [updatestate_assign_eliminate]: 1.148e-05
        [updatestate_loads_eliminate]: 1.003e-05
        [parameter_eliminate]: 3.06999e-06
        [specialize_transform]: 2.023e-05
        [updatestate_useless_node_eliminater]: 2.431e-05
        [accelerated_algorithm]: 1.962e-05
        [meta_shard_fg_expand]: 5.33002e-06
        [get_grad_eliminate_]: 1.844e-05
        [merge_forward]: 1.208e-05
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.918e-05
        [j_node_and_user_rematch]: 3.705e-05
        [meta_fg_expand]: 0.00703026
        [replace_old_param]: 0.00011531
        [inline_without_move]: 0.00011663
        [renormalize]: 0.0121713
        [add_forward_monad_depend]: 2.078e-05
        [auto_monad_grad]: 1.533e-05
        [auto_monad_eliminator]: 9.257e-05
        [cse]: 0.00034988
        [replace_applicator]: 0.00020736
    [Cycle 2]: 0.00323244, [27]
        [switch_simplify]: 6.181e-05
        [loop_unroll]: 5.744e-05
        [a_1]: 0.00102267
        [with_stream_mark]: 2.344e-05
        [recompute_prepare]: 1.378e-05
        [updatestate_depend_eliminate]: 2.682e-05
        [updatestate_assign_eliminate]: 6.52001e-06
        [updatestate_loads_eliminate]: 4.65999e-06
        [parameter_eliminate]: 2.29001e-06
        [specialize_transform]: 1.031e-05
        [updatestate_useless_node_eliminater]: 1.175e-05
        [accelerated_algorithm]: 9.17999e-06
        [meta_shard_fg_expand]: 2.99999e-06
        [get_grad_eliminate_]: 8.27e-06
        [merge_forward]: 6.55002e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.07e-05
        [j_node_and_user_rematch]: 1.394e-05
        [meta_fg_expand]: 0.00010493
        [replace_old_param]: 1.351e-05
        [inline_without_move]: 8.38999e-06
        [renormalize]: 0.00152088
        [add_forward_monad_depend]: 5.02e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.807e-05
        [cse]: 6.578e-05
        [replace_applicator]: 1.819e-05
    [Cycle 3]: 0.00064218, [27]
        [switch_simplify]: 1.007e-05
        [loop_unroll]: 8.32e-06
        [a_1]: 0.00017362
        [with_stream_mark]: 1.412e-05
        [recompute_prepare]: 8.48999e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 5.00999e-06
        [updatestate_loads_eliminate]: 4.32003e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 8.21002e-06
        [updatestate_useless_node_eliminater]: 1.168e-05
        [accelerated_algorithm]: 7.95e-06
        [meta_shard_fg_expand]: 2.40002e-06
        [get_grad_eliminate_]: 7.78001e-06
        [merge_forward]: 4.20999e-06
        [cell_reuse_recompute_pass]: 2.38998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.788e-05
        [j_node_and_user_rematch]: 1.349e-05
        [meta_fg_expand]: 3.78001e-06
        [replace_old_param]: 1.117e-05
        [inline_without_move]: 7.8e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 9.10019e-07
        [auto_monad_eliminator]: 1.051e-05
        [cse]: 3.488e-05
        [replace_applicator]: 1.477e-05
[py_interpret_to_execute_after_opt_a]: 1.97e-05
[rewriter_after_opt_a]: 6.932e-05
[convert_after_rewriter]: 1.057e-05
[order_py_execute_after_rewriter]: 6.57002e-06
[mutable_eliminate]: 0.00087244
[jit_opt_b]: 7.01e-05, [1]
    [Cycle 1]: 6.241e-05, [2]
        [frontend_op_eliminate]: 2.536e-05
        [inline_after_opt_a]: 2.521e-05
[cconv]: 2.703e-05
[loop_unroll]: 0.0004413
[jit_opt_after_cconv]: 0.00020819, [1]
    [Cycle 1]: 0.00020167, [11]
        [c_1]: 4.451e-05
        [parameter_eliminate]: 2.25002e-06
        [updatestate_depend_eliminate]: 7.85e-06
        [updatestate_assign_eliminate]: 4.35999e-06
        [updatestate_loads_eliminate]: 4.15999e-06
        [cse]: 3.481e-05
        [call_graph_tuple_transform]: 2.773e-05
        [tuple_list_get_item_eliminator]: 9.05001e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 9.36e-06
[remove_dup_value]: 2.649e-05
[partial_unused_args_eliminate]: 2.52001e-06
[environ_conv]: 1.088e-05
[add_recomputation]: 7.299e-05
[cse_after_recomputation]: 3.543e-05, [1]
    [Cycle 1]: 2.915e-05, [1]
        [cse]: 2.265e-05
[auto_monad_reorder]: 2.897e-05
[get_jit_bprop_graph]: 2.31e-06
[rewriter_after_jit_bprop_graph]: 8.25e-06
[opt_after_jit_grad]: 0.00057755
[symbol_engine_optimizer]: 0.00011686, [1]
    [Cycle 1]: 0.00010995, [6]
        [build]: 1.536e-05
        [elim_shapecalc]: 1.316e-05
        [elim_not_effective]: 2.055e-05
        [opt_reshape]: 1.009e-05
        [fold_const_symbol]: 1.592e-05
        [renormalize]: 6.59988e-07
[validate]: 5.617e-05
[backend_pass]: 1.49e-06
[task_emit]: 0.00824808
[execute]: 8.42e-06
Sums
    bootstrap                            :     0.000219s :  0.33%
    type_inference                       :     0.030157s : 45.13%
    event_method                         :     0.000142s :  0.21%
    auto_monad                           :     0.000202s :  0.30%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000034s :  0.05%
    rewriter_before_opt_a                :     0.000140s :  0.21%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000148s :  0.22%
    jit_opt_a.loop_unroll                :     0.000116s :  0.17%
    jit_opt_a.a_1                        :     0.002345s :  3.51%
    jit_opt_a.with_stream_mark           :     0.000071s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000049s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000045s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000023s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000019s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000007s :  0.01%
    jit_opt_a.specialize_transform       :     0.000039s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000048s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000037s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000011s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000034s :  0.05%
    jit_opt_a.merge_forward              :     0.000023s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000078s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000064s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.007139s : 10.68%
    jit_opt_a.replace_old_param          :     0.000140s :  0.21%
    jit_opt_a.inline_without_move        :     0.000133s :  0.20%
    jit_opt_a.renormalize                :     0.013692s : 20.49%
    jit_opt_a.add_forward_monad_depend   :     0.000027s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000019s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000121s :  0.18%
    jit_opt_a.cse                        :     0.000451s :  0.67%
    jit_opt_a.replace_applicator         :     0.000240s :  0.36%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.03%
    rewriter_after_opt_a                 :     0.000069s :  0.10%
    convert_after_rewriter               :     0.000011s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000872s :  1.31%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.04%
    cconv                                :     0.000027s :  0.04%
    loop_unroll                          :     0.000441s :  0.66%
    jit_opt_after_cconv.c_1              :     0.000045s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.01%
    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 :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.01%
    remove_dup_value                     :     0.000026s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000011s :  0.02%
    add_recomputation                    :     0.000073s :  0.11%
    cse_after_recomputation.cse          :     0.000023s :  0.03%
    auto_monad_reorder                   :     0.000029s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.01%
    opt_after_jit_grad                   :     0.000578s :  0.86%
    symbol_engine_optimizer.build        :     0.000015s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000056s :  0.08%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008248s : 12.34%
    execute                              :     0.000008s :  0.01%


Time group info:
------[substitution.]   0.001576   159
 1.38% :     0.000022s :      8: substitution.depend_value_elim
 0.18% :     0.000003s :      4: substitution.elim_not_effective
 0.16% :     0.000002s :      4: substitution.fold_const_symbol
55.56% :     0.000876s :      4: substitution.getattr_setattr_resolve
 0.52% :     0.000008s :      5: substitution.graph_param_transform
25.25% :     0.000398s :     10: substitution.inline
 2.03% :     0.000032s :      4: substitution.inline_without_move
 0.83% :     0.000013s :     20: substitution.j_node_and_user_rematch
 0.61% :     0.000010s :      5: substitution.minmaximum_grad
 0.27% :     0.000004s :      3: substitution.partial_eliminate
 0.97% :     0.000015s :     20: substitution.remove_not_recompute_node
 3.53% :     0.000056s :     13: substitution.replace_applicator
 0.83% :     0.000013s :     17: substitution.replace_old_param
 0.22% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.37% :     0.000022s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.84% :     0.000013s :      5: substitution.tuple_list_get_item_depend_reorder
 2.42% :     0.000038s :     10: substitution.tuple_list_get_item_eliminator
 1.12% :     0.000018s :      8: substitution.updatestate_pure_node_eliminater
 1.91% :     0.000030s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.030074     2
94.46% :     0.028408s :      1: type_inference.infer
 5.54% :     0.001666s :      1: type_inference.specialize
------[replace.]   0.000299    20
16.30% :     0.000049s :      3: replace.getattr_setattr_resolve
30.10% :     0.000090s :     10: replace.inline
16.39% :     0.000049s :      1: replace.replace_applicator
29.54% :     0.000088s :      5: replace.tuple_list_get_item_eliminator
 7.67% :     0.000023s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001260    20
65.67% :     0.000827s :      3: match.getattr_setattr_resolve
31.07% :     0.000391s :     10: match.inline
 1.01% :     0.000013s :      1: match.replace_applicator
 1.45% :     0.000018s :      5: match.tuple_list_get_item_eliminator
 0.80% :     0.000010s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000459  2975
 1.48% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.22% :     0.000006s :     46: predicate.addn_check_dump
 1.48% :     0.000007s :     46: predicate.addn_zero_filter
 2.17% :     0.000010s :     46: predicate.arithmetic_simplify
 1.53% :     0.000007s :     46: predicate.cast_eliminate
 0.24% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.30% :     0.000006s :     46: predicate.compare_switch_simplify
 1.41% :     0.000006s :     46: predicate.depend_value_elim
 1.30% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.52% :     0.000007s :     46: predicate.dict_get_item_eliminator
 1.36% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.34% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      5: predicate.elim_not_effective
 0.28% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.36% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.30% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.38% :     0.000006s :     46: predicate.environ_get_eliminate
 1.27% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 1.05% :     0.000005s :     27: predicate.get_grad_eliminate
 1.35% :     0.000006s :     20: predicate.getattr_setattr_resolve
 0.15% :     0.000001s :      5: predicate.graph_param_transform
 4.06% :     0.000019s :     72: predicate.inline
 3.24% :     0.000015s :     89: predicate.inline_without_move
 0.43% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.19% :     0.000005s :     27: predicate.less_batch_normalization
 1.66% :     0.000008s :     51: predicate.list_to_tuple_eliminator_
 1.75% :     0.000008s :     56: predicate.load_eliminater
 0.42% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.65% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.79% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000006s :     46: predicate.merge_addn
 1.47% :     0.000007s :     46: predicate.minmaximum_grad
 0.48% :     0.000002s :      5: predicate.mutable_eliminate
 0.23% :     0.000001s :      5: predicate.opt_reshape
 2.10% :     0.000010s :     56: predicate.partial_eliminate
 1.38% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.94% :     0.000009s :     46: predicate.reduce_eliminate
 1.66% :     0.000008s :     51: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000002s :     27: predicate.remove_not_recompute_node
 2.80% :     0.000013s :    126: predicate.replace_applicator
 1.60% :     0.000007s :     89: predicate.replace_old_param
 0.12% :     0.000001s :      5: predicate.reset_defer_inline
 1.55% :     0.000007s :     46: predicate.reshape_eliminate
 1.38% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.34% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.40% :     0.000006s :     46: predicate.same_eliminate
 0.54% :     0.000002s :     29: predicate.set_cell_output_no_recompute
 0.50% :     0.000002s :     10: predicate.special_op_eliminate
 1.09% :     0.000005s :     27: predicate.specialize_transform
 1.66% :     0.000008s :     46: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.23% :     0.000010s :     62: predicate.switch_defer_inline
 2.00% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.30% :     0.000029s :    188: predicate.switch_simplify
 1.40% :     0.000006s :     46: predicate.tile_eliminate
 1.46% :     0.000007s :     46: predicate.transpose_eliminate
 1.76% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.75% :     0.000008s :     46: predicate.tuple_list_get_item_depend_reorder
 3.27% :     0.000015s :     61: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000008s :     46: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000008s :     51: predicate.tuple_to_list_eliminator_
 1.68% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000013s :     84: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000009s :     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.]   0.003036    48
66.57% :     0.002021s :     27: func_graph_cloner_run.FuncGraphClonerGraph
33.43% :     0.001015s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.089884    93
 0.08% :     0.000076s :      1: add_recomputation
 0.23% :     0.000211s :      1: auto_monad
 0.04% :     0.000032s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.26% :     0.000234s :      1: bootstrap
 0.03% :     0.000030s :      1: cconv
 0.01% :     0.000013s :      1: convert_after_rewriter
 0.04% :     0.000038s :      1: cse_after_recomputation
 0.01% :     0.000013s :      1: environ_conv
 0.17% :     0.000150s :      1: event_method
 0.02% :     0.000014s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000011s :      1: graph_reusing
32.88% :     0.029555s :      1: jit_opt_a
 0.23% :     0.000211s :      1: jit_opt_after_cconv
 0.08% :     0.000073s :      1: jit_opt_b
 0.50% :     0.000450s :      1: loop_unroll
 0.98% :     0.000882s :      1: mutable_eliminate
 3.80% :     0.003415s :     39: opt.transform.jit_opt_a
 0.10% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000043s :      4: opt.transform.jit_opt_b
 0.02% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 1.09% :     0.000975s :      2: opt.transform.opt_resolve
 0.06% :     0.000056s :      4: opt.transform.symbol_engine_opt
 0.65% :     0.000587s :      1: opt_after_jit_grad
 0.01% :     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
 0.04% :     0.000037s :      1: py_interpret_to_execute
 0.03% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000029s :      1: remove_dup_value
11.32% :     0.010174s :      2: renormalize.infer
 3.89% :     0.003497s :      2: renormalize.specialize
 0.01% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000073s :      1: rewriter_after_opt_a
 0.16% :     0.000144s :      1: rewriter_before_opt_a
 0.13% :     0.000120s :      1: symbol_engine_optimizer
 9.20% :     0.008265s :      1: task_emit
33.57% :     0.030173s :      1: type_inference
 0.09% :     0.000085s :      1: validate

[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.679.769 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdc8479dc0> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.680.448 [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 = 0.0410653, [33]
[bootstrap]: 0.00029548
[type_inference]: 0.0220174
[event_method]: 1.915e-05
[auto_monad]: 0.00014569
[graph_reusing]: 4.87e-06
[pre_auto_parallel]: 2.19001e-06
[py_interpret_to_execute]: 0.00015915
[rewriter_before_opt_a]: 0.00010121
[expand_dump_flag]: 2.93e-06
[jit_opt_a]: 0.00580246, [2]
    [Cycle 1]: 0.00215215, [27]
        [switch_simplify]: 5.343e-05
        [loop_unroll]: 3.563e-05
        [a_1]: 0.00073861
        [with_stream_mark]: 1.607e-05
        [recompute_prepare]: 1.1e-05
        [updatestate_depend_eliminate]: 5.69e-06
        [updatestate_assign_eliminate]: 5.18002e-06
        [updatestate_loads_eliminate]: 4.52998e-06
        [parameter_eliminate]: 1.62999e-06
        [specialize_transform]: 9.20999e-06
        [updatestate_useless_node_eliminater]: 1.118e-05
        [accelerated_algorithm]: 8.40999e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 8.20999e-06
        [merge_forward]: 4.25999e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.771e-05
        [j_node_and_user_rematch]: 1.338e-05
        [meta_fg_expand]: 3.03e-06
        [replace_old_param]: 1.07e-05
        [inline_without_move]: 8.3e-06
        [renormalize]: 0.00094722
        [add_forward_monad_depend]: 5.14e-06
        [auto_monad_grad]: 1.29998e-06
        [auto_monad_eliminator]: 1.778e-05
        [cse]: 2.681e-05
        [replace_applicator]: 1.596e-05
    [Cycle 2]: 0.00045234, [27]
        [switch_simplify]: 8.97e-06
        [loop_unroll]: 7.82e-06
        [a_1]: 0.00016228
        [with_stream_mark]: 1.122e-05
        [recompute_prepare]: 8.09002e-06
        [updatestate_depend_eliminate]: 4.85001e-06
        [updatestate_assign_eliminate]: 4.22e-06
        [updatestate_loads_eliminate]: 3.75e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 8.15999e-06
        [updatestate_useless_node_eliminater]: 1.029e-05
        [accelerated_algorithm]: 7.41999e-06
        [meta_shard_fg_expand]: 1.73002e-06
        [get_grad_eliminate_]: 7.23e-06
        [merge_forward]: 3.94002e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.567e-05
        [j_node_and_user_rematch]: 1.247e-05
        [meta_fg_expand]: 2.72001e-06
        [replace_old_param]: 9.67999e-06
        [inline_without_move]: 7.58999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 9.79e-06
        [cse]: 1.98e-05
        [replace_applicator]: 7.60998e-06
[py_interpret_to_execute_after_opt_a]: 1.083e-05
[rewriter_after_opt_a]: 0.00027794
[convert_after_rewriter]: 1.007e-05
[order_py_execute_after_rewriter]: 5.70001e-06
[mutable_eliminate]: 0.00058694
[jit_opt_b]: 6.642e-05, [1]
    [Cycle 1]: 6.025e-05, [2]
        [frontend_op_eliminate]: 2.393e-05
        [inline_after_opt_a]: 2.454e-05
[cconv]: 1.759e-05
[loop_unroll]: 0.00043683
[jit_opt_after_cconv]: 0.00018539, [1]
    [Cycle 1]: 0.00017861, [11]
        [c_1]: 4.364e-05
        [parameter_eliminate]: 2.72001e-06
        [updatestate_depend_eliminate]: 8.09002e-06
        [updatestate_assign_eliminate]: 4.23999e-06
        [updatestate_loads_eliminate]: 4.71002e-06
        [cse]: 2.622e-05
        [call_graph_tuple_transform]: 1.929e-05
        [tuple_list_get_item_eliminator]: 8.20999e-06
        [none_parameter_eliminate]: 9.10019e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 8.35001e-06
[remove_dup_value]: 1.226e-05
[partial_unused_args_eliminate]: 1.44e-06
[environ_conv]: 9.09e-06
[add_recomputation]: 5.156e-05
[cse_after_recomputation]: 2.919e-05, [1]
    [Cycle 1]: 2.352e-05, [1]
        [cse]: 1.748e-05
[auto_monad_reorder]: 1.948e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 4.60999e-06
[opt_after_jit_grad]: 0.00056522
[symbol_engine_optimizer]: 9.784e-05, [1]
    [Cycle 1]: 9.131e-05, [6]
        [build]: 9.63002e-06
        [elim_shapecalc]: 1.164e-05
        [elim_not_effective]: 1.819e-05
        [opt_reshape]: 8.63001e-06
        [fold_const_symbol]: 1.32e-05
        [renormalize]: 6.09987e-07
[validate]: 4.452e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00986805
[execute]: 5.83997e-06
Sums
    bootstrap                            :     0.000295s :  0.79%
    type_inference                       :     0.022017s : 59.10%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000146s :  0.39%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000159s :  0.43%
    rewriter_before_opt_a                :     0.000101s :  0.27%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000062s :  0.17%
    jit_opt_a.loop_unroll                :     0.000043s :  0.12%
    jit_opt_a.a_1                        :     0.000901s :  2.42%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000021s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000947s :  2.54%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.07%
    jit_opt_a.cse                        :     0.000047s :  0.13%
    jit_opt_a.replace_applicator         :     0.000024s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000278s :  0.75%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000587s :  1.58%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.07%
    cconv                                :     0.000018s :  0.05%
    loop_unroll                          :     0.000437s :  1.17%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.12%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000012s :  0.03%
    partial_unused_args_eliminate        :     0.000001s :  0.00%
    environ_conv                         :     0.000009s :  0.02%
    add_recomputation                    :     0.000052s :  0.14%
    cse_after_recomputation.cse          :     0.000017s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000565s :  1.52%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009868s : 26.49%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000223    47
 3.53% :     0.000008s :      2: substitution.depend_value_elim
 1.07% :     0.000002s :      4: substitution.elim_not_effective
 0.75% :     0.000002s :      4: substitution.fold_const_symbol
 2.12% :     0.000005s :      5: substitution.graph_param_transform
73.92% :     0.000165s :      4: substitution.inline
 2.01% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.93% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.46% :     0.000003s :      2: substitution.replace_old_param
 4.86% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
 3.80% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.55% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021956     2
93.91% :     0.020619s :      1: type_inference.infer
 6.09% :     0.001337s :      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.]   0.000172     6
94.38% :     0.000162s :      4: match.inline
 5.62% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1021
 1.27% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.27% :     0.000002s :     15: predicate.addn_zero_filter
 2.06% :     0.000003s :     15: predicate.arithmetic_simplify
 1.43% :     0.000002s :     15: predicate.cast_eliminate
 0.61% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     15: predicate.compare_switch_simplify
 1.42% :     0.000002s :     15: predicate.depend_value_elim
 1.22% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.56% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.25% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     15: predicate.environ_get_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.05% :     0.000002s :     10: predicate.get_grad_eliminate
 0.30% :     0.000001s :      5: predicate.graph_param_transform
 5.12% :     0.000009s :     31: predicate.inline
 1.10% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :     10: predicate.less_batch_normalization
 1.41% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.99% :     0.000003s :     22: predicate.load_eliminater
 1.34% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.58% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     15: predicate.merge_addn
 1.21% :     0.000002s :     15: predicate.minmaximum_grad
 1.43% :     0.000002s :      5: predicate.mutable_eliminate
 0.52% :     0.000001s :      5: predicate.opt_reshape
 2.35% :     0.000004s :     22: predicate.partial_eliminate
 1.28% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.93% :     0.000003s :     15: predicate.reduce_eliminate
 1.52% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.83% :     0.000003s :     27: predicate.replace_applicator
 0.58% :     0.000001s :     10: predicate.replace_old_param
 0.37% :     0.000001s :      5: predicate.reset_defer_inline
 1.24% :     0.000002s :     15: predicate.reshape_eliminate
 1.58% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     15: predicate.same_eliminate
 0.71% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.05% :     0.000002s :     10: predicate.special_op_eliminate
 1.14% :     0.000002s :     10: predicate.specialize_transform
 1.40% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.14% :     0.000004s :     21: predicate.switch_defer_inline
 1.96% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.51% :     0.000011s :     68: predicate.switch_simplify
 1.31% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000002s :     15: predicate.transpose_eliminate
 1.51% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.99% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.31% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000003s :     15: predicate.value_based_eliminate
 0.47% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000876    13
49.46% :     0.000433s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.21% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.33% :     0.000406s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043383    76
 0.13% :     0.000055s :      1: add_recomputation
 0.35% :     0.000152s :      1: auto_monad
 0.05% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.71% :     0.000310s :      1: bootstrap
 0.05% :     0.000020s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.05% :     0.000023s :      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% :     0.000008s :      1: graph_reusing
13.38% :     0.005805s :      1: jit_opt_a
 0.43% :     0.000188s :      1: jit_opt_after_cconv
 0.16% :     0.000069s :      1: jit_opt_b
 1.02% :     0.000444s :      1: loop_unroll
 1.37% :     0.000596s :      1: mutable_eliminate
 2.74% :     0.001189s :     26: opt.transform.jit_opt_a
 0.18% :     0.000076s :      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% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000048s :      4: opt.transform.symbol_engine_opt
 1.32% :     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
 0.38% :     0.000165s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000015s :      1: remove_dup_value
 1.27% :     0.000549s :      1: renormalize.infer
 0.90% :     0.000390s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.65% :     0.000282s :      1: rewriter_after_opt_a
 0.24% :     0.000106s :      1: rewriter_before_opt_a
 0.23% :     0.000101s :      1: symbol_engine_optimizer
22.77% :     0.009878s :      1: task_emit
50.78% :     0.022031s :      1: type_inference
 0.15% :     0.000065s :      1: validate


TotalTime = 0.0403988, [33]
[bootstrap]: 0.0001884
[type_inference]: 0.0149933
[event_method]: 5.003e-05
[auto_monad]: 9.559e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.38002e-06
[py_interpret_to_execute]: 2.498e-05
[rewriter_before_opt_a]: 0.00010427
[expand_dump_flag]: 2.48e-06
[jit_opt_a]: 0.0157804, [3]
    [Cycle 1]: 0.00998094, [27]
        [switch_simplify]: 6.039e-05
        [loop_unroll]: 4.369e-05
        [a_1]: 0.00094607
        [with_stream_mark]: 2.247e-05
        [recompute_prepare]: 2.055e-05
        [updatestate_depend_eliminate]: 7.80998e-06
        [updatestate_assign_eliminate]: 6.63998e-06
        [updatestate_loads_eliminate]: 6.12001e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 1.473e-05
        [updatestate_useless_node_eliminater]: 1.375e-05
        [accelerated_algorithm]: 1.472e-05
        [meta_shard_fg_expand]: 3.79002e-06
        [get_grad_eliminate_]: 1.352e-05
        [merge_forward]: 7.5e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.558e-05
        [j_node_and_user_rematch]: 2.37e-05
        [meta_fg_expand]: 0.00153392
        [replace_old_param]: 6.559e-05
        [inline_without_move]: 5.891e-05
        [renormalize]: 0.00655647
        [add_forward_monad_depend]: 1.135e-05
        [auto_monad_grad]: 6.38e-06
        [auto_monad_eliminator]: 5.188e-05
        [cse]: 0.00018424
        [replace_applicator]: 7.709e-05
    [Cycle 2]: 0.00216665, [27]
        [switch_simplify]: 4.176e-05
        [loop_unroll]: 4.061e-05
        [a_1]: 0.00115921
        [with_stream_mark]: 1.377e-05
        [recompute_prepare]: 8.25e-06
        [updatestate_depend_eliminate]: 3.98999e-06
        [updatestate_assign_eliminate]: 3.86001e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 7.16001e-06
        [updatestate_useless_node_eliminater]: 6.02001e-06
        [accelerated_algorithm]: 5.99999e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 6.68e-06
        [merge_forward]: 3.35e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.374e-05
        [j_node_and_user_rematch]: 9.14e-06
        [meta_fg_expand]: 3.641e-05
        [replace_old_param]: 1.123e-05
        [inline_without_move]: 5.66e-06
        [renormalize]: 0.00058205
        [add_forward_monad_depend]: 4.42e-06
        [auto_monad_grad]: 1.43002e-06
        [auto_monad_eliminator]: 1.159e-05
        [cse]: 2.202e-05
        [replace_applicator]: 1.281e-05
    [Cycle 3]: 0.00035715, [27]
        [switch_simplify]: 6.63998e-06
        [loop_unroll]: 6.07001e-06
        [a_1]: 0.00011385
        [with_stream_mark]: 9.94001e-06
        [recompute_prepare]: 5.81e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 6.34999e-06
        [updatestate_useless_node_eliminater]: 5.95002e-06
        [accelerated_algorithm]: 5.90002e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 2.68998e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.314e-05
        [j_node_and_user_rematch]: 9.12001e-06
        [meta_fg_expand]: 1.88002e-06
        [replace_old_param]: 9.52001e-06
        [inline_without_move]: 5.67001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.92001e-06
        [cse]: 1.326e-05
        [replace_applicator]: 6.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.215e-05
[rewriter_after_opt_a]: 4.112e-05
[convert_after_rewriter]: 7.33e-06
[order_py_execute_after_rewriter]: 6.12001e-06
[mutable_eliminate]: 0.00060789
[jit_opt_b]: 5.674e-05, [1]
    [Cycle 1]: 4.949e-05, [2]
        [frontend_op_eliminate]: 1.974e-05
        [inline_after_opt_a]: 1.806e-05
[cconv]: 2.363e-05
[loop_unroll]: 0.00043393
[jit_opt_after_cconv]: 0.0001575, [1]
    [Cycle 1]: 0.00015122, [11]
        [c_1]: 2.568e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 6.29999e-06
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 2.501e-05
        [call_graph_tuple_transform]: 2.176e-05
        [tuple_list_get_item_eliminator]: 6.49001e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 6.29999e-06
[remove_dup_value]: 2.265e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 6.46999e-06
[add_recomputation]: 4.689e-05
[cse_after_recomputation]: 2.633e-05, [1]
    [Cycle 1]: 2.078e-05, [1]
        [cse]: 1.458e-05
[auto_monad_reorder]: 1.824e-05
[get_jit_bprop_graph]: 1.84998e-06
[rewriter_after_jit_bprop_graph]: 5.39998e-06
[opt_after_jit_grad]: 0.00047423
[symbol_engine_optimizer]: 7.657e-05, [1]
    [Cycle 1]: 6.992e-05, [6]
        [build]: 2.99001e-06
        [elim_shapecalc]: 9.14e-06
        [elim_not_effective]: 1.396e-05
        [opt_reshape]: 6.65002e-06
        [fold_const_symbol]: 9.10001e-06
        [renormalize]: 4.30009e-07
[validate]: 8.402e-05
[backend_pass]: 1.04003e-06
[task_emit]: 0.00681624
[execute]: 6.92002e-06
Sums
    bootstrap                            :     0.000188s :  0.52%
    type_inference                       :     0.014993s : 41.27%
    event_method                         :     0.000050s :  0.14%
    auto_monad                           :     0.000096s :  0.26%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.07%
    rewriter_before_opt_a                :     0.000104s :  0.29%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000109s :  0.30%
    jit_opt_a.loop_unroll                :     0.000090s :  0.25%
    jit_opt_a.a_1                        :     0.002219s :  6.11%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.04%
    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        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000028s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  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_        :     0.000026s :  0.07%
    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 :     0.000052s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.001572s :  4.33%
    jit_opt_a.replace_old_param          :     0.000086s :  0.24%
    jit_opt_a.inline_without_move        :     0.000070s :  0.19%
    jit_opt_a.renormalize                :     0.007139s : 19.65%
    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      :     0.000069s :  0.19%
    jit_opt_a.cse                        :     0.000220s :  0.60%
    jit_opt_a.replace_applicator         :     0.000096s :  0.26%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000041s :  0.11%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000608s :  1.67%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.05%
    cconv                                :     0.000024s :  0.07%
    loop_unroll                          :     0.000434s :  1.19%
    jit_opt_after_cconv.c_1              :     0.000026s :  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              :     0.000025s :  0.07%
    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                     :     0.000023s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000047s :  0.13%
    cse_after_recomputation.cse          :     0.000015s :  0.04%
    auto_monad_reorder                   :     0.000018s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000474s :  1.31%
    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 :  0.04%
    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                             :     0.000084s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006816s : 18.76%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000610   113
 0.33% :     0.000002s :      2: substitution.elim_not_effective
 0.20% :     0.000001s :      2: substitution.fold_const_symbol
 0.97% :     0.000006s :      4: substitution.graph_param_transform
69.90% :     0.000426s :     12: substitution.inline
 2.79% :     0.000017s :      2: substitution.inline_without_move
 1.31% :     0.000008s :     12: substitution.j_node_and_user_rematch
 1.81% :     0.000011s :      7: substitution.minmaximum_grad
 1.09% :     0.000007s :      5: substitution.partial_eliminate
 1.53% :     0.000009s :     12: substitution.remove_not_recompute_node
 4.66% :     0.000028s :     10: substitution.replace_applicator
 1.70% :     0.000010s :     14: substitution.replace_old_param
 0.51% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.57% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.48% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.16% :     0.000044s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.014923     2
91.30% :     0.013624s :      1: type_inference.infer
 8.70% :     0.001299s :      1: type_inference.specialize
------[replace.]   0.000168    21
58.78% :     0.000099s :     12: replace.inline
41.22% :     0.000069s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000442    21
94.94% :     0.000420s :     12: match.inline
 5.06% :     0.000022s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000379  2828
 1.46% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.46% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.41% :     0.000005s :     47: predicate.addn_check_dump
 1.51% :     0.000006s :     47: predicate.addn_zero_filter
 1.95% :     0.000007s :     47: predicate.arithmetic_simplify
 1.58% :     0.000006s :     47: predicate.cast_eliminate
 0.22% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.39% :     0.000005s :     47: predicate.compare_switch_simplify
 1.43% :     0.000005s :     47: predicate.depend_value_elim
 1.43% :     0.000005s :     47: predicate.dict_get_item_const_eliminator
 1.55% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.47% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.34% :     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
 1.44% :     0.000005s :     47: predicate.environ_add_const_eliminate
 1.41% :     0.000005s :     47: predicate.environ_get_add_eliminate
 1.44% :     0.000005s :     47: predicate.environ_get_depend_swap
 1.66% :     0.000006s :     47: predicate.environ_get_eliminate
 1.43% :     0.000005s :     47: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.83% :     0.000003s :     21: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.41% :     0.000017s :     76: predicate.inline
 1.96% :     0.000007s :     47: predicate.inline_without_move
 0.38% :     0.000001s :     21: predicate.j_node_and_user_rematch
 0.95% :     0.000004s :     21: predicate.less_batch_normalization
 1.82% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 2.01% :     0.000008s :     60: predicate.load_eliminater
 0.60% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.33% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.79% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.42% :     0.000005s :     47: predicate.merge_addn
 1.46% :     0.000006s :     47: predicate.minmaximum_grad
 0.51% :     0.000002s :      4: predicate.mutable_eliminate
 0.17% :     0.000001s :      4: predicate.opt_reshape
 2.51% :     0.000010s :     60: predicate.partial_eliminate
 1.45% :     0.000006s :     47: predicate.print_const_string_wrapper
 1.81% :     0.000007s :     47: predicate.reduce_eliminate
 1.85% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.87% :     0.000011s :    108: predicate.replace_applicator
 1.06% :     0.000004s :     47: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.46% :     0.000006s :     47: predicate.reshape_eliminate
 1.47% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.25% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.49% :     0.000006s :     47: predicate.same_eliminate
 0.50% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.86% :     0.000003s :     21: predicate.specialize_transform
 1.67% :     0.000006s :     47: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.26% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.69% :     0.000010s :     68: predicate.switch_defer_inline
 2.57% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.47% :     0.000025s :    169: predicate.switch_simplify
 1.51% :     0.000006s :     47: predicate.tile_eliminate
 1.48% :     0.000006s :     47: predicate.transpose_eliminate
 1.95% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.71% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.28% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.93% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000007s :     47: predicate.value_based_eliminate
 0.13% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.22% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001705    31
61.29% :     0.001045s :     15: func_graph_cloner_run.FuncGraphClonerGraph
38.71% :     0.000660s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050530    91
 0.10% :     0.000050s :      1: add_recomputation
 0.20% :     0.000099s :      1: auto_monad
 0.04% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.41% :     0.000208s :      1: bootstrap
 0.05% :     0.000026s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000028s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.11% :     0.000054s :      1: event_method
 0.02% :     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
31.24% :     0.015784s :      1: jit_opt_a
 0.32% :     0.000160s :      1: jit_opt_after_cconv
 0.12% :     0.000059s :      1: jit_opt_b
 0.87% :     0.000441s :      1: loop_unroll
 1.22% :     0.000617s :      1: mutable_eliminate
 5.67% :     0.002864s :     39: opt.transform.jit_opt_a
 0.11% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000032s :      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
 0.05% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000036s :      4: opt.transform.symbol_engine_opt
 0.95% :     0.000482s :      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% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000025s :      1: remove_dup_value
10.95% :     0.005533s :      2: renormalize.infer
 3.15% :     0.001591s :      2: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000044s :      1: rewriter_after_opt_a
 0.21% :     0.000107s :      1: rewriter_before_opt_a
 0.16% :     0.000079s :      1: symbol_engine_optimizer
13.51% :     0.006826s :      1: task_emit
29.75% :     0.015032s :      1: type_inference
 0.20% :     0.000103s :      1: validate

[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.716.134 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe94940> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.716.743 [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 = 0.039911, [33]
[bootstrap]: 0.0002874
[type_inference]: 0.0211616
[event_method]: 1.959e-05
[auto_monad]: 0.00012894
[graph_reusing]: 4.08001e-06
[pre_auto_parallel]: 2.19001e-06
[py_interpret_to_execute]: 0.00011076
[rewriter_before_opt_a]: 0.00011978
[expand_dump_flag]: 2.66999e-06
[jit_opt_a]: 0.00558798, [2]
    [Cycle 1]: 0.00205262, [27]
        [switch_simplify]: 5.335e-05
        [loop_unroll]: 3.617e-05
        [a_1]: 0.00069379
        [with_stream_mark]: 1.513e-05
        [recompute_prepare]: 1.04e-05
        [updatestate_depend_eliminate]: 5.38002e-06
        [updatestate_assign_eliminate]: 5.04e-06
        [updatestate_loads_eliminate]: 4.08999e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 9.09e-06
        [updatestate_useless_node_eliminater]: 1.153e-05
        [accelerated_algorithm]: 8.77e-06
        [meta_shard_fg_expand]: 1.98002e-06
        [get_grad_eliminate_]: 8.77e-06
        [merge_forward]: 4.47998e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.698e-05
        [j_node_and_user_rematch]: 1.33e-05
        [meta_fg_expand]: 3.12002e-06
        [replace_old_param]: 1.052e-05
        [inline_without_move]: 8.06001e-06
        [renormalize]: 0.00089918
        [add_forward_monad_depend]: 5.12999e-06
        [auto_monad_grad]: 1.55999e-06
        [auto_monad_eliminator]: 1.758e-05
        [cse]: 2.555e-05
        [replace_applicator]: 1.444e-05
    [Cycle 2]: 0.0004598, [27]
        [switch_simplify]: 9.17001e-06
        [loop_unroll]: 7.92e-06
        [a_1]: 0.00016607
        [with_stream_mark]: 1.119e-05
        [recompute_prepare]: 8.45001e-06
        [updatestate_depend_eliminate]: 5.12999e-06
        [updatestate_assign_eliminate]: 4.65001e-06
        [updatestate_loads_eliminate]: 3.88001e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 8.37e-06
        [updatestate_useless_node_eliminater]: 1.043e-05
        [accelerated_algorithm]: 7.52002e-06
        [meta_shard_fg_expand]: 1.67001e-06
        [get_grad_eliminate_]: 7.25e-06
        [merge_forward]: 4.11001e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.604e-05
        [j_node_and_user_rematch]: 1.266e-05
        [meta_fg_expand]: 2.89999e-06
        [replace_old_param]: 9.52999e-06
        [inline_without_move]: 7.41999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 8.99978e-07
        [auto_monad_eliminator]: 9.36e-06
        [cse]: 1.943e-05
        [replace_applicator]: 7.77e-06
[py_interpret_to_execute_after_opt_a]: 1.103e-05
[rewriter_after_opt_a]: 0.00027771
[convert_after_rewriter]: 1.118e-05
[order_py_execute_after_rewriter]: 6.83998e-06
[mutable_eliminate]: 0.00053641
[jit_opt_b]: 6.628e-05, [1]
    [Cycle 1]: 5.921e-05, [2]
        [frontend_op_eliminate]: 2.345e-05
        [inline_after_opt_a]: 2.369e-05
[cconv]: 2.16e-05
[loop_unroll]: 0.00043506
[jit_opt_after_cconv]: 0.00018695, [1]
    [Cycle 1]: 0.0001799, [11]
        [c_1]: 4.411e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 8.59e-06
        [updatestate_assign_eliminate]: 4.48001e-06
        [updatestate_loads_eliminate]: 4.02e-06
        [cse]: 2.493e-05
        [call_graph_tuple_transform]: 2.158e-05
        [tuple_list_get_item_eliminator]: 8.77999e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 7.94002e-06
[remove_dup_value]: 1.861e-05
[partial_unused_args_eliminate]: 2.13998e-06
[environ_conv]: 1.027e-05
[add_recomputation]: 6.21e-05
[cse_after_recomputation]: 2.899e-05, [1]
    [Cycle 1]: 2.287e-05, [1]
        [cse]: 1.716e-05
[auto_monad_reorder]: 2.646e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 4.90999e-06
[opt_after_jit_grad]: 0.00055631
[symbol_engine_optimizer]: 0.00010898, [1]
    [Cycle 1]: 0.000102, [6]
        [build]: 1.023e-05
        [elim_shapecalc]: 1.302e-05
        [elim_not_effective]: 2.128e-05
        [opt_reshape]: 9.51998e-06
        [fold_const_symbol]: 1.474e-05
        [renormalize]: 4.30009e-07
[validate]: 5.527e-05
[backend_pass]: 1.48002e-06
[task_emit]: 0.00983786
[execute]: 7.29001e-06
Sums
    bootstrap                            :     0.000287s :  0.79%
    type_inference                       :     0.021162s : 58.43%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000129s :  0.36%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000111s :  0.31%
    rewriter_before_opt_a                :     0.000120s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000063s :  0.17%
    jit_opt_a.loop_unroll                :     0.000044s :  0.12%
    jit_opt_a.a_1                        :     0.000860s :  2.37%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.000899s :  2.48%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.07%
    jit_opt_a.cse                        :     0.000045s :  0.12%
    jit_opt_a.replace_applicator         :     0.000022s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000278s :  0.77%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000536s :  1.48%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000022s :  0.06%
    loop_unroll                          :     0.000435s :  1.20%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000025s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000062s :  0.17%
    cse_after_recomputation.cse          :     0.000017s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000556s :  1.54%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000055s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009838s : 27.16%
    execute                              :     0.000007s :  0.02%


TotalTime = 0.0672657, [33]
[bootstrap]: 0.00021428
[type_inference]: 0.0295392
[event_method]: 0.00014128
[auto_monad]: 0.00020075
[graph_reusing]: 7.66999e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 3.857e-05
[rewriter_before_opt_a]: 0.00012625
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.0272111, [3]
    [Cycle 1]: 0.0204624, [27]
        [switch_simplify]: 7.54e-05
        [loop_unroll]: 4.996e-05
        [a_1]: 0.00112048
        [with_stream_mark]: 3.317e-05
        [recompute_prepare]: 2.618e-05
        [updatestate_depend_eliminate]: 1.247e-05
        [updatestate_assign_eliminate]: 1.116e-05
        [updatestate_loads_eliminate]: 9.72001e-06
        [parameter_eliminate]: 3.20998e-06
        [specialize_transform]: 1.983e-05
        [updatestate_useless_node_eliminater]: 2.41e-05
        [accelerated_algorithm]: 1.98e-05
        [meta_shard_fg_expand]: 5.09998e-06
        [get_grad_eliminate_]: 1.869e-05
        [merge_forward]: 1.194e-05
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.858e-05
        [j_node_and_user_rematch]: 3.378e-05
        [meta_fg_expand]: 0.00543714
        [replace_old_param]: 0.00011497
        [inline_without_move]: 0.00011621
        [renormalize]: 0.0125239
        [add_forward_monad_depend]: 1.531e-05
        [auto_monad_grad]: 9.09998e-06
        [auto_monad_eliminator]: 7.489e-05
        [cse]: 0.00027038
        [replace_applicator]: 0.00015533
    [Cycle 2]: 0.00274209, [27]
        [switch_simplify]: 6.016e-05
        [loop_unroll]: 5.732e-05
        [a_1]: 0.00096892
        [with_stream_mark]: 1.53e-05
        [recompute_prepare]: 1.287e-05
        [updatestate_depend_eliminate]: 2.397e-05
        [updatestate_assign_eliminate]: 5.18002e-06
        [updatestate_loads_eliminate]: 3.99002e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 1.065e-05
        [updatestate_useless_node_eliminater]: 1.29e-05
        [accelerated_algorithm]: 1.032e-05
        [meta_shard_fg_expand]: 1.93002e-06
        [get_grad_eliminate_]: 8.47e-06
        [merge_forward]: 4.90999e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.737e-05
        [j_node_and_user_rematch]: 1.432e-05
        [meta_fg_expand]: 7.441e-05
        [replace_old_param]: 1.177e-05
        [inline_without_move]: 8.40999e-06
        [renormalize]: 0.00113561
        [add_forward_monad_depend]: 5.48002e-06
        [auto_monad_grad]: 1.34e-06
        [auto_monad_eliminator]: 1.787e-05
        [cse]: 6.55e-05
        [replace_applicator]: 1.891e-05
    [Cycle 3]: 0.00053671, [27]
        [switch_simplify]: 9.67001e-06
        [loop_unroll]: 8.88002e-06
        [a_1]: 0.00018903
        [with_stream_mark]: 1.345e-05
        [recompute_prepare]: 9.31e-06
        [updatestate_depend_eliminate]: 5.89999e-06
        [updatestate_assign_eliminate]: 4.87e-06
        [updatestate_loads_eliminate]: 4.38999e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 9.69e-06
        [updatestate_useless_node_eliminater]: 1.302e-05
        [accelerated_algorithm]: 1.003e-05
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 8.55001e-06
        [merge_forward]: 4.68999e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.813e-05
        [j_node_and_user_rematch]: 1.356e-05
        [meta_fg_expand]: 3.06999e-06
        [replace_old_param]: 1.171e-05
        [inline_without_move]: 8.23001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 1.232e-05
        [cse]: 3.353e-05
        [replace_applicator]: 8.95001e-06
[py_interpret_to_execute_after_opt_a]: 1.326e-05
[rewriter_after_opt_a]: 5.688e-05
[convert_after_rewriter]: 8.77e-06
[order_py_execute_after_rewriter]: 6.56999e-06
[mutable_eliminate]: 0.00068772
[jit_opt_b]: 7.421e-05, [1]
    [Cycle 1]: 6.695e-05, [2]
        [frontend_op_eliminate]: 2.729e-05
        [inline_after_opt_a]: 2.609e-05
[cconv]: 1.974e-05
[loop_unroll]: 0.00047908
[jit_opt_after_cconv]: 0.00020717, [1]
    [Cycle 1]: 0.00020034, [11]
        [c_1]: 4.876e-05
        [parameter_eliminate]: 2.81999e-06
        [updatestate_depend_eliminate]: 8.21002e-06
        [updatestate_assign_eliminate]: 4.57e-06
        [updatestate_loads_eliminate]: 4.2e-06
        [cse]: 3.248e-05
        [call_graph_tuple_transform]: 2.36e-05
        [tuple_list_get_item_eliminator]: 8.94e-06
        [none_parameter_eliminate]: 1.71998e-06
        [renormalize]: 5.09986e-07
        [switch_simplify]: 8.89e-06
[remove_dup_value]: 1.859e-05
[partial_unused_args_eliminate]: 1.85001e-06
[environ_conv]: 9.07001e-06
[add_recomputation]: 5.177e-05
[cse_after_recomputation]: 3.254e-05, [1]
    [Cycle 1]: 2.628e-05, [1]
        [cse]: 1.984e-05
[auto_monad_reorder]: 1.858e-05
[get_jit_bprop_graph]: 1.27e-06
[rewriter_after_jit_bprop_graph]: 4.37003e-06
[opt_after_jit_grad]: 0.00053193
[symbol_engine_optimizer]: 0.00010971, [1]
    [Cycle 1]: 0.00010294, [6]
        [build]: 9.29998e-06
        [elim_shapecalc]: 1.293e-05
        [elim_not_effective]: 2.191e-05
        [opt_reshape]: 1.039e-05
        [fold_const_symbol]: 1.529e-05
        [renormalize]: 6.30011e-07
[validate]: 4.568e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.0071775
[execute]: 5.12e-06
Sums
    bootstrap                            :     0.000214s :  0.34%
    type_inference                       :     0.029539s : 46.95%
    event_method                         :     0.000141s :  0.22%
    auto_monad                           :     0.000201s :  0.32%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000039s :  0.06%
    rewriter_before_opt_a                :     0.000126s :  0.20%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000145s :  0.23%
    jit_opt_a.loop_unroll                :     0.000116s :  0.18%
    jit_opt_a.a_1                        :     0.002278s :  3.62%
    jit_opt_a.with_stream_mark           :     0.000062s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000048s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000042s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000021s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000018s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.01%
    jit_opt_a.specialize_transform       :     0.000040s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000050s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000040s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000036s :  0.06%
    jit_opt_a.merge_forward              :     0.000022s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000074s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000062s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.005515s :  8.77%
    jit_opt_a.replace_old_param          :     0.000138s :  0.22%
    jit_opt_a.inline_without_move        :     0.000133s :  0.21%
    jit_opt_a.renormalize                :     0.013660s : 21.71%
    jit_opt_a.add_forward_monad_depend   :     0.000022s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000012s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000105s :  0.17%
    jit_opt_a.cse                        :     0.000369s :  0.59%
    jit_opt_a.replace_applicator         :     0.000183s :  0.29%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.02%
    rewriter_after_opt_a                 :     0.000057s :  0.09%
    convert_after_rewriter               :     0.000009s :  0.01%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000688s :  1.09%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.04%
    cconv                                :     0.000020s :  0.03%
    loop_unroll                          :     0.000479s :  0.76%
    jit_opt_after_cconv.c_1              :     0.000049s :  0.08%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.01%
    remove_dup_value                     :     0.000019s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000009s :  0.01%
    add_recomputation                    :     0.000052s :  0.08%
    cse_after_recomputation.cse          :     0.000020s :  0.03%
    auto_monad_reorder                   :     0.000019s :  0.03%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000532s :  0.85%
    symbol_engine_optimizer.build        :     0.000009s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.07%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007177s : 11.41%
    execute                              :     0.000005s :  0.01%


Time group info:
------[substitution.]   0.000264    47
 3.26% :     0.000009s :      2: substitution.depend_value_elim
 1.15% :     0.000003s :      4: substitution.elim_not_effective
 0.79% :     0.000002s :      4: substitution.fold_const_symbol
 2.48% :     0.000007s :      5: substitution.graph_param_transform
77.88% :     0.000206s :      4: substitution.inline
 1.58% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.53% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.13% :     0.000003s :      2: substitution.replace_old_param
 3.17% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
 2.88% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.15% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021106     2
93.78% :     0.019793s :      1: type_inference.infer
 6.22% :     0.001313s :      1: type_inference.specialize
------[replace.]   0.000087     6
66.21% :     0.000057s :      4: replace.inline
33.79% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
96.54% :     0.000203s :      4: match.inline
 3.46% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1021
 1.26% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.71% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.29% :     0.000002s :     15: predicate.addn_check_dump
 1.35% :     0.000002s :     15: predicate.addn_zero_filter
 2.26% :     0.000004s :     15: predicate.arithmetic_simplify
 1.33% :     0.000002s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     15: predicate.compare_switch_simplify
 1.49% :     0.000003s :     15: predicate.depend_value_elim
 1.13% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.97% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      5: predicate.elim_not_effective
 0.74% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     15: predicate.environ_get_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      5: predicate.fold_const_symbol
 1.12% :     0.000002s :     10: predicate.get_grad_eliminate
 0.30% :     0.000001s :      5: predicate.graph_param_transform
 4.95% :     0.000008s :     31: predicate.inline
 1.08% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :     10: predicate.less_batch_normalization
 1.52% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     22: predicate.load_eliminater
 1.39% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.39% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.48% :     0.000002s :      5: predicate.mutable_eliminate
 0.56% :     0.000001s :      5: predicate.opt_reshape
 2.33% :     0.000004s :     22: predicate.partial_eliminate
 1.23% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.74% :     0.000003s :     15: predicate.reduce_eliminate
 1.50% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.68% :     0.000003s :     27: predicate.replace_applicator
 0.63% :     0.000001s :     10: predicate.replace_old_param
 0.38% :     0.000001s :      5: predicate.reset_defer_inline
 1.38% :     0.000002s :     15: predicate.reshape_eliminate
 1.24% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.10% :     0.000002s :     10: predicate.special_op_eliminate
 1.17% :     0.000002s :     10: predicate.specialize_transform
 1.45% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.25% :     0.000004s :     21: predicate.switch_defer_inline
 1.95% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.42% :     0.000011s :     68: predicate.switch_simplify
 1.26% :     0.000002s :     15: predicate.tile_eliminate
 1.25% :     0.000002s :     15: predicate.transpose_eliminate
 1.56% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.94% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.41% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000003s :     15: predicate.value_based_eliminate
 0.47% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000790    13
42.53% :     0.000336s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.07% :     0.000072s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.40% :     0.000382s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042152    76
 0.15% :     0.000065s :      1: add_recomputation
 0.32% :     0.000134s :      1: auto_monad
 0.07% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.72% :     0.000302s :      1: bootstrap
 0.06% :     0.000024s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.06% :     0.000023s :      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% :     0.000007s :      1: graph_reusing
13.26% :     0.005591s :      1: jit_opt_a
 0.45% :     0.000190s :      1: jit_opt_after_cconv
 0.16% :     0.000069s :      1: jit_opt_b
 1.05% :     0.000443s :      1: loop_unroll
 1.29% :     0.000545s :      1: mutable_eliminate
 2.72% :     0.001148s :     26: opt.transform.jit_opt_a
 0.19% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000041s :      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% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000054s :      4: opt.transform.symbol_engine_opt
 1.35% :     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% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000115s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.23% :     0.000518s :      1: renormalize.infer
 0.89% :     0.000374s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.67% :     0.000282s :      1: rewriter_after_opt_a
 0.30% :     0.000125s :      1: rewriter_before_opt_a
 0.27% :     0.000112s :      1: symbol_engine_optimizer
23.37% :     0.009849s :      1: task_emit
50.23% :     0.021173s :      1: type_inference
 0.19% :     0.000079s :      1: validate


Time group info:
------[substitution.]   0.001383   160
 1.33% :     0.000018s :      8: substitution.depend_value_elim
 0.21% :     0.000003s :      4: substitution.elim_not_effective
 0.14% :     0.000002s :      4: substitution.fold_const_symbol
56.46% :     0.000781s :      4: substitution.getattr_setattr_resolve
 0.45% :     0.000006s :      5: substitution.graph_param_transform
24.67% :     0.000341s :     10: substitution.inline
 2.30% :     0.000032s :      4: substitution.inline_without_move
 0.81% :     0.000011s :     20: substitution.j_node_and_user_rematch
 0.58% :     0.000008s :      5: substitution.minmaximum_grad
 0.30% :     0.000004s :      3: substitution.partial_eliminate
 1.07% :     0.000015s :     20: substitution.remove_not_recompute_node
 3.09% :     0.000043s :     13: substitution.replace_applicator
 0.95% :     0.000013s :     18: substitution.replace_old_param
 0.20% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.27% :     0.000018s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.83% :     0.000011s :      5: substitution.tuple_list_get_item_depend_reorder
 2.33% :     0.000032s :     10: substitution.tuple_list_get_item_eliminator
 1.14% :     0.000016s :      8: substitution.updatestate_pure_node_eliminater
 1.86% :     0.000026s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.029459     2
93.78% :     0.027625s :      1: type_inference.infer
 6.22% :     0.001834s :      1: type_inference.specialize
------[replace.]   0.000281    20
17.11% :     0.000048s :      3: replace.getattr_setattr_resolve
32.24% :     0.000090s :     10: replace.inline
10.18% :     0.000029s :      1: replace.replace_applicator
32.40% :     0.000091s :      5: replace.tuple_list_get_item_eliminator
 8.07% :     0.000023s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001103    20
66.49% :     0.000733s :      3: match.getattr_setattr_resolve
30.33% :     0.000334s :     10: match.inline
 0.90% :     0.000010s :      1: match.replace_applicator
 1.56% :     0.000017s :      5: match.tuple_list_get_item_eliminator
 0.72% :     0.000008s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000461  2979
 1.45% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.25% :     0.000006s :     46: predicate.addn_check_dump
 1.56% :     0.000007s :     46: predicate.addn_zero_filter
 2.07% :     0.000010s :     46: predicate.arithmetic_simplify
 1.45% :     0.000007s :     46: predicate.cast_eliminate
 0.25% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.33% :     0.000006s :     46: predicate.compare_switch_simplify
 1.66% :     0.000008s :     46: predicate.depend_value_elim
 1.31% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.44% :     0.000007s :     46: predicate.dict_get_item_eliminator
 1.41% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.34% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.12% :     0.000001s :      5: predicate.elim_not_effective
 0.26% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.30% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.33% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.47% :     0.000007s :     46: predicate.environ_get_eliminate
 1.28% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 1.03% :     0.000005s :     27: predicate.get_grad_eliminate
 1.31% :     0.000006s :     20: predicate.getattr_setattr_resolve
 0.11% :     0.000001s :      5: predicate.graph_param_transform
 3.93% :     0.000018s :     72: predicate.inline
 3.19% :     0.000015s :     91: predicate.inline_without_move
 0.42% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.30% :     0.000006s :     27: predicate.less_batch_normalization
 1.62% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.76% :     0.000008s :     56: predicate.load_eliminater
 0.59% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.72% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.70% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.27% :     0.000006s :     46: predicate.merge_addn
 1.33% :     0.000006s :     46: predicate.minmaximum_grad
 0.59% :     0.000003s :      5: predicate.mutable_eliminate
 0.24% :     0.000001s :      5: predicate.opt_reshape
 2.12% :     0.000010s :     56: predicate.partial_eliminate
 1.41% :     0.000007s :     46: predicate.print_const_string_wrapper
 1.85% :     0.000009s :     46: predicate.reduce_eliminate
 1.64% :     0.000008s :     51: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000002s :     27: predicate.remove_not_recompute_node
 2.92% :     0.000013s :    126: predicate.replace_applicator
 1.63% :     0.000007s :     91: predicate.replace_old_param
 0.14% :     0.000001s :      5: predicate.reset_defer_inline
 1.46% :     0.000007s :     46: predicate.reshape_eliminate
 1.50% :     0.000007s :     46: predicate.row_tensor_add_zeros_like
 0.33% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.53% :     0.000007s :     46: predicate.same_eliminate
 0.57% :     0.000003s :     29: predicate.set_cell_output_no_recompute
 0.57% :     0.000003s :     10: predicate.special_op_eliminate
 1.04% :     0.000005s :     27: predicate.specialize_transform
 1.70% :     0.000008s :     46: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.32% :     0.000011s :     62: predicate.switch_defer_inline
 2.10% :     0.000010s :     62: predicate.switch_layer_defer_inline
 6.10% :     0.000028s :    188: predicate.switch_simplify
 1.37% :     0.000006s :     46: predicate.tile_eliminate
 1.43% :     0.000007s :     46: predicate.transpose_eliminate
 1.75% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.60% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 2.95% :     0.000014s :     61: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000008s :     46: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.75% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000014s :     84: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000008s :     46: predicate.value_based_eliminate
 0.17% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.34% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003191    48
65.75% :     0.002098s :     27: func_graph_cloner_run.FuncGraphClonerGraph
34.25% :     0.001093s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.085294    93
 0.06% :     0.000055s :      1: add_recomputation
 0.24% :     0.000209s :      1: auto_monad
 0.02% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.27% :     0.000227s :      1: bootstrap
 0.03% :     0.000023s :      1: cconv
 0.01% :     0.000012s :      1: convert_after_rewriter
 0.04% :     0.000035s :      1: cse_after_recomputation
 0.01% :     0.000011s :      1: environ_conv
 0.17% :     0.000149s :      1: event_method
 0.01% :     0.000009s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000003s :      1: get_jit_bprop_graph
 0.01% :     0.000011s :      1: graph_reusing
31.91% :     0.027215s :      1: jit_opt_a
 0.25% :     0.000210s :      1: jit_opt_after_cconv
 0.09% :     0.000077s :      1: jit_opt_b
 0.57% :     0.000488s :      1: loop_unroll
 0.82% :     0.000698s :      1: mutable_eliminate
 3.86% :     0.003294s :     39: opt.transform.jit_opt_a
 0.10% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000046s :      4: opt.transform.jit_opt_b
 0.02% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 1.03% :     0.000877s :      2: opt.transform.opt_resolve
 0.07% :     0.000056s :      4: opt.transform.symbol_engine_opt
 0.63% :     0.000541s :      1: opt_after_jit_grad
 0.01% :     0.000009s :      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% :     0.000041s :      1: py_interpret_to_execute
 0.02% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000021s :      1: remove_dup_value
12.44% :     0.010614s :      2: renormalize.infer
 3.55% :     0.003029s :      2: renormalize.specialize
 0.01% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000061s :      1: rewriter_after_opt_a
 0.15% :     0.000129s :      1: rewriter_before_opt_a
 0.13% :     0.000113s :      1: symbol_engine_optimizer
 8.43% :     0.007188s :      1: task_emit
34.65% :     0.029551s :      1: type_inference
 0.08% :     0.000070s :      1: validate


TotalTime = 0.0337423, [33]
[bootstrap]: 0.00018016
[type_inference]: 0.00993762
[event_method]: 4.585e-05
[auto_monad]: 9.395e-05
[graph_reusing]: 6.93e-06
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 2.546e-05
[rewriter_before_opt_a]: 0.00010801
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.0153964, [3]
    [Cycle 1]: 0.00979008, [27]
        [switch_simplify]: 6.474e-05
        [loop_unroll]: 4.421e-05
        [a_1]: 0.00091081
        [with_stream_mark]: 2.497e-05
        [recompute_prepare]: 2.032e-05
        [updatestate_depend_eliminate]: 8.18001e-06
        [updatestate_assign_eliminate]: 7.19001e-06
        [updatestate_loads_eliminate]: 7.16001e-06
        [parameter_eliminate]: 3.49001e-06
        [specialize_transform]: 1.566e-05
        [updatestate_useless_node_eliminater]: 1.338e-05
        [accelerated_algorithm]: 1.403e-05
        [meta_shard_fg_expand]: 3.93001e-06
        [get_grad_eliminate_]: 1.358e-05
        [merge_forward]: 8.75999e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.77e-05
        [j_node_and_user_rematch]: 2.45e-05
        [meta_fg_expand]: 0.0014849
        [replace_old_param]: 8.353e-05
        [inline_without_move]: 5.644e-05
        [renormalize]: 0.00642203
        [add_forward_monad_depend]: 9.28002e-06
        [auto_monad_grad]: 5.08002e-06
        [auto_monad_eliminator]: 4.851e-05
        [cse]: 0.00018754
        [replace_applicator]: 7.522e-05
    [Cycle 2]: 0.00213379, [27]
        [switch_simplify]: 4.389e-05
        [loop_unroll]: 4.198e-05
        [a_1]: 0.00115457
        [with_stream_mark]: 1.297e-05
        [recompute_prepare]: 7.93999e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 6.94001e-06
        [updatestate_useless_node_eliminater]: 6.09999e-06
        [accelerated_algorithm]: 6.24001e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 5.96e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.163e-05
        [j_node_and_user_rematch]: 8.85999e-06
        [meta_fg_expand]: 3.594e-05
        [replace_old_param]: 1.116e-05
        [inline_without_move]: 6.02999e-06
        [renormalize]: 0.00056114
        [add_forward_monad_depend]: 4.60999e-06
        [auto_monad_grad]: 1.42999e-06
        [auto_monad_eliminator]: 1.036e-05
        [cse]: 2.149e-05
        [replace_applicator]: 1.229e-05
    [Cycle 3]: 0.00035767, [27]
        [switch_simplify]: 6.69999e-06
        [loop_unroll]: 5.91e-06
        [a_1]: 0.00011619
        [with_stream_mark]: 8.28999e-06
        [recompute_prepare]: 5.64e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 6.08998e-06
        [updatestate_useless_node_eliminater]: 5.79999e-06
        [accelerated_algorithm]: 5.84999e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.51998e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.246e-05
        [j_node_and_user_rematch]: 9.15001e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 9.78998e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.19003e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.84e-06
        [cse]: 1.437e-05
        [replace_applicator]: 5.91e-06
[py_interpret_to_execute_after_opt_a]: 8.88002e-06
[rewriter_after_opt_a]: 2.941e-05
[convert_after_rewriter]: 5.81e-06
[order_py_execute_after_rewriter]: 5.27999e-06
[mutable_eliminate]: 0.00052568
[jit_opt_b]: 5.533e-05, [1]
    [Cycle 1]: 4.875e-05, [2]
        [frontend_op_eliminate]: 1.903e-05
        [inline_after_opt_a]: 1.828e-05
[cconv]: 1.779e-05
[loop_unroll]: 0.00042112
[jit_opt_after_cconv]: 0.0001596, [1]
    [Cycle 1]: 0.00015329, [11]
        [c_1]: 2.64e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 6.28e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.77002e-06
        [cse]: 2.127e-05
        [call_graph_tuple_transform]: 1.876e-05
        [tuple_list_get_item_eliminator]: 6.46e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 3.50003e-07
        [switch_simplify]: 6.12001e-06
[remove_dup_value]: 1.15e-05
[partial_unused_args_eliminate]: 1.94e-06
[environ_conv]: 4.85001e-06
[add_recomputation]: 3.424e-05
[cse_after_recomputation]: 2.529e-05, [1]
    [Cycle 1]: 2.006e-05, [1]
        [cse]: 1.431e-05
[auto_monad_reorder]: 1.195e-05
[get_jit_bprop_graph]: 1.22e-06
[rewriter_after_jit_bprop_graph]: 4.26001e-06
[opt_after_jit_grad]: 0.00045935
[symbol_engine_optimizer]: 7.347e-05, [1]
    [Cycle 1]: 6.764e-05, [6]
        [build]: 2.32999e-06
        [elim_shapecalc]: 8.59e-06
        [elim_not_effective]: 1.31e-05
        [opt_reshape]: 6.99001e-06
        [fold_const_symbol]: 9.24e-06
        [renormalize]: 3.39991e-07
[validate]: 2.97e-05
[backend_pass]: 7.60017e-07
[task_emit]: 0.00587325
[execute]: 4.68999e-06
Sums
    bootstrap                            :     0.000180s :  0.60%
    type_inference                       :     0.009938s : 33.28%
    event_method                         :     0.000046s :  0.15%
    auto_monad                           :     0.000094s :  0.31%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.09%
    rewriter_before_opt_a                :     0.000108s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000115s :  0.39%
    jit_opt_a.loop_unroll                :     0.000092s :  0.31%
    jit_opt_a.a_1                        :     0.002182s :  7.30%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000034s :  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        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000029s :  0.10%
    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       :     0.000007s :  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 :     0.000052s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001523s :  5.10%
    jit_opt_a.replace_old_param          :     0.000104s :  0.35%
    jit_opt_a.inline_without_move        :     0.000068s :  0.23%
    jit_opt_a.renormalize                :     0.006983s : 23.38%
    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      :     0.000065s :  0.22%
    jit_opt_a.cse                        :     0.000223s :  0.75%
    jit_opt_a.replace_applicator         :     0.000093s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.03%
    rewriter_after_opt_a                 :     0.000029s :  0.10%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000526s :  1.76%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000018s :  0.06%
    loop_unroll                          :     0.000421s :  1.41%
    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              :     0.000021s :  0.07%
    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 :     0.000001s :  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                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000034s :  0.11%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000012s :  0.04%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000459s :  1.54%
    symbol_engine_optimizer.build        :     0.000002s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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 :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005873s : 19.67%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000526   113
 0.31% :     0.000002s :      2: substitution.elim_not_effective
 0.21% :     0.000001s :      2: substitution.fold_const_symbol
 0.81% :     0.000004s :      4: substitution.graph_param_transform
65.85% :     0.000347s :     12: substitution.inline
 2.99% :     0.000016s :      2: substitution.inline_without_move
 1.60% :     0.000008s :     12: substitution.j_node_and_user_rematch
 1.92% :     0.000010s :      7: substitution.minmaximum_grad
 1.43% :     0.000008s :      5: substitution.partial_eliminate
 1.75% :     0.000009s :     12: substitution.remove_not_recompute_node
 5.25% :     0.000028s :     10: substitution.replace_applicator
 1.94% :     0.000010s :     14: substitution.replace_old_param
 0.53% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.09% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.81% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.53% :     0.000045s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009865     2
88.64% :     0.008745s :      1: type_inference.infer
11.36% :     0.001121s :      1: type_inference.specialize
------[replace.]   0.000169    21
59.23% :     0.000100s :     12: replace.inline
40.77% :     0.000069s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000365    21
93.06% :     0.000340s :     12: match.inline
 6.94% :     0.000025s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000387  2828
 1.52% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.36% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000006s :     47: predicate.addn_check_dump
 1.61% :     0.000006s :     47: predicate.addn_zero_filter
 1.90% :     0.000007s :     47: predicate.arithmetic_simplify
 1.56% :     0.000006s :     47: predicate.cast_eliminate
 0.13% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.47% :     0.000006s :     47: predicate.compare_switch_simplify
 1.41% :     0.000005s :     47: predicate.depend_value_elim
 1.46% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.52% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.47% :     0.000006s :     47: 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
 1.54% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.44% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.45% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.49% :     0.000006s :     47: predicate.environ_get_eliminate
 1.42% :     0.000005s :     47: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.80% :     0.000003s :     21: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.41% :     0.000017s :     76: predicate.inline
 1.93% :     0.000007s :     47: predicate.inline_without_move
 0.41% :     0.000002s :     21: predicate.j_node_and_user_rematch
 0.92% :     0.000004s :     21: predicate.less_batch_normalization
 1.84% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 2.04% :     0.000008s :     60: predicate.load_eliminater
 0.40% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.43% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.88% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.42% :     0.000006s :     47: predicate.merge_addn
 1.50% :     0.000006s :     47: predicate.minmaximum_grad
 0.45% :     0.000002s :      4: predicate.mutable_eliminate
 0.17% :     0.000001s :      4: predicate.opt_reshape
 2.51% :     0.000010s :     60: predicate.partial_eliminate
 1.48% :     0.000006s :     47: predicate.print_const_string_wrapper
 1.81% :     0.000007s :     47: predicate.reduce_eliminate
 1.89% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.74% :     0.000011s :    108: predicate.replace_applicator
 1.11% :     0.000004s :     47: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.53% :     0.000006s :     47: predicate.reshape_eliminate
 1.50% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.49% :     0.000006s :     47: predicate.same_eliminate
 0.47% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.38% :     0.000001s :      8: predicate.special_op_eliminate
 0.87% :     0.000003s :     21: predicate.specialize_transform
 1.75% :     0.000007s :     47: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.75% :     0.000011s :     68: predicate.switch_defer_inline
 2.49% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.40% :     0.000025s :    169: predicate.switch_simplify
 1.49% :     0.000006s :     47: predicate.tile_eliminate
 1.49% :     0.000006s :     47: predicate.transpose_eliminate
 1.88% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.75% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 3.36% :     0.000013s :     64: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000008s :     47: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.99% :     0.000008s :     60: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000007s :     47: 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.]   0.001619    31
61.92% :     0.001002s :     15: func_graph_cloner_run.FuncGraphClonerGraph
38.08% :     0.000616s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043700    91
 0.08% :     0.000037s :      1: add_recomputation
 0.22% :     0.000098s :      1: auto_monad
 0.03% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.44% :     0.000193s :      1: bootstrap
 0.05% :     0.000020s :      1: cconv
 0.02% :     0.000008s :      1: convert_after_rewriter
 0.06% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.11% :     0.000050s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
35.24% :     0.015400s :      1: jit_opt_a
 0.37% :     0.000162s :      1: jit_opt_after_cconv
 0.13% :     0.000058s :      1: jit_opt_b
 0.98% :     0.000429s :      1: loop_unroll
 1.22% :     0.000535s :      1: mutable_eliminate
 6.52% :     0.002847s :     39: opt.transform.jit_opt_a
 0.12% :     0.000055s :      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
 0.03% :     0.000014s :      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
 1.07% :     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
 0.06% :     0.000028s :      1: py_interpret_to_execute
 0.03% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000014s :      1: remove_dup_value
12.38% :     0.005411s :      2: renormalize.infer
 3.57% :     0.001559s :      2: renormalize.specialize
 0.01% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000032s :      1: rewriter_after_opt_a
 0.25% :     0.000111s :      1: rewriter_before_opt_a
 0.17% :     0.000076s :      1: symbol_engine_optimizer
13.46% :     0.005883s :      1: task_emit
22.77% :     0.009951s :      1: type_inference
 0.11% :     0.000046s :      1: validate

[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.760.019 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe94dc0> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.760.583 [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 = 0.0125035, [30]
[bootstrap]: 0.00027839
[type_inference]: 0.00480654
[event_method]: 2.349e-05
[auto_monad]: 6.638e-05
[graph_reusing]: 5.77999e-06
[pre_auto_parallel]: 2.18998e-06
[py_interpret_to_execute]: 3.368e-05
[rewriter_before_opt_a]: 8.682e-05
[expand_dump_flag]: 3.13998e-06
[jit_opt_a]: 0.00517304, [2]
    [Cycle 1]: 0.00161915, [27]
        [switch_simplify]: 5.732e-05
        [loop_unroll]: 3.535e-05
        [a_1]: 0.00060225
        [with_stream_mark]: 1.567e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 2.39e-05
        [updatestate_assign_eliminate]: 3.38999e-06
        [updatestate_loads_eliminate]: 3.09001e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 8.01001e-06
        [updatestate_useless_node_eliminater]: 6.39999e-06
        [accelerated_algorithm]: 6.37001e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 5.71998e-06
        [merge_forward]: 3.83999e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.507e-05
        [j_node_and_user_rematch]: 9.19998e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 1.075e-05
        [inline_without_move]: 6.05002e-06
        [renormalize]: 0.00055943
        [add_forward_monad_depend]: 5.25999e-06
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 1.39e-05
        [cse]: 3.311e-05
        [replace_applicator]: 1.264e-05
    [Cycle 2]: 0.00034567, [27]
        [switch_simplify]: 6.71e-06
        [loop_unroll]: 5.66998e-06
        [a_1]: 0.00011244
        [with_stream_mark]: 8.67998e-06
        [recompute_prepare]: 5.46998e-06
        [updatestate_depend_eliminate]: 3.14001e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 5.62999e-06
        [accelerated_algorithm]: 5.68002e-06
        [meta_shard_fg_expand]: 1.26997e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 2.69999e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.268e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 9.13002e-06
        [inline_without_move]: 5.40999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.35999e-06
        [cse]: 1.293e-05
        [replace_applicator]: 5.77001e-06
[py_interpret_to_execute_after_opt_a]: 1.788e-05
[rewriter_after_opt_a]: 3.754e-05
[convert_after_rewriter]: 7.52002e-06
[order_py_execute_after_rewriter]: 5.28002e-06
[mutable_eliminate]: 0.00045738
[jit_opt_b]: 5.318e-05, [1]
    [Cycle 1]: 4.697e-05, [2]
        [frontend_op_eliminate]: 1.81e-05
        [inline_after_opt_a]: 1.725e-05
[cconv]: 2.369e-05
[loop_unroll]: 0.00039667
[jit_opt_after_cconv]: 0.00018394, [1]
    [Cycle 1]: 0.00017775, [11]
        [c_1]: 5.637e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 6.04999e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.53998e-06
        [cse]: 2.141e-05
        [call_graph_tuple_transform]: 2.054e-05
        [tuple_list_get_item_eliminator]: 6.24999e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 6.21e-06
[remove_dup_value]: 1.449e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 5.69e-06
[add_recomputation]: 4.905e-05
[cse_after_recomputation]: 2.266e-05, [1]
    [Cycle 1]: 1.75e-05, [1]
        [cse]: 1.192e-05
[auto_monad_reorder]: 1.765e-05
[get_jit_bprop_graph]: 1.66e-06
[rewriter_after_jit_bprop_graph]: 4.60999e-06
[opt_after_jit_grad]: 0.00043954
[symbol_engine_optimizer]: 8.312e-05, [1]
    [Cycle 1]: 7.733e-05, [6]
        [build]: 2.94001e-06
        [elim_shapecalc]: 9.20999e-06
        [elim_not_effective]: 1.36e-05
        [opt_reshape]: 6.04001e-06
        [fold_const_symbol]: 9.21998e-06
        [renormalize]: 4.10015e-07
[validate]: 3.397e-05
Sums
    bootstrap                            :     0.000278s :  3.19%
    type_inference                       :     0.004807s : 55.06%
    event_method                         :     0.000023s :  0.27%
    auto_monad                           :     0.000066s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.39%
    rewriter_before_opt_a                :     0.000087s :  0.99%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000064s :  0.73%
    jit_opt_a.loop_unroll                :     0.000041s :  0.47%
    jit_opt_a.a_1                        :     0.000715s :  8.19%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000027s :  0.31%
    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       :     0.000014s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  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 :  0.32%
    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          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000559s :  6.41%
    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 :  0.22%
    jit_opt_a.cse                        :     0.000046s :  0.53%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.20%
    rewriter_after_opt_a                 :     0.000038s :  0.43%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000457s :  5.24%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000024s :  0.27%
    loop_unroll                          :     0.000397s :  4.54%
    jit_opt_after_cconv.c_1              :     0.000056s :  0.65%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  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                     :     0.000014s :  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          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000440s :  5.03%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    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  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.39%


Time group info:
------[substitution.]   0.000199    28
 1.18% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.97% :     0.000006s :      4: substitution.graph_param_transform
81.16% :     0.000162s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000005s :      4: substitution.replace_old_param
 8.02% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004746     2
81.10% :     0.003849s :      1: type_inference.infer
18.90% :     0.000897s :      1: type_inference.specialize
------[replace.]   0.000082     8
57.34% :     0.000047s :      4: replace.inline
42.66% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000173     8
91.90% :     0.000159s :      4: match.inline
 8.10% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     13: predicate.addn_check_dump
 1.24% :     0.000002s :     13: predicate.addn_zero_filter
 1.95% :     0.000003s :     13: predicate.arithmetic_simplify
 1.36% :     0.000002s :     13: predicate.cast_eliminate
 0.47% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     13: predicate.compare_switch_simplify
 1.24% :     0.000002s :     13: predicate.depend_value_elim
 1.17% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.61% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     13: predicate.environ_get_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.92% :     0.000001s :      8: predicate.get_grad_eliminate
 0.31% :     0.000000s :      4: predicate.graph_param_transform
 5.65% :     0.000008s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.46% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :      8: predicate.less_batch_normalization
 1.71% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     21: predicate.load_eliminater
 1.32% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.35% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.44% :     0.000002s :      4: predicate.mutable_eliminate
 0.50% :     0.000001s :      4: predicate.opt_reshape
 2.73% :     0.000004s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     13: predicate.reduce_eliminate
 1.65% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.30% :     0.000000s :      4: predicate.reset_defer_inline
 1.26% :     0.000002s :     13: predicate.reshape_eliminate
 1.18% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      8: predicate.special_op_eliminate
 1.03% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     21: predicate.switch_defer_inline
 2.23% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.77% :     0.000010s :     67: predicate.switch_simplify
 1.12% :     0.000002s :     13: predicate.tile_eliminate
 1.24% :     0.000002s :     13: predicate.transpose_eliminate
 1.44% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000577    12
50.32% :     0.000290s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.62% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.06% :     0.000260s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014120    72
 0.37% :     0.000052s :      1: add_recomputation
 0.49% :     0.000070s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 2.07% :     0.000292s :      1: bootstrap
 0.19% :     0.000026s :      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
 0.19% :     0.000027s :      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
36.66% :     0.005176s :      1: jit_opt_a
 1.32% :     0.000187s :      1: jit_opt_after_cconv
 0.40% :     0.000056s :      1: jit_opt_b
 2.91% :     0.000412s :      1: loop_unroll
 3.29% :     0.000465s :      1: mutable_eliminate
 6.73% :     0.000950s :     26: opt.transform.jit_opt_a
 0.61% :     0.000086s :      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% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.16% :     0.000447s :      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% :     0.000036s :      1: py_interpret_to_execute
 0.15% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.10% :     0.000296s :      1: renormalize.infer
 1.81% :     0.000256s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000041s :      1: rewriter_after_opt_a
 0.63% :     0.000090s :      1: rewriter_before_opt_a
 0.61% :     0.000086s :      1: symbol_engine_optimizer
34.12% :     0.004818s :      1: type_inference


TotalTime = 0.0341188, [33]
[bootstrap]: 0.00016889
[type_inference]: 0.00945949
[event_method]: 4.273e-05
[auto_monad]: 8.932e-05
[graph_reusing]: 6.27001e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 2.455e-05
[rewriter_before_opt_a]: 0.00010558
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.0153918, [3]
    [Cycle 1]: 0.00991766, [27]
        [switch_simplify]: 6.689e-05
        [loop_unroll]: 4.389e-05
        [a_1]: 0.00090579
        [with_stream_mark]: 2.462e-05
        [recompute_prepare]: 1.997e-05
        [updatestate_depend_eliminate]: 8.48999e-06
        [updatestate_assign_eliminate]: 6.96001e-06
        [updatestate_loads_eliminate]: 6.91999e-06
        [parameter_eliminate]: 2.08998e-06
        [specialize_transform]: 1.484e-05
        [updatestate_useless_node_eliminater]: 1.35e-05
        [accelerated_algorithm]: 1.378e-05
        [meta_shard_fg_expand]: 3.31999e-06
        [get_grad_eliminate_]: 1.312e-05
        [merge_forward]: 7.93999e-06
        [cell_reuse_recompute_pass]: 8.00006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.788e-05
        [j_node_and_user_rematch]: 2.367e-05
        [meta_fg_expand]: 0.00149076
        [replace_old_param]: 6.201e-05
        [inline_without_move]: 5.64e-05
        [renormalize]: 0.00657178
        [add_forward_monad_depend]: 1.017e-05
        [auto_monad_grad]: 5.80002e-06
        [auto_monad_eliminator]: 5.228e-05
        [cse]: 0.00018981
        [replace_applicator]: 7.597e-05
    [Cycle 2]: 0.00209584, [27]
        [switch_simplify]: 4.278e-05
        [loop_unroll]: 3.934e-05
        [a_1]: 0.00108854
        [with_stream_mark]: 1.164e-05
        [recompute_prepare]: 8e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 6.74999e-06
        [updatestate_useless_node_eliminater]: 6.12999e-06
        [accelerated_algorithm]: 6.31998e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 5.96e-06
        [merge_forward]: 2.77002e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.106e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 3.38e-05
        [replace_old_param]: 1.063e-05
        [inline_without_move]: 5.72001e-06
        [renormalize]: 0.00060111
        [add_forward_monad_depend]: 4.4e-06
        [auto_monad_grad]: 1.41002e-06
        [auto_monad_eliminator]: 1.085e-05
        [cse]: 2.148e-05
        [replace_applicator]: 1.256e-05
    [Cycle 3]: 0.00035232, [27]
        [switch_simplify]: 6.93998e-06
        [loop_unroll]: 6.26998e-06
        [a_1]: 0.00011282
        [with_stream_mark]: 8.40999e-06
        [recompute_prepare]: 5.69999e-06
        [updatestate_depend_eliminate]: 3.2e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.67001e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 5.82001e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.34e-05
        [j_node_and_user_rematch]: 8.75999e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 9.07001e-06
        [inline_without_move]: 5.76e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.99e-06
        [cse]: 1.372e-05
        [replace_applicator]: 6.09999e-06
[py_interpret_to_execute_after_opt_a]: 8.87e-06
[rewriter_after_opt_a]: 3.591e-05
[convert_after_rewriter]: 7.43e-06
[order_py_execute_after_rewriter]: 5.39e-06
[mutable_eliminate]: 0.00052039
[jit_opt_b]: 5.348e-05, [1]
    [Cycle 1]: 4.731e-05, [2]
        [frontend_op_eliminate]: 1.801e-05
        [inline_after_opt_a]: 1.817e-05
[cconv]: 2.372e-05
[loop_unroll]: 0.00041549
[jit_opt_after_cconv]: 0.00015113, [1]
    [Cycle 1]: 0.00014476, [11]
        [c_1]: 2.491e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 5.89999e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 3.05998e-06
        [cse]: 2.211e-05
        [call_graph_tuple_transform]: 1.977e-05
        [tuple_list_get_item_eliminator]: 6.94999e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 6.19001e-06
[remove_dup_value]: 1.645e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.41002e-06
[add_recomputation]: 4.306e-05
[cse_after_recomputation]: 2.497e-05, [1]
    [Cycle 1]: 2.002e-05, [1]
        [cse]: 1.396e-05
[auto_monad_reorder]: 1.775e-05
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 4.57e-06
[opt_after_jit_grad]: 0.00046
[symbol_engine_optimizer]: 7.636e-05, [1]
    [Cycle 1]: 6.977e-05, [6]
        [build]: 2.96999e-06
        [elim_shapecalc]: 8.75001e-06
        [elim_not_effective]: 1.362e-05
        [opt_reshape]: 6.61e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 5.89993e-07
[validate]: 3.588e-05
[backend_pass]: 1.17e-06
[task_emit]: 0.0067197
[execute]: 7.88999e-06
Sums
    bootstrap                            :     0.000169s :  0.56%
    type_inference                       :     0.009459s : 31.18%
    event_method                         :     0.000043s :  0.14%
    auto_monad                           :     0.000089s :  0.29%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.08%
    rewriter_before_opt_a                :     0.000106s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000117s :  0.38%
    jit_opt_a.loop_unroll                :     0.000089s :  0.29%
    jit_opt_a.a_1                        :     0.002107s :  6.95%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000034s :  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        :     0.000004s :  0.01%
    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      :     0.000026s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.08%
    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 :     0.000052s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000041s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001526s :  5.03%
    jit_opt_a.replace_old_param          :     0.000082s :  0.27%
    jit_opt_a.inline_without_move        :     0.000068s :  0.22%
    jit_opt_a.renormalize                :     0.007173s : 23.64%
    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      :     0.000069s :  0.23%
    jit_opt_a.cse                        :     0.000225s :  0.74%
    jit_opt_a.replace_applicator         :     0.000095s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.03%
    rewriter_after_opt_a                 :     0.000036s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000520s :  1.72%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000024s :  0.08%
    loop_unroll                          :     0.000415s :  1.37%
    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              :     0.000022s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  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  :     0.000006s :  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                    :     0.000043s :  0.14%
    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       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000460s :  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 :  0.04%
    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                             :     0.000036s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006720s : 22.15%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000495   113
 0.37% :     0.000002s :      2: substitution.elim_not_effective
 0.26% :     0.000001s :      2: substitution.fold_const_symbol
 1.01% :     0.000005s :      4: substitution.graph_param_transform
64.22% :     0.000318s :     12: substitution.inline
 3.21% :     0.000016s :      2: substitution.inline_without_move
 1.57% :     0.000008s :     12: substitution.j_node_and_user_rematch
 2.06% :     0.000010s :      7: substitution.minmaximum_grad
 1.33% :     0.000007s :      5: substitution.partial_eliminate
 1.93% :     0.000010s :     12: substitution.remove_not_recompute_node
 5.49% :     0.000027s :     10: substitution.replace_applicator
 1.97% :     0.000010s :     14: substitution.replace_old_param
 0.56% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.20% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.14% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 8.67% :     0.000043s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009396     2
88.17% :     0.008285s :      1: type_inference.infer
11.83% :     0.001111s :      1: type_inference.specialize
------[replace.]   0.000162    21
58.40% :     0.000094s :     12: replace.inline
41.60% :     0.000067s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000334    21
93.18% :     0.000311s :     12: match.inline
 6.82% :     0.000023s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000376  2828
 1.59% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.35% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.42% :     0.000005s :     47: predicate.addn_check_dump
 1.65% :     0.000006s :     47: predicate.addn_zero_filter
 1.94% :     0.000007s :     47: predicate.arithmetic_simplify
 1.72% :     0.000006s :     47: predicate.cast_eliminate
 0.14% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.45% :     0.000005s :     47: predicate.compare_switch_simplify
 1.47% :     0.000006s :     47: predicate.depend_value_elim
 1.45% :     0.000005s :     47: predicate.dict_get_item_const_eliminator
 1.53% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.47% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.28% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000005s :     47: predicate.environ_add_const_eliminate
 1.43% :     0.000005s :     47: predicate.environ_get_add_eliminate
 1.47% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.44% :     0.000005s :     47: predicate.environ_get_eliminate
 1.54% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.81% :     0.000003s :     21: predicate.get_grad_eliminate
 0.09% :     0.000000s :      4: predicate.graph_param_transform
 4.36% :     0.000016s :     76: predicate.inline
 1.87% :     0.000007s :     47: predicate.inline_without_move
 0.40% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.01% :     0.000004s :     21: predicate.less_batch_normalization
 1.82% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 2.02% :     0.000008s :     60: predicate.load_eliminater
 0.39% :     0.000001s :      4: predicate.loop_unroll_after_grad
 3.40% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.73% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000005s :     47: predicate.merge_addn
 1.50% :     0.000006s :     47: predicate.minmaximum_grad
 0.52% :     0.000002s :      4: predicate.mutable_eliminate
 0.19% :     0.000001s :      4: predicate.opt_reshape
 2.64% :     0.000010s :     60: predicate.partial_eliminate
 1.47% :     0.000006s :     47: predicate.print_const_string_wrapper
 1.86% :     0.000007s :     47: predicate.reduce_eliminate
 1.85% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.85% :     0.000011s :    108: predicate.replace_applicator
 1.01% :     0.000004s :     47: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.49% :     0.000006s :     47: predicate.reshape_eliminate
 1.48% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.33% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.55% :     0.000006s :     47: predicate.same_eliminate
 0.53% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.34% :     0.000001s :      8: predicate.special_op_eliminate
 0.85% :     0.000003s :     21: predicate.specialize_transform
 1.66% :     0.000006s :     47: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000005s :     47: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.68% :     0.000010s :     68: predicate.switch_defer_inline
 2.54% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.33% :     0.000024s :    169: predicate.switch_simplify
 1.48% :     0.000006s :     47: predicate.tile_eliminate
 1.46% :     0.000005s :     47: predicate.transpose_eliminate
 1.95% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.74% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000013s :     64: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.90% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000007s :     47: predicate.value_based_eliminate
 0.13% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.17% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001666    31
60.20% :     0.001003s :     15: func_graph_cloner_run.FuncGraphClonerGraph
39.80% :     0.000663s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044165    91
 0.10% :     0.000046s :      1: add_recomputation
 0.21% :     0.000094s :      1: auto_monad
 0.06% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.41% :     0.000180s :      1: bootstrap
 0.06% :     0.000026s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.11% :     0.000047s :      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
34.86% :     0.015395s :      1: jit_opt_a
 0.35% :     0.000154s :      1: jit_opt_after_cconv
 0.13% :     0.000056s :      1: jit_opt_b
 0.96% :     0.000423s :      1: loop_unroll
 1.20% :     0.000528s :      1: mutable_eliminate
 6.22% :     0.002748s :     39: opt.transform.jit_opt_a
 0.12% :     0.000054s :      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
 0.03% :     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
 1.06% :     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
 0.06% :     0.000027s :      1: py_interpret_to_execute
 0.03% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
12.68% :     0.005599s :      2: renormalize.infer
 3.53% :     0.001560s :      2: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000039s :      1: rewriter_after_opt_a
 0.25% :     0.000108s :      1: rewriter_before_opt_a
 0.18% :     0.000079s :      1: symbol_engine_optimizer
15.25% :     0.006733s :      1: task_emit
21.45% :     0.009472s :      1: type_inference
 0.12% :     0.000053s :      1: validate


TotalTime = 0.0387991, [33]
[bootstrap]: 0.00029308
[type_inference]: 0.0210568
[event_method]: 2.045e-05
[auto_monad]: 0.00014561
[graph_reusing]: 6.12999e-06
[pre_auto_parallel]: 2.80997e-06
[py_interpret_to_execute]: 0.00012391
[rewriter_before_opt_a]: 9.995e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.00558531, [2]
    [Cycle 1]: 0.0019993, [27]
        [switch_simplify]: 5.207e-05
        [loop_unroll]: 3.555e-05
        [a_1]: 0.00064247
        [with_stream_mark]: 1.687e-05
        [recompute_prepare]: 1.013e-05
        [updatestate_depend_eliminate]: 5.91e-06
        [updatestate_assign_eliminate]: 5.10999e-06
        [updatestate_loads_eliminate]: 4.89998e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 9.10001e-06
        [updatestate_useless_node_eliminater]: 1.145e-05
        [accelerated_algorithm]: 8.05e-06
        [meta_shard_fg_expand]: 2.60002e-06
        [get_grad_eliminate_]: 8.02998e-06
        [merge_forward]: 4.79e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.891e-05
        [j_node_and_user_rematch]: 1.295e-05
        [meta_fg_expand]: 3.39001e-06
        [replace_old_param]: 1.093e-05
        [inline_without_move]: 8.13999e-06
        [renormalize]: 0.00089649
        [add_forward_monad_depend]: 4.93001e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.61e-05
        [cse]: 2.523e-05
        [replace_applicator]: 1.569e-05
    [Cycle 2]: 0.00045481, [27]
        [switch_simplify]: 8.92999e-06
        [loop_unroll]: 7.95998e-06
        [a_1]: 0.00016257
        [with_stream_mark]: 1.05e-05
        [recompute_prepare]: 7.92e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 4.57e-06
        [updatestate_loads_eliminate]: 4.01001e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 8.25999e-06
        [updatestate_useless_node_eliminater]: 1.076e-05
        [accelerated_algorithm]: 7.6e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 7.54002e-06
        [merge_forward]: 4.11001e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.576e-05
        [j_node_and_user_rematch]: 1.269e-05
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 9.74e-06
        [inline_without_move]: 7.4e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 9.67001e-06
        [cse]: 1.886e-05
        [replace_applicator]: 7.98999e-06
[py_interpret_to_execute_after_opt_a]: 9.87999e-06
[rewriter_after_opt_a]: 0.00027694
[convert_after_rewriter]: 9.71e-06
[order_py_execute_after_rewriter]: 5.65001e-06
[mutable_eliminate]: 0.00049183
[jit_opt_b]: 6.438e-05, [1]
    [Cycle 1]: 5.81e-05, [2]
        [frontend_op_eliminate]: 2.3e-05
        [inline_after_opt_a]: 2.32e-05
[cconv]: 1.627e-05
[loop_unroll]: 0.00043372
[jit_opt_after_cconv]: 0.00018482, [1]
    [Cycle 1]: 0.0001782, [11]
        [c_1]: 4.305e-05
        [parameter_eliminate]: 2.75002e-06
        [updatestate_depend_eliminate]: 8.11002e-06
        [updatestate_assign_eliminate]: 4.37e-06
        [updatestate_loads_eliminate]: 3.98001e-06
        [cse]: 2.538e-05
        [call_graph_tuple_transform]: 2.071e-05
        [tuple_list_get_item_eliminator]: 9.00999e-06
        [none_parameter_eliminate]: 9.5999e-07
        [renormalize]: 4.69998e-07
        [switch_simplify]: 8.84e-06
[remove_dup_value]: 1.129e-05
[partial_unused_args_eliminate]: 1.39e-06
[environ_conv]: 8.50999e-06
[add_recomputation]: 5.217e-05
[cse_after_recomputation]: 2.869e-05, [1]
    [Cycle 1]: 2.341e-05, [1]
        [cse]: 1.754e-05
[auto_monad_reorder]: 1.857e-05
[get_jit_bprop_graph]: 1.51002e-06
[rewriter_after_jit_bprop_graph]: 3.95e-06
[opt_after_jit_grad]: 0.00047729
[symbol_engine_optimizer]: 9.368e-05, [1]
    [Cycle 1]: 8.765e-05, [6]
        [build]: 8.82e-06
        [elim_shapecalc]: 1.132e-05
        [elim_not_effective]: 1.78e-05
        [opt_reshape]: 8.3e-06
        [fold_const_symbol]: 1.266e-05
        [renormalize]: 5.39992e-07
[validate]: 4.163e-05
[backend_pass]: 8.09989e-07
[task_emit]: 0.0090211
[execute]: 5.47999e-06
Sums
    bootstrap                            :     0.000293s :  0.84%
    type_inference                       :     0.021057s : 60.05%
    event_method                         :     0.000020s :  0.06%
    auto_monad                           :     0.000146s :  0.42%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000124s :  0.35%
    rewriter_before_opt_a                :     0.000100s :  0.29%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000061s :  0.17%
    jit_opt_a.loop_unroll                :     0.000044s :  0.12%
    jit_opt_a.a_1                        :     0.000805s :  2.30%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000897s :  2.56%
    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      :     0.000026s :  0.07%
    jit_opt_a.cse                        :     0.000044s :  0.13%
    jit_opt_a.replace_applicator         :     0.000024s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000277s :  0.79%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000492s :  1.40%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.07%
    cconv                                :     0.000016s :  0.05%
    loop_unroll                          :     0.000434s :  1.24%
    jit_opt_after_cconv.c_1              :     0.000043s :  0.12%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    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  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000011s :  0.03%
    partial_unused_args_eliminate        :     0.000001s :  0.00%
    environ_conv                         :     0.000009s :  0.02%
    add_recomputation                    :     0.000052s :  0.15%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000477s :  1.36%
    symbol_engine_optimizer.build        :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009021s : 25.73%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000224    47
 3.62% :     0.000008s :      2: substitution.depend_value_elim
 1.08% :     0.000002s :      4: substitution.elim_not_effective
 0.73% :     0.000002s :      4: substitution.fold_const_symbol
 2.22% :     0.000005s :      5: substitution.graph_param_transform
75.12% :     0.000168s :      4: substitution.inline
 1.95% :     0.000004s :      8: substitution.j_node_and_user_rematch
 3.09% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.57% :     0.000004s :      2: substitution.replace_old_param
 3.53% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
 3.41% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.68% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.020992     2
93.46% :     0.019619s :      1: type_inference.infer
 6.54% :     0.001373s :      1: type_inference.specialize
------[replace.]   0.000082     6
65.52% :     0.000054s :      4: replace.inline
34.48% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     6
96.02% :     0.000165s :      4: match.inline
 3.98% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000164  1021
 1.40% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.09% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.40% :     0.000002s :     15: predicate.addn_zero_filter
 1.96% :     0.000003s :     15: predicate.arithmetic_simplify
 1.27% :     0.000002s :     15: predicate.cast_eliminate
 0.48% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     15: predicate.compare_switch_simplify
 1.44% :     0.000002s :     15: predicate.depend_value_elim
 1.20% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     15: predicate.environ_get_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.08% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 5.19% :     0.000009s :     31: predicate.inline
 1.07% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :     10: predicate.less_batch_normalization
 1.54% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.90% :     0.000003s :     22: predicate.load_eliminater
 1.43% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.42% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     15: predicate.merge_addn
 1.20% :     0.000002s :     15: predicate.minmaximum_grad
 1.62% :     0.000003s :      5: predicate.mutable_eliminate
 0.51% :     0.000001s :      5: predicate.opt_reshape
 2.40% :     0.000004s :     22: predicate.partial_eliminate
 1.29% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.71% :     0.000003s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.81% :     0.000003s :     27: predicate.replace_applicator
 0.62% :     0.000001s :     10: predicate.replace_old_param
 0.35% :     0.000001s :      5: predicate.reset_defer_inline
 1.34% :     0.000002s :     15: predicate.reshape_eliminate
 1.31% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.00% :     0.000002s :     10: predicate.special_op_eliminate
 1.21% :     0.000002s :     10: predicate.specialize_transform
 1.41% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.23% :     0.000004s :     21: predicate.switch_defer_inline
 1.93% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.49% :     0.000011s :     68: predicate.switch_simplify
 1.24% :     0.000002s :     15: predicate.tile_eliminate
 1.31% :     0.000002s :     15: predicate.transpose_eliminate
 1.65% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.40% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000003s :     15: predicate.value_based_eliminate
 0.67% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000780    13
44.07% :     0.000344s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.90% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.03% :     0.000398s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.040966    76
 0.13% :     0.000055s :      1: add_recomputation
 0.37% :     0.000151s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.75% :     0.000306s :      1: bootstrap
 0.05% :     0.000019s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.06% :     0.000025s :      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
13.64% :     0.005588s :      1: jit_opt_a
 0.46% :     0.000188s :      1: jit_opt_after_cconv
 0.16% :     0.000067s :      1: jit_opt_b
 1.08% :     0.000442s :      1: loop_unroll
 1.22% :     0.000501s :      1: mutable_eliminate
 2.67% :     0.001092s :     26: opt.transform.jit_opt_a
 0.19% :     0.000078s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000040s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000047s :      4: opt.transform.symbol_engine_opt
 1.18% :     0.000485s :      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
 0.31% :     0.000128s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000014s :      1: remove_dup_value
 1.27% :     0.000520s :      1: renormalize.infer
 0.90% :     0.000369s :      1: renormalize.specialize
 0.01% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.69% :     0.000281s :      1: rewriter_after_opt_a
 0.25% :     0.000104s :      1: rewriter_before_opt_a
 0.24% :     0.000096s :      1: symbol_engine_optimizer
22.05% :     0.009034s :      1: task_emit
51.43% :     0.021070s :      1: type_inference
 0.15% :     0.000060s :      1: validate

[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.803.385 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffe2caf6820> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.804.045 [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(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.804.179 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe9b280> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.804.716 [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 = 0.0125111, [30]
[bootstrap]: 0.00026631
[type_inference]: 0.00466134
[event_method]: 2.051e-05
[auto_monad]: 6.189e-05
[graph_reusing]: 5.59e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 3.258e-05
[rewriter_before_opt_a]: 8.906e-05
[expand_dump_flag]: 3.00002e-06
[jit_opt_a]: 0.00537101, [2]
    [Cycle 1]: 0.00172293, [27]
        [switch_simplify]: 5.902e-05
        [loop_unroll]: 3.592e-05
        [a_1]: 0.00061655
        [with_stream_mark]: 1.868e-05
        [recompute_prepare]: 8.45001e-06
        [updatestate_depend_eliminate]: 3.84002e-06
        [updatestate_assign_eliminate]: 3.85e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 1.84998e-06
        [specialize_transform]: 6.95998e-06
        [updatestate_useless_node_eliminater]: 6.07999e-06
        [accelerated_algorithm]: 6.33998e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.85002e-06
        [merge_forward]: 3.81001e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.472e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 2.39999e-06
        [replace_old_param]: 1.086e-05
        [inline_without_move]: 6.01998e-06
        [renormalize]: 0.00066241
        [add_forward_monad_depend]: 5.77001e-06
        [auto_monad_grad]: 1.96e-06
        [auto_monad_eliminator]: 1.491e-05
        [cse]: 3.243e-05
        [replace_applicator]: 1.192e-05
    [Cycle 2]: 0.00036097, [27]
        [switch_simplify]: 6.42001e-06
        [loop_unroll]: 5.52001e-06
        [a_1]: 0.00011267
        [with_stream_mark]: 9.15999e-06
        [recompute_prepare]: 5.62999e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 5.92999e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.369e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 8.77e-06
        [inline_without_move]: 5.36998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.41002e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 5.69999e-06
        [cse]: 1.381e-05
        [replace_applicator]: 5.82001e-06
[py_interpret_to_execute_after_opt_a]: 1.086e-05
[rewriter_after_opt_a]: 3.49e-05
[convert_after_rewriter]: 7.92998e-06
[order_py_execute_after_rewriter]: 5.44998e-06
[mutable_eliminate]: 0.00048189
[jit_opt_b]: 5.243e-05, [1]
    [Cycle 1]: 4.609e-05, [2]
        [frontend_op_eliminate]: 1.755e-05
        [inline_after_opt_a]: 1.755e-05
[cconv]: 2.346e-05
[loop_unroll]: 0.00040548
[jit_opt_after_cconv]: 0.00014757, [1]
    [Cycle 1]: 0.00014131, [11]
        [c_1]: 2.434e-05
        [parameter_eliminate]: 2.37001e-06
        [updatestate_depend_eliminate]: 5.82001e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [cse]: 2.024e-05
        [call_graph_tuple_transform]: 1.978e-05
        [tuple_list_get_item_eliminator]: 6.11e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 3.39991e-07
        [switch_simplify]: 6.02001e-06
[remove_dup_value]: 1.507e-05
[partial_unused_args_eliminate]: 2.49999e-06
[environ_conv]: 6.11998e-06
[add_recomputation]: 4.954e-05
[cse_after_recomputation]: 2.344e-05, [1]
    [Cycle 1]: 1.815e-05, [1]
        [cse]: 1.243e-05
[auto_monad_reorder]: 1.79e-05
[get_jit_bprop_graph]: 1.97001e-06
[rewriter_after_jit_bprop_graph]: 5.61003e-06
[opt_after_jit_grad]: 0.00043952
[symbol_engine_optimizer]: 7.393e-05, [1]
    [Cycle 1]: 6.797e-05, [6]
        [build]: 3.57002e-06
        [elim_shapecalc]: 8.35999e-06
        [elim_not_effective]: 1.329e-05
        [opt_reshape]: 6.31e-06
        [fold_const_symbol]: 9.10999e-06
        [renormalize]: 3.9002e-07
[validate]: 3.456e-05
Sums
    bootstrap                            :     0.000266s :  3.07%
    type_inference                       :     0.004661s : 53.79%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000062s :  0.71%
    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                :     0.000089s :  1.03%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.76%
    jit_opt_a.loop_unroll                :     0.000041s :  0.48%
    jit_opt_a.a_1                        :     0.000729s :  8.42%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.32%
    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 :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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_        :     0.000012s :  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 :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000662s :  7.64%
    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      :     0.000021s :  0.24%
    jit_opt_a.cse                        :     0.000046s :  0.53%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  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                    :     0.000482s :  5.56%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000023s :  0.27%
    loop_unroll                          :     0.000405s :  4.68%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.28%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.23%
    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      :     0.000000s :  0.00%
    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 :  0.07%
    add_recomputation                    :     0.000050s :  0.57%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000440s :  5.07%
    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 :  0.00%
    validate                             :     0.000035s :  0.40%


Time group info:
------[substitution.]   0.000210    28
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.81% :     0.000006s :      4: substitution.graph_param_transform
82.07% :     0.000172s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000005s :      4: substitution.replace_old_param
 7.61% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004601     2
79.93% :     0.003677s :      1: type_inference.infer
20.07% :     0.000923s :      1: type_inference.specialize
------[replace.]   0.000082     8
57.64% :     0.000047s :      4: replace.inline
42.36% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     8
92.41% :     0.000170s :      4: match.inline
 7.59% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   912
 1.23% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.97% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     13: predicate.addn_check_dump
 1.27% :     0.000002s :     13: predicate.addn_zero_filter
 2.24% :     0.000003s :     13: predicate.arithmetic_simplify
 1.31% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.21% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      4: predicate.elim_not_effective
 0.51% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     13: predicate.environ_get_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 1.09% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.48% :     0.000007s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      8: predicate.less_batch_normalization
 1.84% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     21: predicate.load_eliminater
 1.48% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.07% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     13: predicate.merge_addn
 1.11% :     0.000002s :     13: predicate.minmaximum_grad
 1.33% :     0.000002s :      4: predicate.mutable_eliminate
 0.41% :     0.000001s :      4: predicate.opt_reshape
 2.62% :     0.000004s :     21: predicate.partial_eliminate
 1.20% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     13: predicate.reduce_eliminate
 1.66% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.96% :     0.000003s :     25: predicate.replace_applicator
 0.78% :     0.000001s :      8: predicate.replace_old_param
 0.37% :     0.000001s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.26% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.78% :     0.000001s :      8: predicate.special_op_eliminate
 1.03% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     21: predicate.switch_defer_inline
 2.13% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.67% :     0.000010s :     67: predicate.switch_simplify
 1.18% :     0.000002s :     13: predicate.tile_eliminate
 1.30% :     0.000002s :     13: predicate.transpose_eliminate
 1.43% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     13: predicate.value_based_eliminate
 0.41% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000592    12
49.21% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.16% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.63% :     0.000270s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014214    72
 0.37% :     0.000052s :      1: add_recomputation
 0.46% :     0.000065s :      1: auto_monad
 0.15% :     0.000021s :      1: auto_monad_reorder
 1.98% :     0.000282s :      1: bootstrap
 0.18% :     0.000026s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000026s :      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
37.81% :     0.005374s :      1: jit_opt_a
 1.06% :     0.000151s :      1: jit_opt_after_cconv
 0.39% :     0.000055s :      1: jit_opt_b
 2.90% :     0.000413s :      1: loop_unroll
 3.44% :     0.000489s :      1: mutable_eliminate
 6.80% :     0.000967s :     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
 0.09% :     0.000013s :      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
 0.24% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.14% :     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% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000035s :      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
 2.58% :     0.000367s :      1: renormalize.infer
 2.02% :     0.000287s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000038s :      1: rewriter_after_opt_a
 0.65% :     0.000092s :      1: rewriter_before_opt_a
 0.54% :     0.000076s :      1: symbol_engine_optimizer
32.87% :     0.004672s :      1: type_inference


TotalTime = 0.0361445, [33]
[bootstrap]: 0.00017527
[type_inference]: 0.00973601
[event_method]: 4.312e-05
[auto_monad]: 9.518e-05
[graph_reusing]: 7.19001e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 2.834e-05
[rewriter_before_opt_a]: 0.00012074
[expand_dump_flag]: 3.13998e-06
[jit_opt_a]: 0.0158716, [3]
    [Cycle 1]: 0.0101265, [27]
        [switch_simplify]: 6.913e-05
        [loop_unroll]: 4.637e-05
        [a_1]: 0.00090889
        [with_stream_mark]: 2.664e-05
        [recompute_prepare]: 1.962e-05
        [updatestate_depend_eliminate]: 8.11002e-06
        [updatestate_assign_eliminate]: 7.06001e-06
        [updatestate_loads_eliminate]: 6.61e-06
        [parameter_eliminate]: 2.74001e-06
        [specialize_transform]: 1.444e-05
        [updatestate_useless_node_eliminater]: 1.44e-05
        [accelerated_algorithm]: 1.368e-05
        [meta_shard_fg_expand]: 3.66999e-06
        [get_grad_eliminate_]: 1.307e-05
        [merge_forward]: 8.35001e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.851e-05
        [j_node_and_user_rematch]: 2.384e-05
        [meta_fg_expand]: 0.0014935
        [replace_old_param]: 6.415e-05
        [inline_without_move]: 5.652e-05
        [renormalize]: 0.00670807
        [add_forward_monad_depend]: 1.164e-05
        [auto_monad_grad]: 7.24001e-06
        [auto_monad_eliminator]: 5.771e-05
        [cse]: 0.00020411
        [replace_applicator]: 8.532e-05
    [Cycle 2]: 0.00222492, [27]
        [switch_simplify]: 4.455e-05
        [loop_unroll]: 4.214e-05
        [a_1]: 0.00118382
        [with_stream_mark]: 1.564e-05
        [recompute_prepare]: 8.52e-06
        [updatestate_depend_eliminate]: 4.13999e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 3.38999e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 6.92002e-06
        [updatestate_useless_node_eliminater]: 6.53e-06
        [accelerated_algorithm]: 6.73e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 6.60002e-06
        [merge_forward]: 3.36001e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.392e-05
        [j_node_and_user_rematch]: 9.76e-06
        [meta_fg_expand]: 3.936e-05
        [replace_old_param]: 1.111e-05
        [inline_without_move]: 6.21998e-06
        [renormalize]: 0.00059349
        [add_forward_monad_depend]: 5.00999e-06
        [auto_monad_grad]: 1.97001e-06
        [auto_monad_eliminator]: 1.188e-05
        [cse]: 2.432e-05
        [replace_applicator]: 1.335e-05
    [Cycle 3]: 0.00037017, [27]
        [switch_simplify]: 7.60998e-06
        [loop_unroll]: 6.55997e-06
        [a_1]: 0.00012248
        [with_stream_mark]: 9.02999e-06
        [recompute_prepare]: 5.95002e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 6.34001e-06
        [updatestate_useless_node_eliminater]: 6.41e-06
        [accelerated_algorithm]: 6.15002e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 6.01e-06
        [merge_forward]: 2.65002e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.345e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 9.54999e-06
        [inline_without_move]: 5.91e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.26002e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.77001e-06
        [cse]: 1.33e-05
        [replace_applicator]: 6.38e-06
[py_interpret_to_execute_after_opt_a]: 1.2e-05
[rewriter_after_opt_a]: 3.839e-05
[convert_after_rewriter]: 7.25998e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.00064608
[jit_opt_b]: 5.694e-05, [1]
    [Cycle 1]: 4.895e-05, [2]
        [frontend_op_eliminate]: 1.861e-05
        [inline_after_opt_a]: 1.832e-05
[cconv]: 2.306e-05
[loop_unroll]: 0.00043026
[jit_opt_after_cconv]: 0.00015344, [1]
    [Cycle 1]: 0.00014744, [11]
        [c_1]: 2.572e-05
        [parameter_eliminate]: 2.69999e-06
        [updatestate_depend_eliminate]: 6.12999e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.93998e-06
        [cse]: 2.398e-05
        [call_graph_tuple_transform]: 2.026e-05
        [tuple_list_get_item_eliminator]: 6.81999e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 6.18998e-06
[remove_dup_value]: 1.615e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 6.06e-06
[add_recomputation]: 4.506e-05
[cse_after_recomputation]: 2.641e-05, [1]
    [Cycle 1]: 2.09e-05, [1]
        [cse]: 1.477e-05
[auto_monad_reorder]: 1.852e-05
[get_jit_bprop_graph]: 1.65001e-06
[rewriter_after_jit_bprop_graph]: 4.93001e-06
[opt_after_jit_grad]: 0.00047428
[symbol_engine_optimizer]: 7.618e-05, [1]
    [Cycle 1]: 6.996e-05, [6]
        [build]: 3.70998e-06
        [elim_shapecalc]: 8.82e-06
        [elim_not_effective]: 1.368e-05
        [opt_reshape]: 6.73e-06
        [fold_const_symbol]: 9.82001e-06
        [renormalize]: 4.69998e-07
[validate]: 3.55e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00778222
[execute]: 6.27001e-06
Sums
    bootstrap                            :     0.000175s :  0.54%
    type_inference                       :     0.009736s : 30.24%
    event_method                         :     0.000043s :  0.13%
    auto_monad                           :     0.000095s :  0.30%
    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                :     0.000121s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000121s :  0.38%
    jit_opt_a.loop_unroll                :     0.000095s :  0.30%
    jit_opt_a.a_1                        :     0.002215s :  6.88%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000034s :  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 :     0.000012s :  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 :     0.000027s :  0.08%
    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_        :     0.000026s :  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 :     0.000056s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.001535s :  4.77%
    jit_opt_a.replace_old_param          :     0.000085s :  0.26%
    jit_opt_a.inline_without_move        :     0.000069s :  0.21%
    jit_opt_a.renormalize                :     0.007302s : 22.68%
    jit_opt_a.add_forward_monad_depend   :     0.000018s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000075s :  0.23%
    jit_opt_a.cse                        :     0.000242s :  0.75%
    jit_opt_a.replace_applicator         :     0.000105s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000646s :  2.01%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000023s :  0.07%
    loop_unroll                          :     0.000430s :  1.34%
    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 :  0.07%
    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      :     0.000001s :  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                    :     0.000045s :  0.14%
    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                   :     0.000474s :  1.47%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    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 :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007782s : 24.17%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000518   113
 0.36% :     0.000002s :      2: substitution.elim_not_effective
 0.26% :     0.000001s :      2: substitution.fold_const_symbol
 1.07% :     0.000006s :      4: substitution.graph_param_transform
63.12% :     0.000327s :     12: substitution.inline
 3.19% :     0.000017s :      2: substitution.inline_without_move
 1.57% :     0.000008s :     12: substitution.j_node_and_user_rematch
 2.07% :     0.000011s :      7: substitution.minmaximum_grad
 1.37% :     0.000007s :      5: substitution.partial_eliminate
 1.94% :     0.000010s :     12: substitution.remove_not_recompute_node
 5.99% :     0.000031s :     10: substitution.replace_applicator
 2.02% :     0.000010s :     14: substitution.replace_old_param
 0.48% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.39% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.97% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 9.19% :     0.000048s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009661     2
88.67% :     0.008566s :      1: type_inference.infer
11.33% :     0.001094s :      1: type_inference.specialize
------[replace.]   0.000170    21
59.60% :     0.000102s :     12: replace.inline
40.40% :     0.000069s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000346    21
92.57% :     0.000320s :     12: match.inline
 7.43% :     0.000026s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000416  2828
 1.42% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.38% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.37% :     0.000006s :     47: predicate.addn_check_dump
 1.51% :     0.000006s :     47: predicate.addn_zero_filter
 1.82% :     0.000008s :     47: predicate.arithmetic_simplify
 1.50% :     0.000006s :     47: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.34% :     0.000006s :     47: predicate.compare_switch_simplify
 1.30% :     0.000005s :     47: predicate.depend_value_elim
 1.33% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.46% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.34% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.33% :     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
 1.37% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.31% :     0.000005s :     47: predicate.environ_get_add_eliminate
 1.35% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.34% :     0.000006s :     47: predicate.environ_get_eliminate
 1.31% :     0.000005s :     47: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 0.81% :     0.000003s :     21: predicate.get_grad_eliminate
 0.08% :     0.000000s :      4: predicate.graph_param_transform
 4.25% :     0.000018s :     76: predicate.inline
 1.78% :     0.000007s :     47: predicate.inline_without_move
 0.36% :     0.000001s :     21: predicate.j_node_and_user_rematch
 0.88% :     0.000004s :     21: predicate.less_batch_normalization
 1.70% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 1.88% :     0.000008s :     60: predicate.load_eliminater
 0.43% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.33% :     0.000014s :     97: predicate.loop_unroll_before_grad
 1.70% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.37% :     0.000006s :     47: predicate.merge_addn
 1.43% :     0.000006s :     47: predicate.minmaximum_grad
 0.61% :     0.000003s :      4: predicate.mutable_eliminate
 0.18% :     0.000001s :      4: predicate.opt_reshape
 2.24% :     0.000009s :     60: predicate.partial_eliminate
 1.41% :     0.000006s :     47: predicate.print_const_string_wrapper
 1.74% :     0.000007s :     47: predicate.reduce_eliminate
 1.72% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.61% :     0.000011s :    108: predicate.replace_applicator
 0.93% :     0.000004s :     47: predicate.replace_old_param
 0.10% :     0.000000s :      4: predicate.reset_defer_inline
 1.39% :     0.000006s :     47: predicate.reshape_eliminate
 1.44% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.29% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.37% :     0.000006s :     47: predicate.same_eliminate
 0.45% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.28% :     0.000001s :      8: predicate.special_op_eliminate
 0.80% :     0.000003s :     21: predicate.specialize_transform
 1.51% :     0.000006s :     47: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.13% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.60% :     0.000011s :     68: predicate.switch_defer_inline
 2.29% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.17% :     0.000026s :    169: predicate.switch_simplify
 1.39% :     0.000006s :     47: predicate.tile_eliminate
 1.38% :     0.000006s :     47: predicate.transpose_eliminate
 1.70% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.62% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 9.70% :     0.000040s :     64: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.78% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000007s :     47: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.19% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001627    31
62.16% :     0.001011s :     15: func_graph_cloner_run.FuncGraphClonerGraph
37.84% :     0.000615s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046456    91
 0.10% :     0.000048s :      1: add_recomputation
 0.21% :     0.000100s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.40% :     0.000187s :      1: bootstrap
 0.06% :     0.000026s :      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
 0.10% :     0.000047s :      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% :     0.000010s :      1: graph_reusing
34.17% :     0.015875s :      1: jit_opt_a
 0.36% :     0.000165s :      1: jit_opt_after_cconv
 0.13% :     0.000060s :      1: jit_opt_b
 0.94% :     0.000438s :      1: loop_unroll
 1.41% :     0.000656s :      1: mutable_eliminate
 6.21% :     0.002884s :     39: opt.transform.jit_opt_a
 0.12% :     0.000056s :      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% :     0.000017s :      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
 1.04% :     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% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000031s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
12.27% :     0.005700s :      2: renormalize.infer
 3.41% :     0.001585s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000041s :      1: rewriter_after_opt_a
 0.27% :     0.000124s :      1: rewriter_before_opt_a
 0.17% :     0.000079s :      1: symbol_engine_optimizer
16.77% :     0.007793s :      1: task_emit
20.98% :     0.009748s :      1: type_inference
 0.12% :     0.000055s :      1: validate

[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.851.208 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe9b700> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.851.843 [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 = 0.038553, [33]
[bootstrap]: 0.00026679
[type_inference]: 0.0203234
[event_method]: 1.78e-05
[auto_monad]: 0.00014661
[graph_reusing]: 4.02e-06
[pre_auto_parallel]: 1.86998e-06
[py_interpret_to_execute]: 0.00011941
[rewriter_before_opt_a]: 8.719e-05
[expand_dump_flag]: 3.9e-06
[jit_opt_a]: 0.005903, [2]
    [Cycle 1]: 0.00214322, [27]
        [switch_simplify]: 5.575e-05
        [loop_unroll]: 3.517e-05
        [a_1]: 0.0006477
        [with_stream_mark]: 1.414e-05
        [recompute_prepare]: 9.96998e-06
        [updatestate_depend_eliminate]: 5.67999e-06
        [updatestate_assign_eliminate]: 4.16001e-06
        [updatestate_loads_eliminate]: 4.00998e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 9.45001e-06
        [updatestate_useless_node_eliminater]: 1.144e-05
        [accelerated_algorithm]: 8.52e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 7.88999e-06
        [merge_forward]: 4.24002e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.619e-05
        [j_node_and_user_rematch]: 1.316e-05
        [meta_fg_expand]: 3.06999e-06
        [replace_old_param]: 1.005e-05
        [inline_without_move]: 7.92998e-06
        [renormalize]: 0.00103122
        [add_forward_monad_depend]: 6.24999e-06
        [auto_monad_grad]: 2.59001e-06
        [auto_monad_eliminator]: 2.007e-05
        [cse]: 2.837e-05
        [replace_applicator]: 1.501e-05
    [Cycle 2]: 0.0004601, [27]
        [switch_simplify]: 9.37001e-06
        [loop_unroll]: 8.12998e-06
        [a_1]: 0.00016486
        [with_stream_mark]: 1.108e-05
        [recompute_prepare]: 8.17998e-06
        [updatestate_depend_eliminate]: 4.74e-06
        [updatestate_assign_eliminate]: 4.82e-06
        [updatestate_loads_eliminate]: 3.78001e-06
        [parameter_eliminate]: 1.44998e-06
        [specialize_transform]: 8e-06
        [updatestate_useless_node_eliminater]: 1.048e-05
        [accelerated_algorithm]: 7.92003e-06
        [meta_shard_fg_expand]: 1.79e-06
        [get_grad_eliminate_]: 7.18e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.613e-05
        [j_node_and_user_rematch]: 1.294e-05
        [meta_fg_expand]: 2.91999e-06
        [replace_old_param]: 1.006e-05
        [inline_without_move]: 7.51999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 9.52999e-06
        [cse]: 1.949e-05
        [replace_applicator]: 7.83999e-06
[py_interpret_to_execute_after_opt_a]: 1.258e-05
[rewriter_after_opt_a]: 0.00028805
[convert_after_rewriter]: 1.084e-05
[order_py_execute_after_rewriter]: 7.77e-06
[mutable_eliminate]: 0.00056983
[jit_opt_b]: 6.692e-05, [1]
    [Cycle 1]: 5.998e-05, [2]
        [frontend_op_eliminate]: 2.365e-05
        [inline_after_opt_a]: 2.394e-05
[cconv]: 4.166e-05
[loop_unroll]: 0.00045618
[jit_opt_after_cconv]: 0.00018195, [1]
    [Cycle 1]: 0.00017557, [11]
        [c_1]: 4.334e-05
        [parameter_eliminate]: 1.72001e-06
        [updatestate_depend_eliminate]: 6.63e-06
        [updatestate_assign_eliminate]: 4.23999e-06
        [updatestate_loads_eliminate]: 3.85e-06
        [cse]: 2.458e-05
        [call_graph_tuple_transform]: 2.126e-05
        [tuple_list_get_item_eliminator]: 8.37998e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 8.36002e-06
[remove_dup_value]: 1.859e-05
[partial_unused_args_eliminate]: 2.52001e-06
[environ_conv]: 9.56e-06
[add_recomputation]: 6.134e-05
[cse_after_recomputation]: 2.985e-05, [1]
    [Cycle 1]: 2.44e-05, [1]
        [cse]: 1.826e-05
[auto_monad_reorder]: 2.561e-05
[get_jit_bprop_graph]: 1.76003e-06
[rewriter_after_jit_bprop_graph]: 4.63999e-06
[opt_after_jit_grad]: 0.00048541
[symbol_engine_optimizer]: 9.654e-05, [1]
    [Cycle 1]: 9.045e-05, [6]
        [build]: 9.50001e-06
        [elim_shapecalc]: 1.175e-05
        [elim_not_effective]: 1.825e-05
        [opt_reshape]: 8.64e-06
        [fold_const_symbol]: 1.342e-05
        [renormalize]: 4.10015e-07
[validate]: 4.999e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00905217
[execute]: 5.07e-06
Sums
    bootstrap                            :     0.000267s :  0.77%
    type_inference                       :     0.020323s : 58.65%
    event_method                         :     0.000018s :  0.05%
    auto_monad                           :     0.000147s :  0.42%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000119s :  0.34%
    rewriter_before_opt_a                :     0.000087s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.19%
    jit_opt_a.loop_unroll                :     0.000043s :  0.12%
    jit_opt_a.a_1                        :     0.000813s :  2.35%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  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 :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001031s :  2.98%
    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      :     0.000030s :  0.09%
    jit_opt_a.cse                        :     0.000048s :  0.14%
    jit_opt_a.replace_applicator         :     0.000023s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000288s :  0.83%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000570s :  1.64%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000042s :  0.12%
    loop_unroll                          :     0.000456s :  1.32%
    jit_opt_after_cconv.c_1              :     0.000043s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.00%
    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              :     0.000025s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000061s :  0.18%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000485s :  1.40%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000050s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009052s : 26.13%
    execute                              :     0.000005s :  0.01%


Time group info:
------[substitution.]   0.000224    47
 3.64% :     0.000008s :      2: substitution.depend_value_elim
 1.14% :     0.000003s :      4: substitution.elim_not_effective
 0.98% :     0.000002s :      4: substitution.fold_const_symbol
 2.89% :     0.000006s :      5: substitution.graph_param_transform
74.97% :     0.000168s :      4: substitution.inline
 1.93% :     0.000004s :      8: substitution.j_node_and_user_rematch
 3.12% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.43% :     0.000003s :      2: substitution.replace_old_param
 3.23% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
 3.29% :     0.000007s :      3: substitution.updatestate_pure_node_eliminater
 3.38% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.020271     2
93.60% :     0.018974s :      1: type_inference.infer
 6.40% :     0.001297s :      1: type_inference.specialize
------[replace.]   0.000088     6
67.15% :     0.000059s :      4: replace.inline
32.85% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     6
96.43% :     0.000165s :      4: match.inline
 3.57% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1021
 1.30% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.34% :     0.000002s :     15: predicate.addn_zero_filter
 2.02% :     0.000003s :     15: predicate.arithmetic_simplify
 1.39% :     0.000002s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.43% :     0.000002s :     15: predicate.compare_switch_simplify
 1.45% :     0.000002s :     15: predicate.depend_value_elim
 1.20% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.97% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.31% :     0.000001s :      5: predicate.elim_not_effective
 0.62% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.30% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 4.85% :     0.000008s :     31: predicate.inline
 1.07% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :     10: predicate.less_batch_normalization
 1.50% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     22: predicate.load_eliminater
 1.35% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.45% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.16% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     15: predicate.merge_addn
 1.16% :     0.000002s :     15: predicate.minmaximum_grad
 1.37% :     0.000002s :      5: predicate.mutable_eliminate
 0.51% :     0.000001s :      5: predicate.opt_reshape
 2.37% :     0.000004s :     22: predicate.partial_eliminate
 1.34% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.66% :     0.000003s :     15: predicate.reduce_eliminate
 1.55% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.80% :     0.000003s :     27: predicate.replace_applicator
 0.58% :     0.000001s :     10: predicate.replace_old_param
 0.37% :     0.000001s :      5: predicate.reset_defer_inline
 1.40% :     0.000002s :     15: predicate.reshape_eliminate
 1.24% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.92% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.29% :     0.000002s :     10: predicate.special_op_eliminate
 1.09% :     0.000002s :     10: predicate.specialize_transform
 1.47% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.17% :     0.000004s :     21: predicate.switch_defer_inline
 1.94% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.49% :     0.000011s :     68: predicate.switch_simplify
 1.31% :     0.000002s :     15: predicate.tile_eliminate
 1.30% :     0.000002s :     15: predicate.transpose_eliminate
 1.65% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.50% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.83% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.27% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000736    13
44.00% :     0.000324s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.12% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.88% :     0.000375s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.040867    76
 0.16% :     0.000064s :      1: add_recomputation
 0.37% :     0.000152s :      1: auto_monad
 0.07% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000281s :      1: bootstrap
 0.11% :     0.000045s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.08% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.05% :     0.000021s :      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
 0.02% :     0.000007s :      1: graph_reusing
14.45% :     0.005907s :      1: jit_opt_a
 0.45% :     0.000185s :      1: jit_opt_after_cconv
 0.17% :     0.000070s :      1: jit_opt_b
 1.13% :     0.000464s :      1: loop_unroll
 1.42% :     0.000578s :      1: mutable_eliminate
 2.69% :     0.001101s :     26: opt.transform.jit_opt_a
 0.19% :     0.000078s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000041s :      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% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.21% :     0.000493s :      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
 0.30% :     0.000123s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.48% :     0.000604s :      1: renormalize.infer
 1.03% :     0.000421s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.71% :     0.000292s :      1: rewriter_after_opt_a
 0.22% :     0.000091s :      1: rewriter_before_opt_a
 0.24% :     0.000099s :      1: symbol_engine_optimizer
22.18% :     0.009063s :      1: task_emit
49.76% :     0.020335s :      1: type_inference
 0.18% :     0.000072s :      1: validate


TotalTime = 0.0123525, [30]
[bootstrap]: 0.00027726
[type_inference]: 0.00461391
[event_method]: 2.044e-05
[auto_monad]: 6.002e-05
[graph_reusing]: 4.99e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 3.221e-05
[rewriter_before_opt_a]: 8.559e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00522109, [2]
    [Cycle 1]: 0.00172967, [27]
        [switch_simplify]: 5.497e-05
        [loop_unroll]: 3.52e-05
        [a_1]: 0.0006463
        [with_stream_mark]: 1.858e-05
        [recompute_prepare]: 8.2e-06
        [updatestate_depend_eliminate]: 3.55998e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 6.79001e-06
        [updatestate_useless_node_eliminater]: 6.21998e-06
        [accelerated_algorithm]: 6.11998e-06
        [meta_shard_fg_expand]: 2.11003e-06
        [get_grad_eliminate_]: 6.04001e-06
        [merge_forward]: 3.04999e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.346e-05
        [j_node_and_user_rematch]: 9.69999e-06
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 1.013e-05
        [inline_without_move]: 6.17999e-06
        [renormalize]: 0.00065731
        [add_forward_monad_depend]: 5.99e-06
        [auto_monad_grad]: 2.07001e-06
        [auto_monad_eliminator]: 1.329e-05
        [cse]: 2.64e-05
        [replace_applicator]: 1.269e-05
    [Cycle 2]: 0.00036543, [27]
        [switch_simplify]: 6.83e-06
        [loop_unroll]: 5.79e-06
        [a_1]: 0.00011218
        [with_stream_mark]: 9.35001e-06
        [recompute_prepare]: 5.83002e-06
        [updatestate_depend_eliminate]: 3.03998e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 5.53002e-06
        [accelerated_algorithm]: 5.77999e-06
        [meta_shard_fg_expand]: 1.49998e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.18e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.449e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 8.76002e-06
        [inline_without_move]: 5.66e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.57999e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 6.60002e-06
        [cse]: 1.437e-05
        [replace_applicator]: 6.04999e-06
[py_interpret_to_execute_after_opt_a]: 1.247e-05
[rewriter_after_opt_a]: 3.294e-05
[convert_after_rewriter]: 6.73998e-06
[order_py_execute_after_rewriter]: 5.22999e-06
[mutable_eliminate]: 0.00051597
[jit_opt_b]: 5.349e-05, [1]
    [Cycle 1]: 4.68e-05, [2]
        [frontend_op_eliminate]: 1.833e-05
        [inline_after_opt_a]: 1.682e-05
[cconv]: 2.134e-05
[loop_unroll]: 0.00041302
[jit_opt_after_cconv]: 0.00014638, [1]
    [Cycle 1]: 0.00013974, [11]
        [c_1]: 2.427e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 5.98002e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.064e-05
        [call_graph_tuple_transform]: 1.938e-05
        [tuple_list_get_item_eliminator]: 5.86e-06
        [none_parameter_eliminate]: 1.17e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.92999e-06
[remove_dup_value]: 1.272e-05
[partial_unused_args_eliminate]: 2.01e-06
[environ_conv]: 5.05001e-06
[add_recomputation]: 4.247e-05
[cse_after_recomputation]: 2.367e-05, [1]
    [Cycle 1]: 1.84e-05, [1]
        [cse]: 1.26e-05
[auto_monad_reorder]: 1.495e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 4.83001e-06
[opt_after_jit_grad]: 0.0004473
[symbol_engine_optimizer]: 7.327e-05, [1]
    [Cycle 1]: 6.733e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 8.30999e-06
        [elim_not_effective]: 1.297e-05
        [opt_reshape]: 6.04999e-06
        [fold_const_symbol]: 9.12999e-06
        [renormalize]: 4.00003e-07
[validate]: 3.704e-05
Sums
    bootstrap                            :     0.000277s :  3.20%
    type_inference                       :     0.004614s : 53.25%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000060s :  0.69%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.37%
    rewriter_before_opt_a                :     0.000086s :  0.99%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000062s :  0.71%
    jit_opt_a.loop_unroll                :     0.000041s :  0.47%
    jit_opt_a.a_1                        :     0.000758s :  8.75%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.32%
    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        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000011s :  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 :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000657s :  7.59%
    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 :  0.23%
    jit_opt_a.cse                        :     0.000041s :  0.47%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.14%
    rewriter_after_opt_a                 :     0.000033s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000516s :  5.95%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000021s :  0.25%
    loop_unroll                          :     0.000413s :  4.77%
    jit_opt_after_cconv.c_1              :     0.000024s :  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 :     0.000003s :  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 :     0.000019s :  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      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000013s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000042s :  0.49%
    cse_after_recomputation.cse          :     0.000013s :  0.15%
    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                   :     0.000447s :  5.16%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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 :  0.00%
    validate                             :     0.000037s :  0.43%


Time group info:
------[substitution.]   0.000240    28
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.14% :     0.000005s :      4: substitution.graph_param_transform
85.32% :     0.000205s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.87% :     0.000004s :      4: substitution.replace_old_param
 6.17% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004557     2
80.42% :     0.003665s :      1: type_inference.infer
19.58% :     0.000892s :      1: type_inference.specialize
------[replace.]   0.000083     8
56.46% :     0.000047s :      4: replace.inline
43.54% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000215     8
94.01% :     0.000202s :      4: match.inline
 5.99% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     13: predicate.addn_check_dump
 1.44% :     0.000002s :     13: predicate.addn_zero_filter
 1.73% :     0.000002s :     13: predicate.arithmetic_simplify
 1.58% :     0.000002s :     13: predicate.cast_eliminate
 0.50% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     13: predicate.compare_switch_simplify
 1.20% :     0.000002s :     13: predicate.depend_value_elim
 1.16% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      4: predicate.elim_not_effective
 0.56% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     13: predicate.environ_get_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.90% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.75% :     0.000008s :     29: predicate.inline
 0.96% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      8: predicate.less_batch_normalization
 1.61% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.94% :     0.000003s :     21: predicate.load_eliminater
 1.34% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.18% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     13: predicate.merge_addn
 1.13% :     0.000001s :     13: predicate.minmaximum_grad
 1.56% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.48% :     0.000003s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     13: predicate.reduce_eliminate
 1.67% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.75% :     0.000001s :      8: predicate.replace_old_param
 0.32% :     0.000000s :      4: predicate.reset_defer_inline
 1.23% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     13: predicate.same_eliminate
 0.76% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      8: predicate.special_op_eliminate
 1.01% :     0.000001s :      8: predicate.specialize_transform
 1.31% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     21: predicate.switch_defer_inline
 2.20% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.61% :     0.000010s :     67: predicate.switch_simplify
 1.20% :     0.000002s :     13: predicate.tile_eliminate
 1.26% :     0.000002s :     13: predicate.transpose_eliminate
 1.52% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.85% :     0.000002s :     21: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000582    12
46.42% :     0.000270s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.06% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.52% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014077    72
 0.32% :     0.000046s :      1: add_recomputation
 0.45% :     0.000064s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 2.08% :     0.000292s :      1: bootstrap
 0.17% :     0.000024s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000007s :      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
 0.05% :     0.000007s :      1: graph_reusing
37.11% :     0.005224s :      1: jit_opt_a
 1.06% :     0.000149s :      1: jit_opt_after_cconv
 0.40% :     0.000056s :      1: jit_opt_b
 2.99% :     0.000421s :      1: loop_unroll
 3.73% :     0.000525s :      1: mutable_eliminate
 7.05% :     0.000993s :     26: opt.transform.jit_opt_a
 0.37% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000029s :      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
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.24% :     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
 0.25% :     0.000035s :      1: py_interpret_to_execute
 0.11% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000015s :      1: remove_dup_value
 2.63% :     0.000371s :      1: renormalize.infer
 1.98% :     0.000279s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000036s :      1: rewriter_after_opt_a
 0.63% :     0.000088s :      1: rewriter_before_opt_a
 0.54% :     0.000076s :      1: symbol_engine_optimizer
32.85% :     0.004625s :      1: type_inference


TotalTime = 0.0689847, [33]
[bootstrap]: 0.00021499
[type_inference]: 0.0293747
[event_method]: 0.00018894
[auto_monad]: 0.00017262
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.17001e-06
[py_interpret_to_execute]: 3.733e-05
[rewriter_before_opt_a]: 0.00012383
[expand_dump_flag]: 2.19999e-06
[jit_opt_a]: 0.0282682, [3]
    [Cycle 1]: 0.0212573, [27]
        [switch_simplify]: 7.044e-05
        [loop_unroll]: 5.041e-05
        [a_1]: 0.0010749
        [with_stream_mark]: 2.629e-05
        [recompute_prepare]: 2.605e-05
        [updatestate_depend_eliminate]: 1.194e-05
        [updatestate_assign_eliminate]: 1.05e-05
        [updatestate_loads_eliminate]: 9.38002e-06
        [parameter_eliminate]: 2.30002e-06
        [specialize_transform]: 2.009e-05
        [updatestate_useless_node_eliminater]: 2.31e-05
        [accelerated_algorithm]: 1.982e-05
        [meta_shard_fg_expand]: 4.85001e-06
        [get_grad_eliminate_]: 1.853e-05
        [merge_forward]: 1.056e-05
        [cell_reuse_recompute_pass]: 8.39995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 3.531e-05
        [j_node_and_user_rematch]: 3.361e-05
        [meta_fg_expand]: 0.00590155
        [replace_old_param]: 0.00011649
        [inline_without_move]: 0.00011516
        [renormalize]: 0.0128887
        [add_forward_monad_depend]: 1.534e-05
        [auto_monad_grad]: 1.034e-05
        [auto_monad_eliminator]: 7.91e-05
        [cse]: 0.00028085
        [replace_applicator]: 0.00016552
    [Cycle 2]: 0.0030369, [27]
        [switch_simplify]: 6.087e-05
        [loop_unroll]: 5.838e-05
        [a_1]: 0.0009934
        [with_stream_mark]: 1.649e-05
        [recompute_prepare]: 1.18e-05
        [updatestate_depend_eliminate]: 2.58e-05
        [updatestate_assign_eliminate]: 5.54998e-06
        [updatestate_loads_eliminate]: 4.3e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 1.036e-05
        [updatestate_useless_node_eliminater]: 1.181e-05
        [accelerated_algorithm]: 8.83001e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 8.28001e-06
        [merge_forward]: 4.95001e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.65e-05
        [j_node_and_user_rematch]: 1.371e-05
        [meta_fg_expand]: 7.679e-05
        [replace_old_param]: 1.308e-05
        [inline_without_move]: 8.54e-06
        [renormalize]: 0.00140155
        [add_forward_monad_depend]: 5.49e-06
        [auto_monad_grad]: 1.71e-06
        [auto_monad_eliminator]: 1.848e-05
        [cse]: 7.445e-05
        [replace_applicator]: 1.734e-05
    [Cycle 3]: 0.00049132, [27]
        [switch_simplify]: 9.47001e-06
        [loop_unroll]: 8.52e-06
        [a_1]: 0.00017588
        [with_stream_mark]: 1.385e-05
        [recompute_prepare]: 8.33999e-06
        [updatestate_depend_eliminate]: 5.76e-06
        [updatestate_assign_eliminate]: 5.07999e-06
        [updatestate_loads_eliminate]: 4.3e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 8.54e-06
        [updatestate_useless_node_eliminater]: 1.116e-05
        [accelerated_algorithm]: 8.37998e-06
        [meta_shard_fg_expand]: 2.38998e-06
        [get_grad_eliminate_]: 7.66999e-06
        [merge_forward]: 4.22e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.653e-05
        [j_node_and_user_rematch]: 1.324e-05
        [meta_fg_expand]: 3.13e-06
        [replace_old_param]: 1.085e-05
        [inline_without_move]: 7.52002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 9.49978e-07
        [auto_monad_eliminator]: 1.079e-05
        [cse]: 2.381e-05
        [replace_applicator]: 7.93001e-06
[py_interpret_to_execute_after_opt_a]: 1.552e-05
[rewriter_after_opt_a]: 5.825e-05
[convert_after_rewriter]: 9.14998e-06
[order_py_execute_after_rewriter]: 6.19999e-06
[mutable_eliminate]: 0.00077996
[jit_opt_b]: 7.006e-05, [1]
    [Cycle 1]: 6.297e-05, [2]
        [frontend_op_eliminate]: 2.611e-05
        [inline_after_opt_a]: 2.513e-05
[cconv]: 2.477e-05
[loop_unroll]: 0.00047545
[jit_opt_after_cconv]: 0.00020595, [1]
    [Cycle 1]: 0.00019884, [11]
        [c_1]: 4.876e-05
        [parameter_eliminate]: 2.86999e-06
        [updatestate_depend_eliminate]: 7.06999e-06
        [updatestate_assign_eliminate]: 4.69998e-06
        [updatestate_loads_eliminate]: 4.2e-06
        [cse]: 3.46e-05
        [call_graph_tuple_transform]: 2.455e-05
        [tuple_list_get_item_eliminator]: 8.74e-06
        [none_parameter_eliminate]: 1.26002e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 8.17e-06
[remove_dup_value]: 2.145e-05
[partial_unused_args_eliminate]: 2.84999e-06
[environ_conv]: 9.64e-06
[add_recomputation]: 7.341e-05
[cse_after_recomputation]: 3.798e-05, [1]
    [Cycle 1]: 3.131e-05, [1]
        [cse]: 2.44e-05
[auto_monad_reorder]: 2.436e-05
[get_jit_bprop_graph]: 1.35999e-06
[rewriter_after_jit_bprop_graph]: 5.81e-06
[opt_after_jit_grad]: 0.00052564
[symbol_engine_optimizer]: 0.00010836, [1]
    [Cycle 1]: 0.00010177, [6]
        [build]: 1.148e-05
        [elim_shapecalc]: 1.304e-05
        [elim_not_effective]: 2.063e-05
        [opt_reshape]: 1.041e-05
        [fold_const_symbol]: 1.545e-05
        [renormalize]: 4.59986e-07
[validate]: 5.137e-05
[backend_pass]: 1.25001e-06
[task_emit]: 0.00783556
[execute]: 5.53002e-06
Sums
    bootstrap                            :     0.000215s :  0.33%
    type_inference                       :     0.029375s : 45.46%
    event_method                         :     0.000189s :  0.29%
    auto_monad                           :     0.000173s :  0.27%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000037s :  0.06%
    rewriter_before_opt_a                :     0.000124s :  0.19%
    expand_dump_flag                     :     0.000002s :  0.00%
    jit_opt_a.switch_simplify            :     0.000141s :  0.22%
    jit_opt_a.loop_unroll                :     0.000117s :  0.18%
    jit_opt_a.a_1                        :     0.002244s :  3.47%
    jit_opt_a.with_stream_mark           :     0.000057s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000046s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000043s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000021s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000018s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000039s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000046s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000037s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000034s :  0.05%
    jit_opt_a.merge_forward              :     0.000020s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000068s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000061s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.005981s :  9.26%
    jit_opt_a.replace_old_param          :     0.000140s :  0.22%
    jit_opt_a.inline_without_move        :     0.000131s :  0.20%
    jit_opt_a.renormalize                :     0.014290s : 22.12%
    jit_opt_a.add_forward_monad_depend   :     0.000022s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000013s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000108s :  0.17%
    jit_opt_a.cse                        :     0.000379s :  0.59%
    jit_opt_a.replace_applicator         :     0.000191s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.02%
    rewriter_after_opt_a                 :     0.000058s :  0.09%
    convert_after_rewriter               :     0.000009s :  0.01%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000780s :  1.21%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.04%
    cconv                                :     0.000025s :  0.04%
    loop_unroll                          :     0.000475s :  0.74%
    jit_opt_after_cconv.c_1              :     0.000049s :  0.08%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  0.01%
    remove_dup_value                     :     0.000021s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000010s :  0.01%
    add_recomputation                    :     0.000073s :  0.11%
    cse_after_recomputation.cse          :     0.000024s :  0.04%
    auto_monad_reorder                   :     0.000024s :  0.04%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000526s :  0.81%
    symbol_engine_optimizer.build        :     0.000011s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000051s :  0.08%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007836s : 12.13%
    execute                              :     0.000006s :  0.01%


Time group info:
------[substitution.]   0.001416   160
 1.34% :     0.000019s :      8: substitution.depend_value_elim
 0.21% :     0.000003s :      4: substitution.elim_not_effective
 0.16% :     0.000002s :      4: substitution.fold_const_symbol
57.40% :     0.000813s :      4: substitution.getattr_setattr_resolve
 0.46% :     0.000006s :      5: substitution.graph_param_transform
23.86% :     0.000338s :     10: substitution.inline
 2.19% :     0.000031s :      4: substitution.inline_without_move
 0.80% :     0.000011s :     20: substitution.j_node_and_user_rematch
 0.66% :     0.000009s :      5: substitution.minmaximum_grad
 0.29% :     0.000004s :      3: substitution.partial_eliminate
 0.95% :     0.000013s :     20: substitution.remove_not_recompute_node
 3.24% :     0.000046s :     13: substitution.replace_applicator
 0.90% :     0.000013s :     18: substitution.replace_old_param
 0.20% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.34% :     0.000019s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.85% :     0.000012s :      5: substitution.tuple_list_get_item_depend_reorder
 2.18% :     0.000031s :     10: substitution.tuple_list_get_item_eliminator
 1.10% :     0.000016s :      8: substitution.updatestate_pure_node_eliminater
 1.89% :     0.000027s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.029293     2
94.10% :     0.027565s :      1: type_inference.infer
 5.90% :     0.001729s :      1: type_inference.specialize
------[replace.]   0.000286    20
16.51% :     0.000047s :      3: replace.getattr_setattr_resolve
30.59% :     0.000087s :     10: replace.inline
12.27% :     0.000035s :      1: replace.replace_applicator
32.31% :     0.000092s :      5: replace.tuple_list_get_item_eliminator
 8.32% :     0.000024s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001127    20
67.85% :     0.000765s :      3: match.getattr_setattr_resolve
29.35% :     0.000331s :     10: match.inline
 0.84% :     0.000009s :      1: match.replace_applicator
 1.24% :     0.000014s :      5: match.tuple_list_get_item_eliminator
 0.72% :     0.000008s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000453  2979
 1.43% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.51% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.25% :     0.000006s :     46: predicate.addn_check_dump
 1.77% :     0.000008s :     46: predicate.addn_zero_filter
 2.05% :     0.000009s :     46: predicate.arithmetic_simplify
 1.62% :     0.000007s :     46: predicate.cast_eliminate
 0.22% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.25% :     0.000006s :     46: predicate.compare_switch_simplify
 1.68% :     0.000008s :     46: predicate.depend_value_elim
 1.39% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.42% :     0.000006s :     46: predicate.dict_get_item_eliminator
 1.50% :     0.000007s :     46: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      5: predicate.elim_not_effective
 0.31% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.35% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.23% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.35% :     0.000006s :     46: predicate.environ_get_eliminate
 1.26% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 1.08% :     0.000005s :     27: predicate.get_grad_eliminate
 1.30% :     0.000006s :     20: predicate.getattr_setattr_resolve
 0.10% :     0.000000s :      5: predicate.graph_param_transform
 3.79% :     0.000017s :     72: predicate.inline
 3.32% :     0.000015s :     91: predicate.inline_without_move
 0.43% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.35% :     0.000006s :     27: predicate.less_batch_normalization
 1.69% :     0.000008s :     51: predicate.list_to_tuple_eliminator_
 1.82% :     0.000008s :     56: predicate.load_eliminater
 0.51% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.78% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.74% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000006s :     46: predicate.merge_addn
 1.37% :     0.000006s :     46: predicate.minmaximum_grad
 0.59% :     0.000003s :      5: predicate.mutable_eliminate
 0.21% :     0.000001s :      5: predicate.opt_reshape
 2.09% :     0.000009s :     56: predicate.partial_eliminate
 1.56% :     0.000007s :     46: predicate.print_const_string_wrapper
 1.81% :     0.000008s :     46: predicate.reduce_eliminate
 1.60% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000002s :     27: predicate.remove_not_recompute_node
 2.77% :     0.000013s :    126: predicate.replace_applicator
 1.65% :     0.000007s :     91: predicate.replace_old_param
 0.13% :     0.000001s :      5: predicate.reset_defer_inline
 1.37% :     0.000006s :     46: predicate.reshape_eliminate
 1.46% :     0.000007s :     46: predicate.row_tensor_add_zeros_like
 0.39% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.39% :     0.000006s :     46: predicate.same_eliminate
 0.56% :     0.000003s :     29: predicate.set_cell_output_no_recompute
 0.54% :     0.000002s :     10: predicate.special_op_eliminate
 1.02% :     0.000005s :     27: predicate.specialize_transform
 1.75% :     0.000008s :     46: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.22% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.37% :     0.000011s :     62: predicate.switch_defer_inline
 2.05% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.11% :     0.000028s :    188: predicate.switch_simplify
 1.40% :     0.000006s :     46: predicate.tile_eliminate
 1.44% :     0.000007s :     46: predicate.transpose_eliminate
 1.76% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.60% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 2.71% :     0.000012s :     61: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000009s :     46: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.68% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000014s :     84: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000008s :     46: predicate.value_based_eliminate
 0.16% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003445    48
68.10% :     0.002346s :     27: func_graph_cloner_run.FuncGraphClonerGraph
31.90% :     0.001099s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.087625    93
 0.09% :     0.000077s :      1: add_recomputation
 0.21% :     0.000181s :      1: auto_monad
 0.03% :     0.000027s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.26% :     0.000232s :      1: bootstrap
 0.03% :     0.000027s :      1: cconv
 0.01% :     0.000012s :      1: convert_after_rewriter
 0.05% :     0.000040s :      1: cse_after_recomputation
 0.01% :     0.000012s :      1: environ_conv
 0.23% :     0.000200s :      1: event_method
 0.01% :     0.000009s :      1: execute
 0.00% :     0.000004s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
32.27% :     0.028272s :      1: jit_opt_a
 0.24% :     0.000209s :      1: jit_opt_after_cconv
 0.08% :     0.000073s :      1: jit_opt_b
 0.55% :     0.000483s :      1: loop_unroll
 0.90% :     0.000789s :      1: mutable_eliminate
 3.71% :     0.003250s :     39: opt.transform.jit_opt_a
 0.10% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000044s :      4: opt.transform.jit_opt_b
 0.02% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 1.04% :     0.000909s :      2: opt.transform.opt_resolve
 0.06% :     0.000055s :      4: opt.transform.symbol_engine_opt
 0.61% :     0.000533s :      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
 0.05% :     0.000040s :      1: py_interpret_to_execute
 0.02% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000024s :      1: remove_dup_value
12.50% :     0.010950s :      2: renormalize.infer
 3.79% :     0.003323s :      2: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000062s :      1: rewriter_after_opt_a
 0.15% :     0.000127s :      1: rewriter_before_opt_a
 0.13% :     0.000111s :      1: symbol_engine_optimizer
 8.96% :     0.007848s :      1: task_emit
33.54% :     0.029387s :      1: type_inference
 0.11% :     0.000098s :      1: validate

[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.885.387 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffe2caf6280> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.886.077 [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 = 0.0120778, [30]
[bootstrap]: 0.00027115
[type_inference]: 0.00445842
[event_method]: 2.137e-05
[auto_monad]: 6.156e-05
[graph_reusing]: 5.47999e-06
[pre_auto_parallel]: 2.19001e-06
[py_interpret_to_execute]: 3.18e-05
[rewriter_before_opt_a]: 8.384e-05
[expand_dump_flag]: 2.59999e-06
[jit_opt_a]: 0.00513443, [2]
    [Cycle 1]: 0.00161715, [27]
        [switch_simplify]: 5.27e-05
        [loop_unroll]: 3.452e-05
        [a_1]: 0.00059857
        [with_stream_mark]: 1.383e-05
        [recompute_prepare]: 7.88001e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 7.08e-06
        [updatestate_useless_node_eliminater]: 5.74e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 6.07999e-06
        [merge_forward]: 2.84999e-06
        [cell_reuse_recompute_pass]: 8.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.249e-05
        [j_node_and_user_rematch]: 9.46e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 1.035e-05
        [inline_without_move]: 5.91998e-06
        [renormalize]: 0.00061652
        [add_forward_monad_depend]: 4.50001e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.157e-05
        [cse]: 2e-05
        [replace_applicator]: 1.309e-05
    [Cycle 2]: 0.00034471, [27]
        [switch_simplify]: 6.31998e-06
        [loop_unroll]: 5.57999e-06
        [a_1]: 0.00011237
        [with_stream_mark]: 8.83001e-06
        [recompute_prepare]: 5.54e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.58002e-06
        [updatestate_useless_node_eliminater]: 5.55001e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 5.26002e-06
        [merge_forward]: 2.47001e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.273e-05
        [j_node_and_user_rematch]: 8.50001e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 8.48999e-06
        [inline_without_move]: 5.34998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 6.21e-06
        [cse]: 1.291e-05
        [replace_applicator]: 5.51e-06
[py_interpret_to_execute_after_opt_a]: 9.38002e-06
[rewriter_after_opt_a]: 3.187e-05
[convert_after_rewriter]: 6.53e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00049624
[jit_opt_b]: 5.63e-05, [1]
    [Cycle 1]: 4.974e-05, [2]
        [frontend_op_eliminate]: 1.986e-05
        [inline_after_opt_a]: 1.735e-05
[cconv]: 1.982e-05
[loop_unroll]: 0.00041007
[jit_opt_after_cconv]: 0.00015806, [1]
    [Cycle 1]: 0.00015212, [11]
        [c_1]: 2.489e-05
        [parameter_eliminate]: 2.14999e-06
        [updatestate_depend_eliminate]: 6.03998e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.011e-05
        [call_graph_tuple_transform]: 2.005e-05
        [tuple_list_get_item_eliminator]: 6.12001e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 6.02001e-06
[remove_dup_value]: 1.352e-05
[partial_unused_args_eliminate]: 1.92999e-06
[environ_conv]: 5.25001e-06
[add_recomputation]: 3.84e-05
[cse_after_recomputation]: 2.269e-05, [1]
    [Cycle 1]: 1.739e-05, [1]
        [cse]: 1.17e-05
[auto_monad_reorder]: 1.639e-05
[get_jit_bprop_graph]: 1.40001e-06
[rewriter_after_jit_bprop_graph]: 5.11997e-06
[opt_after_jit_grad]: 0.00044724
[symbol_engine_optimizer]: 7.491e-05, [1]
    [Cycle 1]: 6.92e-05, [6]
        [build]: 2.98e-06
        [elim_shapecalc]: 8.69e-06
        [elim_not_effective]: 1.344e-05
        [opt_reshape]: 6.17001e-06
        [fold_const_symbol]: 9.41e-06
        [renormalize]: 4.30009e-07
[validate]: 3.282e-05
Sums
    bootstrap                            :     0.000271s :  3.24%
    type_inference                       :     0.004458s : 53.35%
    event_method                         :     0.000021s :  0.26%
    auto_monad                           :     0.000062s :  0.74%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.38%
    rewriter_before_opt_a                :     0.000084s :  1.00%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000059s :  0.71%
    jit_opt_a.loop_unroll                :     0.000040s :  0.48%
    jit_opt_a.a_1                        :     0.000711s :  8.51%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.27%
    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 :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  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_        :     0.000011s :  0.14%
    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 :     0.000025s :  0.30%
    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          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000617s :  7.38%
    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                        :     0.000033s :  0.39%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000032s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000496s :  5.94%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000020s :  0.24%
    loop_unroll                          :     0.000410s :  4.91%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.30%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.24%
    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  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000038s :  0.46%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000016s :  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 :  5.35%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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                             :     0.000033s :  0.39%


Time group info:
------[substitution.]   0.000198    28
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000005s :      4: substitution.graph_param_transform
83.13% :     0.000165s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.28% :     0.000005s :      4: substitution.replace_old_param
 7.26% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004405     2
80.53% :     0.003547s :      1: type_inference.infer
19.47% :     0.000858s :      1: type_inference.specialize
------[replace.]   0.000078     8
56.41% :     0.000044s :      4: replace.inline
43.59% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     8
92.94% :     0.000162s :      4: match.inline
 7.06% :     0.000012s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   912
 1.22% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     13: predicate.addn_check_dump
 1.29% :     0.000002s :     13: predicate.addn_zero_filter
 1.80% :     0.000002s :     13: predicate.arithmetic_simplify
 1.29% :     0.000002s :     13: predicate.cast_eliminate
 0.47% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     13: predicate.compare_switch_simplify
 1.20% :     0.000002s :     13: predicate.depend_value_elim
 1.30% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      4: predicate.elim_not_effective
 0.53% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     13: predicate.environ_get_eliminate
 1.18% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 1.14% :     0.000002s :      8: predicate.get_grad_eliminate
 0.40% :     0.000001s :      4: predicate.graph_param_transform
 5.34% :     0.000007s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      8: predicate.less_batch_normalization
 1.60% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     21: predicate.load_eliminater
 1.44% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     42: predicate.loop_unroll_before_grad
 2.19% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.50% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.57% :     0.000003s :     21: predicate.partial_eliminate
 1.19% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.87% :     0.000002s :     25: predicate.replace_applicator
 0.63% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.29% :     0.000002s :     13: predicate.reshape_eliminate
 1.17% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      8: predicate.special_op_eliminate
 1.07% :     0.000001s :      8: predicate.specialize_transform
 1.33% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     21: predicate.switch_defer_inline
 2.44% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.48% :     0.000010s :     67: predicate.switch_simplify
 1.34% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.42% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000562    12
46.78% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.48% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.74% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013719    72
 0.30% :     0.000041s :      1: add_recomputation
 0.47% :     0.000065s :      1: auto_monad
 0.14% :     0.000019s :      1: auto_monad_reorder
 2.07% :     0.000285s :      1: bootstrap
 0.16% :     0.000022s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.18% :     0.000025s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.45% :     0.005137s :      1: jit_opt_a
 1.17% :     0.000161s :      1: jit_opt_after_cconv
 0.43% :     0.000059s :      1: jit_opt_b
 3.05% :     0.000418s :      1: loop_unroll
 3.68% :     0.000505s :      1: mutable_eliminate
 6.84% :     0.000939s :     26: opt.transform.jit_opt_a
 0.39% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000031s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.32% :     0.000455s :      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% :     0.000035s :      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
 2.52% :     0.000346s :      1: renormalize.infer
 1.92% :     0.000264s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000035s :      1: rewriter_after_opt_a
 0.63% :     0.000087s :      1: rewriter_before_opt_a
 0.57% :     0.000078s :      1: symbol_engine_optimizer
32.58% :     0.004470s :      1: type_inference


TotalTime = 0.03792, [33]
[bootstrap]: 0.0002046
[type_inference]: 0.0110091
[event_method]: 4.927e-05
[auto_monad]: 9.862e-05
[graph_reusing]: 6.44001e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 2.554e-05
[rewriter_before_opt_a]: 0.00010863
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.016948, [3]
    [Cycle 1]: 0.0108139, [27]
        [switch_simplify]: 6.498e-05
        [loop_unroll]: 4.41e-05
        [a_1]: 0.00093445
        [with_stream_mark]: 2.775e-05
        [recompute_prepare]: 2.043e-05
        [updatestate_depend_eliminate]: 8.35001e-06
        [updatestate_assign_eliminate]: 7.23e-06
        [updatestate_loads_eliminate]: 6.71999e-06
        [parameter_eliminate]: 2.79001e-06
        [specialize_transform]: 1.522e-05
        [updatestate_useless_node_eliminater]: 1.386e-05
        [accelerated_algorithm]: 1.46e-05
        [meta_shard_fg_expand]: 3.58e-06
        [get_grad_eliminate_]: 1.425e-05
        [merge_forward]: 8.12e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.905e-05
        [j_node_and_user_rematch]: 2.411e-05
        [meta_fg_expand]: 0.00162925
        [replace_old_param]: 6.591e-05
        [inline_without_move]: 5.777e-05
        [renormalize]: 0.00690187
        [add_forward_monad_depend]: 1.162e-05
        [auto_monad_grad]: 6.54999e-06
        [auto_monad_eliminator]: 5.773e-05
        [cse]: 0.00051814
        [replace_applicator]: 8.737e-05
    [Cycle 2]: 0.00227219, [27]
        [switch_simplify]: 4.198e-05
        [loop_unroll]: 3.95e-05
        [a_1]: 0.00111656
        [with_stream_mark]: 1.71e-05
        [recompute_prepare]: 8.06001e-06
        [updatestate_depend_eliminate]: 4.26001e-06
        [updatestate_assign_eliminate]: 3.20998e-06
        [updatestate_loads_eliminate]: 3.71001e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 6.92002e-06
        [updatestate_useless_node_eliminater]: 6.01998e-06
        [accelerated_algorithm]: 6.39001e-06
        [meta_shard_fg_expand]: 2.24999e-06
        [get_grad_eliminate_]: 6.23998e-06
        [merge_forward]: 4.27e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.537e-05
        [j_node_and_user_rematch]: 9.61e-06
        [meta_fg_expand]: 4.817e-05
        [replace_old_param]: 1.21e-05
        [inline_without_move]: 6.21e-06
        [renormalize]: 0.0006998
        [add_forward_monad_depend]: 5.62001e-06
        [auto_monad_grad]: 1.62999e-06
        [auto_monad_eliminator]: 1.213e-05
        [cse]: 2.502e-05
        [replace_applicator]: 1.405e-05
    [Cycle 3]: 0.00035628, [27]
        [switch_simplify]: 6.83998e-06
        [loop_unroll]: 6.23998e-06
        [a_1]: 0.00011492
        [with_stream_mark]: 8.77e-06
        [recompute_prepare]: 6.02001e-06
        [updatestate_depend_eliminate]: 3.11999e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.93002e-06
        [updatestate_useless_node_eliminater]: 5.63002e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 1.36998e-06
        [get_grad_eliminate_]: 5.56998e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.42e-05
        [j_node_and_user_rematch]: 9.04e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 9.25001e-06
        [inline_without_move]: 5.67001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 6.39993e-07
        [auto_monad_eliminator]: 6.31e-06
        [cse]: 1.399e-05
        [replace_applicator]: 6.19001e-06
[py_interpret_to_execute_after_opt_a]: 1.268e-05
[rewriter_after_opt_a]: 4.028e-05
[convert_after_rewriter]: 7.19001e-06
[order_py_execute_after_rewriter]: 5.27001e-06
[mutable_eliminate]: 0.00075195
[jit_opt_b]: 5.775e-05, [1]
    [Cycle 1]: 4.904e-05, [2]
        [frontend_op_eliminate]: 1.876e-05
        [inline_after_opt_a]: 1.843e-05
[cconv]: 2.593e-05
[loop_unroll]: 0.00046315
[jit_opt_after_cconv]: 0.00015914, [1]
    [Cycle 1]: 0.00015228, [11]
        [c_1]: 2.524e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 6.39001e-06
        [updatestate_assign_eliminate]: 3.29001e-06
        [updatestate_loads_eliminate]: 3.01001e-06
        [cse]: 2.523e-05
        [call_graph_tuple_transform]: 2.207e-05
        [tuple_list_get_item_eliminator]: 7.26999e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 6.29001e-06
[remove_dup_value]: 1.692e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 5.97999e-06
[add_recomputation]: 4.671e-05
[cse_after_recomputation]: 2.725e-05, [1]
    [Cycle 1]: 2.14e-05, [1]
        [cse]: 1.54e-05
[auto_monad_reorder]: 1.903e-05
[get_jit_bprop_graph]: 1.97001e-06
[rewriter_after_jit_bprop_graph]: 6.23e-06
[opt_after_jit_grad]: 0.00049231
[symbol_engine_optimizer]: 7.828e-05, [1]
    [Cycle 1]: 7.189e-05, [6]
        [build]: 4.01001e-06
        [elim_shapecalc]: 8.98002e-06
        [elim_not_effective]: 1.455e-05
        [opt_reshape]: 6.43e-06
        [fold_const_symbol]: 9.61998e-06
        [renormalize]: 5.09986e-07
[validate]: 4.08e-05
[backend_pass]: 1.01997e-06
[task_emit]: 0.00698773
[execute]: 6.79999e-06
Sums
    bootstrap                            :     0.000205s :  0.61%
    type_inference                       :     0.011009s : 32.78%
    event_method                         :     0.000049s :  0.15%
    auto_monad                           :     0.000099s :  0.29%
    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                :     0.000109s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000114s :  0.34%
    jit_opt_a.loop_unroll                :     0.000090s :  0.27%
    jit_opt_a.a_1                        :     0.002166s :  6.45%
    jit_opt_a.with_stream_mark           :     0.000054s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.10%
    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 :     0.000013s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.08%
    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_        :     0.000026s :  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 :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.001679s :  5.00%
    jit_opt_a.replace_old_param          :     0.000087s :  0.26%
    jit_opt_a.inline_without_move        :     0.000070s :  0.21%
    jit_opt_a.renormalize                :     0.007602s : 22.63%
    jit_opt_a.add_forward_monad_depend   :     0.000018s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000076s :  0.23%
    jit_opt_a.cse                        :     0.000557s :  1.66%
    jit_opt_a.replace_applicator         :     0.000108s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000752s :  2.24%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.05%
    cconv                                :     0.000026s :  0.08%
    loop_unroll                          :     0.000463s :  1.38%
    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              :     0.000025s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  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  :     0.000006s :  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                    :     0.000047s :  0.14%
    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       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000492s :  1.47%
    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 :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006988s : 20.80%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000552   113
 0.36% :     0.000002s :      2: substitution.elim_not_effective
 0.24% :     0.000001s :      2: substitution.fold_const_symbol
 1.02% :     0.000006s :      4: substitution.graph_param_transform
64.38% :     0.000355s :     12: substitution.inline
 3.10% :     0.000017s :      2: substitution.inline_without_move
 1.51% :     0.000008s :     12: substitution.j_node_and_user_rematch
 2.00% :     0.000011s :      7: substitution.minmaximum_grad
 1.36% :     0.000007s :      5: substitution.partial_eliminate
 1.85% :     0.000010s :     12: substitution.remove_not_recompute_node
 5.67% :     0.000031s :     10: substitution.replace_applicator
 1.98% :     0.000011s :     14: substitution.replace_old_param
 0.55% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.47% :     0.000025s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.80% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.71% :     0.000048s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010931     2
89.34% :     0.009767s :      1: type_inference.infer
10.66% :     0.001165s :      1: type_inference.specialize
------[replace.]   0.000179    21
60.21% :     0.000108s :     12: replace.inline
39.79% :     0.000071s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000373    21
93.36% :     0.000348s :     12: match.inline
 6.64% :     0.000025s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000382  2828
 1.54% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.39% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000005s :     47: predicate.addn_check_dump
 1.54% :     0.000006s :     47: predicate.addn_zero_filter
 2.09% :     0.000008s :     47: predicate.arithmetic_simplify
 1.60% :     0.000006s :     47: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.55% :     0.000006s :     47: predicate.compare_switch_simplify
 1.45% :     0.000006s :     47: predicate.depend_value_elim
 1.42% :     0.000005s :     47: predicate.dict_get_item_const_eliminator
 1.55% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.49% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.34% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      4: predicate.elim_not_effective
 0.18% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.41% :     0.000005s :     47: predicate.environ_get_add_eliminate
 1.43% :     0.000005s :     47: predicate.environ_get_depend_swap
 1.49% :     0.000006s :     47: predicate.environ_get_eliminate
 1.42% :     0.000005s :     47: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.86% :     0.000003s :     21: predicate.get_grad_eliminate
 0.15% :     0.000001s :      4: predicate.graph_param_transform
 4.44% :     0.000017s :     76: predicate.inline
 1.94% :     0.000007s :     47: predicate.inline_without_move
 0.40% :     0.000002s :     21: predicate.j_node_and_user_rematch
 1.03% :     0.000004s :     21: predicate.less_batch_normalization
 1.82% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 1.96% :     0.000008s :     60: predicate.load_eliminater
 0.43% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.29% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.72% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.41% :     0.000005s :     47: predicate.merge_addn
 1.52% :     0.000006s :     47: predicate.minmaximum_grad
 0.68% :     0.000003s :      4: predicate.mutable_eliminate
 0.15% :     0.000001s :      4: predicate.opt_reshape
 2.49% :     0.000009s :     60: predicate.partial_eliminate
 1.52% :     0.000006s :     47: predicate.print_const_string_wrapper
 1.84% :     0.000007s :     47: predicate.reduce_eliminate
 1.81% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.85% :     0.000011s :    108: predicate.replace_applicator
 1.03% :     0.000004s :     47: predicate.replace_old_param
 0.13% :     0.000000s :      4: predicate.reset_defer_inline
 1.47% :     0.000006s :     47: predicate.reshape_eliminate
 1.49% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.26% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.57% :     0.000006s :     47: predicate.same_eliminate
 0.51% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.83% :     0.000003s :     21: predicate.specialize_transform
 1.64% :     0.000006s :     47: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.68% :     0.000010s :     68: predicate.switch_defer_inline
 2.50% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.52% :     0.000025s :    169: predicate.switch_simplify
 1.46% :     0.000006s :     47: predicate.tile_eliminate
 1.47% :     0.000006s :     47: predicate.transpose_eliminate
 1.83% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.66% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000014s :     64: predicate.tuple_list_get_item_eliminator
 1.95% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.88% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000006s :     47: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.19% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001803    31
61.31% :     0.001105s :     15: func_graph_cloner_run.FuncGraphClonerGraph
38.69% :     0.000697s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048481    91
 0.12% :     0.000060s :      1: add_recomputation
 0.21% :     0.000103s :      1: auto_monad
 0.04% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.45% :     0.000219s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.11% :     0.000054s :      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
34.96% :     0.016951s :      1: jit_opt_a
 0.33% :     0.000162s :      1: jit_opt_after_cconv
 0.12% :     0.000060s :      1: jit_opt_b
 0.97% :     0.000472s :      1: loop_unroll
 1.57% :     0.000763s :      1: mutable_eliminate
 5.84% :     0.002830s :     39: opt.transform.jit_opt_a
 0.12% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000031s :      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
 0.05% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.03% :     0.000501s :      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
 0.06% :     0.000028s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
12.00% :     0.005817s :      2: renormalize.infer
 3.65% :     0.001767s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000043s :      1: rewriter_after_opt_a
 0.23% :     0.000112s :      1: rewriter_before_opt_a
 0.17% :     0.000081s :      1: symbol_engine_optimizer
14.44% :     0.007000s :      1: task_emit
22.74% :     0.011023s :      1: type_inference
 0.13% :     0.000063s :      1: validate

[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.900.817 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe9bb80> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.901.611 [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 = 0.0120531, [30]
[bootstrap]: 0.00025867
[type_inference]: 0.00440029
[event_method]: 2.011e-05
[auto_monad]: 5.304e-05
[graph_reusing]: 5.49998e-06
[pre_auto_parallel]: 2.01e-06
[py_interpret_to_execute]: 3.422e-05
[rewriter_before_opt_a]: 8.435e-05
[expand_dump_flag]: 2.46e-06
[jit_opt_a]: 0.00514527, [2]
    [Cycle 1]: 0.00160933, [27]
        [switch_simplify]: 5.253e-05
        [loop_unroll]: 3.509e-05
        [a_1]: 0.00059001
        [with_stream_mark]: 1.444e-05
        [recompute_prepare]: 8.1e-06
        [updatestate_depend_eliminate]: 3.05002e-06
        [updatestate_assign_eliminate]: 2.38998e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 6.66999e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 6.26998e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 5.86e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 8.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.179e-05
        [j_node_and_user_rematch]: 9.17999e-06
        [meta_fg_expand]: 1.92999e-06
        [replace_old_param]: 9.64e-06
        [inline_without_move]: 5.97001e-06
        [renormalize]: 0.00061835
        [add_forward_monad_depend]: 4.33001e-06
        [auto_monad_grad]: 1.81003e-06
        [auto_monad_eliminator]: 1.281e-05
        [cse]: 2.139e-05
        [replace_applicator]: 1.223e-05
    [Cycle 2]: 0.00035893, [27]
        [switch_simplify]: 6.51999e-06
        [loop_unroll]: 5.66e-06
        [a_1]: 0.00011131
        [with_stream_mark]: 9.32999e-06
        [recompute_prepare]: 5.60001e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 6.20002e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 5.09998e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.16002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.597e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 1.66998e-06
        [replace_old_param]: 9.57001e-06
        [inline_without_move]: 5.82001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 9.80013e-07
        [auto_monad_eliminator]: 6.69001e-06
        [cse]: 1.392e-05
        [replace_applicator]: 5.76998e-06
[py_interpret_to_execute_after_opt_a]: 1.161e-05
[rewriter_after_opt_a]: 3.328e-05
[convert_after_rewriter]: 6.71999e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00050605
[jit_opt_b]: 5.363e-05, [1]
    [Cycle 1]: 4.642e-05, [2]
        [frontend_op_eliminate]: 1.781e-05
        [inline_after_opt_a]: 1.716e-05
[cconv]: 2.028e-05
[loop_unroll]: 0.00041718
[jit_opt_after_cconv]: 0.00014981, [1]
    [Cycle 1]: 0.00014356, [11]
        [c_1]: 2.481e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.89999e-06
        [updatestate_assign_eliminate]: 2.96999e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [cse]: 2.248e-05
        [call_graph_tuple_transform]: 1.932e-05
        [tuple_list_get_item_eliminator]: 6.20002e-06
        [none_parameter_eliminate]: 1.04e-06
        [renormalize]: 5.49975e-07
        [switch_simplify]: 5.92001e-06
[remove_dup_value]: 1.137e-05
[partial_unused_args_eliminate]: 1.85001e-06
[environ_conv]: 5.25001e-06
[add_recomputation]: 3.849e-05
[cse_after_recomputation]: 3.586e-05, [1]
    [Cycle 1]: 3.072e-05, [1]
        [cse]: 2.471e-05
[auto_monad_reorder]: 1.524e-05
[get_jit_bprop_graph]: 1.62001e-06
[rewriter_after_jit_bprop_graph]: 5.26002e-06
[opt_after_jit_grad]: 0.0004695
[symbol_engine_optimizer]: 7.404e-05, [1]
    [Cycle 1]: 6.808e-05, [6]
        [build]: 3.3e-06
        [elim_shapecalc]: 8.47e-06
        [elim_not_effective]: 1.297e-05
        [opt_reshape]: 6.23e-06
        [fold_const_symbol]: 8.99998e-06
        [renormalize]: 4.30009e-07
[validate]: 3.407e-05
Sums
    bootstrap                            :     0.000259s :  3.11%
    type_inference                       :     0.004400s : 52.82%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000053s :  0.64%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.41%
    rewriter_before_opt_a                :     0.000084s :  1.01%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000059s :  0.71%
    jit_opt_a.loop_unroll                :     0.000041s :  0.49%
    jit_opt_a.a_1                        :     0.000701s :  8.42%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.29%
    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 :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    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       :     0.000003s :  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 :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.22%
    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        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000618s :  7.42%
    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      :     0.000020s :  0.23%
    jit_opt_a.cse                        :     0.000035s :  0.42%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.14%
    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                    :     0.000506s :  6.07%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000020s :  0.24%
    loop_unroll                          :     0.000417s :  5.01%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.30%
    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 :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000022s :  0.27%
    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 :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  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                    :     0.000038s :  0.46%
    cse_after_recomputation.cse          :     0.000025s :  0.30%
    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                   :     0.000469s :  5.64%
    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 :  0.41%


Time group info:
------[substitution.]   0.000191    28
 0.76% :     0.000001s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000005s :      4: substitution.graph_param_transform
81.87% :     0.000157s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 3.16% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000004s :      4: substitution.replace_old_param
 7.12% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004343     2
80.11% :     0.003480s :      1: type_inference.infer
19.89% :     0.000864s :      1: type_inference.specialize
------[replace.]   0.000080     8
55.90% :     0.000045s :      4: replace.inline
44.10% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     8
92.88% :     0.000154s :      4: match.inline
 7.12% :     0.000012s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   912
 1.32% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     13: predicate.addn_check_dump
 1.23% :     0.000002s :     13: predicate.addn_zero_filter
 1.74% :     0.000002s :     13: predicate.arithmetic_simplify
 1.28% :     0.000002s :     13: predicate.cast_eliminate
 0.45% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     13: predicate.compare_switch_simplify
 1.14% :     0.000001s :     13: predicate.depend_value_elim
 1.19% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.46% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     13: predicate.environ_get_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000001s :      8: predicate.get_grad_eliminate
 0.29% :     0.000000s :      4: predicate.graph_param_transform
 5.40% :     0.000007s :     29: predicate.inline
 1.00% :     0.000001s :      8: predicate.inline_without_move
 0.45% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      8: predicate.less_batch_normalization
 1.68% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     21: predicate.load_eliminater
 1.32% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.38% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     13: predicate.merge_addn
 1.11% :     0.000001s :     13: predicate.minmaximum_grad
 1.44% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.41% :     0.000003s :     21: predicate.partial_eliminate
 1.20% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     13: predicate.reduce_eliminate
 1.95% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.00% :     0.000003s :     25: predicate.replace_applicator
 0.66% :     0.000001s :      8: predicate.replace_old_param
 0.41% :     0.000001s :      4: predicate.reset_defer_inline
 1.23% :     0.000002s :     13: predicate.reshape_eliminate
 1.20% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     13: predicate.same_eliminate
 0.65% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      8: predicate.special_op_eliminate
 0.95% :     0.000001s :      8: predicate.specialize_transform
 1.29% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     21: predicate.switch_defer_inline
 2.24% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.56% :     0.000010s :     67: predicate.switch_simplify
 1.19% :     0.000002s :     13: predicate.tile_eliminate
 1.25% :     0.000002s :     13: predicate.transpose_eliminate
 1.51% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000552    12
46.50% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.77% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.73% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013686    72
 0.30% :     0.000041s :      1: add_recomputation
 0.41% :     0.000056s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 1.98% :     0.000271s :      1: bootstrap
 0.17% :     0.000023s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.28% :     0.000038s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.17% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.61% :     0.005148s :      1: jit_opt_a
 1.11% :     0.000152s :      1: jit_opt_after_cconv
 0.41% :     0.000057s :      1: jit_opt_b
 3.11% :     0.000425s :      1: loop_unroll
 3.76% :     0.000514s :      1: mutable_eliminate
 6.82% :     0.000933s :     26: opt.transform.jit_opt_a
 0.39% :     0.000053s :      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
 0.11% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.49% :     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% :     0.000037s :      1: py_interpret_to_execute
 0.10% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000014s :      1: remove_dup_value
 2.45% :     0.000335s :      1: renormalize.infer
 2.02% :     0.000276s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000037s :      1: rewriter_after_opt_a
 0.64% :     0.000087s :      1: rewriter_before_opt_a
 0.56% :     0.000077s :      1: symbol_engine_optimizer
32.23% :     0.004411s :      1: type_inference


TotalTime = 0.0437589, [33]
[bootstrap]: 0.0002814
[type_inference]: 0.0210117
[event_method]: 1.98e-05
[auto_monad]: 0.00015602
[graph_reusing]: 6.66e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 0.00012475
[rewriter_before_opt_a]: 9.796e-05
[expand_dump_flag]: 3.68e-06
[jit_opt_a]: 0.00625997, [2]
    [Cycle 1]: 0.00240335, [27]
        [switch_simplify]: 6.486e-05
        [loop_unroll]: 3.651e-05
        [a_1]: 0.00070996
        [with_stream_mark]: 2.134e-05
        [recompute_prepare]: 1.124e-05
        [updatestate_depend_eliminate]: 6.39999e-06
        [updatestate_assign_eliminate]: 5.25999e-06
        [updatestate_loads_eliminate]: 4.75001e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 9.20999e-06
        [updatestate_useless_node_eliminater]: 1.17e-05
        [accelerated_algorithm]: 8.89e-06
        [meta_shard_fg_expand]: 3.18e-06
        [get_grad_eliminate_]: 7.97e-06
        [merge_forward]: 5.87999e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.055e-05
        [j_node_and_user_rematch]: 1.35e-05
        [meta_fg_expand]: 3.56999e-06
        [replace_old_param]: 1.171e-05
        [inline_without_move]: 7.8e-06
        [renormalize]: 0.00116629
        [add_forward_monad_depend]: 5.85002e-06
        [auto_monad_grad]: 3.41001e-06
        [auto_monad_eliminator]: 2.122e-05
        [cse]: 3.985e-05
        [replace_applicator]: 1.776e-05
    [Cycle 2]: 0.00048536, [27]
        [switch_simplify]: 1.01e-05
        [loop_unroll]: 8.95001e-06
        [a_1]: 0.0001774
        [with_stream_mark]: 1.277e-05
        [recompute_prepare]: 8.26002e-06
        [updatestate_depend_eliminate]: 5.56998e-06
        [updatestate_assign_eliminate]: 4.40999e-06
        [updatestate_loads_eliminate]: 4.15999e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 8.88002e-06
        [updatestate_useless_node_eliminater]: 1.106e-05
        [accelerated_algorithm]: 8.35001e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 8.08001e-06
        [merge_forward]: 4.90001e-06
        [cell_reuse_recompute_pass]: 2.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.683e-05
        [j_node_and_user_rematch]: 1.303e-05
        [meta_fg_expand]: 2.99001e-06
        [replace_old_param]: 1.014e-05
        [inline_without_move]: 7.61999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 1.41002e-06
        [auto_monad_eliminator]: 9.98002e-06
        [cse]: 1.933e-05
        [replace_applicator]: 7.68001e-06
[py_interpret_to_execute_after_opt_a]: 1.389e-05
[rewriter_after_opt_a]: 0.00028665
[convert_after_rewriter]: 1.176e-05
[order_py_execute_after_rewriter]: 7.85e-06
[mutable_eliminate]: 0.00070236
[jit_opt_b]: 7.072e-05, [1]
    [Cycle 1]: 6.282e-05, [2]
        [frontend_op_eliminate]: 2.418e-05
        [inline_after_opt_a]: 2.571e-05
[cconv]: 2.659e-05
[loop_unroll]: 0.00047171
[jit_opt_after_cconv]: 0.0002066, [1]
    [Cycle 1]: 0.0001988, [11]
        [c_1]: 4.551e-05
        [parameter_eliminate]: 3.48999e-06
        [updatestate_depend_eliminate]: 8.06001e-06
        [updatestate_assign_eliminate]: 4.49998e-06
        [updatestate_loads_eliminate]: 4.02998e-06
        [cse]: 3.204e-05
        [call_graph_tuple_transform]: 2.482e-05
        [tuple_list_get_item_eliminator]: 9.20001e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 1.021e-05
[remove_dup_value]: 1.996e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 1.374e-05
[add_recomputation]: 6.906e-05
[cse_after_recomputation]: 3.081e-05, [1]
    [Cycle 1]: 2.496e-05, [1]
        [cse]: 1.853e-05
[auto_monad_reorder]: 2.564e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 6.69999e-06
[opt_after_jit_grad]: 0.0005455
[symbol_engine_optimizer]: 0.00010705, [1]
    [Cycle 1]: 0.00010037, [6]
        [build]: 1.462e-05
        [elim_shapecalc]: 1.169e-05
        [elim_not_effective]: 2.04e-05
        [opt_reshape]: 9.54e-06
        [fold_const_symbol]: 1.443e-05
        [renormalize]: 4.00003e-07
[validate]: 9.037e-05
[backend_pass]: 1.09e-06
[task_emit]: 0.0128643
[execute]: 7.51001e-06
Sums
    bootstrap                            :     0.000281s :  0.71%
    type_inference                       :     0.021012s : 52.86%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000156s :  0.39%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000125s :  0.31%
    rewriter_before_opt_a                :     0.000098s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000887s :  2.23%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    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  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.05%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001166s :  2.93%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000031s :  0.08%
    jit_opt_a.cse                        :     0.000059s :  0.15%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000287s :  0.72%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000702s :  1.77%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000027s :  0.07%
    loop_unroll                          :     0.000472s :  1.19%
    jit_opt_after_cconv.c_1              :     0.000046s :  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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.03%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000014s :  0.03%
    add_recomputation                    :     0.000069s :  0.17%
    cse_after_recomputation.cse          :     0.000019s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000545s :  1.37%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000090s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012864s : 32.36%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000278    47
 3.55% :     0.000010s :      2: substitution.depend_value_elim
 1.03% :     0.000003s :      4: substitution.elim_not_effective
 0.76% :     0.000002s :      4: substitution.fold_const_symbol
 2.54% :     0.000007s :      5: substitution.graph_param_transform
74.96% :     0.000208s :      4: substitution.inline
 1.61% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.66% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.56% :     0.000004s :      2: substitution.replace_old_param
 4.74% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.26% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.33% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.020943     2
92.80% :     0.019436s :      1: type_inference.infer
 7.20% :     0.001507s :      1: type_inference.specialize
------[replace.]   0.000092     6
66.93% :     0.000062s :      4: replace.inline
33.07% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
94.44% :     0.000205s :      4: match.inline
 5.56% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1021
 1.36% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     15: predicate.addn_check_dump
 1.26% :     0.000002s :     15: predicate.addn_zero_filter
 2.30% :     0.000004s :     15: predicate.arithmetic_simplify
 1.34% :     0.000002s :     15: predicate.cast_eliminate
 0.46% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     15: predicate.compare_switch_simplify
 1.35% :     0.000002s :     15: predicate.depend_value_elim
 1.11% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.87% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      5: predicate.elim_not_effective
 0.55% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     15: predicate.environ_get_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.31% :     0.000001s :      5: predicate.graph_param_transform
 4.98% :     0.000009s :     31: predicate.inline
 1.05% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :     10: predicate.less_batch_normalization
 1.37% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.91% :     0.000003s :     22: predicate.load_eliminater
 1.32% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.64% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.13% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000002s :     15: predicate.merge_addn
 1.14% :     0.000002s :     15: predicate.minmaximum_grad
 1.67% :     0.000003s :      5: predicate.mutable_eliminate
 0.57% :     0.000001s :      5: predicate.opt_reshape
 2.49% :     0.000004s :     22: predicate.partial_eliminate
 1.18% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.63% :     0.000003s :     15: predicate.reduce_eliminate
 1.40% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.73% :     0.000003s :     27: predicate.replace_applicator
 0.57% :     0.000001s :     10: predicate.replace_old_param
 0.42% :     0.000001s :      5: predicate.reset_defer_inline
 1.39% :     0.000002s :     15: predicate.reshape_eliminate
 1.31% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.03% :     0.000002s :     10: predicate.special_op_eliminate
 1.13% :     0.000002s :     10: predicate.specialize_transform
 1.57% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.58% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.23% :     0.000004s :     21: predicate.switch_defer_inline
 1.92% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.35% :     0.000011s :     68: predicate.switch_simplify
 1.40% :     0.000002s :     15: predicate.tile_eliminate
 1.31% :     0.000002s :     15: predicate.transpose_eliminate
 1.51% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.33% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.18% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000003s :     15: predicate.value_based_eliminate
 0.46% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000787    13
42.47% :     0.000334s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.44% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.09% :     0.000410s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046319    76
 0.16% :     0.000072s :      1: add_recomputation
 0.35% :     0.000161s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.63% :     0.000293s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000016s :      1: environ_conv
 0.05% :     0.000024s :      1: event_method
 0.02% :     0.000012s :      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
13.52% :     0.006263s :      1: jit_opt_a
 0.45% :     0.000210s :      1: jit_opt_after_cconv
 0.16% :     0.000074s :      1: jit_opt_b
 1.04% :     0.000480s :      1: loop_unroll
 1.54% :     0.000712s :      1: mutable_eliminate
 2.58% :     0.001196s :     26: opt.transform.jit_opt_a
 0.18% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      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
 0.11% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.20% :     0.000555s :      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
 0.28% :     0.000128s :      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
 1.53% :     0.000707s :      1: renormalize.infer
 0.97% :     0.000451s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000292s :      1: rewriter_after_opt_a
 0.22% :     0.000102s :      1: rewriter_before_opt_a
 0.24% :     0.000110s :      1: symbol_engine_optimizer
27.80% :     0.012875s :      1: task_emit
45.39% :     0.021025s :      1: type_inference
 0.25% :     0.000117s :      1: validate


TotalTime = 0.0385494, [33]
[bootstrap]: 0.00021389
[type_inference]: 0.010963
[event_method]: 4.446e-05
[auto_monad]: 9.676e-05
[graph_reusing]: 7.09001e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 2.635e-05
[rewriter_before_opt_a]: 0.00039794
[expand_dump_flag]: 4.22e-06
[jit_opt_a]: 0.0166399, [3]
    [Cycle 1]: 0.0109721, [27]
        [switch_simplify]: 6.65e-05
        [loop_unroll]: 4.377e-05
        [a_1]: 0.00094313
        [with_stream_mark]: 2.574e-05
        [recompute_prepare]: 2.044e-05
        [updatestate_depend_eliminate]: 8.38999e-06
        [updatestate_assign_eliminate]: 7.2e-06
        [updatestate_loads_eliminate]: 6.81999e-06
        [parameter_eliminate]: 2.79999e-06
        [specialize_transform]: 1.549e-05
        [updatestate_useless_node_eliminater]: 1.335e-05
        [accelerated_algorithm]: 1.395e-05
        [meta_shard_fg_expand]: 3.91999e-06
        [get_grad_eliminate_]: 1.341e-05
        [merge_forward]: 8.79e-06
        [cell_reuse_recompute_pass]: 1.58997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.891e-05
        [j_node_and_user_rematch]: 2.55e-05
        [meta_fg_expand]: 0.00154082
        [replace_old_param]: 6.658e-05
        [inline_without_move]: 5.978e-05
        [renormalize]: 0.0074933
        [add_forward_monad_depend]: 1.088e-05
        [auto_monad_grad]: 6.84001e-06
        [auto_monad_eliminator]: 5.51e-05
        [cse]: 0.00019526
        [replace_applicator]: 7.313e-05
    [Cycle 2]: 0.00215185, [27]
        [switch_simplify]: 4.127e-05
        [loop_unroll]: 3.943e-05
        [a_1]: 0.00107532
        [with_stream_mark]: 1.176e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.30001e-06
        [specialize_transform]: 6.53998e-06
        [updatestate_useless_node_eliminater]: 6.21998e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 1.69998e-06
        [get_grad_eliminate_]: 5.87999e-06
        [merge_forward]: 3.24001e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.178e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 3.91e-05
        [replace_old_param]: 1.166e-05
        [inline_without_move]: 5.95002e-06
        [renormalize]: 0.00065524
        [add_forward_monad_depend]: 4.46002e-06
        [auto_monad_grad]: 1.46002e-06
        [auto_monad_eliminator]: 1.167e-05
        [cse]: 2.459e-05
        [replace_applicator]: 1.195e-05
    [Cycle 3]: 0.00035627, [27]
        [switch_simplify]: 6.83998e-06
        [loop_unroll]: 6.61999e-06
        [a_1]: 0.00011303
        [with_stream_mark]: 8.77999e-06
        [recompute_prepare]: 5.99999e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.82002e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.68997e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.295e-05
        [j_node_and_user_rematch]: 8.99e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 9.56003e-06
        [inline_without_move]: 5.61e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.01997e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 6.26e-06
        [cse]: 1.431e-05
        [replace_applicator]: 5.75001e-06
[py_interpret_to_execute_after_opt_a]: 1.184e-05
[rewriter_after_opt_a]: 3.822e-05
[convert_after_rewriter]: 6.90002e-06
[order_py_execute_after_rewriter]: 5.42001e-06
[mutable_eliminate]: 0.0005705
[jit_opt_b]: 5.562e-05, [1]
    [Cycle 1]: 4.827e-05, [2]
        [frontend_op_eliminate]: 1.891e-05
        [inline_after_opt_a]: 1.809e-05
[cconv]: 2.288e-05
[loop_unroll]: 0.0004266
[jit_opt_after_cconv]: 0.00015446, [1]
    [Cycle 1]: 0.00014806, [11]
        [c_1]: 2.544e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 5.82999e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [cse]: 2.405e-05
        [call_graph_tuple_transform]: 2.119e-05
        [tuple_list_get_item_eliminator]: 6.48003e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 2.69996e-07
        [switch_simplify]: 6.14001e-06
[remove_dup_value]: 1.697e-05
[partial_unused_args_eliminate]: 2.67001e-06
[environ_conv]: 5.22e-06
[add_recomputation]: 5.099e-05
[cse_after_recomputation]: 2.545e-05, [1]
    [Cycle 1]: 2.001e-05, [1]
        [cse]: 1.418e-05
[auto_monad_reorder]: 1.891e-05
[get_jit_bprop_graph]: 1.72001e-06
[rewriter_after_jit_bprop_graph]: 4.68999e-06
[opt_after_jit_grad]: 0.00046425
[symbol_engine_optimizer]: 7.609e-05, [1]
    [Cycle 1]: 6.943e-05, [6]
        [build]: 2.91e-06
        [elim_shapecalc]: 8.71002e-06
        [elim_not_effective]: 1.365e-05
        [opt_reshape]: 6.61e-06
        [fold_const_symbol]: 9.58002e-06
        [renormalize]: 6.80011e-07
[validate]: 3.659e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.00794701
[execute]: 7.82e-06
Sums
    bootstrap                            :     0.000214s :  0.62%
    type_inference                       :     0.010963s : 31.69%
    event_method                         :     0.000044s :  0.13%
    auto_monad                           :     0.000097s :  0.28%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.08%
    rewriter_before_opt_a                :     0.000398s :  1.15%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000115s :  0.33%
    jit_opt_a.loop_unroll                :     0.000090s :  0.26%
    jit_opt_a.a_1                        :     0.002131s :  6.16%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.04%
    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        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.08%
    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              :     0.000015s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.001582s :  4.57%
    jit_opt_a.replace_old_param          :     0.000088s :  0.25%
    jit_opt_a.inline_without_move        :     0.000071s :  0.21%
    jit_opt_a.renormalize                :     0.008149s : 23.56%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000073s :  0.21%
    jit_opt_a.cse                        :     0.000234s :  0.68%
    jit_opt_a.replace_applicator         :     0.000091s :  0.26%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000038s :  0.11%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000571s :  1.65%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.05%
    cconv                                :     0.000023s :  0.07%
    loop_unroll                          :     0.000427s :  1.23%
    jit_opt_after_cconv.c_1              :     0.000025s :  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 :     0.000001s :  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        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000051s :  0.15%
    cse_after_recomputation.cse          :     0.000014s :  0.04%
    auto_monad_reorder                   :     0.000019s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000464s :  1.34%
    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 :  0.04%
    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                             :     0.000037s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007947s : 22.97%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000498   113
 0.39% :     0.000002s :      2: substitution.elim_not_effective
 0.25% :     0.000001s :      2: substitution.fold_const_symbol
 1.27% :     0.000006s :      4: substitution.graph_param_transform
62.93% :     0.000314s :     12: substitution.inline
 3.71% :     0.000019s :      2: substitution.inline_without_move
 1.87% :     0.000009s :     12: substitution.j_node_and_user_rematch
 2.05% :     0.000010s :      7: substitution.minmaximum_grad
 1.33% :     0.000007s :      5: substitution.partial_eliminate
 2.03% :     0.000010s :     12: substitution.remove_not_recompute_node
 5.36% :     0.000027s :     10: substitution.replace_applicator
 2.31% :     0.000012s :     14: substitution.replace_old_param
 0.51% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.02% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.04% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.92% :     0.000044s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010888     2
89.31% :     0.009724s :      1: type_inference.infer
10.69% :     0.001164s :      1: type_inference.specialize
------[replace.]   0.000162    21
58.46% :     0.000095s :     12: replace.inline
41.54% :     0.000067s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000332    21
92.42% :     0.000307s :     12: match.inline
 7.58% :     0.000025s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000417  2828
 1.40% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.39% :     0.000006s :     47: predicate.addn_check_dump
 1.39% :     0.000006s :     47: predicate.addn_zero_filter
 1.70% :     0.000007s :     47: predicate.arithmetic_simplify
 1.41% :     0.000006s :     47: predicate.cast_eliminate
 0.14% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.29% :     0.000005s :     47: predicate.compare_switch_simplify
 1.31% :     0.000005s :     47: predicate.depend_value_elim
 1.28% :     0.000005s :     47: predicate.dict_get_item_const_eliminator
 1.35% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.37% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.16% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.34% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.32% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.40% :     0.000006s :     47: predicate.environ_get_eliminate
 1.29% :     0.000005s :     47: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 0.75% :     0.000003s :     21: predicate.get_grad_eliminate
 0.13% :     0.000001s :      4: predicate.graph_param_transform
 3.99% :     0.000017s :     76: predicate.inline
 1.82% :     0.000008s :     47: predicate.inline_without_move
 0.36% :     0.000002s :     21: predicate.j_node_and_user_rematch
 0.88% :     0.000004s :     21: predicate.less_batch_normalization
 1.66% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 1.82% :     0.000008s :     60: predicate.load_eliminater
 0.44% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.10% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.58% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000005s :     47: predicate.merge_addn
 1.36% :     0.000006s :     47: predicate.minmaximum_grad
 0.56% :     0.000002s :      4: predicate.mutable_eliminate
 0.14% :     0.000001s :      4: predicate.opt_reshape
 2.19% :     0.000009s :     60: predicate.partial_eliminate
 1.35% :     0.000006s :     47: predicate.print_const_string_wrapper
 1.68% :     0.000007s :     47: predicate.reduce_eliminate
 1.66% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.48% :     0.000010s :    108: predicate.replace_applicator
 1.02% :     0.000004s :     47: predicate.replace_old_param
 0.12% :     0.000001s :      4: predicate.reset_defer_inline
 1.35% :     0.000006s :     47: predicate.reshape_eliminate
 1.41% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.43% :     0.000006s :     47: predicate.same_eliminate
 0.46% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.32% :     0.000001s :      8: predicate.special_op_eliminate
 0.80% :     0.000003s :     21: predicate.specialize_transform
 1.51% :     0.000006s :     47: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.14% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.42% :     0.000010s :     68: predicate.switch_defer_inline
 2.26% :     0.000009s :     68: predicate.switch_layer_defer_inline
 5.86% :     0.000024s :    169: predicate.switch_simplify
 1.30% :     0.000005s :     47: predicate.tile_eliminate
 1.40% :     0.000006s :     47: predicate.transpose_eliminate
 1.71% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.50% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000013s :     64: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.74% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
11.74% :     0.000049s :     81: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000007s :     47: predicate.value_based_eliminate
 0.12% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.18% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001737    31
63.11% :     0.001096s :     15: func_graph_cloner_run.FuncGraphClonerGraph
36.89% :     0.000641s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049600    91
 0.11% :     0.000054s :      1: add_recomputation
 0.20% :     0.000101s :      1: auto_monad
 0.04% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.46% :     0.000228s :      1: bootstrap
 0.05% :     0.000026s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000028s :      1: cse_after_recomputation
 0.01% :     0.000007s :      1: environ_conv
 0.10% :     0.000049s :      1: event_method
 0.02% :     0.000012s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
33.55% :     0.016643s :      1: jit_opt_a
 0.32% :     0.000157s :      1: jit_opt_after_cconv
 0.12% :     0.000058s :      1: jit_opt_b
 0.88% :     0.000434s :      1: loop_unroll
 1.17% :     0.000578s :      1: mutable_eliminate
 5.60% :     0.002778s :     39: opt.transform.jit_opt_a
 0.11% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000031s :      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
 0.05% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000035s :      4: opt.transform.symbol_engine_opt
 0.95% :     0.000472s :      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
 0.06% :     0.000029s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
13.16% :     0.006526s :      2: renormalize.infer
 3.24% :     0.001607s :      2: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000041s :      1: rewriter_after_opt_a
 0.82% :     0.000406s :      1: rewriter_before_opt_a
 0.16% :     0.000079s :      1: symbol_engine_optimizer
16.05% :     0.007963s :      1: task_emit
22.13% :     0.010979s :      1: type_inference
 0.11% :     0.000055s :      1: validate


TotalTime = 0.0130664, [30]
[bootstrap]: 0.00029882
[type_inference]: 0.00502579
[event_method]: 2.188e-05
[auto_monad]: 6.042e-05
[graph_reusing]: 6.18998e-06
[pre_auto_parallel]: 2.21e-06
[py_interpret_to_execute]: 3.255e-05
[rewriter_before_opt_a]: 8.598e-05
[expand_dump_flag]: 2.78e-06
[jit_opt_a]: 0.00548997, [2]
    [Cycle 1]: 0.00168448, [27]
        [switch_simplify]: 5.407e-05
        [loop_unroll]: 3.449e-05
        [a_1]: 0.00061103
        [with_stream_mark]: 1.812e-05
        [recompute_prepare]: 7.77e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 6.73e-06
        [updatestate_useless_node_eliminater]: 5.84e-06
        [accelerated_algorithm]: 6.05002e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 5.80002e-06
        [merge_forward]: 3.80998e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.413e-05
        [j_node_and_user_rematch]: 9.12001e-06
        [meta_fg_expand]: 1.94999e-06
        [replace_old_param]: 1.034e-05
        [inline_without_move]: 6.36e-06
        [renormalize]: 0.000659
        [add_forward_monad_depend]: 5.05001e-06
        [auto_monad_grad]: 2.20002e-06
        [auto_monad_eliminator]: 1.354e-05
        [cse]: 2.015e-05
        [replace_applicator]: 1.288e-05
    [Cycle 2]: 0.00040172, [27]
        [switch_simplify]: 6.53e-06
        [loop_unroll]: 5.78997e-06
        [a_1]: 0.00015048
        [with_stream_mark]: 9.09e-06
        [recompute_prepare]: 5.77999e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 5.53002e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 2.58998e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.366e-05
        [j_node_and_user_rematch]: 8.85001e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 9.54e-06
        [inline_without_move]: 5.84999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 6.84001e-06
        [cse]: 1.433e-05
        [replace_applicator]: 5.98998e-06
[py_interpret_to_execute_after_opt_a]: 1.135e-05
[rewriter_after_opt_a]: 3.399e-05
[convert_after_rewriter]: 6.63998e-06
[order_py_execute_after_rewriter]: 5.53002e-06
[mutable_eliminate]: 0.00050592
[jit_opt_b]: 5.276e-05, [1]
    [Cycle 1]: 4.649e-05, [2]
        [frontend_op_eliminate]: 1.805e-05
        [inline_after_opt_a]: 1.702e-05
[cconv]: 2.279e-05
[loop_unroll]: 0.00042717
[jit_opt_after_cconv]: 0.00014483, [1]
    [Cycle 1]: 0.0001389, [11]
        [c_1]: 2.411e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.67999e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 1.894e-05
        [call_graph_tuple_transform]: 1.931e-05
        [tuple_list_get_item_eliminator]: 6.05002e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 2.89991e-07
        [switch_simplify]: 5.80002e-06
[remove_dup_value]: 1.394e-05
[partial_unused_args_eliminate]: 1.97001e-06
[environ_conv]: 5.56e-06
[add_recomputation]: 4.217e-05
[cse_after_recomputation]: 2.29e-05, [1]
    [Cycle 1]: 1.761e-05, [1]
        [cse]: 1.197e-05
[auto_monad_reorder]: 1.679e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 4.35e-06
[opt_after_jit_grad]: 0.00044977
[symbol_engine_optimizer]: 7.578e-05, [1]
    [Cycle 1]: 6.971e-05, [6]
        [build]: 3.93999e-06
        [elim_shapecalc]: 8.92e-06
        [elim_not_effective]: 1.365e-05
        [opt_reshape]: 6.22001e-06
        [fold_const_symbol]: 9.29998e-06
        [renormalize]: 3.69997e-07
[validate]: 3.445e-05
Sums
    bootstrap                            :     0.000299s :  3.28%
    type_inference                       :     0.005026s : 55.20%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000060s :  0.66%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.36%
    rewriter_before_opt_a                :     0.000086s :  0.94%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.67%
    jit_opt_a.loop_unroll                :     0.000040s :  0.44%
    jit_opt_a.a_1                        :     0.000762s :  8.36%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000014s :  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 :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    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      :     0.000012s :  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  :     0.000002s :  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          :     0.000020s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000659s :  7.24%
    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      :     0.000020s :  0.22%
    jit_opt_a.cse                        :     0.000034s :  0.38%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000034s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000506s :  5.56%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000023s :  0.25%
    loop_unroll                          :     0.000427s :  4.69%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.26%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  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 :  0.07%
    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 :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000042s :  0.46%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000017s :  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 :  4.94%
    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 :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.38%


Time group info:
------[substitution.]   0.000208    28
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.38% :     0.000005s :      4: substitution.graph_param_transform
82.82% :     0.000173s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000004s :      4: substitution.replace_old_param
 7.35% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004963     2
81.35% :     0.004037s :      1: type_inference.infer
18.65% :     0.000926s :      1: type_inference.specialize
------[replace.]   0.000081     8
57.48% :     0.000046s :      4: replace.inline
42.52% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     8
92.69% :     0.000170s :      4: match.inline
 7.31% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   912
 1.27% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.98% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     13: predicate.addn_check_dump
 1.26% :     0.000002s :     13: predicate.addn_zero_filter
 1.78% :     0.000002s :     13: predicate.arithmetic_simplify
 1.42% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     13: predicate.compare_switch_simplify
 1.18% :     0.000002s :     13: predicate.depend_value_elim
 1.17% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.43% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.45% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.91% :     0.000001s :      8: predicate.get_grad_eliminate
 0.40% :     0.000001s :      4: predicate.graph_param_transform
 5.54% :     0.000007s :     29: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      8: predicate.less_batch_normalization
 1.67% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.01% :     0.000003s :     21: predicate.load_eliminater
 1.14% :     0.000001s :      4: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.92% :     0.000003s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.58% :     0.000003s :     21: predicate.partial_eliminate
 1.23% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     13: predicate.reduce_eliminate
 1.68% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.02% :     0.000003s :     25: predicate.replace_applicator
 0.67% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     13: predicate.same_eliminate
 0.58% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      8: predicate.special_op_eliminate
 0.97% :     0.000001s :      8: predicate.specialize_transform
 1.30% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     21: predicate.switch_defer_inline
 2.30% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.79% :     0.000010s :     67: predicate.switch_simplify
 1.17% :     0.000002s :     13: predicate.tile_eliminate
 1.23% :     0.000002s :     13: predicate.transpose_eliminate
 1.46% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000596    12
47.00% :     0.000280s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.48% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.53% :     0.000277s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014799    72
 0.30% :     0.000045s :      1: add_recomputation
 0.43% :     0.000064s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.12% :     0.000314s :      1: bootstrap
 0.17% :     0.000025s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.17% :     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
37.12% :     0.005493s :      1: jit_opt_a
 1.00% :     0.000148s :      1: jit_opt_after_cconv
 0.37% :     0.000055s :      1: jit_opt_b
 2.94% :     0.000435s :      1: loop_unroll
 3.48% :     0.000514s :      1: mutable_eliminate
 6.72% :     0.000994s :     26: opt.transform.jit_opt_a
 0.35% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000029s :      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% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.10% :     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% :     0.000004s :      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
 2.48% :     0.000366s :      1: renormalize.infer
 1.93% :     0.000285s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000037s :      1: rewriter_after_opt_a
 0.60% :     0.000090s :      1: rewriter_before_opt_a
 0.53% :     0.000078s :      1: symbol_engine_optimizer
34.04% :     0.005038s :      1: type_inference


TotalTime = 0.0700738, [33]
[bootstrap]: 0.00022745
[type_inference]: 0.0313784
[event_method]: 0.00015272
[auto_monad]: 0.00018128
[graph_reusing]: 6.48e-06
[pre_auto_parallel]: 2.18002e-06
[py_interpret_to_execute]: 3.435e-05
[rewriter_before_opt_a]: 0.00011299
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.0265182, [3]
    [Cycle 1]: 0.0195621, [27]
        [switch_simplify]: 8.083e-05
        [loop_unroll]: 4.996e-05
        [a_1]: 0.0011198
        [with_stream_mark]: 2.832e-05
        [recompute_prepare]: 2.597e-05
        [updatestate_depend_eliminate]: 1.215e-05
        [updatestate_assign_eliminate]: 1.054e-05
        [updatestate_loads_eliminate]: 1.039e-05
        [parameter_eliminate]: 2.54999e-06
        [specialize_transform]: 1.987e-05
        [updatestate_useless_node_eliminater]: 2.385e-05
        [accelerated_algorithm]: 1.913e-05
        [meta_shard_fg_expand]: 5.04998e-06
        [get_grad_eliminate_]: 1.879e-05
        [merge_forward]: 1.144e-05
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.873e-05
        [j_node_and_user_rematch]: 3.527e-05
        [meta_fg_expand]: 0.00549408
        [replace_old_param]: 0.00011174
        [inline_without_move]: 0.00017371
        [renormalize]: 0.0114762
        [add_forward_monad_depend]: 1.657e-05
        [auto_monad_grad]: 9.59e-06
        [auto_monad_eliminator]: 7.48e-05
        [cse]: 0.00025154
        [replace_applicator]: 0.00018056
    [Cycle 2]: 0.00289137, [27]
        [switch_simplify]: 6.144e-05
        [loop_unroll]: 5.742e-05
        [a_1]: 0.00098461
        [with_stream_mark]: 2.074e-05
        [recompute_prepare]: 1.356e-05
        [updatestate_depend_eliminate]: 2.643e-05
        [updatestate_assign_eliminate]: 6.16e-06
        [updatestate_loads_eliminate]: 4.67e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 9.76e-06
        [updatestate_useless_node_eliminater]: 1.169e-05
        [accelerated_algorithm]: 8.87e-06
        [meta_shard_fg_expand]: 2.79001e-06
        [get_grad_eliminate_]: 8.77e-06
        [merge_forward]: 5.97001e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.93e-05
        [j_node_and_user_rematch]: 1.343e-05
        [meta_fg_expand]: 7.711e-05
        [replace_old_param]: 1.371e-05
        [inline_without_move]: 8.42e-06
        [renormalize]: 0.00123611
        [add_forward_monad_depend]: 5.89e-06
        [auto_monad_grad]: 2.84001e-06
        [auto_monad_eliminator]: 2.07e-05
        [cse]: 8e-05
        [replace_applicator]: 1.858e-05
    [Cycle 3]: 0.00050936, [27]
        [switch_simplify]: 9.89001e-06
        [loop_unroll]: 8.97e-06
        [a_1]: 0.00017888
        [with_stream_mark]: 1.502e-05
        [recompute_prepare]: 8.25999e-06
        [updatestate_depend_eliminate]: 6.86999e-06
        [updatestate_assign_eliminate]: 5.22e-06
        [updatestate_loads_eliminate]: 4.58001e-06
        [parameter_eliminate]: 1.62001e-06
        [specialize_transform]: 8.07e-06
        [updatestate_useless_node_eliminater]: 1.214e-05
        [accelerated_algorithm]: 9.08002e-06
        [meta_shard_fg_expand]: 2.55002e-06
        [get_grad_eliminate_]: 7.93001e-06
        [merge_forward]: 4.94e-06
        [cell_reuse_recompute_pass]: 2.91e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.839e-05
        [j_node_and_user_rematch]: 1.307e-05
        [meta_fg_expand]: 3.48e-06
        [replace_old_param]: 1.09e-05
        [inline_without_move]: 7.54002e-06
        [renormalize]: 1.59984e-07
        [add_forward_monad_depend]: 2.24001e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 1.125e-05
        [cse]: 2.465e-05
        [replace_applicator]: 7.97e-06
[py_interpret_to_execute_after_opt_a]: 1.7e-05
[rewriter_after_opt_a]: 6.145e-05
[convert_after_rewriter]: 9.61e-06
[order_py_execute_after_rewriter]: 6.58e-06
[mutable_eliminate]: 0.00088529
[jit_opt_b]: 7.528e-05, [1]
    [Cycle 1]: 6.682e-05, [2]
        [frontend_op_eliminate]: 2.749e-05
        [inline_after_opt_a]: 2.638e-05
[cconv]: 2.885e-05
[loop_unroll]: 0.00050139
[jit_opt_after_cconv]: 0.00022839, [1]
    [Cycle 1]: 0.00022169, [11]
        [c_1]: 5.275e-05
        [parameter_eliminate]: 3.38999e-06
        [updatestate_depend_eliminate]: 8.85001e-06
        [updatestate_assign_eliminate]: 5.49e-06
        [updatestate_loads_eliminate]: 4.34997e-06
        [cse]: 4.081e-05
        [call_graph_tuple_transform]: 2.707e-05
        [tuple_list_get_item_eliminator]: 9.52999e-06
        [none_parameter_eliminate]: 1.77999e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 9.09e-06
[remove_dup_value]: 2.512e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 1.074e-05
[add_recomputation]: 6.829e-05
[cse_after_recomputation]: 3.588e-05, [1]
    [Cycle 1]: 2.88e-05, [1]
        [cse]: 2.132e-05
[auto_monad_reorder]: 2.605e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 7.80998e-06
[opt_after_jit_grad]: 0.00054049
[symbol_engine_optimizer]: 0.00011117, [1]
    [Cycle 1]: 0.00010474, [6]
        [build]: 1.127e-05
        [elim_shapecalc]: 1.31e-05
        [elim_not_effective]: 2.222e-05
        [opt_reshape]: 1.026e-05
        [fold_const_symbol]: 1.537e-05
        [renormalize]: 6.50005e-07
[validate]: 5.504e-05
[backend_pass]: 1.15001e-06
[task_emit]: 0.00852286
[execute]: 7.49002e-06
Sums
    bootstrap                            :     0.000227s :  0.35%
    type_inference                       :     0.031378s : 47.83%
    event_method                         :     0.000153s :  0.23%
    auto_monad                           :     0.000181s :  0.28%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000034s :  0.05%
    rewriter_before_opt_a                :     0.000113s :  0.17%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000152s :  0.23%
    jit_opt_a.loop_unroll                :     0.000116s :  0.18%
    jit_opt_a.a_1                        :     0.002283s :  3.48%
    jit_opt_a.with_stream_mark           :     0.000064s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000048s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000045s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000022s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000020s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.01%
    jit_opt_a.specialize_transform       :     0.000038s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000048s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000037s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000010s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000035s :  0.05%
    jit_opt_a.merge_forward              :     0.000022s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000076s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000062s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.005575s :  8.50%
    jit_opt_a.replace_old_param          :     0.000136s :  0.21%
    jit_opt_a.inline_without_move        :     0.000190s :  0.29%
    jit_opt_a.renormalize                :     0.012712s : 19.38%
    jit_opt_a.add_forward_monad_depend   :     0.000025s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000014s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000107s :  0.16%
    jit_opt_a.cse                        :     0.000356s :  0.54%
    jit_opt_a.replace_applicator         :     0.000207s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.03%
    rewriter_after_opt_a                 :     0.000061s :  0.09%
    convert_after_rewriter               :     0.000010s :  0.01%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000885s :  1.35%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.04%
    cconv                                :     0.000029s :  0.04%
    loop_unroll                          :     0.000501s :  0.76%
    jit_opt_after_cconv.c_1              :     0.000053s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    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              :     0.000041s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000010s :  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  :     0.000009s :  0.01%
    remove_dup_value                     :     0.000025s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000011s :  0.02%
    add_recomputation                    :     0.000068s :  0.10%
    cse_after_recomputation.cse          :     0.000021s :  0.03%
    auto_monad_reorder                   :     0.000026s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.01%
    opt_after_jit_grad                   :     0.000540s :  0.82%
    symbol_engine_optimizer.build        :     0.000011s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000055s :  0.08%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008523s : 12.99%
    execute                              :     0.000007s :  0.01%


Time group info:
------[substitution.]   0.001483   159
 1.49% :     0.000022s :      8: substitution.depend_value_elim
 0.23% :     0.000003s :      4: substitution.elim_not_effective
 0.16% :     0.000002s :      4: substitution.fold_const_symbol
57.27% :     0.000849s :      4: substitution.getattr_setattr_resolve
 0.51% :     0.000008s :      5: substitution.graph_param_transform
24.01% :     0.000356s :     10: substitution.inline
 1.87% :     0.000028s :      4: substitution.inline_without_move
 0.79% :     0.000012s :     20: substitution.j_node_and_user_rematch
 0.62% :     0.000009s :      5: substitution.minmaximum_grad
 0.24% :     0.000004s :      3: substitution.partial_eliminate
 1.02% :     0.000015s :     20: substitution.remove_not_recompute_node
 3.21% :     0.000048s :     13: substitution.replace_applicator
 0.90% :     0.000013s :     17: substitution.replace_old_param
 0.18% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.25% :     0.000018s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.81% :     0.000012s :      5: substitution.tuple_list_get_item_depend_reorder
 2.31% :     0.000034s :     10: substitution.tuple_list_get_item_eliminator
 1.18% :     0.000018s :      8: substitution.updatestate_pure_node_eliminater
 1.97% :     0.000029s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.031307     2
94.65% :     0.029631s :      1: type_inference.infer
 5.35% :     0.001676s :      1: type_inference.specialize
------[replace.]   0.000284    20
16.18% :     0.000046s :      3: replace.getattr_setattr_resolve
30.28% :     0.000086s :     10: replace.inline
13.86% :     0.000039s :      1: replace.replace_applicator
31.70% :     0.000090s :      5: replace.tuple_list_get_item_eliminator
 7.98% :     0.000023s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001189    20
67.58% :     0.000803s :      3: match.getattr_setattr_resolve
29.41% :     0.000350s :     10: match.inline
 0.88% :     0.000011s :      1: match.replace_applicator
 1.42% :     0.000017s :      5: match.tuple_list_get_item_eliminator
 0.71% :     0.000008s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000458  2975
 1.57% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.57% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.28% :     0.000006s :     46: predicate.addn_check_dump
 1.49% :     0.000007s :     46: predicate.addn_zero_filter
 2.20% :     0.000010s :     46: predicate.arithmetic_simplify
 1.59% :     0.000007s :     46: predicate.cast_eliminate
 0.25% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.28% :     0.000006s :     46: predicate.compare_switch_simplify
 1.47% :     0.000007s :     46: predicate.depend_value_elim
 1.32% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.42% :     0.000006s :     46: predicate.dict_get_item_eliminator
 1.37% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.16% :     0.000001s :      5: predicate.elim_not_effective
 0.33% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.26% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.25% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.42% :     0.000007s :     46: predicate.environ_get_eliminate
 1.29% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      5: predicate.fold_const_symbol
 1.03% :     0.000005s :     27: predicate.get_grad_eliminate
 1.39% :     0.000006s :     20: predicate.getattr_setattr_resolve
 0.11% :     0.000001s :      5: predicate.graph_param_transform
 4.00% :     0.000018s :     72: predicate.inline
 3.20% :     0.000015s :     89: predicate.inline_without_move
 0.43% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.29% :     0.000006s :     27: predicate.less_batch_normalization
 1.53% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.73% :     0.000008s :     56: predicate.load_eliminater
 0.55% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.73% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.71% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000006s :     46: predicate.merge_addn
 1.30% :     0.000006s :     46: predicate.minmaximum_grad
 0.69% :     0.000003s :      5: predicate.mutable_eliminate
 0.23% :     0.000001s :      5: predicate.opt_reshape
 2.12% :     0.000010s :     56: predicate.partial_eliminate
 1.36% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.89% :     0.000009s :     46: predicate.reduce_eliminate
 1.56% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000002s :     27: predicate.remove_not_recompute_node
 2.82% :     0.000013s :    126: predicate.replace_applicator
 1.57% :     0.000007s :     89: predicate.replace_old_param
 0.11% :     0.000001s :      5: predicate.reset_defer_inline
 1.36% :     0.000006s :     46: predicate.reshape_eliminate
 1.41% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.38% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.45% :     0.000007s :     46: predicate.same_eliminate
 0.55% :     0.000003s :     29: predicate.set_cell_output_no_recompute
 0.45% :     0.000002s :     10: predicate.special_op_eliminate
 1.01% :     0.000005s :     27: predicate.specialize_transform
 1.66% :     0.000008s :     46: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.24% :     0.000010s :     62: predicate.switch_defer_inline
 2.00% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.50% :     0.000030s :    188: predicate.switch_simplify
 1.31% :     0.000006s :     46: predicate.tile_eliminate
 1.56% :     0.000007s :     46: predicate.transpose_eliminate
 1.73% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.66% :     0.000008s :     46: predicate.tuple_list_get_item_depend_reorder
 3.09% :     0.000014s :     61: predicate.tuple_list_get_item_eliminator
 2.14% :     0.000010s :     46: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.75% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000013s :     84: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000008s :     46: predicate.value_based_eliminate
 0.19% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.29% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002975    48
65.85% :     0.001959s :     27: func_graph_cloner_run.FuncGraphClonerGraph
34.15% :     0.001016s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.087313    93
 0.08% :     0.000071s :      1: add_recomputation
 0.22% :     0.000189s :      1: auto_monad
 0.03% :     0.000029s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.28% :     0.000241s :      1: bootstrap
 0.04% :     0.000032s :      1: cconv
 0.01% :     0.000012s :      1: convert_after_rewriter
 0.04% :     0.000038s :      1: cse_after_recomputation
 0.02% :     0.000013s :      1: environ_conv
 0.18% :     0.000159s :      1: event_method
 0.01% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
30.38% :     0.026522s :      1: jit_opt_a
 0.26% :     0.000231s :      1: jit_opt_after_cconv
 0.09% :     0.000078s :      1: jit_opt_b
 0.58% :     0.000510s :      1: loop_unroll
 1.03% :     0.000897s :      1: mutable_eliminate
 3.87% :     0.003376s :     39: opt.transform.jit_opt_a
 0.11% :     0.000094s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000046s :      4: opt.transform.jit_opt_b
 0.02% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 1.08% :     0.000945s :      2: opt.transform.opt_resolve
 0.06% :     0.000056s :      4: opt.transform.symbol_engine_opt
 0.63% :     0.000551s :      1: opt_after_jit_grad
 0.01% :     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
 0.04% :     0.000037s :      1: py_interpret_to_execute
 0.02% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000028s :      1: remove_dup_value
11.07% :     0.009664s :      2: renormalize.infer
 3.47% :     0.003029s :      2: renormalize.specialize
 0.01% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000065s :      1: rewriter_after_opt_a
 0.13% :     0.000116s :      1: rewriter_before_opt_a
 0.13% :     0.000114s :      1: symbol_engine_optimizer
 9.78% :     0.008537s :      1: task_emit
35.95% :     0.031390s :      1: type_inference
 0.09% :     0.000082s :      1: validate

[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.958.736 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe8fa60> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:02.959.422 [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(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.965.837 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffe2caf6c10> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:02.966.526 [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 = 0.0444978, [33]
[bootstrap]: 0.00026191
[type_inference]: 0.0219985
[event_method]: 1.903e-05
[auto_monad]: 0.00015347
[graph_reusing]: 6.76e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 0.00012603
[rewriter_before_opt_a]: 0.00010083
[expand_dump_flag]: 4.27e-06
[jit_opt_a]: 0.00656809, [2]
    [Cycle 1]: 0.00238768, [27]
        [switch_simplify]: 6.063e-05
        [loop_unroll]: 3.656e-05
        [a_1]: 0.00072517
        [with_stream_mark]: 2.06e-05
        [recompute_prepare]: 1.038e-05
        [updatestate_depend_eliminate]: 5.58002e-06
        [updatestate_assign_eliminate]: 4.84e-06
        [updatestate_loads_eliminate]: 4.28001e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 9.32001e-06
        [updatestate_useless_node_eliminater]: 1.127e-05
        [accelerated_algorithm]: 9.12001e-06
        [meta_shard_fg_expand]: 2.66e-06
        [get_grad_eliminate_]: 8.16002e-06
        [merge_forward]: 4.65999e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.703e-05
        [j_node_and_user_rematch]: 1.335e-05
        [meta_fg_expand]: 3.33998e-06
        [replace_old_param]: 1.065e-05
        [inline_without_move]: 8.26002e-06
        [renormalize]: 0.00116994
        [add_forward_monad_depend]: 6.35002e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 2.025e-05
        [cse]: 2.847e-05
        [replace_applicator]: 1.575e-05
    [Cycle 2]: 0.00046144, [27]
        [switch_simplify]: 8.72e-06
        [loop_unroll]: 7.8e-06
        [a_1]: 0.000166
        [with_stream_mark]: 1.221e-05
        [recompute_prepare]: 8.14002e-06
        [updatestate_depend_eliminate]: 4.79e-06
        [updatestate_assign_eliminate]: 5.23002e-06
        [updatestate_loads_eliminate]: 3.95e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 7.93999e-06
        [updatestate_useless_node_eliminater]: 1.03e-05
        [accelerated_algorithm]: 7.88999e-06
        [meta_shard_fg_expand]: 2.04999e-06
        [get_grad_eliminate_]: 7.47998e-06
        [merge_forward]: 4.42e-06
        [cell_reuse_recompute_pass]: 1.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.718e-05
        [j_node_and_user_rematch]: 1.242e-05
        [meta_fg_expand]: 3.04999e-06
        [replace_old_param]: 9.61998e-06
        [inline_without_move]: 7.3e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 1.12e-06
        [auto_monad_eliminator]: 1.01e-05
        [cse]: 1.975e-05
        [replace_applicator]: 7.81001e-06
[py_interpret_to_execute_after_opt_a]: 1.239e-05
[rewriter_after_opt_a]: 0.00027939
[convert_after_rewriter]: 1.093e-05
[order_py_execute_after_rewriter]: 7.25e-06
[mutable_eliminate]: 0.00070026
[jit_opt_b]: 6.664e-05, [1]
    [Cycle 1]: 6.002e-05, [2]
        [frontend_op_eliminate]: 2.427e-05
        [inline_after_opt_a]: 2.401e-05
[cconv]: 2.364e-05
[loop_unroll]: 0.00046175
[jit_opt_after_cconv]: 0.0001946, [1]
    [Cycle 1]: 0.00018778, [11]
        [c_1]: 4.452e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 7.3e-06
        [updatestate_assign_eliminate]: 4.53001e-06
        [updatestate_loads_eliminate]: 4.60001e-06
        [cse]: 3.034e-05
        [call_graph_tuple_transform]: 2.3e-05
        [tuple_list_get_item_eliminator]: 8.25999e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 8.80001e-06
[remove_dup_value]: 1.996e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 1.076e-05
[add_recomputation]: 6.257e-05
[cse_after_recomputation]: 2.94e-05, [1]
    [Cycle 1]: 2.374e-05, [1]
        [cse]: 1.743e-05
[auto_monad_reorder]: 2.344e-05
[get_jit_bprop_graph]: 1.91998e-06
[rewriter_after_jit_bprop_graph]: 5.20001e-06
[opt_after_jit_grad]: 0.00052284
[symbol_engine_optimizer]: 0.00010573, [1]
    [Cycle 1]: 9.854e-05, [6]
        [build]: 1.398e-05
        [elim_shapecalc]: 1.197e-05
        [elim_not_effective]: 1.926e-05
        [opt_reshape]: 8.99e-06
        [fold_const_symbol]: 1.376e-05
        [renormalize]: 4.30009e-07
[validate]: 5.728e-05
[backend_pass]: 1.17999e-06
[task_emit]: 0.0124444
[execute]: 8.05e-06
Sums
    bootstrap                            :     0.000262s :  0.65%
    type_inference                       :     0.021999s : 54.78%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000153s :  0.38%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000126s :  0.31%
    rewriter_before_opt_a                :     0.000101s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.17%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000891s :  2.22%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001170s :  2.91%
    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      :     0.000030s :  0.08%
    jit_opt_a.cse                        :     0.000048s :  0.12%
    jit_opt_a.replace_applicator         :     0.000024s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000279s :  0.70%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000700s :  1.74%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000024s :  0.06%
    loop_unroll                          :     0.000462s :  1.15%
    jit_opt_after_cconv.c_1              :     0.000045s :  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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000063s :  0.16%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000023s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000523s :  1.30%
    symbol_engine_optimizer.build        :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000057s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012444s : 30.99%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000291    47
 3.15% :     0.000009s :      2: substitution.depend_value_elim
 0.94% :     0.000003s :      4: substitution.elim_not_effective
 0.70% :     0.000002s :      4: substitution.fold_const_symbol
 2.34% :     0.000007s :      5: substitution.graph_param_transform
77.38% :     0.000225s :      4: substitution.inline
 1.55% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.51% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.15% :     0.000003s :      2: substitution.replace_old_param
 4.39% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 2.98% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.91% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021929     2
93.16% :     0.020430s :      1: type_inference.infer
 6.84% :     0.001499s :      1: type_inference.specialize
------[replace.]   0.000096     6
66.19% :     0.000063s :      4: replace.inline
33.81% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000234     6
95.02% :     0.000222s :      4: match.inline
 4.98% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1021
 1.46% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.13% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.46% :     0.000002s :     15: predicate.addn_zero_filter
 1.94% :     0.000003s :     15: predicate.arithmetic_simplify
 1.39% :     0.000002s :     15: predicate.cast_eliminate
 0.55% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     15: predicate.compare_switch_simplify
 1.48% :     0.000002s :     15: predicate.depend_value_elim
 1.18% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.91% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.55% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     15: predicate.environ_get_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 5.19% :     0.000009s :     31: predicate.inline
 1.05% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.50% :     0.000003s :     10: predicate.less_batch_normalization
 1.56% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     22: predicate.load_eliminater
 1.36% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.37% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.09% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     15: predicate.merge_addn
 1.14% :     0.000002s :     15: predicate.minmaximum_grad
 1.24% :     0.000002s :      5: predicate.mutable_eliminate
 0.57% :     0.000001s :      5: predicate.opt_reshape
 2.40% :     0.000004s :     22: predicate.partial_eliminate
 1.24% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.93% :     0.000003s :     15: predicate.reduce_eliminate
 1.43% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.74% :     0.000003s :     27: predicate.replace_applicator
 0.73% :     0.000001s :     10: predicate.replace_old_param
 0.35% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000002s :     15: predicate.reshape_eliminate
 1.26% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     15: predicate.same_eliminate
 0.69% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.11% :     0.000002s :     10: predicate.special_op_eliminate
 1.08% :     0.000002s :     10: predicate.specialize_transform
 1.45% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.62% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.36% :     0.000004s :     21: predicate.switch_defer_inline
 1.93% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.60% :     0.000011s :     68: predicate.switch_simplify
 1.30% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000002s :     15: predicate.transpose_eliminate
 1.53% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.78% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000777    13
42.73% :     0.000332s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.22% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.04% :     0.000397s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047046    76
 0.14% :     0.000065s :      1: add_recomputation
 0.34% :     0.000158s :      1: auto_monad
 0.06% :     0.000026s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.58% :     0.000273s :      1: bootstrap
 0.06% :     0.000026s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.02% :     0.000012s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
13.97% :     0.006572s :      1: jit_opt_a
 0.42% :     0.000197s :      1: jit_opt_after_cconv
 0.15% :     0.000069s :      1: jit_opt_b
 1.00% :     0.000470s :      1: loop_unroll
 1.51% :     0.000709s :      1: mutable_eliminate
 2.52% :     0.001187s :     26: opt.transform.jit_opt_a
 0.17% :     0.000081s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.13% :     0.000532s :      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
 0.28% :     0.000130s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.51% :     0.000710s :      1: renormalize.infer
 0.96% :     0.000452s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.60% :     0.000284s :      1: rewriter_after_opt_a
 0.22% :     0.000105s :      1: rewriter_before_opt_a
 0.23% :     0.000109s :      1: symbol_engine_optimizer
26.48% :     0.012457s :      1: task_emit
46.79% :     0.022011s :      1: type_inference
 0.17% :     0.000081s :      1: validate


TotalTime = 0.0143101, [30]
[bootstrap]: 0.00035096
[type_inference]: 0.00580955
[event_method]: 2.387e-05
[auto_monad]: 6.859e-05
[graph_reusing]: 5.78002e-06
[pre_auto_parallel]: 2.78998e-06
[py_interpret_to_execute]: 3.735e-05
[rewriter_before_opt_a]: 9.206e-05
[expand_dump_flag]: 2.87002e-06
[jit_opt_a]: 0.00554098, [2]
    [Cycle 1]: 0.00170229, [27]
        [switch_simplify]: 6.017e-05
        [loop_unroll]: 5.025e-05
        [a_1]: 0.00061156
        [with_stream_mark]: 1.983e-05
        [recompute_prepare]: 8.42e-06
        [updatestate_depend_eliminate]: 4.07e-06
        [updatestate_assign_eliminate]: 3.50003e-06
        [updatestate_loads_eliminate]: 2.97002e-06
        [parameter_eliminate]: 2.22999e-06
        [specialize_transform]: 6.64999e-06
        [updatestate_useless_node_eliminater]: 6.01e-06
        [accelerated_algorithm]: 6.61e-06
        [meta_shard_fg_expand]: 2.03002e-06
        [get_grad_eliminate_]: 6.02001e-06
        [merge_forward]: 3.78999e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.555e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 1.099e-05
        [inline_without_move]: 5.99999e-06
        [renormalize]: 0.0006016
        [add_forward_monad_depend]: 5.16002e-06
        [auto_monad_grad]: 2.20002e-06
        [auto_monad_eliminator]: 1.647e-05
        [cse]: 3.457e-05
        [replace_applicator]: 1.449e-05
    [Cycle 2]: 0.00037524, [27]
        [switch_simplify]: 6.51999e-06
        [loop_unroll]: 5.86e-06
        [a_1]: 0.0001116
        [with_stream_mark]: 1.071e-05
        [recompute_prepare]: 5.87999e-06
        [updatestate_depend_eliminate]: 3.15998e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 5.60001e-06
        [accelerated_algorithm]: 5.81998e-06
        [meta_shard_fg_expand]: 1.62999e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 3.31001e-06
        [cell_reuse_recompute_pass]: 1.51998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.478e-05
        [j_node_and_user_rematch]: 9.66998e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 1.028e-05
        [inline_without_move]: 5.47001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 9.80013e-07
        [auto_monad_eliminator]: 7.5e-06
        [cse]: 1.486e-05
        [replace_applicator]: 7.25e-06
[py_interpret_to_execute_after_opt_a]: 1.026e-05
[rewriter_after_opt_a]: 3.675e-05
[convert_after_rewriter]: 7.53999e-06
[order_py_execute_after_rewriter]: 5.34e-06
[mutable_eliminate]: 0.00056932
[jit_opt_b]: 5.565e-05, [1]
    [Cycle 1]: 4.802e-05, [2]
        [frontend_op_eliminate]: 1.798e-05
        [inline_after_opt_a]: 1.721e-05
[cconv]: 2.734e-05
[loop_unroll]: 0.00048775
[jit_opt_after_cconv]: 0.00016468, [1]
    [Cycle 1]: 0.00015687, [11]
        [c_1]: 2.467e-05
        [parameter_eliminate]: 3.03e-06
        [updatestate_depend_eliminate]: 7.28999e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 2.779e-05
        [call_graph_tuple_transform]: 2.161e-05
        [tuple_list_get_item_eliminator]: 6.39999e-06
        [none_parameter_eliminate]: 2.19001e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 6.03998e-06
[remove_dup_value]: 1.614e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 6.16998e-06
[add_recomputation]: 5.299e-05
[cse_after_recomputation]: 2.534e-05, [1]
    [Cycle 1]: 1.925e-05, [1]
        [cse]: 1.253e-05
[auto_monad_reorder]: 1.909e-05
[get_jit_bprop_graph]: 1.71998e-06
[rewriter_after_jit_bprop_graph]: 6.59999e-06
[opt_after_jit_grad]: 0.00056643
[symbol_engine_optimizer]: 7.973e-05, [1]
    [Cycle 1]: 7.182e-05, [6]
        [build]: 4.37e-06
        [elim_shapecalc]: 8.84e-06
        [elim_not_effective]: 1.406e-05
        [opt_reshape]: 6.58998e-06
        [fold_const_symbol]: 9.02999e-06
        [renormalize]: 4.00003e-07
[validate]: 4.307e-05
Sums
    bootstrap                            :     0.000351s :  3.44%
    type_inference                       :     0.005810s : 56.87%
    event_method                         :     0.000024s :  0.23%
    auto_monad                           :     0.000069s :  0.67%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.37%
    rewriter_before_opt_a                :     0.000092s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000067s :  0.65%
    jit_opt_a.loop_unroll                :     0.000056s :  0.55%
    jit_opt_a.a_1                        :     0.000723s :  7.08%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.30%
    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        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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 :     0.000030s :  0.30%
    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 :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000602s :  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      :     0.000024s :  0.23%
    jit_opt_a.cse                        :     0.000049s :  0.48%
    jit_opt_a.replace_applicator         :     0.000022s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000037s :  0.36%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000569s :  5.57%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.17%
    cconv                                :     0.000027s :  0.27%
    loop_unroll                          :     0.000488s :  4.77%
    jit_opt_after_cconv.c_1              :     0.000025s :  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 :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000028s :  0.27%
    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                     :     0.000016s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000053s :  0.52%
    cse_after_recomputation.cse          :     0.000013s :  0.12%
    auto_monad_reorder                   :     0.000019s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000566s :  5.54%
    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 :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.42%


Time group info:
------[substitution.]   0.000207    28
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000006s :      4: substitution.graph_param_transform
80.78% :     0.000167s :      4: substitution.inline
 1.76% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.83% :     0.000006s :      4: substitution.replace_old_param
 7.91% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005733     2
82.51% :     0.004730s :      1: type_inference.infer
17.49% :     0.001003s :      1: type_inference.specialize
------[replace.]   0.000084     8
59.41% :     0.000050s :      4: replace.inline
40.59% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     8
91.90% :     0.000165s :      4: match.inline
 8.10% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   912
 1.20% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     13: predicate.addn_check_dump
 1.17% :     0.000002s :     13: predicate.addn_zero_filter
 1.61% :     0.000002s :     13: predicate.arithmetic_simplify
 1.31% :     0.000002s :     13: predicate.cast_eliminate
 0.42% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     13: predicate.compare_switch_simplify
 1.16% :     0.000002s :     13: predicate.depend_value_elim
 1.14% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.64% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     13: predicate.environ_get_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      4: predicate.fold_const_symbol
 0.93% :     0.000001s :      8: predicate.get_grad_eliminate
 0.40% :     0.000001s :      4: predicate.graph_param_transform
 5.37% :     0.000007s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.66% :     0.000002s :      8: predicate.less_batch_normalization
 1.58% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     21: predicate.load_eliminater
 1.70% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.18% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     13: predicate.merge_addn
 1.10% :     0.000001s :     13: predicate.minmaximum_grad
 1.65% :     0.000002s :      4: predicate.mutable_eliminate
 0.46% :     0.000001s :      4: predicate.opt_reshape
 2.45% :     0.000003s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     13: predicate.reduce_eliminate
 1.59% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.98% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.14% :     0.000002s :     13: predicate.reshape_eliminate
 1.16% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.54% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      8: predicate.special_op_eliminate
 0.95% :     0.000001s :      8: predicate.specialize_transform
 1.31% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     21: predicate.switch_defer_inline
 2.21% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.87% :     0.000011s :     67: predicate.switch_simplify
 1.13% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.49% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.25% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000666    12
54.55% :     0.000363s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.01% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.44% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015970    72
 0.35% :     0.000056s :      1: add_recomputation
 0.46% :     0.000073s :      1: auto_monad
 0.14% :     0.000022s :      1: auto_monad_reorder
 2.33% :     0.000372s :      1: bootstrap
 0.19% :     0.000030s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000027s :      1: cse_after_recomputation
 0.05% :     0.000009s :      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% :     0.000008s :      1: graph_reusing
34.72% :     0.005544s :      1: jit_opt_a
 1.05% :     0.000168s :      1: jit_opt_after_cconv
 0.37% :     0.000058s :      1: jit_opt_b
 3.12% :     0.000498s :      1: loop_unroll
 3.62% :     0.000578s :      1: mutable_eliminate
 6.16% :     0.000984s :     26: opt.transform.jit_opt_a
 0.35% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000029s :      4: opt.transform.jit_opt_b
 0.11% :     0.000017s :      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
 0.22% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.61% :     0.000577s :      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
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.03% :     0.000325s :      1: renormalize.infer
 1.69% :     0.000270s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000040s :      1: rewriter_after_opt_a
 0.59% :     0.000095s :      1: rewriter_before_opt_a
 0.52% :     0.000082s :      1: symbol_engine_optimizer
36.50% :     0.005828s :      1: type_inference


TotalTime = 0.0398408, [33]
[bootstrap]: 0.0002164
[type_inference]: 0.0110626
[event_method]: 4.072e-05
[auto_monad]: 9.956e-05
[graph_reusing]: 6.78998e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 2.76e-05
[rewriter_before_opt_a]: 0.00011002
[expand_dump_flag]: 3.00998e-06
[jit_opt_a]: 0.017785, [3]
    [Cycle 1]: 0.0114291, [27]
        [switch_simplify]: 7.531e-05
        [loop_unroll]: 4.419e-05
        [a_1]: 0.00097017
        [with_stream_mark]: 3.24e-05
        [recompute_prepare]: 2.014e-05
        [updatestate_depend_eliminate]: 9.34e-06
        [updatestate_assign_eliminate]: 7.48e-06
        [updatestate_loads_eliminate]: 6.52001e-06
        [parameter_eliminate]: 3.33e-06
        [specialize_transform]: 1.575e-05
        [updatestate_useless_node_eliminater]: 1.319e-05
        [accelerated_algorithm]: 1.384e-05
        [meta_shard_fg_expand]: 3.73001e-06
        [get_grad_eliminate_]: 1.448e-05
        [merge_forward]: 8.60001e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.849e-05
        [j_node_and_user_rematch]: 2.418e-05
        [meta_fg_expand]: 0.00179219
        [replace_old_param]: 6.5e-05
        [inline_without_move]: 5.894e-05
        [renormalize]: 0.00759637
        [add_forward_monad_depend]: 1.404e-05
        [auto_monad_grad]: 7.13e-06
        [auto_monad_eliminator]: 6.013e-05
        [cse]: 0.00021288
        [replace_applicator]: 8.477e-05
    [Cycle 2]: 0.00253226, [27]
        [switch_simplify]: 4.16e-05
        [loop_unroll]: 4.02e-05
        [a_1]: 0.00116476
        [with_stream_mark]: 2.364e-05
        [recompute_prepare]: 1.012e-05
        [updatestate_depend_eliminate]: 4.43001e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 3.07002e-06
        [parameter_eliminate]: 2.60002e-06
        [specialize_transform]: 7.61999e-06
        [updatestate_useless_node_eliminater]: 6.51999e-06
        [accelerated_algorithm]: 6.84999e-06
        [meta_shard_fg_expand]: 2.46998e-06
        [get_grad_eliminate_]: 7.55e-06
        [merge_forward]: 4.05998e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.753e-05
        [j_node_and_user_rematch]: 9.96998e-06
        [meta_fg_expand]: 6.364e-05
        [replace_old_param]: 1.625e-05
        [inline_without_move]: 6.72002e-06
        [renormalize]: 0.00084051
        [add_forward_monad_depend]: 6.64999e-06
        [auto_monad_grad]: 3.26999e-06
        [auto_monad_eliminator]: 1.422e-05
        [cse]: 3.536e-05
        [replace_applicator]: 1.701e-05
    [Cycle 3]: 0.00037687, [27]
        [switch_simplify]: 7.28e-06
        [loop_unroll]: 6.49001e-06
        [a_1]: 0.00011751
        [with_stream_mark]: 1.192e-05
        [recompute_prepare]: 6.77002e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.36998e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 6.39001e-06
        [updatestate_useless_node_eliminater]: 6.01e-06
        [accelerated_algorithm]: 6.89999e-06
        [meta_shard_fg_expand]: 1.75001e-06
        [get_grad_eliminate_]: 5.62999e-06
        [merge_forward]: 3.24001e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.576e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 1.078e-05
        [inline_without_move]: 5.84e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 1.40999e-06
        [auto_monad_eliminator]: 6.41998e-06
        [cse]: 1.523e-05
        [replace_applicator]: 5.83002e-06
[py_interpret_to_execute_after_opt_a]: 1.536e-05
[rewriter_after_opt_a]: 3.944e-05
[convert_after_rewriter]: 7.35e-06
[order_py_execute_after_rewriter]: 5.21998e-06
[mutable_eliminate]: 0.00105458
[jit_opt_b]: 6.024e-05, [1]
    [Cycle 1]: 5.094e-05, [2]
        [frontend_op_eliminate]: 1.936e-05
        [inline_after_opt_a]: 1.919e-05
[cconv]: 3.175e-05
[loop_unroll]: 0.00050979
[jit_opt_after_cconv]: 0.00017402, [1]
    [Cycle 1]: 0.00016608, [11]
        [c_1]: 2.565e-05
        [parameter_eliminate]: 4.1e-06
        [updatestate_depend_eliminate]: 7.4e-06
        [updatestate_assign_eliminate]: 3.41999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [cse]: 3.285e-05
        [call_graph_tuple_transform]: 2.405e-05
        [tuple_list_get_item_eliminator]: 6.46e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 6.99976e-07
        [switch_simplify]: 6.84001e-06
[remove_dup_value]: 1.83e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 6.44001e-06
[add_recomputation]: 5.293e-05
[cse_after_recomputation]: 2.592e-05, [1]
    [Cycle 1]: 2.05e-05, [1]
        [cse]: 1.418e-05
[auto_monad_reorder]: 1.885e-05
[get_jit_bprop_graph]: 1.84e-06
[rewriter_after_jit_bprop_graph]: 6.71e-06
[opt_after_jit_grad]: 0.00055504
[symbol_engine_optimizer]: 8.495e-05, [1]
    [Cycle 1]: 7.768e-05, [6]
        [build]: 3.42002e-06
        [elim_shapecalc]: 1.004e-05
        [elim_not_effective]: 1.705e-05
        [opt_reshape]: 7.31001e-06
        [fold_const_symbol]: 1.032e-05
        [renormalize]: 8.99978e-07
[validate]: 4.814e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.0075523
[execute]: 8.38999e-06
Sums
    bootstrap                            :     0.000216s :  0.61%
    type_inference                       :     0.011063s : 31.13%
    event_method                         :     0.000041s :  0.11%
    auto_monad                           :     0.000100s :  0.28%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.08%
    rewriter_before_opt_a                :     0.000110s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000124s :  0.35%
    jit_opt_a.loop_unroll                :     0.000091s :  0.26%
    jit_opt_a.a_1                        :     0.002252s :  6.34%
    jit_opt_a.with_stream_mark           :     0.000068s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  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        :     0.000007s :  0.02%
    jit_opt_a.specialize_transform       :     0.000030s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000028s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000028s :  0.08%
    jit_opt_a.merge_forward              :     0.000016s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000062s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.001858s :  5.23%
    jit_opt_a.replace_old_param          :     0.000092s :  0.26%
    jit_opt_a.inline_without_move        :     0.000072s :  0.20%
    jit_opt_a.renormalize                :     0.008437s : 23.74%
    jit_opt_a.add_forward_monad_depend   :     0.000022s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000012s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000081s :  0.23%
    jit_opt_a.cse                        :     0.000263s :  0.74%
    jit_opt_a.replace_applicator         :     0.000108s :  0.30%
    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                    :     0.001055s :  2.97%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.05%
    cconv                                :     0.000032s :  0.09%
    loop_unroll                          :     0.000510s :  1.43%
    jit_opt_after_cconv.c_1              :     0.000026s :  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              :     0.000033s :  0.09%
    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                     :     0.000018s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000053s :  0.15%
    cse_after_recomputation.cse          :     0.000014s :  0.04%
    auto_monad_reorder                   :     0.000019s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000555s :  1.56%
    symbol_engine_optimizer.build        :     0.000003s :  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  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007552s : 21.25%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000601   113
 0.45% :     0.000003s :      2: substitution.elim_not_effective
 0.25% :     0.000002s :      2: substitution.fold_const_symbol
 1.00% :     0.000006s :      4: substitution.graph_param_transform
64.93% :     0.000390s :     12: substitution.inline
 2.97% :     0.000018s :      2: substitution.inline_without_move
 1.51% :     0.000009s :     12: substitution.j_node_and_user_rematch
 1.93% :     0.000012s :      7: substitution.minmaximum_grad
 1.32% :     0.000008s :      5: substitution.partial_eliminate
 1.81% :     0.000011s :     12: substitution.remove_not_recompute_node
 5.53% :     0.000033s :     10: substitution.replace_applicator
 2.26% :     0.000014s :     14: substitution.replace_old_param
 0.50% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.23% :     0.000025s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.77% :     0.000017s :      7: substitution.tuple_list_get_item_depend_reorder
 8.53% :     0.000051s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010986     2
89.39% :     0.009820s :      1: type_inference.infer
10.61% :     0.001165s :      1: type_inference.specialize
------[replace.]   0.000186    21
60.15% :     0.000112s :     12: replace.inline
39.85% :     0.000074s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000410    21
93.18% :     0.000382s :     12: match.inline
 6.82% :     0.000028s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000395  2828
 1.54% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.55% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.41% :     0.000006s :     47: predicate.addn_check_dump
 1.70% :     0.000007s :     47: predicate.addn_zero_filter
 1.96% :     0.000008s :     47: predicate.arithmetic_simplify
 1.70% :     0.000007s :     47: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.36% :     0.000005s :     47: predicate.compare_switch_simplify
 1.39% :     0.000005s :     47: predicate.depend_value_elim
 1.40% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.49% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.55% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.34% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.17% :     0.000001s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.41% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.40% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.44% :     0.000006s :     47: predicate.environ_get_eliminate
 1.40% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 1.12% :     0.000004s :     21: predicate.get_grad_eliminate
 0.15% :     0.000001s :      4: predicate.graph_param_transform
 4.48% :     0.000018s :     76: predicate.inline
 1.85% :     0.000007s :     47: predicate.inline_without_move
 0.36% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.20% :     0.000005s :     21: predicate.less_batch_normalization
 1.84% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 1.92% :     0.000008s :     60: predicate.load_eliminater
 0.61% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.36% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.75% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.38% :     0.000005s :     47: predicate.merge_addn
 1.46% :     0.000006s :     47: predicate.minmaximum_grad
 0.72% :     0.000003s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.36% :     0.000009s :     60: predicate.partial_eliminate
 1.59% :     0.000006s :     47: predicate.print_const_string_wrapper
 1.81% :     0.000007s :     47: predicate.reduce_eliminate
 1.79% :     0.000007s :     56: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.66% :     0.000010s :    108: predicate.replace_applicator
 1.05% :     0.000004s :     47: predicate.replace_old_param
 0.14% :     0.000001s :      4: predicate.reset_defer_inline
 1.55% :     0.000006s :     47: predicate.reshape_eliminate
 1.47% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.31% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.43% :     0.000006s :     47: predicate.same_eliminate
 0.48% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.35% :     0.000001s :      8: predicate.special_op_eliminate
 0.88% :     0.000003s :     21: predicate.specialize_transform
 1.63% :     0.000006s :     47: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.15% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.69% :     0.000011s :     68: predicate.switch_defer_inline
 2.49% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.43% :     0.000025s :    169: predicate.switch_simplify
 1.42% :     0.000006s :     47: predicate.tile_eliminate
 1.46% :     0.000006s :     47: predicate.transpose_eliminate
 1.83% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.69% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000014s :     64: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.89% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000007s :     47: predicate.value_based_eliminate
 0.19% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.22% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001896    31
60.91% :     0.001155s :     15: func_graph_cloner_run.FuncGraphClonerGraph
39.09% :     0.000741s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.051365    91
 0.11% :     0.000056s :      1: add_recomputation
 0.20% :     0.000104s :      1: auto_monad
 0.04% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.45% :     0.000230s :      1: bootstrap
 0.07% :     0.000035s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000028s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.09% :     0.000045s :      1: event_method
 0.02% :     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
34.63% :     0.017789s :      1: jit_opt_a
 0.34% :     0.000177s :      1: jit_opt_after_cconv
 0.12% :     0.000064s :      1: jit_opt_b
 1.01% :     0.000519s :      1: loop_unroll
 2.08% :     0.001068s :      1: mutable_eliminate
 5.73% :     0.002942s :     39: opt.transform.jit_opt_a
 0.12% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000041s :      4: opt.transform.symbol_engine_opt
 1.10% :     0.000565s :      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
 0.06% :     0.000030s :      1: py_interpret_to_execute
 0.03% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
12.67% :     0.006506s :      2: renormalize.infer
 3.72% :     0.001912s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000043s :      1: rewriter_after_opt_a
 0.22% :     0.000113s :      1: rewriter_before_opt_a
 0.17% :     0.000087s :      1: symbol_engine_optimizer
14.73% :     0.007565s :      1: task_emit
21.56% :     0.011077s :      1: type_inference
 0.14% :     0.000070s :      1: validate


TotalTime = 0.0141145, [30]
[bootstrap]: 0.00034872
[type_inference]: 0.00541622
[event_method]: 2.32e-05
[auto_monad]: 7.025e-05
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 3.844e-05
[rewriter_before_opt_a]: 9.242e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00585192, [2]
    [Cycle 1]: 0.00189741, [27]
        [switch_simplify]: 6.076e-05
        [loop_unroll]: 3.503e-05
        [a_1]: 0.00064084
        [with_stream_mark]: 1.769e-05
        [recompute_prepare]: 9.46e-06
        [updatestate_depend_eliminate]: 4.29997e-06
        [updatestate_assign_eliminate]: 3.56001e-06
        [updatestate_loads_eliminate]: 2.84001e-06
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 7.01001e-06
        [updatestate_useless_node_eliminater]: 6.29001e-06
        [accelerated_algorithm]: 5.97999e-06
        [meta_shard_fg_expand]: 2.22999e-06
        [get_grad_eliminate_]: 6.02999e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.638e-05
        [j_node_and_user_rematch]: 1.042e-05
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 1.23e-05
        [inline_without_move]: 6.28002e-06
        [renormalize]: 0.00073535
        [add_forward_monad_depend]: 6.26998e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 1.661e-05
        [cse]: 3.58e-05
        [replace_applicator]: 1.55e-05
    [Cycle 2]: 0.00037412, [27]
        [switch_simplify]: 6.69999e-06
        [loop_unroll]: 6.21e-06
        [a_1]: 0.00011638
        [with_stream_mark]: 9.69e-06
        [recompute_prepare]: 6.28e-06
        [updatestate_depend_eliminate]: 3.88001e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 5.84999e-06
        [updatestate_useless_node_eliminater]: 5.85002e-06
        [accelerated_algorithm]: 1.123e-05
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.407e-05
        [j_node_and_user_rematch]: 8.87e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 1.084e-05
        [inline_without_move]: 6.09001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 6.44001e-06
        [cse]: 1.335e-05
        [replace_applicator]: 6.51999e-06
[py_interpret_to_execute_after_opt_a]: 1.104e-05
[rewriter_after_opt_a]: 3.776e-05
[convert_after_rewriter]: 7.7e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00056245
[jit_opt_b]: 5.632e-05, [1]
    [Cycle 1]: 4.865e-05, [2]
        [frontend_op_eliminate]: 1.911e-05
        [inline_after_opt_a]: 1.722e-05
[cconv]: 2.419e-05
[loop_unroll]: 0.00045821
[jit_opt_after_cconv]: 0.00015668, [1]
    [Cycle 1]: 0.00014939, [11]
        [c_1]: 2.472e-05
        [parameter_eliminate]: 2.83e-06
        [updatestate_depend_eliminate]: 5.89e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [cse]: 2.337e-05
        [call_graph_tuple_transform]: 2.101e-05
        [tuple_list_get_item_eliminator]: 6.37001e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 5.40022e-07
        [switch_simplify]: 6.68998e-06
[remove_dup_value]: 1.506e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.72001e-06
[add_recomputation]: 5.188e-05
[cse_after_recomputation]: 2.364e-05, [1]
    [Cycle 1]: 1.824e-05, [1]
        [cse]: 1.222e-05
[auto_monad_reorder]: 1.835e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 5.35001e-06
[opt_after_jit_grad]: 0.00052089
[symbol_engine_optimizer]: 7.719e-05, [1]
    [Cycle 1]: 7.099e-05, [6]
        [build]: 3.86001e-06
        [elim_shapecalc]: 8.59002e-06
        [elim_not_effective]: 1.356e-05
        [opt_reshape]: 6.46e-06
        [fold_const_symbol]: 9.86998e-06
        [renormalize]: 6.00005e-07
[validate]: 3.958e-05
Sums
    bootstrap                            :     0.000349s :  3.53%
    type_inference                       :     0.005416s : 54.78%
    event_method                         :     0.000023s :  0.23%
    auto_monad                           :     0.000070s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.39%
    rewriter_before_opt_a                :     0.000092s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000067s :  0.68%
    jit_opt_a.loop_unroll                :     0.000041s :  0.42%
    jit_opt_a.a_1                        :     0.000757s :  7.66%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000016s :  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        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.17%
    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 :  0.31%
    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          :     0.000023s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000735s :  7.44%
    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 :  0.23%
    jit_opt_a.cse                        :     0.000049s :  0.50%
    jit_opt_a.replace_applicator         :     0.000022s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.11%
    rewriter_after_opt_a                 :     0.000038s :  0.38%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000562s :  5.69%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.17%
    cconv                                :     0.000024s :  0.24%
    loop_unroll                          :     0.000458s :  4.63%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.25%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000023s :  0.24%
    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  :     0.000007s :  0.07%
    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 :  0.52%
    cse_after_recomputation.cse          :     0.000012s :  0.12%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000521s :  5.27%
    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 :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.40%


Time group info:
------[substitution.]   0.000227    28
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000006s :      4: substitution.graph_param_transform
81.39% :     0.000185s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.99% :     0.000007s :      4: substitution.replace_old_param
 7.69% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005342     2
80.99% :     0.004326s :      1: type_inference.infer
19.01% :     0.001015s :      1: type_inference.specialize
------[replace.]   0.000085     8
57.94% :     0.000049s :      4: replace.inline
42.06% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     8
92.16% :     0.000182s :      4: match.inline
 7.84% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   912
 1.28% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.01% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     13: predicate.addn_check_dump
 1.33% :     0.000002s :     13: predicate.addn_zero_filter
 1.94% :     0.000003s :     13: predicate.arithmetic_simplify
 1.47% :     0.000002s :     13: predicate.cast_eliminate
 0.51% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     13: predicate.compare_switch_simplify
 1.30% :     0.000002s :     13: predicate.depend_value_elim
 1.08% :     0.000001s :     13: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      4: predicate.elim_not_effective
 0.56% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     13: predicate.environ_get_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.91% :     0.000001s :      8: predicate.get_grad_eliminate
 0.34% :     0.000000s :      4: predicate.graph_param_transform
 5.49% :     0.000008s :     29: predicate.inline
 1.10% :     0.000002s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      8: predicate.less_batch_normalization
 1.64% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     21: predicate.load_eliminater
 1.10% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.99% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     13: predicate.merge_addn
 1.19% :     0.000002s :     13: predicate.minmaximum_grad
 1.37% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.44% :     0.000003s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     13: predicate.reduce_eliminate
 1.67% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.88% :     0.000003s :     25: predicate.replace_applicator
 0.76% :     0.000001s :      8: predicate.replace_old_param
 0.27% :     0.000000s :      4: predicate.reset_defer_inline
 1.18% :     0.000002s :     13: predicate.reshape_eliminate
 1.23% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     13: predicate.same_eliminate
 0.65% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      8: predicate.special_op_eliminate
 0.94% :     0.000001s :      8: predicate.specialize_transform
 1.23% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     21: predicate.switch_defer_inline
 2.27% :     0.000003s :     21: predicate.switch_layer_defer_inline
 8.13% :     0.000011s :     67: predicate.switch_simplify
 1.20% :     0.000002s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.38% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.19% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.27% :     0.000005s :     29: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000661    12
49.37% :     0.000326s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.65% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.98% :     0.000297s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015935    72
 0.35% :     0.000055s :      1: add_recomputation
 0.47% :     0.000074s :      1: auto_monad
 0.13% :     0.000021s :      1: auto_monad_reorder
 2.30% :     0.000367s :      1: bootstrap
 0.17% :     0.000027s :      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
 0.17% :     0.000027s :      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
36.74% :     0.005855s :      1: jit_opt_a
 1.00% :     0.000160s :      1: jit_opt_after_cconv
 0.37% :     0.000059s :      1: jit_opt_b
 2.93% :     0.000467s :      1: loop_unroll
 3.58% :     0.000571s :      1: mutable_eliminate
 6.35% :     0.001013s :     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
 0.10% :     0.000015s :      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
 3.33% :     0.000530s :      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% :     0.000042s :      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
 2.51% :     0.000401s :      1: renormalize.infer
 2.05% :     0.000327s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000041s :      1: rewriter_after_opt_a
 0.60% :     0.000096s :      1: rewriter_before_opt_a
 0.50% :     0.000080s :      1: symbol_engine_optimizer
34.10% :     0.005433s :      1: type_inference

[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:03.274.53 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffdafe8f700> should be wrapped by jit only once.
[WARNING] ME(53359:281473513017136,ForkProcess-1):2026-01-29-17:46:03.281.65 [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 = 0.0766009, [33]
[bootstrap]: 0.000213
[type_inference]: 0.0318957
[event_method]: 0.0001535
[auto_monad]: 0.00022159
[graph_reusing]: 8.05999e-06
[pre_auto_parallel]: 3.51001e-06
[py_interpret_to_execute]: 4.389e-05
[rewriter_before_opt_a]: 0.00014052
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.0327198, [3]
    [Cycle 1]: 0.0248728, [27]
        [switch_simplify]: 8.967e-05
        [loop_unroll]: 5.083e-05
        [a_1]: 0.00123019
        [with_stream_mark]: 3.961e-05
        [recompute_prepare]: 2.925e-05
        [updatestate_depend_eliminate]: 1.399e-05
        [updatestate_assign_eliminate]: 1.151e-05
        [updatestate_loads_eliminate]: 1.001e-05
        [parameter_eliminate]: 3.71999e-06
        [specialize_transform]: 2.1e-05
        [updatestate_useless_node_eliminater]: 2.339e-05
        [accelerated_algorithm]: 3.199e-05
        [meta_shard_fg_expand]: 5.50001e-06
        [get_grad_eliminate_]: 1.932e-05
        [merge_forward]: 1.293e-05
        [cell_reuse_recompute_pass]: 1.24998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 4.396e-05
        [j_node_and_user_rematch]: 3.579e-05
        [meta_fg_expand]: 0.00692524
        [replace_old_param]: 0.00013146
        [inline_without_move]: 0.00012546
        [renormalize]: 0.0151677
        [add_forward_monad_depend]: 2.013e-05
        [auto_monad_grad]: 9.92001e-06
        [auto_monad_eliminator]: 8.052e-05
        [cse]: 0.00028841
        [replace_applicator]: 0.00016312
    [Cycle 2]: 0.00309579, [27]
        [switch_simplify]: 6.089e-05
        [loop_unroll]: 5.777e-05
        [a_1]: 0.00103096
        [with_stream_mark]: 1.885e-05
        [recompute_prepare]: 1.383e-05
        [updatestate_depend_eliminate]: 2.752e-05
        [updatestate_assign_eliminate]: 5.49e-06
        [updatestate_loads_eliminate]: 4e-06
        [parameter_eliminate]: 1.29998e-06
        [specialize_transform]: 1.05e-05
        [updatestate_useless_node_eliminater]: 1.22e-05
        [accelerated_algorithm]: 9.07999e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 8.02e-06
        [merge_forward]: 5.57999e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.876e-05
        [j_node_and_user_rematch]: 1.34e-05
        [meta_fg_expand]: 8.569e-05
        [replace_old_param]: 1.371e-05
        [inline_without_move]: 8.27e-06
        [renormalize]: 0.00137929
        [add_forward_monad_depend]: 7.01999e-06
        [auto_monad_grad]: 2.62001e-06
        [auto_monad_eliminator]: 2.018e-05
        [cse]: 7.546e-05
        [replace_applicator]: 2.194e-05
    [Cycle 3]: 0.0005553, [27]
        [switch_simplify]: 1.011e-05
        [loop_unroll]: 8.80001e-06
        [a_1]: 0.00018317
        [with_stream_mark]: 1.74e-05
        [recompute_prepare]: 1.067e-05
        [updatestate_depend_eliminate]: 8.09997e-06
        [updatestate_assign_eliminate]: 5.09e-06
        [updatestate_loads_eliminate]: 4.49998e-06
        [parameter_eliminate]: 2.03002e-06
        [specialize_transform]: 1.119e-05
        [updatestate_useless_node_eliminater]: 1.333e-05
        [accelerated_algorithm]: 9.07999e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 8.41002e-06
        [merge_forward]: 5.42999e-06
        [cell_reuse_recompute_pass]: 2.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.857e-05
        [j_node_and_user_rematch]: 1.374e-05
        [meta_fg_expand]: 3.3e-06
        [replace_old_param]: 1.229e-05
        [inline_without_move]: 8.07e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 2.08998e-06
        [auto_monad_grad]: 1.61002e-06
        [auto_monad_eliminator]: 1.498e-05
        [cse]: 3.657e-05
        [replace_applicator]: 8.99e-06
[py_interpret_to_execute_after_opt_a]: 1.876e-05
[rewriter_after_opt_a]: 6.054e-05
[convert_after_rewriter]: 9.00999e-06
[order_py_execute_after_rewriter]: 5.89e-06
[mutable_eliminate]: 0.00072025
[jit_opt_b]: 6.983e-05, [1]
    [Cycle 1]: 6.321e-05, [2]
        [frontend_op_eliminate]: 2.633e-05
        [inline_after_opt_a]: 2.489e-05
[cconv]: 2.131e-05
[loop_unroll]: 0.00049097
[jit_opt_after_cconv]: 0.00020469, [1]
    [Cycle 1]: 0.00019704, [11]
        [c_1]: 4.746e-05
        [parameter_eliminate]: 2.65002e-06
        [updatestate_depend_eliminate]: 7.86001e-06
        [updatestate_assign_eliminate]: 5.14e-06
        [updatestate_loads_eliminate]: 4.63001e-06
        [cse]: 3.391e-05
        [call_graph_tuple_transform]: 2.28e-05
        [tuple_list_get_item_eliminator]: 8.62e-06
        [none_parameter_eliminate]: 9.30013e-07
        [renormalize]: 5.19998e-07
        [switch_simplify]: 8.3e-06
[remove_dup_value]: 1.728e-05
[partial_unused_args_eliminate]: 1.76e-06
[environ_conv]: 9.02999e-06
[add_recomputation]: 5.376e-05
[cse_after_recomputation]: 3.495e-05, [1]
    [Cycle 1]: 2.879e-05, [1]
        [cse]: 2.205e-05
[auto_monad_reorder]: 2.253e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 6.83e-06
[opt_after_jit_grad]: 0.00054705
[symbol_engine_optimizer]: 0.00010991, [1]
    [Cycle 1]: 0.00010309, [6]
        [build]: 1.19e-05
        [elim_shapecalc]: 1.338e-05
        [elim_not_effective]: 2.146e-05
        [opt_reshape]: 9.97999e-06
        [fold_const_symbol]: 1.419e-05
        [renormalize]: 4.39992e-07
[validate]: 4.629e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.00850471
[execute]: 7.63001e-06
Sums
    bootstrap                            :     0.000213s :  0.30%
    type_inference                       :     0.031896s : 44.65%
    event_method                         :     0.000153s :  0.21%
    auto_monad                           :     0.000222s :  0.31%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000004s :  0.00%
    py_interpret_to_execute              :     0.000044s :  0.06%
    rewriter_before_opt_a                :     0.000141s :  0.20%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000161s :  0.22%
    jit_opt_a.loop_unroll                :     0.000117s :  0.16%
    jit_opt_a.a_1                        :     0.002444s :  3.42%
    jit_opt_a.with_stream_mark           :     0.000076s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000054s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000050s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000022s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000019s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000007s :  0.01%
    jit_opt_a.specialize_transform       :     0.000043s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000049s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000050s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000010s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000036s :  0.05%
    jit_opt_a.merge_forward              :     0.000024s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000081s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000063s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.007014s :  9.82%
    jit_opt_a.replace_old_param          :     0.000157s :  0.22%
    jit_opt_a.inline_without_move        :     0.000142s :  0.20%
    jit_opt_a.renormalize                :     0.016547s : 23.16%
    jit_opt_a.add_forward_monad_depend   :     0.000029s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000014s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000116s :  0.16%
    jit_opt_a.cse                        :     0.000400s :  0.56%
    jit_opt_a.replace_applicator         :     0.000194s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.03%
    rewriter_after_opt_a                 :     0.000061s :  0.08%
    convert_after_rewriter               :     0.000009s :  0.01%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000720s :  1.01%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.03%
    cconv                                :     0.000021s :  0.03%
    loop_unroll                          :     0.000491s :  0.69%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.07%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  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                    :     0.000054s :  0.08%
    cse_after_recomputation.cse          :     0.000022s :  0.03%
    auto_monad_reorder                   :     0.000023s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000547s :  0.77%
    symbol_engine_optimizer.build        :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000046s :  0.06%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008505s : 11.90%
    execute                              :     0.000008s :  0.01%


Time group info:
------[substitution.]   0.001742   160
 1.17% :     0.000020s :      8: substitution.depend_value_elim
 0.17% :     0.000003s :      4: substitution.elim_not_effective
 0.14% :     0.000002s :      4: substitution.fold_const_symbol
60.25% :     0.001050s :      4: substitution.getattr_setattr_resolve
 0.31% :     0.000005s :      5: substitution.graph_param_transform
23.01% :     0.000401s :     10: substitution.inline
 2.04% :     0.000036s :      4: substitution.inline_without_move
 0.66% :     0.000012s :     20: substitution.j_node_and_user_rematch
 0.58% :     0.000010s :      5: substitution.minmaximum_grad
 0.25% :     0.000004s :      3: substitution.partial_eliminate
 0.89% :     0.000015s :     20: substitution.remove_not_recompute_node
 2.70% :     0.000047s :     13: substitution.replace_applicator
 0.88% :     0.000015s :     18: substitution.replace_old_param
 0.20% :     0.000004s :      1: substitution.set_cell_output_no_recompute
 1.22% :     0.000021s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.77% :     0.000013s :      5: substitution.tuple_list_get_item_depend_reorder
 2.02% :     0.000035s :     10: substitution.tuple_list_get_item_eliminator
 1.03% :     0.000018s :      8: substitution.updatestate_pure_node_eliminater
 1.72% :     0.000030s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.031808     2
94.07% :     0.029922s :      1: type_inference.infer
 5.93% :     0.001886s :      1: type_inference.specialize
------[replace.]   0.000315    20
20.94% :     0.000066s :      3: replace.getattr_setattr_resolve
30.94% :     0.000098s :     10: replace.inline
 9.74% :     0.000031s :      1: replace.replace_applicator
31.18% :     0.000098s :      5: replace.tuple_list_get_item_eliminator
 7.21% :     0.000023s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001422    20
69.66% :     0.000990s :      3: match.getattr_setattr_resolve
27.71% :     0.000394s :     10: match.inline
 0.71% :     0.000010s :      1: match.replace_applicator
 1.34% :     0.000019s :      5: match.tuple_list_get_item_eliminator
 0.60% :     0.000008s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000476  2979
 1.43% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.49% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.36% :     0.000007s :     46: predicate.addn_check_dump
 1.59% :     0.000008s :     46: predicate.addn_zero_filter
 2.95% :     0.000014s :     46: predicate.arithmetic_simplify
 1.80% :     0.000009s :     46: predicate.cast_eliminate
 0.21% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.30% :     0.000006s :     46: predicate.compare_switch_simplify
 1.44% :     0.000007s :     46: predicate.depend_value_elim
 1.23% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.43% :     0.000007s :     46: predicate.dict_get_item_eliminator
 1.34% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.36% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.16% :     0.000001s :      5: predicate.elim_not_effective
 0.23% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000007s :     46: predicate.environ_add_const_eliminate
 1.20% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.38% :     0.000007s :     46: predicate.environ_get_depend_swap
 1.46% :     0.000007s :     46: predicate.environ_get_eliminate
 1.18% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      5: predicate.fold_const_symbol
 1.00% :     0.000005s :     27: predicate.get_grad_eliminate
 1.59% :     0.000008s :     20: predicate.getattr_setattr_resolve
 0.11% :     0.000001s :      5: predicate.graph_param_transform
 3.75% :     0.000018s :     72: predicate.inline
 3.30% :     0.000016s :     91: predicate.inline_without_move
 0.42% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.31% :     0.000006s :     27: predicate.less_batch_normalization
 1.73% :     0.000008s :     51: predicate.list_to_tuple_eliminator_
 1.77% :     0.000008s :     56: predicate.load_eliminater
 0.70% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.57% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.73% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000006s :     46: predicate.merge_addn
 1.29% :     0.000006s :     46: predicate.minmaximum_grad
 0.62% :     0.000003s :      5: predicate.mutable_eliminate
 0.21% :     0.000001s :      5: predicate.opt_reshape
 2.08% :     0.000010s :     56: predicate.partial_eliminate
 1.28% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.96% :     0.000009s :     46: predicate.reduce_eliminate
 1.61% :     0.000008s :     51: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000003s :     27: predicate.remove_not_recompute_node
 2.77% :     0.000013s :    126: predicate.replace_applicator
 1.70% :     0.000008s :     91: predicate.replace_old_param
 0.12% :     0.000001s :      5: predicate.reset_defer_inline
 1.42% :     0.000007s :     46: predicate.reshape_eliminate
 1.64% :     0.000008s :     46: predicate.row_tensor_add_zeros_like
 0.32% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.48% :     0.000007s :     46: predicate.same_eliminate
 0.60% :     0.000003s :     29: predicate.set_cell_output_no_recompute
 0.42% :     0.000002s :     10: predicate.special_op_eliminate
 1.00% :     0.000005s :     27: predicate.specialize_transform
 1.77% :     0.000008s :     46: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.21% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.46% :     0.000012s :     62: predicate.switch_defer_inline
 1.97% :     0.000009s :     62: predicate.switch_layer_defer_inline
 5.98% :     0.000028s :    188: predicate.switch_simplify
 1.44% :     0.000007s :     46: predicate.tile_eliminate
 1.35% :     0.000006s :     46: predicate.transpose_eliminate
 1.60% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.46% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 2.84% :     0.000014s :     61: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000009s :     46: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.67% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000014s :     84: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000008s :     46: predicate.value_based_eliminate
 0.15% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003521    48
67.38% :     0.002372s :     27: func_graph_cloner_run.FuncGraphClonerGraph
32.62% :     0.001149s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.098044    93
 0.06% :     0.000057s :      1: add_recomputation
 0.24% :     0.000231s :      1: auto_monad
 0.03% :     0.000025s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.23% :     0.000229s :      1: bootstrap
 0.02% :     0.000024s :      1: cconv
 0.01% :     0.000012s :      1: convert_after_rewriter
 0.04% :     0.000037s :      1: cse_after_recomputation
 0.01% :     0.000011s :      1: environ_conv
 0.16% :     0.000161s :      1: event_method
 0.01% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000012s :      1: graph_reusing
33.38% :     0.032724s :      1: jit_opt_a
 0.21% :     0.000208s :      1: jit_opt_after_cconv
 0.07% :     0.000072s :      1: jit_opt_b
 0.51% :     0.000500s :      1: loop_unroll
 0.74% :     0.000730s :      1: mutable_eliminate
 3.60% :     0.003529s :     39: opt.transform.jit_opt_a
 0.09% :     0.000083s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000044s :      4: opt.transform.jit_opt_b
 0.02% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 1.21% :     0.001182s :      2: opt.transform.opt_resolve
 0.06% :     0.000055s :      4: opt.transform.symbol_engine_opt
 0.57% :     0.000557s :      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% :     0.000006s :      1: pre_auto_parallel
 0.05% :     0.000047s :      1: py_interpret_to_execute
 0.02% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000020s :      1: remove_dup_value
13.31% :     0.013047s :      2: renormalize.infer
 3.55% :     0.003479s :      2: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000065s :      1: rewriter_after_opt_a
 0.15% :     0.000144s :      1: rewriter_before_opt_a
 0.11% :     0.000113s :      1: symbol_engine_optimizer
 8.69% :     0.008520s :      1: task_emit
32.55% :     0.031909s :      1: type_inference
 0.07% :     0.000070s :      1: validate


TotalTime = 0.047032, [33]
[bootstrap]: 0.00035811
[type_inference]: 0.025521
[event_method]: 2.19e-05
[auto_monad]: 0.00016849
[graph_reusing]: 7.36999e-06
[pre_auto_parallel]: 2.94999e-06
[py_interpret_to_execute]: 0.00012175
[rewriter_before_opt_a]: 0.00010701
[expand_dump_flag]: 3.95e-06
[jit_opt_a]: 0.00669463, [2]
    [Cycle 1]: 0.00243673, [27]
        [switch_simplify]: 6.675e-05
        [loop_unroll]: 3.658e-05
        [a_1]: 0.00073213
        [with_stream_mark]: 1.969e-05
        [recompute_prepare]: 1.086e-05
        [updatestate_depend_eliminate]: 5.94e-06
        [updatestate_assign_eliminate]: 9.47001e-06
        [updatestate_loads_eliminate]: 4.77998e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 1.013e-05
        [updatestate_useless_node_eliminater]: 1.169e-05
        [accelerated_algorithm]: 8.23999e-06
        [meta_shard_fg_expand]: 3.14999e-06
        [get_grad_eliminate_]: 7.98999e-06
        [merge_forward]: 5.25001e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.914e-05
        [j_node_and_user_rematch]: 1.354e-05
        [meta_fg_expand]: 4.55001e-06
        [replace_old_param]: 1.2e-05
        [inline_without_move]: 8.12e-06
        [renormalize]: 0.00118757
        [add_forward_monad_depend]: 5.92999e-06
        [auto_monad_grad]: 2.17999e-06
        [auto_monad_eliminator]: 2.041e-05
        [cse]: 3.374e-05
        [replace_applicator]: 1.688e-05
    [Cycle 2]: 0.00047356, [27]
        [switch_simplify]: 9.21002e-06
        [loop_unroll]: 7.73999e-06
        [a_1]: 0.00016507
        [with_stream_mark]: 1.394e-05
        [recompute_prepare]: 8.57e-06
        [updatestate_depend_eliminate]: 5.09e-06
        [updatestate_assign_eliminate]: 5.26002e-06
        [updatestate_loads_eliminate]: 3.71999e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 8.3e-06
        [updatestate_useless_node_eliminater]: 1.047e-05
        [accelerated_algorithm]: 7.7e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 7.30998e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 2.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.696e-05
        [j_node_and_user_rematch]: 1.265e-05
        [meta_fg_expand]: 2.91e-06
        [replace_old_param]: 9.78998e-06
        [inline_without_move]: 7.48e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 1.006e-05
        [cse]: 2.126e-05
        [replace_applicator]: 8.61997e-06
[py_interpret_to_execute_after_opt_a]: 1.343e-05
[rewriter_after_opt_a]: 0.0003003
[convert_after_rewriter]: 1.1e-05
[order_py_execute_after_rewriter]: 7.46001e-06
[mutable_eliminate]: 0.0007875
[jit_opt_b]: 6.948e-05, [1]
    [Cycle 1]: 6.056e-05, [2]
        [frontend_op_eliminate]: 2.405e-05
        [inline_after_opt_a]: 2.408e-05
[cconv]: 2.955e-05
[loop_unroll]: 0.00054107
[jit_opt_after_cconv]: 0.00022092, [1]
    [Cycle 1]: 0.00021315, [11]
        [c_1]: 4.799e-05
        [parameter_eliminate]: 3.23e-06
        [updatestate_depend_eliminate]: 9.25001e-06
        [updatestate_assign_eliminate]: 4.65001e-06
        [updatestate_loads_eliminate]: 4.12998e-06
        [cse]: 2.922e-05
        [call_graph_tuple_transform]: 2.415e-05
        [tuple_list_get_item_eliminator]: 8.57998e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 9.27999e-06
[remove_dup_value]: 1.839e-05
[partial_unused_args_eliminate]: 2.38998e-06
[environ_conv]: 1.218e-05
[add_recomputation]: 7.391e-05
[cse_after_recomputation]: 3.115e-05, [1]
    [Cycle 1]: 2.483e-05, [1]
        [cse]: 1.876e-05
[auto_monad_reorder]: 2.621e-05
[get_jit_bprop_graph]: 2.29001e-06
[rewriter_after_jit_bprop_graph]: 5.61998e-06
[opt_after_jit_grad]: 0.00058941
[symbol_engine_optimizer]: 0.00010291, [1]
    [Cycle 1]: 9.6e-05, [6]
        [build]: 1.324e-05
        [elim_shapecalc]: 1.169e-05
        [elim_not_effective]: 1.837e-05
        [opt_reshape]: 9.12001e-06
        [fold_const_symbol]: 1.336e-05
        [renormalize]: 4.10015e-07
[validate]: 6.472e-05
[backend_pass]: 1.22e-06
[task_emit]: 0.0108762
[execute]: 7.51999e-06
Sums
    bootstrap                            :     0.000358s :  0.84%
    type_inference                       :     0.025521s : 59.94%
    event_method                         :     0.000022s :  0.05%
    auto_monad                           :     0.000168s :  0.40%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000122s :  0.29%
    rewriter_before_opt_a                :     0.000107s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000076s :  0.18%
    jit_opt_a.loop_unroll                :     0.000044s :  0.10%
    jit_opt_a.a_1                        :     0.000897s :  2.11%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    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 :     0.000036s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001188s :  2.79%
    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      :     0.000030s :  0.07%
    jit_opt_a.cse                        :     0.000055s :  0.13%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000300s :  0.71%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000788s :  1.85%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000030s :  0.07%
    loop_unroll                          :     0.000541s :  1.27%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000029s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000018s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000074s :  0.17%
    cse_after_recomputation.cse          :     0.000019s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000589s :  1.38%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    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 :     0.000013s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000065s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010876s : 25.55%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000298    47
 3.22% :     0.000010s :      2: substitution.depend_value_elim
 0.89% :     0.000003s :      4: substitution.elim_not_effective
 0.71% :     0.000002s :      4: substitution.fold_const_symbol
 2.26% :     0.000007s :      5: substitution.graph_param_transform
76.69% :     0.000229s :      4: substitution.inline
 1.56% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.60% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.55% :     0.000005s :      2: substitution.replace_old_param
 4.35% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 2.84% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.32% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.025437     2
93.88% :     0.023880s :      1: type_inference.infer
 6.12% :     0.001557s :      1: type_inference.specialize
------[replace.]   0.000095     6
66.24% :     0.000063s :      4: replace.inline
33.76% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000238     6
95.09% :     0.000226s :      4: match.inline
 4.91% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     15: predicate.addn_check_dump
 1.53% :     0.000003s :     15: predicate.addn_zero_filter
 2.25% :     0.000004s :     15: predicate.arithmetic_simplify
 1.24% :     0.000002s :     15: predicate.cast_eliminate
 0.47% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     15: predicate.compare_switch_simplify
 1.48% :     0.000003s :     15: predicate.depend_value_elim
 1.21% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.88% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.50% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     15: predicate.environ_get_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.09% :     0.000002s :     10: predicate.get_grad_eliminate
 0.36% :     0.000001s :      5: predicate.graph_param_transform
 5.11% :     0.000009s :     31: predicate.inline
 1.07% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.15% :     0.000002s :     10: predicate.less_batch_normalization
 1.38% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.94% :     0.000003s :     22: predicate.load_eliminater
 1.37% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.52% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.20% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     15: predicate.merge_addn
 1.13% :     0.000002s :     15: predicate.minmaximum_grad
 1.49% :     0.000003s :      5: predicate.mutable_eliminate
 0.67% :     0.000001s :      5: predicate.opt_reshape
 2.46% :     0.000004s :     22: predicate.partial_eliminate
 1.14% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.82% :     0.000003s :     15: predicate.reduce_eliminate
 1.57% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.86% :     0.000003s :     27: predicate.replace_applicator
 0.69% :     0.000001s :     10: predicate.replace_old_param
 0.27% :     0.000000s :      5: predicate.reset_defer_inline
 1.26% :     0.000002s :     15: predicate.reshape_eliminate
 1.26% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :     10: predicate.special_op_eliminate
 1.08% :     0.000002s :     10: predicate.specialize_transform
 1.39% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.58% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.43% :     0.000004s :     21: predicate.switch_defer_inline
 1.80% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.56% :     0.000011s :     68: predicate.switch_simplify
 1.37% :     0.000002s :     15: predicate.tile_eliminate
 1.48% :     0.000003s :     15: predicate.transpose_eliminate
 1.54% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.78% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.22% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000003s :     15: predicate.value_based_eliminate
 0.42% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000870    13
41.96% :     0.000365s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.00% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.04% :     0.000453s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049617    76
 0.16% :     0.000077s :      1: add_recomputation
 0.35% :     0.000175s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.75% :     0.000372s :      1: bootstrap
 0.07% :     0.000032s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000027s :      1: event_method
 0.02% :     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
13.50% :     0.006698s :      1: jit_opt_a
 0.45% :     0.000224s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 1.11% :     0.000550s :      1: loop_unroll
 1.61% :     0.000797s :      1: mutable_eliminate
 2.43% :     0.001204s :     26: opt.transform.jit_opt_a
 0.17% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000041s :      4: opt.transform.jit_opt_b
 0.04% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.21% :     0.000598s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000126s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.45% :     0.000717s :      1: renormalize.infer
 0.93% :     0.000462s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.62% :     0.000306s :      1: rewriter_after_opt_a
 0.23% :     0.000113s :      1: rewriter_before_opt_a
 0.21% :     0.000106s :      1: symbol_engine_optimizer
21.95% :     0.010890s :      1: task_emit
51.47% :     0.025538s :      1: type_inference
 0.18% :     0.000090s :      1: validate


TotalTime = 0.0147286, [30]
[bootstrap]: 0.00032113
[type_inference]: 0.00546983
[event_method]: 2.317e-05
[auto_monad]: 6.706e-05
[graph_reusing]: 6.17001e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.797e-05
[rewriter_before_opt_a]: 9.085e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00605936, [2]
    [Cycle 1]: 0.00203512, [27]
        [switch_simplify]: 6.518e-05
        [loop_unroll]: 3.507e-05
        [a_1]: 0.00067267
        [with_stream_mark]: 2.436e-05
        [recompute_prepare]: 1.05e-05
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 3.66999e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [parameter_eliminate]: 2.36e-06
        [specialize_transform]: 6.92002e-06
        [updatestate_useless_node_eliminater]: 6.58e-06
        [accelerated_algorithm]: 6.52001e-06
        [meta_shard_fg_expand]: 2.76999e-06
        [get_grad_eliminate_]: 5.84e-06
        [merge_forward]: 4.19002e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.636e-05
        [j_node_and_user_rematch]: 1.122e-05
        [meta_fg_expand]: 2.40002e-06
        [replace_old_param]: 1.151e-05
        [inline_without_move]: 6.12001e-06
        [renormalize]: 0.00084428
        [add_forward_monad_depend]: 7.56001e-06
        [auto_monad_grad]: 2.97002e-06
        [auto_monad_eliminator]: 1.681e-05
        [cse]: 3.562e-05
        [replace_applicator]: 1.818e-05
    [Cycle 2]: 0.00037576, [27]
        [switch_simplify]: 6.84001e-06
        [loop_unroll]: 6.04001e-06
        [a_1]: 0.00011279
        [with_stream_mark]: 1.246e-05
        [recompute_prepare]: 5.62999e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.38998e-06
        [parameter_eliminate]: 1.57001e-06
        [specialize_transform]: 5.64e-06
        [updatestate_useless_node_eliminater]: 5.66998e-06
        [accelerated_algorithm]: 6.14001e-06
        [meta_shard_fg_expand]: 2.26998e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 3.77998e-06
        [cell_reuse_recompute_pass]: 2.33998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.504e-05
        [j_node_and_user_rematch]: 8.92999e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 1.044e-05
        [inline_without_move]: 5.36998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.81e-06
        [auto_monad_grad]: 1.00001e-06
        [auto_monad_eliminator]: 7.45998e-06
        [cse]: 1.49e-05
        [replace_applicator]: 6.46999e-06
[py_interpret_to_execute_after_opt_a]: 1.604e-05
[rewriter_after_opt_a]: 4.205e-05
[convert_after_rewriter]: 8.19998e-06
[order_py_execute_after_rewriter]: 5.61998e-06
[mutable_eliminate]: 0.00077746
[jit_opt_b]: 6.26e-05, [1]
    [Cycle 1]: 5.318e-05, [2]
        [frontend_op_eliminate]: 1.969e-05
        [inline_after_opt_a]: 1.886e-05
[cconv]: 3.136e-05
[loop_unroll]: 0.00053843
[jit_opt_after_cconv]: 0.00017994, [1]
    [Cycle 1]: 0.00017195, [11]
        [c_1]: 2.677e-05
        [parameter_eliminate]: 4.76997e-06
        [updatestate_depend_eliminate]: 7.52998e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 3.04999e-06
        [cse]: 3.482e-05
        [call_graph_tuple_transform]: 2.361e-05
        [tuple_list_get_item_eliminator]: 6.19999e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 6.18002e-06
[remove_dup_value]: 1.664e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 6.89999e-06
[add_recomputation]: 5.333e-05
[cse_after_recomputation]: 2.628e-05, [1]
    [Cycle 1]: 1.998e-05, [1]
        [cse]: 1.294e-05
[auto_monad_reorder]: 1.743e-05
[get_jit_bprop_graph]: 2.05002e-06
[rewriter_after_jit_bprop_graph]: 7.68001e-06
[opt_after_jit_grad]: 0.00054103
[symbol_engine_optimizer]: 7.968e-05, [1]
    [Cycle 1]: 7.177e-05, [6]
        [build]: 4.01001e-06
        [elim_shapecalc]: 8.37e-06
        [elim_not_effective]: 1.489e-05
        [opt_reshape]: 6.69999e-06
        [fold_const_symbol]: 9.24e-06
        [renormalize]: 6.60017e-07
[validate]: 4.046e-05
Sums
    bootstrap                            :     0.000321s :  3.08%
    type_inference                       :     0.005470s : 52.45%
    event_method                         :     0.000023s :  0.22%
    auto_monad                           :     0.000067s :  0.64%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.36%
    rewriter_before_opt_a                :     0.000091s :  0.87%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000072s :  0.69%
    jit_opt_a.loop_unroll                :     0.000041s :  0.39%
    jit_opt_a.a_1                        :     0.000785s :  7.53%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  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 :     0.000012s :  0.12%
    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 :  0.11%
    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 :     0.000031s :  0.30%
    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          :     0.000022s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000844s :  8.10%
    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      :     0.000024s :  0.23%
    jit_opt_a.cse                        :     0.000051s :  0.48%
    jit_opt_a.replace_applicator         :     0.000025s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.15%
    rewriter_after_opt_a                 :     0.000042s :  0.40%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000777s :  7.45%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.18%
    cconv                                :     0.000031s :  0.30%
    loop_unroll                          :     0.000538s :  5.16%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000035s :  0.33%
    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                     :     0.000017s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000007s :  0.07%
    add_recomputation                    :     0.000053s :  0.51%
    cse_after_recomputation.cse          :     0.000013s :  0.12%
    auto_monad_reorder                   :     0.000017s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.07%
    opt_after_jit_grad                   :     0.000541s :  5.19%
    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 :  0.06%
    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.]   0.000258    28
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000007s :      4: substitution.graph_param_transform
83.41% :     0.000216s :      4: substitution.inline
 1.48% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.78% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.79% :     0.000007s :      4: substitution.replace_old_param
 6.69% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005399     2
81.61% :     0.004406s :      1: type_inference.infer
18.39% :     0.000993s :      1: type_inference.specialize
------[replace.]   0.000089     8
57.64% :     0.000051s :      4: replace.inline
42.36% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     8
93.30% :     0.000213s :      4: match.inline
 6.70% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   912
 1.19% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.06% :     0.000002s :     13: predicate.addn_check_dump
 1.42% :     0.000002s :     13: predicate.addn_zero_filter
 1.68% :     0.000002s :     13: predicate.arithmetic_simplify
 1.26% :     0.000002s :     13: predicate.cast_eliminate
 0.45% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     13: predicate.compare_switch_simplify
 1.17% :     0.000002s :     13: predicate.depend_value_elim
 1.04% :     0.000001s :     13: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.13% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      4: predicate.elim_not_effective
 0.47% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.05% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     13: predicate.environ_get_eliminate
 1.08% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.85% :     0.000001s :      8: predicate.get_grad_eliminate
 0.36% :     0.000001s :      4: predicate.graph_param_transform
 6.38% :     0.000009s :     29: predicate.inline
 0.90% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      8: predicate.less_batch_normalization
 1.63% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     21: predicate.load_eliminater
 1.52% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.84% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.23% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     13: predicate.merge_addn
 1.04% :     0.000001s :     13: predicate.minmaximum_grad
 1.76% :     0.000003s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.34% :     0.000003s :     21: predicate.partial_eliminate
 1.30% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.38% :     0.000002s :     13: predicate.reduce_eliminate
 1.60% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.90% :     0.000003s :     25: predicate.replace_applicator
 0.60% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.21% :     0.000002s :     13: predicate.reshape_eliminate
 1.17% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     13: predicate.same_eliminate
 0.81% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      8: predicate.special_op_eliminate
 0.93% :     0.000001s :      8: predicate.specialize_transform
 1.21% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     21: predicate.switch_defer_inline
 2.25% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.56% :     0.000011s :     67: predicate.switch_simplify
 1.28% :     0.000002s :     13: predicate.tile_eliminate
 1.15% :     0.000002s :     13: predicate.transpose_eliminate
 1.25% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     13: predicate.value_based_eliminate
 0.36% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000689    12
49.15% :     0.000338s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.61% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.25% :     0.000298s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016706    72
 0.34% :     0.000057s :      1: add_recomputation
 0.42% :     0.000070s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 2.01% :     0.000336s :      1: bootstrap
 0.21% :     0.000035s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000028s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.17% :     0.000028s :      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
36.29% :     0.006063s :      1: jit_opt_a
 1.10% :     0.000183s :      1: jit_opt_after_cconv
 0.39% :     0.000066s :      1: jit_opt_b
 3.29% :     0.000549s :      1: loop_unroll
 4.74% :     0.000792s :      1: mutable_eliminate
 6.24% :     0.001042s :     26: opt.transform.jit_opt_a
 0.36% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000032s :      4: opt.transform.jit_opt_b
 0.11% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.29% :     0.000550s :      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% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000019s :      1: remove_dup_value
 2.57% :     0.000429s :      1: renormalize.infer
 2.43% :     0.000406s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000045s :      1: rewriter_after_opt_a
 0.56% :     0.000094s :      1: rewriter_before_opt_a
 0.49% :     0.000082s :      1: symbol_engine_optimizer
32.86% :     0.005489s :      1: type_inference

[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:03.531.56 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffe2cafe0d0> should be wrapped by jit only once.
[WARNING] ME(53639:281473513017136,ForkProcess-4):2026-01-29-17:46:03.539.72 [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. 

TotalTime = 0.0423441, [33]
[bootstrap]: 0.00022205
[type_inference]: 0.0123666
[event_method]: 5.062e-05
[auto_monad]: 0.00010102
[graph_reusing]: 8.21002e-06
[pre_auto_parallel]: 3.18998e-06
[py_interpret_to_execute]: 2.9e-05
[rewriter_before_opt_a]: 0.00015179
[expand_dump_flag]: 3.50998e-06
[jit_opt_a]: 0.0194267, [3]
    [Cycle 1]: 0.0127476, [27]
        [switch_simplify]: 7.953e-05
        [loop_unroll]: 4.605e-05
        [a_1]: 0.00098356
        [with_stream_mark]: 4.131e-05
        [recompute_prepare]: 2.433e-05
        [updatestate_depend_eliminate]: 9.85002e-06
        [updatestate_assign_eliminate]: 7.55998e-06
        [updatestate_loads_eliminate]: 6.83e-06
        [parameter_eliminate]: 3.30998e-06
        [specialize_transform]: 1.639e-05
        [updatestate_useless_node_eliminater]: 1.368e-05
        [accelerated_algorithm]: 1.548e-05
        [meta_shard_fg_expand]: 5.72999e-06
        [get_grad_eliminate_]: 1.417e-05
        [merge_forward]: 9.10999e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.984e-05
        [j_node_and_user_rematch]: 2.576e-05
        [meta_fg_expand]: 0.00199313
        [replace_old_param]: 7.68e-05
        [inline_without_move]: 6.324e-05
        [renormalize]: 0.00838832
        [add_forward_monad_depend]: 1.672e-05
        [auto_monad_grad]: 9.24e-06
        [auto_monad_eliminator]: 7.965e-05
        [cse]: 0.00028391
        [replace_applicator]: 9.718e-05
    [Cycle 2]: 0.00247334, [27]
        [switch_simplify]: 4.312e-05
        [loop_unroll]: 3.975e-05
        [a_1]: 0.00118808
        [with_stream_mark]: 1.939e-05
        [recompute_prepare]: 1.026e-05
        [updatestate_depend_eliminate]: 4.60001e-06
        [updatestate_assign_eliminate]: 4.26001e-06
        [updatestate_loads_eliminate]: 3.35e-06
        [parameter_eliminate]: 1.88002e-06
        [specialize_transform]: 6.89999e-06
        [updatestate_useless_node_eliminater]: 6.21e-06
        [accelerated_algorithm]: 6.46999e-06
        [meta_shard_fg_expand]: 2.86999e-06
        [get_grad_eliminate_]: 6.94999e-06
        [merge_forward]: 5.00999e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.699e-05
        [j_node_and_user_rematch]: 1.004e-05
        [meta_fg_expand]: 5.495e-05
        [replace_old_param]: 1.311e-05
        [inline_without_move]: 6.33e-06
        [renormalize]: 0.0007959
        [add_forward_monad_depend]: 4.68001e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.191e-05
        [cse]: 2.727e-05
        [replace_applicator]: 1.573e-05
    [Cycle 3]: 0.00037562, [27]
        [switch_simplify]: 6.82002e-06
        [loop_unroll]: 6.64999e-06
        [a_1]: 0.00011495
        [with_stream_mark]: 1.03e-05
        [recompute_prepare]: 6.15002e-06
        [updatestate_depend_eliminate]: 3.53999e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 6.40002e-06
        [updatestate_useless_node_eliminater]: 5.81e-06
        [accelerated_algorithm]: 6.44999e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 5.79999e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 2.31e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.657e-05
        [j_node_and_user_rematch]: 8.70999e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.69e-06
        [inline_without_move]: 5.64e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49998e-06
        [auto_monad_grad]: 1.00001e-06
        [auto_monad_eliminator]: 7.26999e-06
        [cse]: 1.712e-05
        [replace_applicator]: 6.83e-06
[py_interpret_to_execute_after_opt_a]: 1.381e-05
[rewriter_after_opt_a]: 4.125e-05
[convert_after_rewriter]: 7.73001e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.00082177
[jit_opt_b]: 5.758e-05, [1]
    [Cycle 1]: 4.927e-05, [2]
        [frontend_op_eliminate]: 1.915e-05
        [inline_after_opt_a]: 1.851e-05
[cconv]: 2.39e-05
[loop_unroll]: 0.00045698
[jit_opt_after_cconv]: 0.00016079, [1]
    [Cycle 1]: 0.00015365, [11]
        [c_1]: 2.499e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 6.97002e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [cse]: 2.643e-05
        [call_graph_tuple_transform]: 2.133e-05
        [tuple_list_get_item_eliminator]: 6.72002e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 6.68998e-06
[remove_dup_value]: 1.697e-05
[partial_unused_args_eliminate]: 2.30002e-06
[environ_conv]: 7.28e-06
[add_recomputation]: 4.716e-05
[cse_after_recomputation]: 2.438e-05, [1]
    [Cycle 1]: 1.923e-05, [1]
        [cse]: 1.333e-05
[auto_monad_reorder]: 1.739e-05
[get_jit_bprop_graph]: 1.61002e-06
[rewriter_after_jit_bprop_graph]: 4.27003e-06
[opt_after_jit_grad]: 0.00049626
[symbol_engine_optimizer]: 7.941e-05, [1]
    [Cycle 1]: 7.275e-05, [6]
        [build]: 3.45e-06
        [elim_shapecalc]: 8.96002e-06
        [elim_not_effective]: 1.465e-05
        [opt_reshape]: 6.98e-06
        [fold_const_symbol]: 9.91e-06
        [renormalize]: 6.50005e-07
[validate]: 3.785e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.00744044
[execute]: 7.53999e-06
Sums
    bootstrap                            :     0.000222s :  0.59%
    type_inference                       :     0.012367s : 32.98%
    event_method                         :     0.000051s :  0.13%
    auto_monad                           :     0.000101s :  0.27%
    graph_reusing                        :     0.000008s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.08%
    rewriter_before_opt_a                :     0.000152s :  0.40%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000129s :  0.35%
    jit_opt_a.loop_unroll                :     0.000092s :  0.25%
    jit_opt_a.a_1                        :     0.002287s :  6.10%
    jit_opt_a.with_stream_mark           :     0.000071s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000041s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000018s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000030s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000028s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000010s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000027s :  0.07%
    jit_opt_a.merge_forward              :     0.000018s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000063s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.002050s :  5.47%
    jit_opt_a.replace_old_param          :     0.000100s :  0.27%
    jit_opt_a.inline_without_move        :     0.000075s :  0.20%
    jit_opt_a.renormalize                :     0.009184s : 24.49%
    jit_opt_a.add_forward_monad_depend   :     0.000023s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000013s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000099s :  0.26%
    jit_opt_a.cse                        :     0.000328s :  0.88%
    jit_opt_a.replace_applicator         :     0.000120s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.04%
    rewriter_after_opt_a                 :     0.000041s :  0.11%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000822s :  2.19%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.05%
    cconv                                :     0.000024s :  0.06%
    loop_unroll                          :     0.000457s :  1.22%
    jit_opt_after_cconv.c_1              :     0.000025s :  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 :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.07%
    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 :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000047s :  0.13%
    cse_after_recomputation.cse          :     0.000013s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000496s :  1.32%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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                             :     0.000038s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007440s : 19.84%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000641   113
 0.31% :     0.000002s :      2: substitution.elim_not_effective
 0.22% :     0.000001s :      2: substitution.fold_const_symbol
 0.91% :     0.000006s :      4: substitution.graph_param_transform
65.80% :     0.000422s :     12: substitution.inline
 3.40% :     0.000022s :      2: substitution.inline_without_move
 1.52% :     0.000010s :     12: substitution.j_node_and_user_rematch
 1.87% :     0.000012s :      7: substitution.minmaximum_grad
 1.28% :     0.000008s :      5: substitution.partial_eliminate
 1.66% :     0.000011s :     12: substitution.remove_not_recompute_node
 5.78% :     0.000037s :     10: substitution.replace_applicator
 1.80% :     0.000012s :     14: substitution.replace_old_param
 0.46% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.18% :     0.000027s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.70% :     0.000017s :      7: substitution.tuple_list_get_item_depend_reorder
 8.13% :     0.000052s :     16: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012279     2
89.28% :     0.010963s :      1: type_inference.infer
10.72% :     0.001316s :      1: type_inference.specialize
------[replace.]   0.000192    21
59.16% :     0.000113s :     12: replace.inline
40.84% :     0.000078s :      9: replace.tuple_list_get_item_eliminator
------[match.]   0.000442    21
93.73% :     0.000415s :     12: match.inline
 6.27% :     0.000028s :      9: match.tuple_list_get_item_eliminator
------[predicate.]   0.000393  2828
 1.50% :     0.000006s :     47: predicate.accumulaten_eliminater
 0.41% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.42% :     0.000006s :     47: predicate.addn_check_dump
 1.71% :     0.000007s :     47: predicate.addn_zero_filter
 2.19% :     0.000009s :     47: predicate.arithmetic_simplify
 1.58% :     0.000006s :     47: predicate.cast_eliminate
 0.14% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.45% :     0.000006s :     47: predicate.compare_switch_simplify
 1.39% :     0.000005s :     47: 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
 1.50% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.30% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.17% :     0.000001s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.39% :     0.000005s :     47: predicate.environ_get_add_eliminate
 1.36% :     0.000005s :     47: predicate.environ_get_depend_swap
 1.41% :     0.000006s :     47: predicate.environ_get_eliminate
 1.38% :     0.000005s :     47: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.97% :     0.000004s :     21: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.72% :     0.000019s :     76: predicate.inline
 1.97% :     0.000008s :     47: predicate.inline_without_move
 0.37% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.18% :     0.000005s :     21: predicate.less_batch_normalization
 1.76% :     0.000007s :     56: predicate.list_to_tuple_eliminator_
 1.89% :     0.000007s :     60: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.33% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.68% :     0.000007s :     51: predicate.make_slice_get_slice_eliminator
 1.40% :     0.000006s :     47: predicate.merge_addn
 1.43% :     0.000006s :     47: predicate.minmaximum_grad
 0.55% :     0.000002s :      4: predicate.mutable_eliminate
 0.15% :     0.000001s :      4: predicate.opt_reshape
 2.32% :     0.000009s :     60: predicate.partial_eliminate
 1.48% :     0.000006s :     47: predicate.print_const_string_wrapper
 1.75% :     0.000007s :     47: predicate.reduce_eliminate
 1.91% :     0.000008s :     56: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.77% :     0.000011s :    108: predicate.replace_applicator
 1.24% :     0.000005s :     47: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.43% :     0.000006s :     47: predicate.reshape_eliminate
 1.52% :     0.000006s :     47: predicate.row_tensor_add_zeros_like
 0.37% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.53% :     0.000006s :     47: predicate.same_eliminate
 0.60% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.82% :     0.000003s :     21: predicate.specialize_transform
 1.63% :     0.000006s :     47: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000006s :     47: predicate.stack_unstack_eliminate
 0.15% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.76% :     0.000011s :     68: predicate.switch_defer_inline
 2.45% :     0.000010s :     68: predicate.switch_layer_defer_inline
 6.61% :     0.000026s :    169: predicate.switch_simplify
 1.48% :     0.000006s :     47: predicate.tile_eliminate
 1.43% :     0.000006s :     47: predicate.transpose_eliminate
 1.78% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.65% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.39% :     0.000013s :     64: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000008s :     47: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000007s :     56: predicate.tuple_to_list_eliminator_
 1.90% :     0.000007s :     60: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000011s :     81: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000007s :     47: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.19% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002053    31
61.40% :     0.001261s :     15: func_graph_cloner_run.FuncGraphClonerGraph
38.60% :     0.000793s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.054659    91
 0.09% :     0.000050s :      1: add_recomputation
 0.19% :     0.000105s :      1: auto_monad
 0.04% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.43% :     0.000236s :      1: bootstrap
 0.05% :     0.000027s :      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
 0.10% :     0.000056s :      1: event_method
 0.02% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000011s :      1: graph_reusing
35.55% :     0.019430s :      1: jit_opt_a
 0.30% :     0.000164s :      1: jit_opt_after_cconv
 0.11% :     0.000060s :      1: jit_opt_b
 0.85% :     0.000466s :      1: loop_unroll
 1.52% :     0.000832s :      1: mutable_eliminate
 5.50% :     0.003008s :     39: opt.transform.jit_opt_a
 0.10% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000031s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000037s :      4: opt.transform.symbol_engine_opt
 0.92% :     0.000504s :      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
 0.06% :     0.000031s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
13.25% :     0.007242s :      2: renormalize.infer
 3.52% :     0.001923s :      2: renormalize.specialize
 0.01% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000044s :      1: rewriter_after_opt_a
 0.28% :     0.000155s :      1: rewriter_before_opt_a
 0.15% :     0.000082s :      1: symbol_engine_optimizer
13.64% :     0.007457s :      1: task_emit
22.66% :     0.012385s :      1: type_inference
 0.11% :     0.000058s :      1: validate


TotalTime = 0.0138515, [30]
[bootstrap]: 0.00034136
[type_inference]: 0.00502834
[event_method]: 2.276e-05
[auto_monad]: 6.674e-05
[graph_reusing]: 6.11e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 3.483e-05
[rewriter_before_opt_a]: 9.054e-05
[expand_dump_flag]: 2.84001e-06
[jit_opt_a]: 0.00596561, [2]
    [Cycle 1]: 0.0019594, [27]
        [switch_simplify]: 6.33e-05
        [loop_unroll]: 3.467e-05
        [a_1]: 0.00072549
        [with_stream_mark]: 2.493e-05
        [recompute_prepare]: 8.94998e-06
        [updatestate_depend_eliminate]: 4.22003e-06
        [updatestate_assign_eliminate]: 3.49001e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 6.93998e-06
        [updatestate_useless_node_eliminater]: 6.17001e-06
        [accelerated_algorithm]: 6.24999e-06
        [meta_shard_fg_expand]: 2.28002e-06
        [get_grad_eliminate_]: 6.33998e-06
        [merge_forward]: 3.71001e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.659e-05
        [j_node_and_user_rematch]: 1.075e-05
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 1.135e-05
        [inline_without_move]: 6.21e-06
        [renormalize]: 0.00073972
        [add_forward_monad_depend]: 5.62999e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.439e-05
        [cse]: 3.557e-05
        [replace_applicator]: 1.789e-05
    [Cycle 2]: 0.00035776, [27]
        [switch_simplify]: 6.82002e-06
        [loop_unroll]: 5.79e-06
        [a_1]: 0.00011234
        [with_stream_mark]: 1.241e-05
        [recompute_prepare]: 5.54998e-06
        [updatestate_depend_eliminate]: 3.32002e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.61998e-06
        [updatestate_useless_node_eliminater]: 5.63002e-06
        [accelerated_algorithm]: 5.94999e-06
        [meta_shard_fg_expand]: 1.82999e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.462e-05
        [j_node_and_user_rematch]: 8.70999e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 8.70999e-06
        [inline_without_move]: 5.57001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.71e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 6.25002e-06
        [cse]: 1.339e-05
        [replace_applicator]: 6.04999e-06
[py_interpret_to_execute_after_opt_a]: 1.356e-05
[rewriter_after_opt_a]: 3.89e-05
[convert_after_rewriter]: 7.25998e-06
[order_py_execute_after_rewriter]: 5.52001e-06
[mutable_eliminate]: 0.00065192
[jit_opt_b]: 5.531e-05, [1]
    [Cycle 1]: 4.75e-05, [2]
        [frontend_op_eliminate]: 1.746e-05
        [inline_after_opt_a]: 1.825e-05
[cconv]: 2.617e-05
[loop_unroll]: 0.00044172
[jit_opt_after_cconv]: 0.00016192, [1]
    [Cycle 1]: 0.00015509, [11]
        [c_1]: 2.396e-05
        [parameter_eliminate]: 3.16999e-06
        [updatestate_depend_eliminate]: 7.48999e-06
        [updatestate_assign_eliminate]: 3.40998e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 2.745e-05
        [call_graph_tuple_transform]: 2.303e-05
        [tuple_list_get_item_eliminator]: 6.02001e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.82999e-06
[remove_dup_value]: 1.517e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 6.66999e-06
[add_recomputation]: 4.887e-05
[cse_after_recomputation]: 2.362e-05, [1]
    [Cycle 1]: 1.84e-05, [1]
        [cse]: 1.216e-05
[auto_monad_reorder]: 1.785e-05
[get_jit_bprop_graph]: 1.78002e-06
[rewriter_after_jit_bprop_graph]: 5.71e-06
[opt_after_jit_grad]: 0.00048486
[symbol_engine_optimizer]: 7.675e-05, [1]
    [Cycle 1]: 7.046e-05, [6]
        [build]: 4.03999e-06
        [elim_shapecalc]: 8.57998e-06
        [elim_not_effective]: 1.482e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 9.05999e-06
        [renormalize]: 4.39992e-07
[validate]: 3.801e-05
Sums
    bootstrap                            :     0.000341s :  3.55%
    type_inference                       :     0.005028s : 52.31%
    event_method                         :     0.000023s :  0.24%
    auto_monad                           :     0.000067s :  0.69%
    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                :     0.000091s :  0.94%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000070s :  0.73%
    jit_opt_a.loop_unroll                :     0.000040s :  0.42%
    jit_opt_a.a_1                        :     0.000838s :  8.72%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.39%
    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       :     0.000013s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.12%
    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_        :     0.000012s :  0.12%
    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 :     0.000031s :  0.32%
    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          :     0.000020s :  0.21%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000740s :  7.70%
    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      :     0.000021s :  0.21%
    jit_opt_a.cse                        :     0.000049s :  0.51%
    jit_opt_a.replace_applicator         :     0.000024s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.14%
    rewriter_after_opt_a                 :     0.000039s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000652s :  6.78%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.19%
    cconv                                :     0.000026s :  0.27%
    loop_unroll                          :     0.000442s :  4.60%
    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 :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000027s :  0.29%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  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  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000007s :  0.07%
    add_recomputation                    :     0.000049s :  0.51%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000485s :  5.04%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    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  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.40%


Time group info:
------[substitution.]   0.000310    28
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000002s :      2: substitution.fold_const_symbol
 2.02% :     0.000006s :      4: substitution.graph_param_transform
69.55% :     0.000215s :      4: substitution.inline
 1.20% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.58% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.67% :     0.000005s :      4: substitution.replace_old_param
22.84% :     0.000071s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004963     2
81.67% :     0.004053s :      1: type_inference.infer
18.33% :     0.000910s :      1: type_inference.specialize
------[replace.]   0.000089     8
56.52% :     0.000050s :      4: replace.inline
43.48% :     0.000039s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000281     8
75.64% :     0.000213s :      4: match.inline
24.36% :     0.000069s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   912
 1.22% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     13: predicate.addn_check_dump
 1.21% :     0.000002s :     13: predicate.addn_zero_filter
 2.10% :     0.000003s :     13: predicate.arithmetic_simplify
 1.79% :     0.000002s :     13: predicate.cast_eliminate
 0.53% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     13: predicate.compare_switch_simplify
 1.17% :     0.000002s :     13: predicate.depend_value_elim
 1.16% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      4: predicate.elim_not_effective
 0.48% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.50% :     0.000002s :     13: predicate.environ_get_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 1.17% :     0.000002s :      8: predicate.get_grad_eliminate
 0.39% :     0.000001s :      4: predicate.graph_param_transform
 5.86% :     0.000008s :     29: predicate.inline
 0.92% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.10% :     0.000002s :      8: predicate.less_batch_normalization
 1.57% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     21: predicate.load_eliminater
 1.33% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.21% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     13: predicate.merge_addn
 1.17% :     0.000002s :     13: predicate.minmaximum_grad
 1.49% :     0.000002s :      4: predicate.mutable_eliminate
 0.42% :     0.000001s :      4: predicate.opt_reshape
 2.46% :     0.000003s :     21: predicate.partial_eliminate
 1.15% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     13: predicate.reduce_eliminate
 1.85% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.75% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.27% :     0.000003s :     25: predicate.replace_applicator
 0.74% :     0.000001s :      8: predicate.replace_old_param
 0.30% :     0.000000s :      4: predicate.reset_defer_inline
 1.16% :     0.000002s :     13: predicate.reshape_eliminate
 1.18% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     13: predicate.same_eliminate
 0.69% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.27% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     21: predicate.switch_defer_inline
 2.14% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.68% :     0.000011s :     67: predicate.switch_simplify
 1.19% :     0.000002s :     13: predicate.tile_eliminate
 1.19% :     0.000002s :     13: predicate.transpose_eliminate
 1.35% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.85% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000630    12
48.53% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.66% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.81% :     0.000270s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015759    72
 0.33% :     0.000052s :      1: add_recomputation
 0.44% :     0.000070s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 2.25% :     0.000355s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.17% :     0.000027s :      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
37.88% :     0.005969s :      1: jit_opt_a
 1.04% :     0.000165s :      1: jit_opt_after_cconv
 0.37% :     0.000058s :      1: jit_opt_b
 2.85% :     0.000449s :      1: loop_unroll
 4.20% :     0.000662s :      1: mutable_eliminate
 6.90% :     0.001088s :     26: opt.transform.jit_opt_a
 0.35% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000029s :      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
 0.15% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.13% :     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% :     0.000004s :      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
 2.68% :     0.000422s :      1: renormalize.infer
 1.97% :     0.000310s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000042s :      1: rewriter_after_opt_a
 0.59% :     0.000093s :      1: rewriter_before_opt_a
 0.51% :     0.000080s :      1: symbol_engine_optimizer
32.00% :     0.005043s :      1: type_inference


TotalTime = 0.0481385, [33]
[bootstrap]: 0.00030455
[type_inference]: 0.0218046
[event_method]: 2.177e-05
[auto_monad]: 0.00015389
[graph_reusing]: 6.17001e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 0.00011727
[rewriter_before_opt_a]: 0.00010008
[expand_dump_flag]: 3.68e-06
[jit_opt_a]: 0.00658131, [2]
    [Cycle 1]: 0.00238287, [27]
        [switch_simplify]: 6.47e-05
        [loop_unroll]: 3.612e-05
        [a_1]: 0.00071694
        [with_stream_mark]: 1.899e-05
        [recompute_prepare]: 1.109e-05
        [updatestate_depend_eliminate]: 6.20002e-06
        [updatestate_assign_eliminate]: 6.24001e-06
        [updatestate_loads_eliminate]: 4.65001e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 9.87999e-06
        [updatestate_useless_node_eliminater]: 1.165e-05
        [accelerated_algorithm]: 8.87e-06
        [meta_shard_fg_expand]: 2.67001e-06
        [get_grad_eliminate_]: 8.40001e-06
        [merge_forward]: 5.34e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.91e-05
        [j_node_and_user_rematch]: 1.389e-05
        [meta_fg_expand]: 3.75e-06
        [replace_old_param]: 1.135e-05
        [inline_without_move]: 7.87e-06
        [renormalize]: 0.00113366
        [add_forward_monad_depend]: 5.81998e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 2.099e-05
        [cse]: 4.456e-05
        [replace_applicator]: 1.796e-05
    [Cycle 2]: 0.00047352, [27]
        [switch_simplify]: 9.37001e-06
        [loop_unroll]: 7.68999e-06
        [a_1]: 0.00016875
        [with_stream_mark]: 1.284e-05
        [recompute_prepare]: 7.95e-06
        [updatestate_depend_eliminate]: 5.67999e-06
        [updatestate_assign_eliminate]: 4.89e-06
        [updatestate_loads_eliminate]: 4.18001e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 8.11002e-06
        [updatestate_useless_node_eliminater]: 1.053e-05
        [accelerated_algorithm]: 8.12e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 7.46001e-06
        [merge_forward]: 4.95001e-06
        [cell_reuse_recompute_pass]: 1.92001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.696e-05
        [j_node_and_user_rematch]: 1.29e-05
        [meta_fg_expand]: 3.20002e-06
        [replace_old_param]: 1.016e-05
        [inline_without_move]: 7.28999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 1.038e-05
        [cse]: 2.225e-05
        [replace_applicator]: 7.53e-06
[py_interpret_to_execute_after_opt_a]: 1.156e-05
[rewriter_after_opt_a]: 0.00028333
[convert_after_rewriter]: 1.213e-05
[order_py_execute_after_rewriter]: 7.38999e-06
[mutable_eliminate]: 0.0006675
[jit_opt_b]: 6.725e-05, [1]
    [Cycle 1]: 5.947e-05, [2]
        [frontend_op_eliminate]: 2.475e-05
        [inline_after_opt_a]: 2.335e-05
[cconv]: 2.976e-05
[loop_unroll]: 0.00046618
[jit_opt_after_cconv]: 0.00020585, [1]
    [Cycle 1]: 0.00019879, [11]
        [c_1]: 4.797e-05
        [parameter_eliminate]: 3.36001e-06
        [updatestate_depend_eliminate]: 9.94001e-06
        [updatestate_assign_eliminate]: 4.40999e-06
        [updatestate_loads_eliminate]: 4.23999e-06
        [cse]: 3.419e-05
        [call_graph_tuple_transform]: 2.315e-05
        [tuple_list_get_item_eliminator]: 8.33999e-06
        [none_parameter_eliminate]: 1.41002e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 8.75001e-06
[remove_dup_value]: 1.85e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 1.289e-05
[add_recomputation]: 6.703e-05
[cse_after_recomputation]: 3.01e-05, [1]
    [Cycle 1]: 2.414e-05, [1]
        [cse]: 1.808e-05
[auto_monad_reorder]: 2.502e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 6.71e-06
[opt_after_jit_grad]: 0.00051297
[symbol_engine_optimizer]: 0.00010612, [1]
    [Cycle 1]: 9.917e-05, [6]
        [build]: 1.475e-05
        [elim_shapecalc]: 1.154e-05
        [elim_not_effective]: 1.962e-05
        [opt_reshape]: 8.78001e-06
        [fold_const_symbol]: 1.499e-05
        [renormalize]: 3.59985e-07
[validate]: 6.042e-05
[backend_pass]: 9.60019e-07
[task_emit]: 0.0162115
[execute]: 7.53999e-06
Sums
    bootstrap                            :     0.000305s :  0.70%
    type_inference                       :     0.021805s : 49.83%
    event_method                         :     0.000022s :  0.05%
    auto_monad                           :     0.000154s :  0.35%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000117s :  0.27%
    rewriter_before_opt_a                :     0.000100s :  0.23%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.17%
    jit_opt_a.loop_unroll                :     0.000044s :  0.10%
    jit_opt_a.a_1                        :     0.000886s :  2.02%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.05%
    jit_opt_a.inline_without_move        :     0.000015s :  0.03%
    jit_opt_a.renormalize                :     0.001134s :  2.59%
    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      :     0.000031s :  0.07%
    jit_opt_a.cse                        :     0.000067s :  0.15%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000283s :  0.65%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000667s :  1.53%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.05%
    cconv                                :     0.000030s :  0.07%
    loop_unroll                          :     0.000466s :  1.07%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  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              :     0.000034s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000067s :  0.15%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000025s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000513s :  1.17%
    symbol_engine_optimizer.build        :     0.000015s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000060s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.016211s : 37.05%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000290    47
 3.23% :     0.000009s :      2: substitution.depend_value_elim
 0.98% :     0.000003s :      4: substitution.elim_not_effective
 0.86% :     0.000002s :      4: substitution.fold_const_symbol
 2.43% :     0.000007s :      5: substitution.graph_param_transform
75.69% :     0.000219s :      4: substitution.inline
 1.60% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.54% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.58% :     0.000005s :      2: substitution.replace_old_param
 4.55% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.35% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 3.20% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021729     2
93.33% :     0.020279s :      1: type_inference.infer
 6.67% :     0.001450s :      1: type_inference.specialize
------[replace.]   0.000090     6
66.92% :     0.000060s :      4: replace.inline
33.08% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000229     6
94.76% :     0.000217s :      4: match.inline
 5.24% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1021
 1.30% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     15: predicate.addn_check_dump
 1.43% :     0.000002s :     15: predicate.addn_zero_filter
 2.20% :     0.000004s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.63% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.38% :     0.000002s :     15: predicate.depend_value_elim
 1.13% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.90% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.54% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     15: predicate.environ_get_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.21% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 5.20% :     0.000009s :     31: predicate.inline
 0.99% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.58% :     0.000003s :     10: predicate.less_batch_normalization
 1.33% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.31% :     0.000004s :     22: predicate.load_eliminater
 1.31% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.35% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.22% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 1.79% :     0.000003s :      5: predicate.mutable_eliminate
 0.48% :     0.000001s :      5: predicate.opt_reshape
 2.55% :     0.000004s :     22: predicate.partial_eliminate
 1.15% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.78% :     0.000003s :     15: predicate.reduce_eliminate
 1.44% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.69% :     0.000003s :     27: predicate.replace_applicator
 0.75% :     0.000001s :     10: predicate.replace_old_param
 0.32% :     0.000001s :      5: predicate.reset_defer_inline
 1.24% :     0.000002s :     15: predicate.reshape_eliminate
 1.24% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.94% :     0.000002s :     10: predicate.special_op_eliminate
 1.12% :     0.000002s :     10: predicate.specialize_transform
 1.55% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.12% :     0.000004s :     21: predicate.switch_defer_inline
 1.81% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.57% :     0.000011s :     68: predicate.switch_simplify
 1.24% :     0.000002s :     15: predicate.tile_eliminate
 1.23% :     0.000002s :     15: predicate.transpose_eliminate
 1.65% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.97% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.81% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.20% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000793    13
43.81% :     0.000348s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.68% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.51% :     0.000401s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050655    76
 0.14% :     0.000070s :      1: add_recomputation
 0.32% :     0.000160s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.63% :     0.000319s :      1: bootstrap
 0.06% :     0.000032s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.06% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.05% :     0.000027s :      1: event_method
 0.02% :     0.000012s :      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
13.00% :     0.006584s :      1: jit_opt_a
 0.41% :     0.000209s :      1: jit_opt_after_cconv
 0.14% :     0.000070s :      1: jit_opt_b
 0.94% :     0.000475s :      1: loop_unroll
 1.34% :     0.000679s :      1: mutable_eliminate
 2.35% :     0.001190s :     26: opt.transform.jit_opt_a
 0.17% :     0.000084s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000041s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.03% :     0.000522s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000122s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.37% :     0.000693s :      1: renormalize.infer
 0.85% :     0.000432s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.57% :     0.000288s :      1: rewriter_after_opt_a
 0.21% :     0.000105s :      1: rewriter_before_opt_a
 0.21% :     0.000109s :      1: symbol_engine_optimizer
32.03% :     0.016227s :      1: task_emit
43.08% :     0.021823s :      1: type_inference
 0.17% :     0.000085s :      1: validate


TotalTime = 0.0134854, [30]
[bootstrap]: 0.00028343
[type_inference]: 0.00481324
[event_method]: 2.217e-05
[auto_monad]: 6.168e-05
[graph_reusing]: 5.81e-06
[pre_auto_parallel]: 2.26e-06
[py_interpret_to_execute]: 3.516e-05
[rewriter_before_opt_a]: 8.87e-05
[expand_dump_flag]: 3.72002e-06
[jit_opt_a]: 0.00577017, [2]
    [Cycle 1]: 0.00189628, [27]
        [switch_simplify]: 6.428e-05
        [loop_unroll]: 3.489e-05
        [a_1]: 0.00065552
        [with_stream_mark]: 2.232e-05
        [recompute_prepare]: 1.051e-05
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 3.73999e-06
        [updatestate_loads_eliminate]: 3.06001e-06
        [parameter_eliminate]: 2.20002e-06
        [specialize_transform]: 7.28999e-06
        [updatestate_useless_node_eliminater]: 5.99e-06
        [accelerated_algorithm]: 6.81999e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 6.11e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.635e-05
        [j_node_and_user_rematch]: 9.32001e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 1.189e-05
        [inline_without_move]: 6.14001e-06
        [renormalize]: 0.00075556
        [add_forward_monad_depend]: 5.64e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.803e-05
        [cse]: 3.479e-05
        [replace_applicator]: 1.559e-05
    [Cycle 2]: 0.00036995, [27]
        [switch_simplify]: 6.40002e-06
        [loop_unroll]: 5.55001e-06
        [a_1]: 0.00011078
        [with_stream_mark]: 1.057e-05
        [recompute_prepare]: 6.14001e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 8.80013e-07
        [specialize_transform]: 5.72999e-06
        [updatestate_useless_node_eliminater]: 6.28e-06
        [accelerated_algorithm]: 5.86003e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 1.038e-05
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.582e-05
        [j_node_and_user_rematch]: 8.30999e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 9.64e-06
        [inline_without_move]: 5.53997e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 7.9e-06
        [cse]: 1.558e-05
        [replace_applicator]: 6.44999e-06
[py_interpret_to_execute_after_opt_a]: 1.364e-05
[rewriter_after_opt_a]: 3.909e-05
[convert_after_rewriter]: 7.9e-06
[order_py_execute_after_rewriter]: 5.89999e-06
[mutable_eliminate]: 0.00061833
[jit_opt_b]: 5.625e-05, [1]
    [Cycle 1]: 4.855e-05, [2]
        [frontend_op_eliminate]: 1.853e-05
        [inline_after_opt_a]: 1.739e-05
[cconv]: 2.553e-05
[loop_unroll]: 0.00046834
[jit_opt_after_cconv]: 0.00016484, [1]
    [Cycle 1]: 0.00015727, [11]
        [c_1]: 2.384e-05
        [parameter_eliminate]: 4.07e-06
        [updatestate_depend_eliminate]: 8.38001e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [cse]: 2.75e-05
        [call_graph_tuple_transform]: 2.145e-05
        [tuple_list_get_item_eliminator]: 5.97001e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 6.06e-06
[remove_dup_value]: 1.578e-05
[partial_unused_args_eliminate]: 2.41998e-06
[environ_conv]: 6.36e-06
[add_recomputation]: 5.118e-05
[cse_after_recomputation]: 2.541e-05, [1]
    [Cycle 1]: 1.97e-05, [1]
        [cse]: 1.263e-05
[auto_monad_reorder]: 1.823e-05
[get_jit_bprop_graph]: 1.88002e-06
[rewriter_after_jit_bprop_graph]: 6.44999e-06
[opt_after_jit_grad]: 0.00057485
[symbol_engine_optimizer]: 8.083e-05, [1]
    [Cycle 1]: 7.359e-05, [6]
        [build]: 5.56e-06
        [elim_shapecalc]: 8.43999e-06
        [elim_not_effective]: 1.557e-05
        [opt_reshape]: 6.34999e-06
        [fold_const_symbol]: 9.49999e-06
        [renormalize]: 5.8001e-07
[validate]: 4.201e-05
Sums
    bootstrap                            :     0.000283s :  3.02%
    type_inference                       :     0.004813s : 51.30%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000062s :  0.66%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.37%
    rewriter_before_opt_a                :     0.000089s :  0.95%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000071s :  0.75%
    jit_opt_a.loop_unroll                :     0.000040s :  0.43%
    jit_opt_a.a_1                        :     0.000766s :  8.17%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.18%
    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 :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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_        :     0.000016s :  0.18%
    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 :  0.34%
    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          :     0.000022s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000756s :  8.05%
    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      :     0.000026s :  0.28%
    jit_opt_a.cse                        :     0.000050s :  0.54%
    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      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000618s :  6.59%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000026s :  0.27%
    loop_unroll                          :     0.000468s :  4.99%
    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 :     0.000008s :  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              :     0.000028s :  0.29%
    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      :     0.000001s :  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 :  0.07%
    add_recomputation                    :     0.000051s :  0.55%
    cse_after_recomputation.cse          :     0.000013s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000575s :  6.13%
    symbol_engine_optimizer.build        :     0.000006s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  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                             :     0.000042s :  0.45%


Time group info:
------[substitution.]   0.000243    28
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.42% :     0.000006s :      4: substitution.graph_param_transform
83.48% :     0.000203s :      4: substitution.inline
 1.15% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.85% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.39% :     0.000006s :      4: substitution.replace_old_param
 7.28% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004753     2
81.11% :     0.003856s :      1: type_inference.infer
18.89% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000089     8
56.20% :     0.000050s :      4: replace.inline
43.80% :     0.000039s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000216     8
92.76% :     0.000200s :      4: match.inline
 7.24% :     0.000016s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.26% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     13: predicate.addn_check_dump
 1.16% :     0.000002s :     13: predicate.addn_zero_filter
 1.85% :     0.000003s :     13: predicate.arithmetic_simplify
 1.30% :     0.000002s :     13: predicate.cast_eliminate
 0.45% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     13: predicate.compare_switch_simplify
 1.13% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.42% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      4: predicate.elim_not_effective
 0.55% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     13: predicate.environ_get_eliminate
 1.08% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.97% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.54% :     0.000008s :     29: predicate.inline
 0.91% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      8: predicate.less_batch_normalization
 1.58% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.86% :     0.000003s :     21: predicate.load_eliminater
 1.54% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.08% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     13: predicate.merge_addn
 1.07% :     0.000001s :     13: predicate.minmaximum_grad
 2.03% :     0.000003s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.49% :     0.000003s :     21: predicate.partial_eliminate
 1.15% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     13: predicate.reduce_eliminate
 1.50% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.90% :     0.000003s :     25: predicate.replace_applicator
 0.93% :     0.000001s :      8: predicate.replace_old_param
 0.36% :     0.000001s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.14% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     13: predicate.same_eliminate
 0.56% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      8: predicate.special_op_eliminate
 1.01% :     0.000001s :      8: predicate.specialize_transform
 1.48% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     21: predicate.switch_defer_inline
 2.13% :     0.000003s :     21: predicate.switch_layer_defer_inline
 8.05% :     0.000011s :     67: predicate.switch_simplify
 1.10% :     0.000002s :     13: predicate.tile_eliminate
 1.30% :     0.000002s :     13: predicate.transpose_eliminate
 1.30% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.93% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.44% :     0.000005s :     29: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     13: predicate.value_based_eliminate
 0.35% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000619    12
47.63% :     0.000295s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.55% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.82% :     0.000277s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015337    72
 0.35% :     0.000054s :      1: add_recomputation
 0.42% :     0.000065s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 1.93% :     0.000297s :      1: bootstrap
 0.18% :     0.000028s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.18% :     0.000028s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.17% :     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
37.64% :     0.005773s :      1: jit_opt_a
 1.09% :     0.000167s :      1: jit_opt_after_cconv
 0.38% :     0.000059s :      1: jit_opt_b
 3.12% :     0.000479s :      1: loop_unroll
 4.09% :     0.000628s :      1: mutable_eliminate
 6.67% :     0.001023s :     26: opt.transform.jit_opt_a
 0.35% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000030s :      4: opt.transform.jit_opt_b
 0.11% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.82% :     0.000585s :      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
 0.11% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.77% :     0.000425s :      1: renormalize.infer
 2.10% :     0.000321s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000043s :      1: rewriter_after_opt_a
 0.60% :     0.000091s :      1: rewriter_before_opt_a
 0.54% :     0.000084s :      1: symbol_engine_optimizer
31.46% :     0.004825s :      1: type_inference


TotalTime = 0.0746526, [33]
[bootstrap]: 0.00025924
[type_inference]: 0.0317377
[event_method]: 0.00015495
[auto_monad]: 0.00020429
[graph_reusing]: 7.46001e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 4.109e-05
[rewriter_before_opt_a]: 0.00013002
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.0304134, [3]
    [Cycle 1]: 0.0227504, [27]
        [switch_simplify]: 7.845e-05
        [loop_unroll]: 5.223e-05
        [a_1]: 0.00116451
        [with_stream_mark]: 3.363e-05
        [recompute_prepare]: 2.655e-05
        [updatestate_depend_eliminate]: 1.227e-05
        [updatestate_assign_eliminate]: 9.73002e-06
        [updatestate_loads_eliminate]: 9.59e-06
        [parameter_eliminate]: 2.09999e-06
        [specialize_transform]: 2.126e-05
        [updatestate_useless_node_eliminater]: 2.371e-05
        [accelerated_algorithm]: 2.089e-05
        [meta_shard_fg_expand]: 4.82e-06
        [get_grad_eliminate_]: 1.854e-05
        [merge_forward]: 1.205e-05
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.696e-05
        [j_node_and_user_rematch]: 3.379e-05
        [meta_fg_expand]: 0.00635165
        [replace_old_param]: 0.00012671
        [inline_without_move]: 0.00012256
        [renormalize]: 0.0137389
        [add_forward_monad_depend]: 1.611e-05
        [auto_monad_grad]: 9.92001e-06
        [auto_monad_eliminator]: 7.994e-05
        [cse]: 0.00030025
        [replace_applicator]: 0.00017545
    [Cycle 2]: 0.00319899, [27]
        [switch_simplify]: 6.241e-05
        [loop_unroll]: 6.043e-05
        [a_1]: 0.00102965
        [with_stream_mark]: 1.862e-05
        [recompute_prepare]: 1.476e-05
        [updatestate_depend_eliminate]: 2.848e-05
        [updatestate_assign_eliminate]: 6.16e-06
        [updatestate_loads_eliminate]: 4.17998e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 1.12e-05
        [updatestate_useless_node_eliminater]: 1.326e-05
        [accelerated_algorithm]: 9.15001e-06
        [meta_shard_fg_expand]: 1.92999e-06
        [get_grad_eliminate_]: 8.53001e-06
        [merge_forward]: 5.00001e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.729e-05
        [j_node_and_user_rematch]: 1.403e-05
        [meta_fg_expand]: 9.014e-05
        [replace_old_param]: 1.451e-05
        [inline_without_move]: 8.43001e-06
        [renormalize]: 0.00147792
        [add_forward_monad_depend]: 5.73002e-06
        [auto_monad_grad]: 2.83998e-06
        [auto_monad_eliminator]: 2.037e-05
        [cse]: 6.901e-05
        [replace_applicator]: 1.982e-05
    [Cycle 3]: 0.00052756, [27]
        [switch_simplify]: 9.24998e-06
        [loop_unroll]: 9.02999e-06
        [a_1]: 0.00017847
        [with_stream_mark]: 1.618e-05
        [recompute_prepare]: 9.21998e-06
        [updatestate_depend_eliminate]: 6.17999e-06
        [updatestate_assign_eliminate]: 4.81002e-06
        [updatestate_loads_eliminate]: 4.18999e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 9.25999e-06
        [updatestate_useless_node_eliminater]: 1.257e-05
        [accelerated_algorithm]: 8.08001e-06
        [meta_shard_fg_expand]: 2.35002e-06
        [get_grad_eliminate_]: 8.12e-06
        [merge_forward]: 5.00999e-06
        [cell_reuse_recompute_pass]: 2.07999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.823e-05
        [j_node_and_user_rematch]: 1.336e-05
        [meta_fg_expand]: 3.06001e-06
        [replace_old_param]: 1.174e-05
        [inline_without_move]: 7.66001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.77001e-06
        [auto_monad_grad]: 1.79e-06
        [auto_monad_eliminator]: 1.264e-05
        [cse]: 3.121e-05
        [replace_applicator]: 9.45001e-06
[py_interpret_to_execute_after_opt_a]: 1.666e-05
[rewriter_after_opt_a]: 7.471e-05
[convert_after_rewriter]: 9.56e-06
[order_py_execute_after_rewriter]: 5.89e-06
[mutable_eliminate]: 0.00080039
[jit_opt_b]: 7.106e-05, [1]
    [Cycle 1]: 6.327e-05, [2]
        [frontend_op_eliminate]: 2.556e-05
        [inline_after_opt_a]: 2.543e-05
[cconv]: 2.48e-05
[loop_unroll]: 0.00053227
[jit_opt_after_cconv]: 0.00023119, [1]
    [Cycle 1]: 0.00022399, [11]
        [c_1]: 4.903e-05
        [parameter_eliminate]: 3.50998e-06
        [updatestate_depend_eliminate]: 8.82e-06
        [updatestate_assign_eliminate]: 4.65001e-06
        [updatestate_loads_eliminate]: 4.35999e-06
        [cse]: 5.095e-05
        [call_graph_tuple_transform]: 2.62e-05
        [tuple_list_get_item_eliminator]: 8.72998e-06
        [none_parameter_eliminate]: 1.22999e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 8.27e-06
[remove_dup_value]: 2.174e-05
[partial_unused_args_eliminate]: 2.07001e-06
[environ_conv]: 1.139e-05
[add_recomputation]: 6.121e-05
[cse_after_recomputation]: 3.472e-05, [1]
    [Cycle 1]: 2.71e-05, [1]
        [cse]: 2.053e-05
[auto_monad_reorder]: 2.612e-05
[get_jit_bprop_graph]: 1.81003e-06
[rewriter_after_jit_bprop_graph]: 6.88e-06
[opt_after_jit_grad]: 0.00059484
[symbol_engine_optimizer]: 0.00012003, [1]
    [Cycle 1]: 0.00011283, [6]
        [build]: 1.511e-05
        [elim_shapecalc]: 1.227e-05
        [elim_not_effective]: 2.335e-05
        [opt_reshape]: 1.153e-05
        [fold_const_symbol]: 1.841e-05
        [renormalize]: 3.59985e-07
[validate]: 5.866e-05
[backend_pass]: 1.35001e-06
[task_emit]: 0.00872976
[execute]: 8.52e-06
Sums
    bootstrap                            :     0.000259s :  0.37%
    type_inference                       :     0.031738s : 45.50%
    event_method                         :     0.000155s :  0.22%
    auto_monad                           :     0.000204s :  0.29%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000041s :  0.06%
    rewriter_before_opt_a                :     0.000130s :  0.19%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000150s :  0.22%
    jit_opt_a.loop_unroll                :     0.000122s :  0.17%
    jit_opt_a.a_1                        :     0.002373s :  3.40%
    jit_opt_a.with_stream_mark           :     0.000068s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000051s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000047s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000021s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000018s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000042s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000050s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000038s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000035s :  0.05%
    jit_opt_a.merge_forward              :     0.000022s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000072s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000061s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.006445s :  9.24%
    jit_opt_a.replace_old_param          :     0.000153s :  0.22%
    jit_opt_a.inline_without_move        :     0.000139s :  0.20%
    jit_opt_a.renormalize                :     0.015217s : 21.82%
    jit_opt_a.add_forward_monad_depend   :     0.000024s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000015s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000113s :  0.16%
    jit_opt_a.cse                        :     0.000400s :  0.57%
    jit_opt_a.replace_applicator         :     0.000205s :  0.29%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.02%
    rewriter_after_opt_a                 :     0.000075s :  0.11%
    convert_after_rewriter               :     0.000010s :  0.01%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000800s :  1.15%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.04%
    cconv                                :     0.000025s :  0.04%
    loop_unroll                          :     0.000532s :  0.76%
    jit_opt_after_cconv.c_1              :     0.000049s :  0.07%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000051s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  0.01%
    remove_dup_value                     :     0.000022s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000011s :  0.02%
    add_recomputation                    :     0.000061s :  0.09%
    cse_after_recomputation.cse          :     0.000021s :  0.03%
    auto_monad_reorder                   :     0.000026s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000595s :  0.85%
    symbol_engine_optimizer.build        :     0.000015s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000012s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000018s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000059s :  0.08%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008730s : 12.52%
    execute                              :     0.000009s :  0.01%


Time group info:
------[substitution.]   0.001621   160
 1.36% :     0.000022s :      8: substitution.depend_value_elim
 0.20% :     0.000003s :      4: substitution.elim_not_effective
 0.19% :     0.000003s :      4: substitution.fold_const_symbol
57.03% :     0.000924s :      4: substitution.getattr_setattr_resolve
 0.38% :     0.000006s :      5: substitution.graph_param_transform
24.74% :     0.000401s :     10: substitution.inline
 2.28% :     0.000037s :      4: substitution.inline_without_move
 0.76% :     0.000012s :     20: substitution.j_node_and_user_rematch
 0.57% :     0.000009s :      5: substitution.minmaximum_grad
 0.30% :     0.000005s :      3: substitution.partial_eliminate
 0.95% :     0.000015s :     20: substitution.remove_not_recompute_node
 3.02% :     0.000049s :     13: substitution.replace_applicator
 0.85% :     0.000014s :     18: substitution.replace_old_param
 0.19% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.31% :     0.000021s :      5: substitution.tuple_list_convert_item_index_to_positive
 0.81% :     0.000013s :      5: substitution.tuple_list_get_item_depend_reorder
 2.14% :     0.000035s :     10: substitution.tuple_list_get_item_eliminator
 1.13% :     0.000018s :      8: substitution.updatestate_pure_node_eliminater
 1.81% :     0.000029s :     13: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.031639     2
93.78% :     0.029672s :      1: type_inference.infer
 6.22% :     0.001967s :      1: type_inference.specialize
------[replace.]   0.000313    20
18.36% :     0.000058s :      3: replace.getattr_setattr_resolve
31.14% :     0.000098s :     10: replace.inline
11.28% :     0.000035s :      1: replace.replace_applicator
31.05% :     0.000097s :      5: replace.tuple_list_get_item_eliminator
 8.16% :     0.000026s :      1: replace.updatestate_useless_node_eliminater
------[match.]   0.001305    20
66.89% :     0.000873s :      3: match.getattr_setattr_resolve
30.13% :     0.000393s :     10: match.inline
 0.83% :     0.000011s :      1: match.replace_applicator
 1.40% :     0.000018s :      5: match.tuple_list_get_item_eliminator
 0.75% :     0.000010s :      1: match.updatestate_useless_node_eliminater
------[predicate.]   0.000463  2979
 1.45% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.47% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.23% :     0.000006s :     46: predicate.addn_check_dump
 1.56% :     0.000007s :     46: predicate.addn_zero_filter
 2.14% :     0.000010s :     46: predicate.arithmetic_simplify
 1.92% :     0.000009s :     46: predicate.cast_eliminate
 0.19% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.18% :     0.000005s :     46: predicate.compare_switch_simplify
 1.45% :     0.000007s :     46: predicate.depend_value_elim
 1.37% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.53% :     0.000007s :     46: predicate.dict_get_item_eliminator
 1.50% :     0.000007s :     46: predicate.dict_set_item_eliminator
 0.35% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.16% :     0.000001s :      5: predicate.elim_not_effective
 0.27% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000007s :     46: predicate.environ_add_const_eliminate
 1.22% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.40% :     0.000007s :     46: predicate.environ_get_depend_swap
 1.44% :     0.000007s :     46: predicate.environ_get_eliminate
 1.20% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 1.03% :     0.000005s :     27: predicate.get_grad_eliminate
 1.40% :     0.000006s :     20: predicate.getattr_setattr_resolve
 0.09% :     0.000000s :      5: predicate.graph_param_transform
 3.72% :     0.000017s :     72: predicate.inline
 3.27% :     0.000015s :     91: predicate.inline_without_move
 0.43% :     0.000002s :     27: predicate.j_node_and_user_rematch
 1.38% :     0.000006s :     27: predicate.less_batch_normalization
 1.66% :     0.000008s :     51: predicate.list_to_tuple_eliminator_
 2.07% :     0.000010s :     56: predicate.load_eliminater
 0.54% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.77% :     0.000017s :    121: predicate.loop_unroll_before_grad
 1.80% :     0.000008s :     51: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000006s :     46: predicate.merge_addn
 1.47% :     0.000007s :     46: predicate.minmaximum_grad
 0.57% :     0.000003s :      5: predicate.mutable_eliminate
 0.33% :     0.000002s :      5: predicate.opt_reshape
 2.03% :     0.000009s :     56: predicate.partial_eliminate
 1.30% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.81% :     0.000008s :     46: predicate.reduce_eliminate
 1.67% :     0.000008s :     51: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     27: predicate.remove_not_recompute_node
 2.79% :     0.000013s :    126: predicate.replace_applicator
 1.61% :     0.000007s :     91: predicate.replace_old_param
 0.14% :     0.000001s :      5: predicate.reset_defer_inline
 1.51% :     0.000007s :     46: predicate.reshape_eliminate
 1.46% :     0.000007s :     46: predicate.row_tensor_add_zeros_like
 0.44% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.32% :     0.000006s :     46: predicate.same_eliminate
 0.58% :     0.000003s :     29: predicate.set_cell_output_no_recompute
 0.47% :     0.000002s :     10: predicate.special_op_eliminate
 1.04% :     0.000005s :     27: predicate.specialize_transform
 1.55% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.45% :     0.000011s :     62: predicate.switch_defer_inline
 2.11% :     0.000010s :     62: predicate.switch_layer_defer_inline
 6.20% :     0.000029s :    188: predicate.switch_simplify
 1.29% :     0.000006s :     46: predicate.tile_eliminate
 1.35% :     0.000006s :     46: predicate.transpose_eliminate
 1.67% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 2.92% :     0.000014s :     61: predicate.tuple_list_get_item_eliminator
 1.96% :     0.000009s :     46: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000008s :     51: predicate.tuple_to_list_eliminator_
 1.66% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000014s :     84: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000008s :     46: predicate.value_based_eliminate
 0.16% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003390    48
66.07% :     0.002240s :     27: func_graph_cloner_run.FuncGraphClonerGraph
33.93% :     0.001150s :     21: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.094543    93
 0.07% :     0.000065s :      1: add_recomputation
 0.22% :     0.000213s :      1: auto_monad
 0.03% :     0.000029s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.30% :     0.000280s :      1: bootstrap
 0.03% :     0.000028s :      1: cconv
 0.01% :     0.000012s :      1: convert_after_rewriter
 0.04% :     0.000037s :      1: cse_after_recomputation
 0.01% :     0.000014s :      1: environ_conv
 0.17% :     0.000162s :      1: event_method
 0.01% :     0.000013s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000011s :      1: graph_reusing
32.17% :     0.030417s :      1: jit_opt_a
 0.25% :     0.000234s :      1: jit_opt_after_cconv
 0.08% :     0.000074s :      1: jit_opt_b
 0.57% :     0.000542s :      1: loop_unroll
 0.86% :     0.000813s :      1: mutable_eliminate
 3.63% :     0.003436s :     39: opt.transform.jit_opt_a
 0.09% :     0.000088s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000044s :      4: opt.transform.jit_opt_b
 0.02% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 1.10% :     0.001040s :      2: opt.transform.opt_resolve
 0.06% :     0.000061s :      4: opt.transform.symbol_engine_opt
 0.64% :     0.000605s :      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
 0.05% :     0.000044s :      1: py_interpret_to_execute
 0.02% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000024s :      1: remove_dup_value
12.63% :     0.011939s :      2: renormalize.infer
 3.45% :     0.003257s :      2: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000079s :      1: rewriter_after_opt_a
 0.14% :     0.000133s :      1: rewriter_before_opt_a
 0.13% :     0.000123s :      1: symbol_engine_optimizer
 9.25% :     0.008748s :      1: task_emit
33.58% :     0.031752s :      1: type_inference
 0.09% :     0.000086s :      1: validate


TotalTime = 0.0141397, [30]
[bootstrap]: 0.00030487
[type_inference]: 0.00524522
[event_method]: 2.064e-05
[auto_monad]: 6.629e-05
[graph_reusing]: 6.19001e-06
[pre_auto_parallel]: 2.45002e-06
[py_interpret_to_execute]: 3.504e-05
[rewriter_before_opt_a]: 9.681e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00607011, [2]
    [Cycle 1]: 0.00199386, [27]
        [switch_simplify]: 6.43e-05
        [loop_unroll]: 3.668e-05
        [a_1]: 0.00067694
        [with_stream_mark]: 2.203e-05
        [recompute_prepare]: 1.019e-05
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 3.91999e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 2.13002e-06
        [specialize_transform]: 7.23e-06
        [updatestate_useless_node_eliminater]: 5.77999e-06
        [accelerated_algorithm]: 6.01998e-06
        [meta_shard_fg_expand]: 2.43e-06
        [get_grad_eliminate_]: 6.00002e-06
        [merge_forward]: 3.83999e-06
        [cell_reuse_recompute_pass]: 1.21997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.554e-05
        [j_node_and_user_rematch]: 9.59999e-06
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 1.104e-05
        [inline_without_move]: 6.09001e-06
        [renormalize]: 0.00084738
        [add_forward_monad_depend]: 6.35002e-06
        [auto_monad_grad]: 2.58998e-06
        [auto_monad_eliminator]: 1.579e-05
        [cse]: 2.376e-05
        [replace_applicator]: 1.743e-05
    [Cycle 2]: 0.00037954, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 5.80002e-06
        [a_1]: 0.00011183
        [with_stream_mark]: 1.192e-05
        [recompute_prepare]: 5.77001e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.38002e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 5.71e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 6.24999e-06
        [meta_shard_fg_expand]: 1.63002e-06
        [get_grad_eliminate_]: 5.44998e-06
        [merge_forward]: 3.21001e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.498e-05
        [j_node_and_user_rematch]: 8.60999e-06
        [meta_fg_expand]: 1.91003e-06
        [replace_old_param]: 1.397e-05
        [inline_without_move]: 1.895e-05
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.74e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 8.19002e-06
        [cse]: 1.382e-05
        [replace_applicator]: 5.65001e-06
[py_interpret_to_execute_after_opt_a]: 1.646e-05
[rewriter_after_opt_a]: 3.682e-05
[convert_after_rewriter]: 8.03999e-06
[order_py_execute_after_rewriter]: 5.52999e-06
[mutable_eliminate]: 0.000678
[jit_opt_b]: 5.63e-05, [1]
    [Cycle 1]: 4.909e-05, [2]
        [frontend_op_eliminate]: 1.866e-05
        [inline_after_opt_a]: 1.838e-05
[cconv]: 2.731e-05
[loop_unroll]: 0.00043522
[jit_opt_after_cconv]: 0.00015617, [1]
    [Cycle 1]: 0.00014956, [11]
        [c_1]: 2.428e-05
        [parameter_eliminate]: 3.53e-06
        [updatestate_depend_eliminate]: 7.25e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [cse]: 2.482e-05
        [call_graph_tuple_transform]: 2.161e-05
        [tuple_list_get_item_eliminator]: 5.94e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.97999e-06
[remove_dup_value]: 1.429e-05
[partial_unused_args_eliminate]: 2.54999e-06
[environ_conv]: 7.1e-06
[add_recomputation]: 4.789e-05
[cse_after_recomputation]: 2.327e-05, [1]
    [Cycle 1]: 1.775e-05, [1]
        [cse]: 1.163e-05
[auto_monad_reorder]: 1.704e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 5.46e-06
[opt_after_jit_grad]: 0.00047471
[symbol_engine_optimizer]: 7.411e-05, [1]
    [Cycle 1]: 6.817e-05, [6]
        [build]: 4.02998e-06
        [elim_shapecalc]: 8.30999e-06
        [elim_not_effective]: 1.333e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 9.15999e-06
        [renormalize]: 5.69999e-07
[validate]: 4.001e-05
Sums
    bootstrap                            :     0.000305s :  3.09%
    type_inference                       :     0.005245s : 53.13%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000066s :  0.67%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.35%
    rewriter_before_opt_a                :     0.000097s :  0.98%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000071s :  0.72%
    jit_opt_a.loop_unroll                :     0.000042s :  0.43%
    jit_opt_a.a_1                        :     0.000789s :  7.99%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.34%
    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 :     0.000006s :  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       :     0.000013s :  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       :     0.000004s :  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  :     0.000002s :  0.02%
    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          :     0.000025s :  0.25%
    jit_opt_a.inline_without_move        :     0.000025s :  0.25%
    jit_opt_a.renormalize                :     0.000847s :  8.58%
    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 :  0.24%
    jit_opt_a.cse                        :     0.000038s :  0.38%
    jit_opt_a.replace_applicator         :     0.000023s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.17%
    rewriter_after_opt_a                 :     0.000037s :  0.37%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000678s :  6.87%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.19%
    cconv                                :     0.000027s :  0.28%
    loop_unroll                          :     0.000435s :  4.41%
    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 :     0.000003s :  0.03%
    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 :     0.000006s :  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 :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.07%
    add_recomputation                    :     0.000048s :  0.49%
    cse_after_recomputation.cse          :     0.000012s :  0.12%
    auto_monad_reorder                   :     0.000017s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000475s :  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 :     0.000013s :  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                             :     0.000040s :  0.41%


Time group info:
------[substitution.]   0.000257    28
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.14% :     0.000005s :      4: substitution.graph_param_transform
84.75% :     0.000218s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.82% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000006s :      4: substitution.replace_old_param
 6.39% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005169     2
82.15% :     0.004246s :      1: type_inference.infer
17.85% :     0.000923s :      1: type_inference.specialize
------[replace.]   0.000089     8
57.86% :     0.000051s :      4: replace.inline
42.14% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000229     8
93.65% :     0.000215s :      4: match.inline
 6.35% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   912
 1.23% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     13: predicate.addn_check_dump
 1.14% :     0.000002s :     13: predicate.addn_zero_filter
 2.01% :     0.000003s :     13: predicate.arithmetic_simplify
 1.30% :     0.000002s :     13: predicate.cast_eliminate
 0.68% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     13: predicate.compare_switch_simplify
 1.20% :     0.000002s :     13: predicate.depend_value_elim
 1.13% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      4: predicate.elim_not_effective
 0.54% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     13: predicate.environ_get_eliminate
 1.21% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.89% :     0.000001s :      8: predicate.get_grad_eliminate
 0.35% :     0.000000s :      4: predicate.graph_param_transform
 5.47% :     0.000008s :     29: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.39% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :      8: predicate.less_batch_normalization
 1.50% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.90% :     0.000003s :     21: predicate.load_eliminater
 1.15% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.04% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     13: predicate.merge_addn
 1.07% :     0.000001s :     13: predicate.minmaximum_grad
 1.43% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.48% :     0.000003s :     21: predicate.partial_eliminate
 1.17% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     13: predicate.reduce_eliminate
 1.52% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.91% :     0.000003s :     25: predicate.replace_applicator
 0.55% :     0.000001s :      8: predicate.replace_old_param
 0.38% :     0.000001s :      4: predicate.reset_defer_inline
 1.17% :     0.000002s :     13: predicate.reshape_eliminate
 1.22% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      8: predicate.special_op_eliminate
 0.91% :     0.000001s :      8: predicate.specialize_transform
 1.25% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.85% :     0.000004s :     21: predicate.switch_defer_inline
 2.23% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.96% :     0.000011s :     67: predicate.switch_simplify
 1.15% :     0.000002s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.35% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.46% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     13: predicate.value_based_eliminate
 0.36% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000646    12
47.94% :     0.000310s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.05% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.01% :     0.000284s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016120    72
 0.31% :     0.000051s :      1: add_recomputation
 0.43% :     0.000070s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 1.97% :     0.000318s :      1: bootstrap
 0.19% :     0.000030s :      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% :     0.000025s :      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
37.68% :     0.006074s :      1: jit_opt_a
 0.99% :     0.000159s :      1: jit_opt_after_cconv
 0.37% :     0.000059s :      1: jit_opt_b
 2.75% :     0.000443s :      1: loop_unroll
 4.26% :     0.000686s :      1: mutable_eliminate
 6.57% :     0.001058s :     26: opt.transform.jit_opt_a
 0.34% :     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
 0.10% :     0.000016s :      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
 3.00% :     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
 0.24% :     0.000038s :      1: py_interpret_to_execute
 0.12% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 3.03% :     0.000489s :      1: renormalize.infer
 2.17% :     0.000350s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000040s :      1: rewriter_after_opt_a
 0.62% :     0.000100s :      1: rewriter_before_opt_a
 0.48% :     0.000077s :      1: symbol_engine_optimizer
32.61% :     0.005257s :      1: type_inference


TotalTime = 0.0131299, [30]
[bootstrap]: 0.00026025
[type_inference]: 0.00466403
[event_method]: 1.958e-05
[auto_monad]: 5.804e-05
[graph_reusing]: 5.25001e-06
[pre_auto_parallel]: 2.11e-06
[py_interpret_to_execute]: 5.917e-05
[rewriter_before_opt_a]: 8.724e-05
[expand_dump_flag]: 2.86e-06
[jit_opt_a]: 0.00582683, [2]
    [Cycle 1]: 0.00188359, [27]
        [switch_simplify]: 5.822e-05
        [loop_unroll]: 3.51e-05
        [a_1]: 0.00064063
        [with_stream_mark]: 1.788e-05
        [recompute_prepare]: 2.362e-05
        [updatestate_depend_eliminate]: 3.71999e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 7.5e-06
        [updatestate_useless_node_eliminater]: 5.92999e-06
        [accelerated_algorithm]: 6.30002e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 6.58998e-06
        [merge_forward]: 3.77998e-06
        [cell_reuse_recompute_pass]: 1.01002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.522e-05
        [j_node_and_user_rematch]: 3.911e-05
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 1.186e-05
        [inline_without_move]: 6.02999e-06
        [renormalize]: 0.00074363
        [add_forward_monad_depend]: 5.26998e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.332e-05
        [cse]: 2.336e-05
        [replace_applicator]: 1.934e-05
    [Cycle 2]: 0.00039392, [27]
        [switch_simplify]: 6.66999e-06
        [loop_unroll]: 5.72999e-06
        [a_1]: 0.00011266
        [with_stream_mark]: 3.085e-05
        [recompute_prepare]: 6.19999e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 9.32999e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.52999e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.06999e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.442e-05
        [j_node_and_user_rematch]: 8.67e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 9.02e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 6.61e-06
        [cse]: 1.431e-05
        [replace_applicator]: 5.92999e-06
[py_interpret_to_execute_after_opt_a]: 1.317e-05
[rewriter_after_opt_a]: 3.803e-05
[convert_after_rewriter]: 7.16001e-06
[order_py_execute_after_rewriter]: 5.48002e-06
[mutable_eliminate]: 0.00057079
[jit_opt_b]: 5.436e-05, [1]
    [Cycle 1]: 4.727e-05, [2]
        [frontend_op_eliminate]: 1.83e-05
        [inline_after_opt_a]: 1.717e-05
[cconv]: 2.42e-05
[loop_unroll]: 0.00044082
[jit_opt_after_cconv]: 0.00014991, [1]
    [Cycle 1]: 0.00014324, [11]
        [c_1]: 2.392e-05
        [parameter_eliminate]: 3.23e-06
        [updatestate_depend_eliminate]: 6.38e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.33998e-06
        [cse]: 2.216e-05
        [call_graph_tuple_transform]: 2.01e-05
        [tuple_list_get_item_eliminator]: 5.99e-06
        [none_parameter_eliminate]: 1.35999e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 6.07999e-06
[remove_dup_value]: 1.312e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 5.28002e-06
[add_recomputation]: 4.564e-05
[cse_after_recomputation]: 2.299e-05, [1]
    [Cycle 1]: 1.75e-05, [1]
        [cse]: 1.159e-05
[auto_monad_reorder]: 1.675e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 4.92e-06
[opt_after_jit_grad]: 0.00046359
[symbol_engine_optimizer]: 7.537e-05, [1]
    [Cycle 1]: 6.918e-05, [6]
        [build]: 2.90998e-06
        [elim_shapecalc]: 8.16002e-06
        [elim_not_effective]: 1.366e-05
        [opt_reshape]: 6.25002e-06
        [fold_const_symbol]: 9.39e-06
        [renormalize]: 2.89991e-07
[validate]: 3.617e-05
Sums
    bootstrap                            :     0.000260s :  2.89%
    type_inference                       :     0.004664s : 51.78%
    event_method                         :     0.000020s :  0.22%
    auto_monad                           :     0.000058s :  0.64%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000059s :  0.66%
    rewriter_before_opt_a                :     0.000087s :  0.97%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.72%
    jit_opt_a.loop_unroll                :     0.000041s :  0.45%
    jit_opt_a.a_1                        :     0.000753s :  8.36%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.54%
    jit_opt_a.recompute_prepare          :     0.000030s :  0.33%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.14%
    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 :  0.15%
    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_        :     0.000012s :  0.13%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000048s :  0.53%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000021s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000744s :  8.26%
    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 :  0.22%
    jit_opt_a.cse                        :     0.000038s :  0.42%
    jit_opt_a.replace_applicator         :     0.000025s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.15%
    rewriter_after_opt_a                 :     0.000038s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000571s :  6.34%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000024s :  0.27%
    loop_unroll                          :     0.000441s :  4.89%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.27%
    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              :     0.000022s :  0.25%
    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 :     0.000001s :  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 :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000046s :  0.51%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000017s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000464s :  5.15%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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 :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.40%


Time group info:
------[substitution.]   0.000227    28
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.57% :     0.000006s :      4: substitution.graph_param_transform
82.86% :     0.000188s :      4: substitution.inline
 1.57% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000005s :      4: substitution.replace_old_param
 7.28% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004605     2
80.43% :     0.003704s :      1: type_inference.infer
19.57% :     0.000901s :      1: type_inference.specialize
------[replace.]   0.000083     8
56.42% :     0.000047s :      4: replace.inline
43.58% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     8
92.69% :     0.000186s :      4: match.inline
 7.31% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   912
 1.22% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     13: predicate.addn_check_dump
 1.40% :     0.000002s :     13: predicate.addn_zero_filter
 1.80% :     0.000002s :     13: predicate.arithmetic_simplify
 1.35% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     13: predicate.compare_switch_simplify
 1.22% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.45% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     13: predicate.environ_get_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 1.31% :     0.000002s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.62% :     0.000008s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      8: predicate.less_batch_normalization
 1.64% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.96% :     0.000003s :     21: predicate.load_eliminater
 1.27% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.22% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     13: predicate.merge_addn
 1.13% :     0.000002s :     13: predicate.minmaximum_grad
 1.27% :     0.000002s :      4: predicate.mutable_eliminate
 0.46% :     0.000001s :      4: predicate.opt_reshape
 2.45% :     0.000003s :     21: predicate.partial_eliminate
 1.16% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     13: predicate.reduce_eliminate
 1.57% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.93% :     0.000003s :     25: predicate.replace_applicator
 0.58% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.17% :     0.000002s :     13: predicate.reshape_eliminate
 1.53% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.54% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.13% :     0.000002s :      8: predicate.special_op_eliminate
 1.26% :     0.000002s :      8: predicate.specialize_transform
 1.35% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.57% :     0.000004s :     21: predicate.switch_defer_inline
 2.14% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.43% :     0.000010s :     67: predicate.switch_simplify
 1.19% :     0.000002s :     13: predicate.tile_eliminate
 1.28% :     0.000002s :     13: predicate.transpose_eliminate
 1.38% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.85% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.83% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000576    12
45.84% :     0.000264s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.10% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.05% :     0.000271s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014997    72
 0.32% :     0.000049s :      1: add_recomputation
 0.41% :     0.000061s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 1.80% :     0.000270s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000025s :      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
 0.05% :     0.000008s :      1: graph_reusing
38.87% :     0.005830s :      1: jit_opt_a
 1.02% :     0.000153s :      1: jit_opt_after_cconv
 0.38% :     0.000057s :      1: jit_opt_b
 3.00% :     0.000449s :      1: loop_unroll
 3.86% :     0.000580s :      1: mutable_eliminate
 6.97% :     0.001045s :     26: opt.transform.jit_opt_a
 0.35% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000029s :      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
 0.15% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.14% :     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% :     0.000004s :      1: pre_auto_parallel
 0.42% :     0.000063s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.80% :     0.000420s :      1: renormalize.infer
 2.11% :     0.000316s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000041s :      1: rewriter_after_opt_a
 0.60% :     0.000090s :      1: rewriter_before_opt_a
 0.52% :     0.000078s :      1: symbol_engine_optimizer
31.18% :     0.004676s :      1: type_inference


TotalTime = 0.0444525, [33]
[bootstrap]: 0.00029459
[type_inference]: 0.0242457
[event_method]: 2.216e-05
[auto_monad]: 0.00016596
[graph_reusing]: 6.76e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 0.00014147
[rewriter_before_opt_a]: 0.00010361
[expand_dump_flag]: 3.18998e-06
[jit_opt_a]: 0.00689321, [2]
    [Cycle 1]: 0.00252663, [27]
        [switch_simplify]: 7.197e-05
        [loop_unroll]: 3.634e-05
        [a_1]: 0.00077135
        [with_stream_mark]: 2.599e-05
        [recompute_prepare]: 1.048e-05
        [updatestate_depend_eliminate]: 6.33e-06
        [updatestate_assign_eliminate]: 5.92999e-06
        [updatestate_loads_eliminate]: 4.53999e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 8.90999e-06
        [updatestate_useless_node_eliminater]: 1.129e-05
        [accelerated_algorithm]: 8.60001e-06
        [meta_shard_fg_expand]: 3.20002e-06
        [get_grad_eliminate_]: 8.33999e-06
        [merge_forward]: 5.55001e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.019e-05
        [j_node_and_user_rematch]: 1.362e-05
        [meta_fg_expand]: 3.55003e-06
        [replace_old_param]: 1.194e-05
        [inline_without_move]: 7.95e-06
        [renormalize]: 0.00121199
        [add_forward_monad_depend]: 9.91e-06
        [auto_monad_grad]: 2.78998e-06
        [auto_monad_eliminator]: 2.153e-05
        [cse]: 3.886e-05
        [replace_applicator]: 1.575e-05
    [Cycle 2]: 0.00045438, [27]
        [switch_simplify]: 8.94e-06
        [loop_unroll]: 7.66999e-06
        [a_1]: 0.00016187
        [with_stream_mark]: 1.164e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 5.09e-06
        [updatestate_assign_eliminate]: 4.71002e-06
        [updatestate_loads_eliminate]: 3.89002e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 7.98001e-06
        [updatestate_useless_node_eliminater]: 1.058e-05
        [accelerated_algorithm]: 7.73001e-06
        [meta_shard_fg_expand]: 1.96998e-06
        [get_grad_eliminate_]: 7.23e-06
        [merge_forward]: 4.26001e-06
        [cell_reuse_recompute_pass]: 1.84998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.594e-05
        [j_node_and_user_rematch]: 1.247e-05
        [meta_fg_expand]: 2.82002e-06
        [replace_old_param]: 9.69999e-06
        [inline_without_move]: 7.66001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 1.007e-05
        [cse]: 1.948e-05
        [replace_applicator]: 7.71001e-06
[py_interpret_to_execute_after_opt_a]: 1.249e-05
[rewriter_after_opt_a]: 0.00029538
[convert_after_rewriter]: 1.19e-05
[order_py_execute_after_rewriter]: 7.61001e-06
[mutable_eliminate]: 0.00062474
[jit_opt_b]: 7.315e-05, [1]
    [Cycle 1]: 6.553e-05, [2]
        [frontend_op_eliminate]: 2.692e-05
        [inline_after_opt_a]: 2.554e-05
[cconv]: 2.453e-05
[loop_unroll]: 0.00044334
[jit_opt_after_cconv]: 0.00018792, [1]
    [Cycle 1]: 0.00018128, [11]
        [c_1]: 4.465e-05
        [parameter_eliminate]: 2.38998e-06
        [updatestate_depend_eliminate]: 7.60998e-06
        [updatestate_assign_eliminate]: 4.39002e-06
        [updatestate_loads_eliminate]: 3.91001e-06
        [cse]: 2.604e-05
        [call_graph_tuple_transform]: 2.188e-05
        [tuple_list_get_item_eliminator]: 8.48999e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 3.30008e-07
        [switch_simplify]: 8.82e-06
[remove_dup_value]: 2.01e-05
[partial_unused_args_eliminate]: 2.59001e-06
[environ_conv]: 1.019e-05
[add_recomputation]: 6.434e-05
[cse_after_recomputation]: 2.931e-05, [1]
    [Cycle 1]: 2.337e-05, [1]
        [cse]: 1.758e-05
[auto_monad_reorder]: 2.499e-05
[get_jit_bprop_graph]: 2.10002e-06
[rewriter_after_jit_bprop_graph]: 5.02e-06
[opt_after_jit_grad]: 0.0004911
[symbol_engine_optimizer]: 0.00010111, [1]
    [Cycle 1]: 9.465e-05, [6]
        [build]: 9.53002e-06
        [elim_shapecalc]: 1.149e-05
        [elim_not_effective]: 1.965e-05
        [opt_reshape]: 8.77999e-06
        [fold_const_symbol]: 1.474e-05
        [renormalize]: 4.09986e-07
[validate]: 5.159e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00986214
[execute]: 7.88001e-06
Sums
    bootstrap                            :     0.000295s :  0.74%
    type_inference                       :     0.024246s : 60.77%
    event_method                         :     0.000022s :  0.06%
    auto_monad                           :     0.000166s :  0.42%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000141s :  0.35%
    rewriter_before_opt_a                :     0.000104s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000081s :  0.20%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000933s :  2.34%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001212s :  3.04%
    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      :     0.000032s :  0.08%
    jit_opt_a.cse                        :     0.000058s :  0.15%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000295s :  0.74%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000625s :  1.57%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000025s :  0.06%
    loop_unroll                          :     0.000443s :  1.11%
    jit_opt_after_cconv.c_1              :     0.000045s :  0.11%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000064s :  0.16%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000025s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000491s :  1.23%
    symbol_engine_optimizer.build        :     0.000010s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    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                             :     0.000052s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009862s : 24.72%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000324    47
 2.41% :     0.000008s :      2: substitution.depend_value_elim
 0.85% :     0.000003s :      4: substitution.elim_not_effective
 0.63% :     0.000002s :      4: substitution.fold_const_symbol
 2.08% :     0.000007s :      5: substitution.graph_param_transform
79.64% :     0.000258s :      4: substitution.inline
 1.42% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.28% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.37% :     0.000004s :      2: substitution.replace_old_param
 4.07% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 2.70% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.55% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024165     2
93.71% :     0.022646s :      1: type_inference.infer
 6.29% :     0.001520s :      1: type_inference.specialize
------[replace.]   0.000099     6
66.84% :     0.000066s :      4: replace.inline
33.16% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000267     6
95.47% :     0.000255s :      4: match.inline
 4.53% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1021
 1.28% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     15: predicate.addn_check_dump
 1.31% :     0.000002s :     15: predicate.addn_zero_filter
 2.27% :     0.000004s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.57% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     15: predicate.compare_switch_simplify
 1.32% :     0.000002s :     15: predicate.depend_value_elim
 1.22% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.60% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.46% :     0.000003s :     15: predicate.environ_get_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.09% :     0.000002s :     10: predicate.get_grad_eliminate
 0.32% :     0.000001s :      5: predicate.graph_param_transform
 5.13% :     0.000009s :     31: predicate.inline
 1.01% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :     10: predicate.less_batch_normalization
 1.48% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     22: predicate.load_eliminater
 1.38% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.44% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.12% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     15: predicate.merge_addn
 1.10% :     0.000002s :     15: predicate.minmaximum_grad
 1.61% :     0.000003s :      5: predicate.mutable_eliminate
 0.46% :     0.000001s :      5: predicate.opt_reshape
 2.36% :     0.000004s :     22: predicate.partial_eliminate
 1.26% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.66% :     0.000003s :     15: predicate.reduce_eliminate
 1.46% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.71% :     0.000003s :     27: predicate.replace_applicator
 0.66% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.26% :     0.000002s :     15: predicate.reshape_eliminate
 1.49% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.53% :     0.000003s :     15: predicate.same_eliminate
 0.55% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.10% :     0.000002s :     10: predicate.special_op_eliminate
 1.07% :     0.000002s :     10: predicate.specialize_transform
 1.49% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.18% :     0.000004s :     21: predicate.switch_defer_inline
 1.86% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.95% :     0.000012s :     68: predicate.switch_simplify
 1.19% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000002s :     15: predicate.transpose_eliminate
 1.73% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.75% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000865    13
42.16% :     0.000365s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.59% :     0.000066s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.25% :     0.000434s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047091    76
 0.14% :     0.000067s :      1: add_recomputation
 0.36% :     0.000171s :      1: auto_monad
 0.06% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000312s :      1: bootstrap
 0.06% :     0.000027s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.06% :     0.000028s :      1: event_method
 0.02% :     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
14.65% :     0.006896s :      1: jit_opt_a
 0.41% :     0.000191s :      1: jit_opt_after_cconv
 0.16% :     0.000076s :      1: jit_opt_b
 0.96% :     0.000452s :      1: loop_unroll
 1.35% :     0.000634s :      1: mutable_eliminate
 2.63% :     0.001239s :     26: opt.transform.jit_opt_a
 0.17% :     0.000081s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000045s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.06% :     0.000499s :      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
 0.31% :     0.000147s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000023s :      1: remove_dup_value
 1.57% :     0.000738s :      1: renormalize.infer
 0.99% :     0.000465s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.64% :     0.000300s :      1: rewriter_after_opt_a
 0.23% :     0.000107s :      1: rewriter_before_opt_a
 0.22% :     0.000104s :      1: symbol_engine_optimizer
20.97% :     0.009873s :      1: task_emit
51.52% :     0.024263s :      1: type_inference
 0.16% :     0.000073s :      1: validate


TotalTime = 0.0120323, [30]
[bootstrap]: 0.00026016
[type_inference]: 0.00435766
[event_method]: 2.061e-05
[auto_monad]: 5.959e-05
[graph_reusing]: 5.37999e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 3.275e-05
[rewriter_before_opt_a]: 8.424e-05
[expand_dump_flag]: 2.93e-06
[jit_opt_a]: 0.00512938, [2]
    [Cycle 1]: 0.00171745, [27]
        [switch_simplify]: 5.482e-05
        [loop_unroll]: 3.419e-05
        [a_1]: 0.00060873
        [with_stream_mark]: 1.587e-05
        [recompute_prepare]: 7.76001e-06
        [updatestate_depend_eliminate]: 3.98999e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 3.04999e-06
        [parameter_eliminate]: 1.44e-06
        [specialize_transform]: 6.96001e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.93002e-06
        [get_grad_eliminate_]: 5.91e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 8.00006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.43e-05
        [j_node_and_user_rematch]: 9.39e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 1.03e-05
        [inline_without_move]: 5.91e-06
        [renormalize]: 0.00068017
        [add_forward_monad_depend]: 5.39998e-06
        [auto_monad_grad]: 2.24001e-06
        [auto_monad_eliminator]: 1.396e-05
        [cse]: 2.847e-05
        [replace_applicator]: 1.321e-05
    [Cycle 2]: 0.00035869, [27]
        [switch_simplify]: 6.75002e-06
        [loop_unroll]: 5.77999e-06
        [a_1]: 0.0001123
        [with_stream_mark]: 8.83001e-06
        [recompute_prepare]: 5.78002e-06
        [updatestate_depend_eliminate]: 3.23998e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.71e-06
        [updatestate_useless_node_eliminater]: 5.56002e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 1.34003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 8.54998e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 9.61003e-06
        [inline_without_move]: 5.67001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 6.98e-06
        [cse]: 1.358e-05
        [replace_applicator]: 5.82999e-06
[py_interpret_to_execute_after_opt_a]: 1.014e-05
[rewriter_after_opt_a]: 3.437e-05
[convert_after_rewriter]: 7.2e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.00049544
[jit_opt_b]: 5.817e-05, [1]
    [Cycle 1]: 5.074e-05, [2]
        [frontend_op_eliminate]: 1.953e-05
        [inline_after_opt_a]: 1.804e-05
[cconv]: 2.359e-05
[loop_unroll]: 0.00043505
[jit_opt_after_cconv]: 0.00015894, [1]
    [Cycle 1]: 0.00015208, [11]
        [c_1]: 2.611e-05
        [parameter_eliminate]: 2.78998e-06
        [updatestate_depend_eliminate]: 6.26e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.355e-05
        [call_graph_tuple_transform]: 2.027e-05
        [tuple_list_get_item_eliminator]: 6.64999e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 5.90022e-07
        [switch_simplify]: 6.48998e-06
[remove_dup_value]: 1.52e-05
[partial_unused_args_eliminate]: 2.79999e-06
[environ_conv]: 5.79e-06
[add_recomputation]: 4.963e-05
[cse_after_recomputation]: 2.355e-05, [1]
    [Cycle 1]: 1.798e-05, [1]
        [cse]: 1.232e-05
[auto_monad_reorder]: 1.749e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 5.31002e-06
[opt_after_jit_grad]: 0.00046031
[symbol_engine_optimizer]: 7.436e-05, [1]
    [Cycle 1]: 6.808e-05, [6]
        [build]: 3.36001e-06
        [elim_shapecalc]: 8.38001e-06
        [elim_not_effective]: 1.344e-05
        [opt_reshape]: 6.10002e-06
        [fold_const_symbol]: 9.19998e-06
        [renormalize]: 4.69998e-07
[validate]: 3.87e-05
Sums
    bootstrap                            :     0.000260s :  3.09%
    type_inference                       :     0.004358s : 51.78%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000060s :  0.71%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.39%
    rewriter_before_opt_a                :     0.000084s :  1.00%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000062s :  0.73%
    jit_opt_a.loop_unroll                :     0.000040s :  0.47%
    jit_opt_a.a_1                        :     0.000721s :  8.57%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.09%
    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       :     0.000013s :  0.15%
    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       :     0.000003s :  0.04%
    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 :     0.000027s :  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          :     0.000020s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000680s :  8.08%
    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      :     0.000021s :  0.25%
    jit_opt_a.cse                        :     0.000042s :  0.50%
    jit_opt_a.replace_applicator         :     0.000019s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000034s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000495s :  5.89%
    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 :  5.17%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.31%
    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              :     0.000024s :  0.28%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.08%
    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 :  0.08%
    remove_dup_value                     :     0.000015s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000050s :  0.59%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000017s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000460s :  5.47%
    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                             :     0.000039s :  0.46%


Time group info:
------[substitution.]   0.000206    28
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000006s :      4: substitution.graph_param_transform
82.14% :     0.000170s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      4: substitution.replace_old_param
 7.90% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004305     2
80.16% :     0.003451s :      1: type_inference.infer
19.84% :     0.000854s :      1: type_inference.specialize
------[replace.]   0.000081     8
56.31% :     0.000045s :      4: replace.inline
43.69% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     8
92.07% :     0.000167s :      4: match.inline
 7.93% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.27% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.01% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     13: predicate.addn_check_dump
 1.22% :     0.000002s :     13: predicate.addn_zero_filter
 1.83% :     0.000002s :     13: predicate.arithmetic_simplify
 1.49% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     13: predicate.compare_switch_simplify
 1.31% :     0.000002s :     13: predicate.depend_value_elim
 1.18% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      4: predicate.elim_not_effective
 0.54% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     13: predicate.environ_get_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.98% :     0.000001s :      8: predicate.get_grad_eliminate
 0.33% :     0.000000s :      4: predicate.graph_param_transform
 5.44% :     0.000007s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      8: predicate.less_batch_normalization
 1.61% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.04% :     0.000003s :     21: predicate.load_eliminater
 1.58% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     42: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     13: predicate.merge_addn
 1.16% :     0.000002s :     13: predicate.minmaximum_grad
 1.50% :     0.000002s :      4: predicate.mutable_eliminate
 0.50% :     0.000001s :      4: predicate.opt_reshape
 2.57% :     0.000003s :     21: predicate.partial_eliminate
 1.23% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     13: predicate.reduce_eliminate
 1.61% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.94% :     0.000003s :     25: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.36% :     0.000000s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.20% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     13: predicate.same_eliminate
 0.65% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      8: predicate.special_op_eliminate
 0.95% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     21: predicate.switch_defer_inline
 2.18% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.77% :     0.000010s :     67: predicate.switch_simplify
 1.19% :     0.000002s :     13: predicate.tile_eliminate
 1.19% :     0.000002s :     13: predicate.transpose_eliminate
 1.43% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.94% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000562    12
46.84% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.06% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.10% :     0.000265s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013744    72
 0.38% :     0.000052s :      1: add_recomputation
 0.46% :     0.000063s :      1: auto_monad
 0.15% :     0.000020s :      1: auto_monad_reorder
 1.97% :     0.000271s :      1: bootstrap
 0.19% :     0.000026s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.19% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.18% :     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
37.34% :     0.005132s :      1: jit_opt_a
 1.18% :     0.000162s :      1: jit_opt_after_cconv
 0.44% :     0.000061s :      1: jit_opt_b
 3.23% :     0.000444s :      1: loop_unroll
 3.67% :     0.000504s :      1: mutable_eliminate
 6.93% :     0.000952s :     26: opt.transform.jit_opt_a
 0.41% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000031s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.40% :     0.000467s :      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
 0.26% :     0.000036s :      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
 2.86% :     0.000394s :      1: renormalize.infer
 2.03% :     0.000279s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000037s :      1: rewriter_after_opt_a
 0.63% :     0.000087s :      1: rewriter_before_opt_a
 0.56% :     0.000077s :      1: symbol_engine_optimizer
31.78% :     0.004368s :      1: type_inference


TotalTime = 0.0126793, [30]
[bootstrap]: 0.00029137
[type_inference]: 0.00489116
[event_method]: 2.159e-05
[auto_monad]: 6.342e-05
[graph_reusing]: 5.76e-06
[pre_auto_parallel]: 2.34001e-06
[py_interpret_to_execute]: 3.605e-05
[rewriter_before_opt_a]: 8.833e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00519767, [2]
    [Cycle 1]: 0.00164209, [27]
        [switch_simplify]: 5.768e-05
        [loop_unroll]: 3.727e-05
        [a_1]: 0.00062606
        [with_stream_mark]: 1.729e-05
        [recompute_prepare]: 8.95999e-06
        [updatestate_depend_eliminate]: 3.92998e-06
        [updatestate_assign_eliminate]: 3.56999e-06
        [updatestate_loads_eliminate]: 3.09999e-06
        [parameter_eliminate]: 2.05002e-06
        [specialize_transform]: 6.93998e-06
        [updatestate_useless_node_eliminater]: 5.84999e-06
        [accelerated_algorithm]: 5.87999e-06
        [meta_shard_fg_expand]: 3.18998e-06
        [get_grad_eliminate_]: 6.01e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.549e-05
        [j_node_and_user_rematch]: 9.34998e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 1.002e-05
        [inline_without_move]: 5.99999e-06
        [renormalize]: 0.00056811
        [add_forward_monad_depend]: 5.44e-06
        [auto_monad_grad]: 2.11e-06
        [auto_monad_eliminator]: 1.414e-05
        [cse]: 3.105e-05
        [replace_applicator]: 1.297e-05
    [Cycle 2]: 0.00035341, [27]
        [switch_simplify]: 6.53003e-06
        [loop_unroll]: 5.87999e-06
        [a_1]: 0.0001106
        [with_stream_mark]: 8.94e-06
        [recompute_prepare]: 5.44e-06
        [updatestate_depend_eliminate]: 3.01999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 5.51998e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 1.47999e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.13e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.433e-05
        [j_node_and_user_rematch]: 8.3e-06
        [meta_fg_expand]: 1.92999e-06
        [replace_old_param]: 8.67998e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 5.99e-06
        [cse]: 1.325e-05
        [replace_applicator]: 5.74999e-06
[py_interpret_to_execute_after_opt_a]: 1.133e-05
[rewriter_after_opt_a]: 3.616e-05
[convert_after_rewriter]: 6.84001e-06
[order_py_execute_after_rewriter]: 5.07e-06
[mutable_eliminate]: 0.00050145
[jit_opt_b]: 5.323e-05, [1]
    [Cycle 1]: 4.624e-05, [2]
        [frontend_op_eliminate]: 1.834e-05
        [inline_after_opt_a]: 1.661e-05
[cconv]: 2.391e-05
[loop_unroll]: 0.0004552
[jit_opt_after_cconv]: 0.00015127, [1]
    [Cycle 1]: 0.00014454, [11]
        [c_1]: 2.461e-05
        [parameter_eliminate]: 2.46e-06
        [updatestate_depend_eliminate]: 5.99e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [cse]: 2.044e-05
        [call_graph_tuple_transform]: 1.982e-05
        [tuple_list_get_item_eliminator]: 6.27001e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.94999e-06
[remove_dup_value]: 1.462e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 5.84999e-06
[add_recomputation]: 4.668e-05
[cse_after_recomputation]: 2.287e-05, [1]
    [Cycle 1]: 1.76e-05, [1]
        [cse]: 1.209e-05
[auto_monad_reorder]: 1.756e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 5.15001e-06
[opt_after_jit_grad]: 0.00045087
[symbol_engine_optimizer]: 7.396e-05, [1]
    [Cycle 1]: 6.809e-05, [6]
        [build]: 3.29001e-06
        [elim_shapecalc]: 8.32998e-06
        [elim_not_effective]: 1.291e-05
        [opt_reshape]: 6.16e-06
        [fold_const_symbol]: 9.24e-06
        [renormalize]: 3.50003e-07
[validate]: 3.374e-05
Sums
    bootstrap                            :     0.000291s :  3.27%
    type_inference                       :     0.004891s : 54.89%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000063s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.40%
    rewriter_before_opt_a                :     0.000088s :  0.99%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000064s :  0.72%
    jit_opt_a.loop_unroll                :     0.000043s :  0.48%
    jit_opt_a.a_1                        :     0.000737s :  8.27%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.29%
    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 :  0.07%
    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      :     0.000012s :  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              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.33%
    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 :  0.13%
    jit_opt_a.renormalize                :     0.000568s :  6.38%
    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                        :     0.000044s :  0.50%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000036s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000501s :  5.63%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000024s :  0.27%
    loop_unroll                          :     0.000455s :  5.11%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.28%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  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 :  0.07%
    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 :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000047s :  0.52%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.20%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000451s :  5.06%
    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 :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.38%


Time group info:
------[substitution.]   0.000208    28
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.51% :     0.000005s :      4: substitution.graph_param_transform
82.01% :     0.000170s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000004s :      4: substitution.replace_old_param
 8.34% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004832     2
81.62% :     0.003944s :      1: type_inference.infer
18.38% :     0.000888s :      1: type_inference.specialize
------[replace.]   0.000081     8
57.55% :     0.000047s :      4: replace.inline
42.45% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     8
91.60% :     0.000168s :      4: match.inline
 8.40% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   912
 1.18% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     13: predicate.addn_check_dump
 1.19% :     0.000002s :     13: predicate.addn_zero_filter
 1.87% :     0.000003s :     13: predicate.arithmetic_simplify
 1.31% :     0.000002s :     13: predicate.cast_eliminate
 0.48% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     13: predicate.compare_switch_simplify
 1.18% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.52% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 1.05% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.53% :     0.000008s :     29: predicate.inline
 0.91% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      8: predicate.less_batch_normalization
 1.62% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     21: predicate.load_eliminater
 1.33% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.46% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     13: predicate.merge_addn
 1.16% :     0.000002s :     13: predicate.minmaximum_grad
 1.25% :     0.000002s :      4: predicate.mutable_eliminate
 0.41% :     0.000001s :      4: predicate.opt_reshape
 2.59% :     0.000004s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.70% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.26% :     0.000002s :     13: predicate.reshape_eliminate
 1.18% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     13: predicate.same_eliminate
 0.55% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      8: predicate.special_op_eliminate
 0.95% :     0.000001s :      8: predicate.specialize_transform
 1.30% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     21: predicate.switch_defer_inline
 2.43% :     0.000003s :     21: predicate.switch_layer_defer_inline
 8.14% :     0.000011s :     67: predicate.switch_simplify
 1.18% :     0.000002s :     13: predicate.tile_eliminate
 1.35% :     0.000002s :     13: predicate.transpose_eliminate
 1.40% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000600    12
51.06% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.08% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.85% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014298    72
 0.35% :     0.000049s :      1: add_recomputation
 0.47% :     0.000067s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 2.16% :     0.000308s :      1: bootstrap
 0.18% :     0.000026s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.18% :     0.000026s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
36.37% :     0.005200s :      1: jit_opt_a
 1.08% :     0.000154s :      1: jit_opt_after_cconv
 0.39% :     0.000056s :      1: jit_opt_b
 3.25% :     0.000464s :      1: loop_unroll
 3.56% :     0.000509s :      1: mutable_eliminate
 6.81% :     0.000973s :     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
 0.09% :     0.000013s :      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
 0.24% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.21% :     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
 0.27% :     0.000039s :      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
 2.10% :     0.000300s :      1: renormalize.infer
 1.83% :     0.000261s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000039s :      1: rewriter_after_opt_a
 0.64% :     0.000091s :      1: rewriter_before_opt_a
 0.53% :     0.000076s :      1: symbol_engine_optimizer
34.29% :     0.004902s :      1: type_inference


TotalTime = 0.0411497, [33]
[bootstrap]: 0.00029986
[type_inference]: 0.0221849
[event_method]: 1.883e-05
[auto_monad]: 0.00012887
[graph_reusing]: 4.09002e-06
[pre_auto_parallel]: 2.17999e-06
[py_interpret_to_execute]: 0.00012469
[rewriter_before_opt_a]: 9.055e-05
[expand_dump_flag]: 2.75002e-06
[jit_opt_a]: 0.00590167, [2]
    [Cycle 1]: 0.00216815, [27]
        [switch_simplify]: 5.091e-05
        [loop_unroll]: 3.516e-05
        [a_1]: 0.00067943
        [with_stream_mark]: 1.507e-05
        [recompute_prepare]: 1.064e-05
        [updatestate_depend_eliminate]: 5.58997e-06
        [updatestate_assign_eliminate]: 4.52e-06
        [updatestate_loads_eliminate]: 4.1e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 9.38002e-06
        [updatestate_useless_node_eliminater]: 1.158e-05
        [accelerated_algorithm]: 8.74e-06
        [meta_shard_fg_expand]: 2.74999e-06
        [get_grad_eliminate_]: 8.45001e-06
        [merge_forward]: 4.75999e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.679e-05
        [j_node_and_user_rematch]: 1.335e-05
        [meta_fg_expand]: 3.27997e-06
        [replace_old_param]: 1.128e-05
        [inline_without_move]: 7.95e-06
        [renormalize]: 0.00102758
        [add_forward_monad_depend]: 5.42001e-06
        [auto_monad_grad]: 2.12999e-06
        [auto_monad_eliminator]: 1.655e-05
        [cse]: 2.994e-05
        [replace_applicator]: 1.489e-05
    [Cycle 2]: 0.00045806, [27]
        [switch_simplify]: 8.79e-06
        [loop_unroll]: 8.10999e-06
        [a_1]: 0.00016401
        [with_stream_mark]: 1.171e-05
        [recompute_prepare]: 8.05e-06
        [updatestate_depend_eliminate]: 4.74998e-06
        [updatestate_assign_eliminate]: 4.22e-06
        [updatestate_loads_eliminate]: 3.83001e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 8.35001e-06
        [updatestate_useless_node_eliminater]: 1.063e-05
        [accelerated_algorithm]: 7.63001e-06
        [meta_shard_fg_expand]: 1.67999e-06
        [get_grad_eliminate_]: 7.76001e-06
        [merge_forward]: 4.43999e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.528e-05
        [j_node_and_user_rematch]: 1.224e-05
        [meta_fg_expand]: 2.80997e-06
        [replace_old_param]: 9.48002e-06
        [inline_without_move]: 7.46001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 9.02999e-06
        [cse]: 2.068e-05
        [replace_applicator]: 8.02003e-06
[py_interpret_to_execute_after_opt_a]: 1.264e-05
[rewriter_after_opt_a]: 0.00027672
[convert_after_rewriter]: 1.063e-05
[order_py_execute_after_rewriter]: 7.2e-06
[mutable_eliminate]: 0.00058064
[jit_opt_b]: 7.157e-05, [1]
    [Cycle 1]: 6.403e-05, [2]
        [frontend_op_eliminate]: 2.615e-05
        [inline_after_opt_a]: 2.535e-05
[cconv]: 2.155e-05
[loop_unroll]: 0.00045705
[jit_opt_after_cconv]: 0.00018254, [1]
    [Cycle 1]: 0.0001762, [11]
        [c_1]: 4.234e-05
        [parameter_eliminate]: 2.56998e-06
        [updatestate_depend_eliminate]: 7.26001e-06
        [updatestate_assign_eliminate]: 4.52e-06
        [updatestate_loads_eliminate]: 4.15e-06
        [cse]: 2.435e-05
        [call_graph_tuple_transform]: 2.147e-05
        [tuple_list_get_item_eliminator]: 8.13001e-06
        [none_parameter_eliminate]: 1.13001e-06
        [renormalize]: 3.80009e-07
        [switch_simplify]: 8.46002e-06
[remove_dup_value]: 1.556e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 1.019e-05
[add_recomputation]: 6.024e-05
[cse_after_recomputation]: 3.02e-05, [1]
    [Cycle 1]: 2.493e-05, [1]
        [cse]: 1.84e-05
[auto_monad_reorder]: 2.489e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 5.37999e-06
[opt_after_jit_grad]: 0.00049495
[symbol_engine_optimizer]: 0.00010046, [1]
    [Cycle 1]: 9.378e-05, [6]
        [build]: 9.71003e-06
        [elim_shapecalc]: 1.186e-05
        [elim_not_effective]: 1.857e-05
        [opt_reshape]: 9.15999e-06
        [fold_const_symbol]: 1.403e-05
        [renormalize]: 3.80009e-07
[validate]: 5.021e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00976243
[execute]: 8.50999e-06
Sums
    bootstrap                            :     0.000300s :  0.80%
    type_inference                       :     0.022185s : 59.53%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000129s :  0.35%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000125s :  0.33%
    rewriter_before_opt_a                :     0.000091s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000060s :  0.16%
    jit_opt_a.loop_unroll                :     0.000043s :  0.12%
    jit_opt_a.a_1                        :     0.000843s :  2.26%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  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 :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001028s :  2.76%
    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      :     0.000026s :  0.07%
    jit_opt_a.cse                        :     0.000051s :  0.14%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000277s :  0.74%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000581s :  1.56%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.07%
    cconv                                :     0.000022s :  0.06%
    loop_unroll                          :     0.000457s :  1.23%
    jit_opt_after_cconv.c_1              :     0.000042s :  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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  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 :     0.000008s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000016s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000060s :  0.16%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000025s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000495s :  1.33%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000050s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009762s : 26.20%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000251    47
 3.20% :     0.000008s :      2: substitution.depend_value_elim
 1.09% :     0.000003s :      4: substitution.elim_not_effective
 0.82% :     0.000002s :      4: substitution.fold_const_symbol
 2.51% :     0.000006s :      5: substitution.graph_param_transform
75.96% :     0.000190s :      4: substitution.inline
 1.77% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.56% :     0.000006s :      8: substitution.remove_not_recompute_node
 1.46% :     0.000004s :      2: substitution.replace_old_param
 4.35% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
 3.11% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.16% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022125     2
93.70% :     0.020731s :      1: type_inference.infer
 6.30% :     0.001393s :      1: type_inference.specialize
------[replace.]   0.000088     6
65.54% :     0.000058s :      4: replace.inline
34.46% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
95.03% :     0.000188s :      4: match.inline
 4.97% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000165  1021
 1.46% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.10% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.17% :     0.000002s :     15: predicate.addn_check_dump
 1.45% :     0.000002s :     15: predicate.addn_zero_filter
 1.92% :     0.000003s :     15: predicate.arithmetic_simplify
 1.32% :     0.000002s :     15: predicate.cast_eliminate
 0.57% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     15: predicate.compare_switch_simplify
 1.54% :     0.000003s :     15: predicate.depend_value_elim
 1.19% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.78% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     15: predicate.environ_get_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.08% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 5.22% :     0.000009s :     31: predicate.inline
 1.08% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :     10: predicate.less_batch_normalization
 1.71% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     22: predicate.load_eliminater
 1.17% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.58% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     15: predicate.merge_addn
 1.24% :     0.000002s :     15: predicate.minmaximum_grad
 1.25% :     0.000002s :      5: predicate.mutable_eliminate
 0.57% :     0.000001s :      5: predicate.opt_reshape
 2.28% :     0.000004s :     22: predicate.partial_eliminate
 1.26% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     15: predicate.reduce_eliminate
 1.73% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.80% :     0.000003s :     27: predicate.replace_applicator
 0.73% :     0.000001s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000002s :     15: predicate.reshape_eliminate
 1.24% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     15: predicate.same_eliminate
 0.63% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.09% :     0.000002s :     10: predicate.special_op_eliminate
 1.14% :     0.000002s :     10: predicate.specialize_transform
 1.60% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.20% :     0.000004s :     21: predicate.switch_defer_inline
 1.97% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.43% :     0.000011s :     68: predicate.switch_simplify
 1.23% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.63% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.45% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.48% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.32% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000786    13
45.52% :     0.000358s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.34% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.14% :     0.000386s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043492    76
 0.15% :     0.000063s :      1: add_recomputation
 0.31% :     0.000133s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.72% :     0.000312s :      1: bootstrap
 0.06% :     0.000024s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000012s :      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% :     0.000007s :      1: graph_reusing
13.58% :     0.005905s :      1: jit_opt_a
 0.43% :     0.000185s :      1: jit_opt_after_cconv
 0.17% :     0.000074s :      1: jit_opt_b
 1.07% :     0.000466s :      1: loop_unroll
 1.35% :     0.000589s :      1: mutable_eliminate
 2.60% :     0.001129s :     26: opt.transform.jit_opt_a
 0.18% :     0.000077s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000044s :      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% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.16% :     0.000504s :      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
 0.30% :     0.000129s :      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
 1.42% :     0.000618s :      1: renormalize.infer
 0.93% :     0.000403s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.65% :     0.000282s :      1: rewriter_after_opt_a
 0.22% :     0.000094s :      1: rewriter_before_opt_a
 0.24% :     0.000103s :      1: symbol_engine_optimizer
22.48% :     0.009775s :      1: task_emit
51.04% :     0.022196s :      1: type_inference
 0.17% :     0.000072s :      1: validate


TotalTime = 0.0121968, [30]
[bootstrap]: 0.00026683
[type_inference]: 0.00455488
[event_method]: 2.338e-05
[auto_monad]: 6.345e-05
[graph_reusing]: 6.29999e-06
[pre_auto_parallel]: 2.56998e-06
[py_interpret_to_execute]: 3.697e-05
[rewriter_before_opt_a]: 8.717e-05
[expand_dump_flag]: 3.34001e-06
[jit_opt_a]: 0.00508122, [2]
    [Cycle 1]: 0.00160976, [27]
        [switch_simplify]: 5.581e-05
        [loop_unroll]: 3.472e-05
        [a_1]: 0.00058754
        [with_stream_mark]: 1.714e-05
        [recompute_prepare]: 8.28001e-06
        [updatestate_depend_eliminate]: 3.93001e-06
        [updatestate_assign_eliminate]: 3.43e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 6.69001e-06
        [updatestate_useless_node_eliminater]: 5.93002e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 2.43002e-06
        [get_grad_eliminate_]: 6.26e-06
        [merge_forward]: 3.97e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.4e-05
        [j_node_and_user_rematch]: 9.69e-06
        [meta_fg_expand]: 2.15002e-06
        [replace_old_param]: 1.08e-05
        [inline_without_move]: 6.28e-06
        [renormalize]: 0.00058494
        [add_forward_monad_depend]: 5.69e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.459e-05
        [cse]: 3.093e-05
        [replace_applicator]: 1.323e-05
    [Cycle 2]: 0.00034829, [27]
        [switch_simplify]: 6.12999e-06
        [loop_unroll]: 5.73997e-06
        [a_1]: 0.00011138
        [with_stream_mark]: 1.04e-05
        [recompute_prepare]: 5.61003e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.87001e-06
        [updatestate_useless_node_eliminater]: 5.57001e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.51998e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.335e-05
        [j_node_and_user_rematch]: 8.1e-06
        [meta_fg_expand]: 1.68002e-06
        [replace_old_param]: 8.96002e-06
        [inline_without_move]: 5.30001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 5.71e-06
        [cse]: 1.241e-05
        [replace_applicator]: 5.52001e-06
[py_interpret_to_execute_after_opt_a]: 1.064e-05
[rewriter_after_opt_a]: 3.381e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00053285
[jit_opt_b]: 5.587e-05, [1]
    [Cycle 1]: 4.957e-05, [2]
        [frontend_op_eliminate]: 1.803e-05
        [inline_after_opt_a]: 1.77e-05
[cconv]: 2.295e-05
[loop_unroll]: 0.00041353
[jit_opt_after_cconv]: 0.00015056, [1]
    [Cycle 1]: 0.00014413, [11]
        [c_1]: 2.448e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 6.24999e-06
        [updatestate_assign_eliminate]: 2.93998e-06
        [updatestate_loads_eliminate]: 2.36998e-06
        [cse]: 2.126e-05
        [call_graph_tuple_transform]: 1.96e-05
        [tuple_list_get_item_eliminator]: 6.09999e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 8.89995e-07
        [switch_simplify]: 6.06e-06
[remove_dup_value]: 1.599e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 5.54e-06
[add_recomputation]: 4.928e-05
[cse_after_recomputation]: 2.3e-05, [1]
    [Cycle 1]: 1.76e-05, [1]
        [cse]: 1.207e-05
[auto_monad_reorder]: 1.785e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 5.06002e-06
[opt_after_jit_grad]: 0.00045054
[symbol_engine_optimizer]: 7.424e-05, [1]
    [Cycle 1]: 6.819e-05, [6]
        [build]: 3.14999e-06
        [elim_shapecalc]: 8.92e-06
        [elim_not_effective]: 1.324e-05
        [opt_reshape]: 6.14001e-06
        [fold_const_symbol]: 9.25001e-06
        [renormalize]: 3.69997e-07
[validate]: 3.633e-05
Sums
    bootstrap                            :     0.000267s :  3.13%
    type_inference                       :     0.004555s : 53.46%
    event_method                         :     0.000023s :  0.27%
    auto_monad                           :     0.000063s :  0.74%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.43%
    rewriter_before_opt_a                :     0.000087s :  1.02%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.73%
    jit_opt_a.loop_unroll                :     0.000040s :  0.47%
    jit_opt_a.a_1                        :     0.000699s :  8.20%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.32%
    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 :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.14%
    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 :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.32%
    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          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000585s :  6.87%
    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 :  0.24%
    jit_opt_a.cse                        :     0.000043s :  0.51%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000034s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000533s :  6.25%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.21%
    cconv                                :     0.000023s :  0.27%
    loop_unroll                          :     0.000414s :  4.85%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.29%
    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 :  0.03%
    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 :     0.000006s :  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                     :     0.000016s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000049s :  0.58%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000451s :  5.29%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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 :  0.00%
    validate                             :     0.000036s :  0.43%


Time group info:
------[substitution.]   0.000191    28
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000005s :      4: substitution.graph_param_transform
80.64% :     0.000154s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.44% :     0.000005s :      4: substitution.replace_old_param
 8.65% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004493     2
80.15% :     0.003601s :      1: type_inference.infer
19.85% :     0.000892s :      1: type_inference.specialize
------[replace.]   0.000076     8
56.81% :     0.000043s :      4: replace.inline
43.19% :     0.000033s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     8
91.19% :     0.000151s :      4: match.inline
 8.81% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.54% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     13: predicate.addn_check_dump
 1.24% :     0.000002s :     13: predicate.addn_zero_filter
 1.76% :     0.000002s :     13: predicate.arithmetic_simplify
 1.37% :     0.000002s :     13: predicate.cast_eliminate
 0.40% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     13: predicate.compare_switch_simplify
 1.15% :     0.000002s :     13: predicate.depend_value_elim
 1.16% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.62% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     13: predicate.environ_get_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 1.14% :     0.000002s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.30% :     0.000007s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :      8: predicate.less_batch_normalization
 1.63% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.02% :     0.000003s :     21: predicate.load_eliminater
 1.41% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.17% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     13: predicate.merge_addn
 1.16% :     0.000002s :     13: predicate.minmaximum_grad
 1.55% :     0.000002s :      4: predicate.mutable_eliminate
 0.42% :     0.000001s :      4: predicate.opt_reshape
 2.57% :     0.000003s :     21: predicate.partial_eliminate
 1.17% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     13: predicate.reduce_eliminate
 1.61% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.86% :     0.000002s :     25: predicate.replace_applicator
 0.79% :     0.000001s :      8: predicate.replace_old_param
 0.38% :     0.000001s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.18% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      8: predicate.special_op_eliminate
 1.04% :     0.000001s :      8: predicate.specialize_transform
 1.35% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     21: predicate.switch_defer_inline
 2.17% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.65% :     0.000010s :     67: predicate.switch_simplify
 1.13% :     0.000002s :     13: predicate.tile_eliminate
 1.35% :     0.000002s :     13: predicate.transpose_eliminate
 1.38% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.19% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000596    12
51.62% :     0.000308s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.63% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.75% :     0.000261s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013790    72
 0.38% :     0.000052s :      1: add_recomputation
 0.48% :     0.000067s :      1: auto_monad
 0.15% :     0.000020s :      1: auto_monad_reorder
 2.04% :     0.000281s :      1: bootstrap
 0.19% :     0.000026s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.20% :     0.000027s :      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
36.87% :     0.005084s :      1: jit_opt_a
 1.11% :     0.000153s :      1: jit_opt_after_cconv
 0.43% :     0.000059s :      1: jit_opt_b
 3.05% :     0.000421s :      1: loop_unroll
 3.92% :     0.000541s :      1: mutable_eliminate
 6.75% :     0.000931s :     26: opt.transform.jit_opt_a
 0.39% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000030s :      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
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.32% :     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
 0.29% :     0.000040s :      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
 2.24% :     0.000309s :      1: renormalize.infer
 1.95% :     0.000269s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000036s :      1: rewriter_after_opt_a
 0.65% :     0.000090s :      1: rewriter_before_opt_a
 0.56% :     0.000077s :      1: symbol_engine_optimizer
33.13% :     0.004568s :      1: type_inference


TotalTime = 0.0125612, [30]
[bootstrap]: 0.00027247
[type_inference]: 0.00461227
[event_method]: 2.144e-05
[auto_monad]: 6.608e-05
[graph_reusing]: 5.84e-06
[pre_auto_parallel]: 2.79999e-06
[py_interpret_to_execute]: 3.474e-05
[rewriter_before_opt_a]: 9.026e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00537183, [2]
    [Cycle 1]: 0.00171323, [27]
        [switch_simplify]: 5.735e-05
        [loop_unroll]: 3.459e-05
        [a_1]: 0.00060499
        [with_stream_mark]: 1.738e-05
        [recompute_prepare]: 8.2e-06
        [updatestate_depend_eliminate]: 4.11001e-06
        [updatestate_assign_eliminate]: 3.98001e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 7.26001e-06
        [updatestate_useless_node_eliminater]: 5.92001e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 2.04999e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.445e-05
        [j_node_and_user_rematch]: 9.42001e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 1.082e-05
        [inline_without_move]: 6.11e-06
        [renormalize]: 0.00066943
        [add_forward_monad_depend]: 5.37999e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.416e-05
        [cse]: 3.241e-05
        [replace_applicator]: 1.311e-05
    [Cycle 2]: 0.00034789, [27]
        [switch_simplify]: 6.71e-06
        [loop_unroll]: 5.84999e-06
        [a_1]: 0.00011208
        [with_stream_mark]: 9.52999e-06
        [recompute_prepare]: 5.44998e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 5.62999e-06
        [accelerated_algorithm]: 5.70001e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.03e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 8.80001e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.40999e-06
        [cse]: 1.303e-05
        [replace_applicator]: 5.52001e-06
[py_interpret_to_execute_after_opt_a]: 1.017e-05
[rewriter_after_opt_a]: 3.562e-05
[convert_after_rewriter]: 7.08998e-06
[order_py_execute_after_rewriter]: 5.19998e-06
[mutable_eliminate]: 0.00050259
[jit_opt_b]: 5.349e-05, [1]
    [Cycle 1]: 4.671e-05, [2]
        [frontend_op_eliminate]: 1.779e-05
        [inline_after_opt_a]: 1.698e-05
[cconv]: 2.409e-05
[loop_unroll]: 0.00041819
[jit_opt_after_cconv]: 0.00014708, [1]
    [Cycle 1]: 0.00014059, [11]
        [c_1]: 2.43e-05
        [parameter_eliminate]: 2.42001e-06
        [updatestate_depend_eliminate]: 5.44998e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [cse]: 2.147e-05
        [call_graph_tuple_transform]: 1.969e-05
        [tuple_list_get_item_eliminator]: 6.67002e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.71e-06
[remove_dup_value]: 1.492e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 6.02999e-06
[add_recomputation]: 4.874e-05
[cse_after_recomputation]: 2.495e-05, [1]
    [Cycle 1]: 1.926e-05, [1]
        [cse]: 1.353e-05
[auto_monad_reorder]: 1.764e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 5.29e-06
[opt_after_jit_grad]: 0.00046902
[symbol_engine_optimizer]: 7.589e-05, [1]
    [Cycle 1]: 6.966e-05, [6]
        [build]: 3.86001e-06
        [elim_shapecalc]: 8.68001e-06
        [elim_not_effective]: 1.348e-05
        [opt_reshape]: 6.46999e-06
        [fold_const_symbol]: 9.24e-06
        [renormalize]: 6.09987e-07
[validate]: 3.828e-05
Sums
    bootstrap                            :     0.000272s :  3.14%
    type_inference                       :     0.004612s : 53.09%
    event_method                         :     0.000021s :  0.25%
    auto_monad                           :     0.000066s :  0.76%
    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                :     0.000090s :  1.04%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000064s :  0.74%
    jit_opt_a.loop_unroll                :     0.000040s :  0.47%
    jit_opt_a.a_1                        :     0.000717s :  8.25%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.09%
    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       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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_        :     0.000011s :  0.13%
    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 :  0.32%
    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          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000669s :  7.71%
    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                        :     0.000045s :  0.52%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000036s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000503s :  5.79%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000024s :  0.28%
    loop_unroll                          :     0.000418s :  4.81%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.28%
    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 :     0.000003s :  0.03%
    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 :     0.000007s :  0.08%
    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 :  0.07%
    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          :     0.000014s :  0.16%
    auto_monad_reorder                   :     0.000018s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000469s :  5.40%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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                             :     0.000038s :  0.44%


Time group info:
------[substitution.]   0.000204    28
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000006s :      4: substitution.graph_param_transform
81.90% :     0.000167s :      4: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.47% :     0.000005s :      4: substitution.replace_old_param
 7.72% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004547     2
79.87% :     0.003631s :      1: type_inference.infer
20.13% :     0.000916s :      1: type_inference.specialize
------[replace.]   0.000081     8
57.91% :     0.000047s :      4: replace.inline
42.09% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     8
92.25% :     0.000164s :      4: match.inline
 7.75% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.16% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.24% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     13: predicate.addn_check_dump
 1.45% :     0.000002s :     13: predicate.addn_zero_filter
 1.82% :     0.000002s :     13: predicate.arithmetic_simplify
 1.38% :     0.000002s :     13: predicate.cast_eliminate
 0.42% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.22% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.90% :     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
 1.21% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     13: predicate.environ_get_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.93% :     0.000001s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 5.31% :     0.000007s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :      8: predicate.less_batch_normalization
 1.56% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     21: predicate.load_eliminater
 1.21% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.32% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     13: predicate.merge_addn
 1.20% :     0.000002s :     13: predicate.minmaximum_grad
 1.43% :     0.000002s :      4: predicate.mutable_eliminate
 0.50% :     0.000001s :      4: predicate.opt_reshape
 2.58% :     0.000003s :     21: predicate.partial_eliminate
 1.16% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     13: predicate.reduce_eliminate
 1.85% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.98% :     0.000003s :     25: predicate.replace_applicator
 0.72% :     0.000001s :      8: predicate.replace_old_param
 0.40% :     0.000001s :      4: predicate.reset_defer_inline
 1.35% :     0.000002s :     13: predicate.reshape_eliminate
 1.23% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      8: predicate.special_op_eliminate
 1.03% :     0.000001s :      8: predicate.specialize_transform
 1.27% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.66% :     0.000010s :     67: predicate.switch_simplify
 1.21% :     0.000002s :     13: predicate.tile_eliminate
 1.20% :     0.000002s :     13: predicate.transpose_eliminate
 1.40% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.91% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000565    12
47.18% :     0.000267s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.12% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.69% :     0.000269s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014254    72
 0.42% :     0.000060s :      1: add_recomputation
 0.49% :     0.000069s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 2.03% :     0.000290s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000027s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.18% :     0.000025s :      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
37.71% :     0.005375s :      1: jit_opt_a
 1.05% :     0.000150s :      1: jit_opt_after_cconv
 0.39% :     0.000056s :      1: jit_opt_b
 2.99% :     0.000426s :      1: loop_unroll
 3.58% :     0.000511s :      1: mutable_eliminate
 6.68% :     0.000952s :     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
 0.09% :     0.000013s :      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
 0.24% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.35% :     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
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.76% :     0.000394s :      1: renormalize.infer
 1.88% :     0.000268s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000039s :      1: rewriter_after_opt_a
 0.65% :     0.000093s :      1: rewriter_before_opt_a
 0.55% :     0.000078s :      1: symbol_engine_optimizer
32.44% :     0.004624s :      1: type_inference


TotalTime = 0.0400749, [33]
[bootstrap]: 0.00029114
[type_inference]: 0.0215784
[event_method]: 1.885e-05
[auto_monad]: 0.00013093
[graph_reusing]: 3.75998e-06
[pre_auto_parallel]: 2.12999e-06
[py_interpret_to_execute]: 0.00011413
[rewriter_before_opt_a]: 8.697e-05
[expand_dump_flag]: 2.52001e-06
[jit_opt_a]: 0.0059609, [2]
    [Cycle 1]: 0.00211086, [27]
        [switch_simplify]: 5.562e-05
        [loop_unroll]: 3.636e-05
        [a_1]: 0.00065823
        [with_stream_mark]: 1.488e-05
        [recompute_prepare]: 1.03e-05
        [updatestate_depend_eliminate]: 5.60001e-06
        [updatestate_assign_eliminate]: 5.00001e-06
        [updatestate_loads_eliminate]: 4.2e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 9.39e-06
        [updatestate_useless_node_eliminater]: 1.169e-05
        [accelerated_algorithm]: 8.45999e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 8.28001e-06
        [merge_forward]: 4.35999e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.689e-05
        [j_node_and_user_rematch]: 1.348e-05
        [meta_fg_expand]: 3.37002e-06
        [replace_old_param]: 1.054e-05
        [inline_without_move]: 8.32e-06
        [renormalize]: 0.00097549
        [add_forward_monad_depend]: 5.37001e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 1.792e-05
        [cse]: 3.132e-05
        [replace_applicator]: 1.517e-05
    [Cycle 2]: 0.00049163, [27]
        [switch_simplify]: 9.94999e-06
        [loop_unroll]: 9.10999e-06
        [a_1]: 0.00018384
        [with_stream_mark]: 1.143e-05
        [recompute_prepare]: 8.79e-06
        [updatestate_depend_eliminate]: 5.60001e-06
        [updatestate_assign_eliminate]: 4.99998e-06
        [updatestate_loads_eliminate]: 3.91999e-06
        [parameter_eliminate]: 1.16002e-06
        [specialize_transform]: 8.77e-06
        [updatestate_useless_node_eliminater]: 1.173e-05
        [accelerated_algorithm]: 8.33001e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 7.31001e-06
        [merge_forward]: 3.91001e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.563e-05
        [j_node_and_user_rematch]: 1.218e-05
        [meta_fg_expand]: 2.87002e-06
        [replace_old_param]: 1.009e-05
        [inline_without_move]: 7.88999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 8.99978e-07
        [auto_monad_eliminator]: 1.035e-05
        [cse]: 2.225e-05
        [replace_applicator]: 8.70999e-06
[py_interpret_to_execute_after_opt_a]: 1.266e-05
[rewriter_after_opt_a]: 0.00027674
[convert_after_rewriter]: 1.101e-05
[order_py_execute_after_rewriter]: 7.72998e-06
[mutable_eliminate]: 0.00056757
[jit_opt_b]: 6.807e-05, [1]
    [Cycle 1]: 6.16e-05, [2]
        [frontend_op_eliminate]: 2.387e-05
        [inline_after_opt_a]: 2.578e-05
[cconv]: 2.37e-05
[loop_unroll]: 0.00046842
[jit_opt_after_cconv]: 0.00020029, [1]
    [Cycle 1]: 0.00019361, [11]
        [c_1]: 4.66e-05
        [parameter_eliminate]: 2.81999e-06
        [updatestate_depend_eliminate]: 8.95999e-06
        [updatestate_assign_eliminate]: 4.70001e-06
        [updatestate_loads_eliminate]: 4.55001e-06
        [cse]: 2.973e-05
        [call_graph_tuple_transform]: 2.155e-05
        [tuple_list_get_item_eliminator]: 8.51002e-06
        [none_parameter_eliminate]: 9.89996e-07
        [renormalize]: 4.09986e-07
        [switch_simplify]: 9.83002e-06
[remove_dup_value]: 1.784e-05
[partial_unused_args_eliminate]: 1.95001e-06
[environ_conv]: 1.005e-05
[add_recomputation]: 5.579e-05
[cse_after_recomputation]: 3.232e-05, [1]
    [Cycle 1]: 2.619e-05, [1]
        [cse]: 2.001e-05
[auto_monad_reorder]: 2.41e-05
[get_jit_bprop_graph]: 1.77001e-06
[rewriter_after_jit_bprop_graph]: 6.21e-06
[opt_after_jit_grad]: 0.00054334
[symbol_engine_optimizer]: 0.00010203, [1]
    [Cycle 1]: 9.498e-05, [6]
        [build]: 9.52999e-06
        [elim_shapecalc]: 1.188e-05
        [elim_not_effective]: 1.817e-05
        [opt_reshape]: 9.41e-06
        [fold_const_symbol]: 1.496e-05
        [renormalize]: 4.09986e-07
[validate]: 5.213e-05
[backend_pass]: 1.49e-06
[task_emit]: 0.00918292
[execute]: 6.14001e-06
Sums
    bootstrap                            :     0.000291s :  0.81%
    type_inference                       :     0.021578s : 59.80%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000131s :  0.36%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000114s :  0.32%
    rewriter_before_opt_a                :     0.000087s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.18%
    jit_opt_a.loop_unroll                :     0.000045s :  0.13%
    jit_opt_a.a_1                        :     0.000842s :  2.33%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000976s :  2.70%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.08%
    jit_opt_a.cse                        :     0.000054s :  0.15%
    jit_opt_a.replace_applicator         :     0.000024s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000277s :  0.77%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000568s :  1.57%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.07%
    cconv                                :     0.000024s :  0.07%
    loop_unroll                          :     0.000468s :  1.30%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.03%
    remove_dup_value                     :     0.000018s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000056s :  0.15%
    cse_after_recomputation.cse          :     0.000020s :  0.06%
    auto_monad_reorder                   :     0.000024s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000543s :  1.51%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000052s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009183s : 25.45%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000232    47
 3.85% :     0.000009s :      2: substitution.depend_value_elim
 1.01% :     0.000002s :      4: substitution.elim_not_effective
 0.80% :     0.000002s :      4: substitution.fold_const_symbol
 2.21% :     0.000005s :      5: substitution.graph_param_transform
74.87% :     0.000174s :      4: substitution.inline
 1.80% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.86% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.42% :     0.000003s :      2: substitution.replace_old_param
 3.96% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
 3.71% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.52% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021520     2
94.00% :     0.020229s :      1: type_inference.infer
 6.00% :     0.001291s :      1: type_inference.specialize
------[replace.]   0.000083     6
66.11% :     0.000055s :      4: replace.inline
33.89% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
95.53% :     0.000171s :      4: match.inline
 4.47% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000172  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.24% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.30% :     0.000002s :     15: predicate.addn_check_dump
 1.30% :     0.000002s :     15: predicate.addn_zero_filter
 1.98% :     0.000003s :     15: predicate.arithmetic_simplify
 1.27% :     0.000002s :     15: predicate.cast_eliminate
 0.58% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.47% :     0.000003s :     15: predicate.compare_switch_simplify
 1.35% :     0.000002s :     15: predicate.depend_value_elim
 1.19% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.00% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.65% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.26% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     15: predicate.environ_get_eliminate
 1.23% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      5: predicate.fold_const_symbol
 1.01% :     0.000002s :     10: predicate.get_grad_eliminate
 0.30% :     0.000001s :      5: predicate.graph_param_transform
 4.99% :     0.000009s :     31: predicate.inline
 1.05% :     0.000002s :     10: predicate.inline_without_move
 0.47% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :     10: predicate.less_batch_normalization
 1.47% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.88% :     0.000003s :     22: predicate.load_eliminater
 1.42% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.53% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.59% :     0.000003s :      5: predicate.mutable_eliminate
 0.55% :     0.000001s :      5: predicate.opt_reshape
 2.27% :     0.000004s :     22: predicate.partial_eliminate
 1.39% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.68% :     0.000003s :     15: predicate.reduce_eliminate
 1.46% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.70% :     0.000003s :     27: predicate.replace_applicator
 0.68% :     0.000001s :     10: predicate.replace_old_param
 0.30% :     0.000001s :      5: predicate.reset_defer_inline
 1.26% :     0.000002s :     15: predicate.reshape_eliminate
 1.33% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.12% :     0.000002s :     10: predicate.special_op_eliminate
 1.16% :     0.000002s :     10: predicate.specialize_transform
 1.51% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.32% :     0.000004s :     21: predicate.switch_defer_inline
 1.89% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.26% :     0.000011s :     68: predicate.switch_simplify
 1.24% :     0.000002s :     15: predicate.tile_eliminate
 1.34% :     0.000002s :     15: predicate.transpose_eliminate
 1.64% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.40% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000003s :     15: predicate.value_based_eliminate
 0.49% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000820    13
48.13% :     0.000395s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.90% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.97% :     0.000385s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042384    76
 0.14% :     0.000059s :      1: add_recomputation
 0.32% :     0.000136s :      1: auto_monad
 0.06% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.72% :     0.000303s :      1: bootstrap
 0.06% :     0.000027s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.08% :     0.000035s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.05% :     0.000023s :      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
 0.01% :     0.000006s :      1: graph_reusing
14.07% :     0.005964s :      1: jit_opt_a
 0.48% :     0.000203s :      1: jit_opt_after_cconv
 0.17% :     0.000071s :      1: jit_opt_b
 1.12% :     0.000477s :      1: loop_unroll
 1.36% :     0.000577s :      1: mutable_eliminate
 2.69% :     0.001139s :     26: opt.transform.jit_opt_a
 0.19% :     0.000083s :      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
 0.05% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.30% :     0.000551s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000118s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.29% :     0.000549s :      1: renormalize.infer
 0.99% :     0.000419s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.66% :     0.000281s :      1: rewriter_after_opt_a
 0.22% :     0.000091s :      1: rewriter_before_opt_a
 0.25% :     0.000105s :      1: symbol_engine_optimizer
21.70% :     0.009196s :      1: task_emit
50.94% :     0.021590s :      1: type_inference
 0.18% :     0.000075s :      1: validate


TotalTime = 0.0120675, [30]
[bootstrap]: 0.00025282
[type_inference]: 0.00439453
[event_method]: 2.042e-05
[auto_monad]: 4.578e-05
[graph_reusing]: 3.97e-06
[pre_auto_parallel]: 1.53002e-06
[py_interpret_to_execute]: 3.334e-05
[rewriter_before_opt_a]: 8.151e-05
[expand_dump_flag]: 2.74999e-06
[jit_opt_a]: 0.00524262, [2]
    [Cycle 1]: 0.00161257, [27]
        [switch_simplify]: 5.197e-05
        [loop_unroll]: 3.47e-05
        [a_1]: 0.00058271
        [with_stream_mark]: 1.451e-05
        [recompute_prepare]: 8.75001e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 6.76999e-06
        [updatestate_useless_node_eliminater]: 5.97999e-06
        [accelerated_algorithm]: 6.85002e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 2.77002e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.208e-05
        [j_node_and_user_rematch]: 1.005e-05
        [meta_fg_expand]: 2.15002e-06
        [replace_old_param]: 1.039e-05
        [inline_without_move]: 6.22001e-06
        [renormalize]: 0.0006184
        [add_forward_monad_depend]: 4.3e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 1.147e-05
        [cse]: 2.36e-05
        [replace_applicator]: 1.229e-05
    [Cycle 2]: 0.0004066, [27]
        [switch_simplify]: 6.73e-06
        [loop_unroll]: 5.56998e-06
        [a_1]: 0.00015963
        [with_stream_mark]: 9.24e-06
        [recompute_prepare]: 5.92999e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 5.68002e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.369e-05
        [j_node_and_user_rematch]: 8.77999e-06
        [meta_fg_expand]: 1.94999e-06
        [replace_old_param]: 8.71002e-06
        [inline_without_move]: 6.10002e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 6.93998e-06
        [cse]: 1.443e-05
        [replace_applicator]: 5.65001e-06
[py_interpret_to_execute_after_opt_a]: 1.048e-05
[rewriter_after_opt_a]: 3.174e-05
[convert_after_rewriter]: 6.31998e-06
[order_py_execute_after_rewriter]: 4.40999e-06
[mutable_eliminate]: 0.00051625
[jit_opt_b]: 5.425e-05, [1]
    [Cycle 1]: 4.786e-05, [2]
        [frontend_op_eliminate]: 1.848e-05
        [inline_after_opt_a]: 1.76e-05
[cconv]: 1.677e-05
[loop_unroll]: 0.00041206
[jit_opt_after_cconv]: 0.00014512, [1]
    [Cycle 1]: 0.00013898, [11]
        [c_1]: 2.44e-05
        [parameter_eliminate]: 2.57001e-06
        [updatestate_depend_eliminate]: 5.71e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [cse]: 2.101e-05
        [call_graph_tuple_transform]: 1.822e-05
        [tuple_list_get_item_eliminator]: 5.84e-06
        [none_parameter_eliminate]: 1.15999e-06
        [renormalize]: 1.90019e-07
        [switch_simplify]: 6.01998e-06
[remove_dup_value]: 9.42999e-06
[partial_unused_args_eliminate]: 1.62001e-06
[environ_conv]: 4.26001e-06
[add_recomputation]: 3.461e-05
[cse_after_recomputation]: 2.246e-05, [1]
    [Cycle 1]: 1.749e-05, [1]
        [cse]: 1.17e-05
[auto_monad_reorder]: 1.213e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.43999e-06
[opt_after_jit_grad]: 0.00044921
[symbol_engine_optimizer]: 7.239e-05, [1]
    [Cycle 1]: 6.64e-05, [6]
        [build]: 2.86e-06
        [elim_shapecalc]: 8.01001e-06
        [elim_not_effective]: 1.264e-05
        [opt_reshape]: 6.28998e-06
        [fold_const_symbol]: 8.99998e-06
        [renormalize]: 4.10015e-07
[validate]: 3.088e-05
Sums
    bootstrap                            :     0.000253s :  3.05%
    type_inference                       :     0.004395s : 52.97%
    event_method                         :     0.000020s :  0.25%
    auto_monad                           :     0.000046s :  0.55%
    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                :     0.000082s :  0.98%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000059s :  0.71%
    jit_opt_a.loop_unroll                :     0.000040s :  0.49%
    jit_opt_a.a_1                        :     0.000742s :  8.95%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000015s :  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 :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.15%
    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              :     0.000006s :  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    :     0.000019s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.15%
    jit_opt_a.renormalize                :     0.000618s :  7.45%
    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      :     0.000018s :  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  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000032s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000516s :  6.22%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.21%
    cconv                                :     0.000017s :  0.20%
    loop_unroll                          :     0.000412s :  4.97%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.29%
    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              :     0.000021s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  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  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000009s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000035s :  0.42%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000012s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000449s :  5.41%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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 :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.37%


Time group info:
------[substitution.]   0.000175    28
 0.70% :     0.000001s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.45% :     0.000004s :      4: substitution.graph_param_transform
83.07% :     0.000145s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000004s :      4: substitution.replace_old_param
 7.01% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004339     2
80.17% :     0.003479s :      1: type_inference.infer
19.83% :     0.000860s :      1: type_inference.specialize
------[replace.]   0.000078     8
55.52% :     0.000043s :      4: replace.inline
44.48% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     8
93.26% :     0.000142s :      4: match.inline
 6.74% :     0.000010s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000180   912
 1.37% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.71% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 0.82% :     0.000001s :     13: predicate.addn_check_dump
 1.18% :     0.000002s :     13: predicate.addn_zero_filter
 1.48% :     0.000003s :     13: predicate.arithmetic_simplify
 0.99% :     0.000002s :     13: predicate.cast_eliminate
 0.32% :     0.000001s :      4: predicate.check_bprop_eliminate
 0.81% :     0.000001s :     13: predicate.compare_switch_simplify
 0.91% :     0.000002s :     13: predicate.depend_value_elim
 1.05% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 0.96% :     0.000002s :     13: predicate.dict_get_item_eliminator
 0.88% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.69% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      4: predicate.elim_not_effective
 0.37% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 0.90% :     0.000002s :     13: predicate.environ_add_const_eliminate
 0.86% :     0.000002s :     13: predicate.environ_get_add_eliminate
 0.85% :     0.000002s :     13: predicate.environ_get_depend_swap
 0.92% :     0.000002s :     13: predicate.environ_get_eliminate
 0.83% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      4: predicate.fold_const_symbol
 0.76% :     0.000001s :      8: predicate.get_grad_eliminate
 0.20% :     0.000000s :      4: predicate.graph_param_transform
 3.72% :     0.000007s :     29: predicate.inline
 0.72% :     0.000001s :      8: predicate.inline_without_move
 0.31% :     0.000001s :      8: predicate.j_node_and_user_rematch
 0.84% :     0.000002s :      8: predicate.less_batch_normalization
 1.23% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.56% :     0.000003s :     21: predicate.load_eliminater
 0.86% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.07% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.36% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 0.82% :     0.000001s :     13: predicate.merge_addn
 0.84% :     0.000002s :     13: predicate.minmaximum_grad
 1.16% :     0.000002s :      4: predicate.mutable_eliminate
 0.33% :     0.000001s :      4: predicate.opt_reshape
 1.82% :     0.000003s :     21: predicate.partial_eliminate
 0.89% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.24% :     0.000002s :     13: predicate.reduce_eliminate
 1.16% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.51% :     0.000003s :     25: predicate.replace_applicator
 0.51% :     0.000001s :      8: predicate.replace_old_param
 0.27% :     0.000000s :      4: predicate.reset_defer_inline
 1.07% :     0.000002s :     13: predicate.reshape_eliminate
 0.92% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      4: predicate.row_tensor_eliminate
 0.90% :     0.000002s :     13: predicate.same_eliminate
 0.50% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.72% :     0.000001s :      8: predicate.special_op_eliminate
 0.71% :     0.000001s :      8: predicate.specialize_transform
 1.00% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 0.89% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.30% :     0.000001s :      4: predicate.switch_call_monad_eliminater
27.53% :     0.000049s :     21: predicate.switch_defer_inline
 1.70% :     0.000003s :     21: predicate.switch_layer_defer_inline
 5.54% :     0.000010s :     67: predicate.switch_simplify
 0.86% :     0.000002s :     13: predicate.tile_eliminate
 0.95% :     0.000002s :     13: predicate.transpose_eliminate
 1.06% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.05% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 2.67% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.13% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.21% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.42% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.27% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.16% :     0.000002s :     13: predicate.value_based_eliminate
 0.28% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.47% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
46.11% :     0.000253s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.14% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.75% :     0.000267s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013743    72
 0.27% :     0.000037s :      1: add_recomputation
 0.36% :     0.000050s :      1: auto_monad
 0.11% :     0.000015s :      1: auto_monad_reorder
 1.92% :     0.000264s :      1: bootstrap
 0.14% :     0.000019s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.17% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
38.17% :     0.005246s :      1: jit_opt_a
 1.08% :     0.000148s :      1: jit_opt_after_cconv
 0.41% :     0.000057s :      1: jit_opt_b
 3.05% :     0.000420s :      1: loop_unroll
 3.82% :     0.000525s :      1: mutable_eliminate
 7.07% :     0.000972s :     26: opt.transform.jit_opt_a
 0.37% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000030s :      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
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.32% :     0.000457s :      1: opt_after_jit_grad
 0.05% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000003s :      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
 0.09% :     0.000012s :      1: remove_dup_value
 2.48% :     0.000341s :      1: renormalize.infer
 1.97% :     0.000271s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000035s :      1: rewriter_after_opt_a
 0.61% :     0.000084s :      1: rewriter_before_opt_a
 0.55% :     0.000075s :      1: symbol_engine_optimizer
32.06% :     0.004406s :      1: type_inference


TotalTime = 0.0123449, [30]
[bootstrap]: 0.00026584
[type_inference]: 0.00451842
[event_method]: 2.062e-05
[auto_monad]: 5.732e-05
[graph_reusing]: 5.45001e-06
[pre_auto_parallel]: 2.01e-06
[py_interpret_to_execute]: 3.756e-05
[rewriter_before_opt_a]: 8.481e-05
[expand_dump_flag]: 2.61999e-06
[jit_opt_a]: 0.00527408, [2]
    [Cycle 1]: 0.00175548, [27]
        [switch_simplify]: 5.426e-05
        [loop_unroll]: 3.645e-05
        [a_1]: 0.0006219
        [with_stream_mark]: 1.7e-05
        [recompute_prepare]: 8.52e-06
        [updatestate_depend_eliminate]: 3.70998e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 7.16999e-06
        [updatestate_useless_node_eliminater]: 6.08002e-06
        [accelerated_algorithm]: 6.44999e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.99999e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 9.60019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.389e-05
        [j_node_and_user_rematch]: 9.41e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 1.029e-05
        [inline_without_move]: 6.15002e-06
        [renormalize]: 0.00070077
        [add_forward_monad_depend]: 5.68002e-06
        [auto_monad_grad]: 3.23e-06
        [auto_monad_eliminator]: 1.34e-05
        [cse]: 2.784e-05
        [replace_applicator]: 1.422e-05
    [Cycle 2]: 0.00035774, [27]
        [switch_simplify]: 6.91001e-06
        [loop_unroll]: 5.70001e-06
        [a_1]: 0.00011323
        [with_stream_mark]: 1.035e-05
        [recompute_prepare]: 5.65001e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.31002e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.63002e-06
        [accelerated_algorithm]: 5.79999e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 8.30999e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 9.18002e-06
        [inline_without_move]: 5.51002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 9.89996e-07
        [auto_monad_grad]: 1.31002e-06
        [auto_monad_eliminator]: 6.01e-06
        [cse]: 1.326e-05
        [replace_applicator]: 5.81e-06
[py_interpret_to_execute_after_opt_a]: 1.283e-05
[rewriter_after_opt_a]: 3.507e-05
[convert_after_rewriter]: 7.66999e-06
[order_py_execute_after_rewriter]: 5.35001e-06
[mutable_eliminate]: 0.00053088
[jit_opt_b]: 5.491e-05, [1]
    [Cycle 1]: 4.744e-05, [2]
        [frontend_op_eliminate]: 1.769e-05
        [inline_after_opt_a]: 1.841e-05
[cconv]: 2.393e-05
[loop_unroll]: 0.00042634
[jit_opt_after_cconv]: 0.00015029, [1]
    [Cycle 1]: 0.0001442, [11]
        [c_1]: 2.539e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.99e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.097e-05
        [call_graph_tuple_transform]: 1.927e-05
        [tuple_list_get_item_eliminator]: 6.48e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 6.44999e-06
[remove_dup_value]: 1.411e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 5.42999e-06
[add_recomputation]: 4.623e-05
[cse_after_recomputation]: 2.35e-05, [1]
    [Cycle 1]: 1.816e-05, [1]
        [cse]: 1.213e-05
[auto_monad_reorder]: 1.623e-05
[get_jit_bprop_graph]: 1.54998e-06
[rewriter_after_jit_bprop_graph]: 4.67e-06
[opt_after_jit_grad]: 0.00045287
[symbol_engine_optimizer]: 7.54e-05, [1]
    [Cycle 1]: 6.945e-05, [6]
        [build]: 3.36001e-06
        [elim_shapecalc]: 8.80001e-06
        [elim_not_effective]: 1.334e-05
        [opt_reshape]: 6.21e-06
        [fold_const_symbol]: 9.34e-06
        [renormalize]: 6.00005e-07
[validate]: 3.513e-05
Sums
    bootstrap                            :     0.000266s :  3.08%
    type_inference                       :     0.004518s : 52.35%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000057s :  0.66%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.44%
    rewriter_before_opt_a                :     0.000085s :  0.98%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.71%
    jit_opt_a.loop_unroll                :     0.000042s :  0.49%
    jit_opt_a.a_1                        :     0.000735s :  8.52%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.32%
    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       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000011s :  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 :     0.000027s :  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          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000701s :  8.12%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000041s :  0.48%
    jit_opt_a.replace_applicator         :     0.000020s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.15%
    rewriter_after_opt_a                 :     0.000035s :  0.41%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000531s :  6.15%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.21%
    cconv                                :     0.000024s :  0.28%
    loop_unroll                          :     0.000426s :  4.94%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.29%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    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      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000046s :  0.54%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000016s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000453s :  5.25%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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                             :     0.000035s :  0.41%


Time group info:
------[substitution.]   0.000209    28
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000006s :      4: substitution.graph_param_transform
82.41% :     0.000172s :      4: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000005s :      4: substitution.replace_old_param
 7.74% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004461     2
80.28% :     0.003582s :      1: type_inference.infer
19.72% :     0.000880s :      1: type_inference.specialize
------[replace.]   0.000084     8
56.31% :     0.000048s :      4: replace.inline
43.69% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     8
92.28% :     0.000169s :      4: match.inline
 7.72% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   912
 1.25% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     13: predicate.addn_check_dump
 1.57% :     0.000002s :     13: predicate.addn_zero_filter
 2.03% :     0.000003s :     13: predicate.arithmetic_simplify
 1.33% :     0.000002s :     13: predicate.cast_eliminate
 0.45% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.18% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      4: predicate.elim_not_effective
 0.57% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     13: predicate.environ_get_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 1.02% :     0.000001s :      8: predicate.get_grad_eliminate
 0.38% :     0.000001s :      4: predicate.graph_param_transform
 5.30% :     0.000007s :     29: predicate.inline
 0.99% :     0.000001s :      8: predicate.inline_without_move
 0.50% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :      8: predicate.less_batch_normalization
 1.66% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     21: predicate.load_eliminater
 1.42% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.32% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     13: predicate.merge_addn
 1.13% :     0.000002s :     13: predicate.minmaximum_grad
 1.33% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.55% :     0.000003s :     21: predicate.partial_eliminate
 1.14% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     13: predicate.reduce_eliminate
 1.61% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.00% :     0.000003s :     25: predicate.replace_applicator
 0.78% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.17% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      8: predicate.special_op_eliminate
 1.02% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     21: predicate.switch_defer_inline
 2.46% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.77% :     0.000010s :     67: predicate.switch_simplify
 1.20% :     0.000002s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.51% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     13: predicate.value_based_eliminate
 0.36% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000569    12
45.86% :     0.000261s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.51% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.63% :     0.000271s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014100    72
 0.35% :     0.000049s :      1: add_recomputation
 0.43% :     0.000061s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 1.97% :     0.000278s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.17% :     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
37.43% :     0.005278s :      1: jit_opt_a
 1.08% :     0.000153s :      1: jit_opt_after_cconv
 0.41% :     0.000058s :      1: jit_opt_b
 3.08% :     0.000434s :      1: loop_unroll
 3.82% :     0.000539s :      1: mutable_eliminate
 6.89% :     0.000972s :     26: opt.transform.jit_opt_a
 0.38% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000030s :      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
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.27% :     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% :     0.000040s :      1: py_interpret_to_execute
 0.11% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.70% :     0.000380s :      1: renormalize.infer
 2.22% :     0.000313s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000038s :      1: rewriter_after_opt_a
 0.62% :     0.000088s :      1: rewriter_before_opt_a
 0.55% :     0.000078s :      1: symbol_engine_optimizer
32.12% :     0.004529s :      1: type_inference


TotalTime = 0.040659, [33]
[bootstrap]: 0.00026823
[type_inference]: 0.0212208
[event_method]: 1.978e-05
[auto_monad]: 0.00015524
[graph_reusing]: 6.61999e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 0.0001164
[rewriter_before_opt_a]: 9.856e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00594893, [2]
    [Cycle 1]: 0.00213677, [27]
        [switch_simplify]: 5.955e-05
        [loop_unroll]: 3.575e-05
        [a_1]: 0.00068261
        [with_stream_mark]: 1.68e-05
        [recompute_prepare]: 1.097e-05
        [updatestate_depend_eliminate]: 6.39001e-06
        [updatestate_assign_eliminate]: 5.69e-06
        [updatestate_loads_eliminate]: 4.94e-06
        [parameter_eliminate]: 1.93002e-06
        [specialize_transform]: 8.98002e-06
        [updatestate_useless_node_eliminater]: 1.151e-05
        [accelerated_algorithm]: 8.17e-06
        [meta_shard_fg_expand]: 3.30003e-06
        [get_grad_eliminate_]: 7.89002e-06
        [merge_forward]: 5.00999e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.049e-05
        [j_node_and_user_rematch]: 1.31e-05
        [meta_fg_expand]: 3.68999e-06
        [replace_old_param]: 1.122e-05
        [inline_without_move]: 8.22e-06
        [renormalize]: 0.00095811
        [add_forward_monad_depend]: 5.69999e-06
        [auto_monad_grad]: 1.69e-06
        [auto_monad_eliminator]: 2.058e-05
        [cse]: 3.692e-05
        [replace_applicator]: 1.589e-05
    [Cycle 2]: 0.00046531, [27]
        [switch_simplify]: 8.84998e-06
        [loop_unroll]: 8.3e-06
        [a_1]: 0.0001679
        [with_stream_mark]: 1.112e-05
        [recompute_prepare]: 8.12e-06
        [updatestate_depend_eliminate]: 5.16002e-06
        [updatestate_assign_eliminate]: 4.67998e-06
        [updatestate_loads_eliminate]: 4.1e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 8e-06
        [updatestate_useless_node_eliminater]: 1.085e-05
        [accelerated_algorithm]: 8.01001e-06
        [meta_shard_fg_expand]: 1.77999e-06
        [get_grad_eliminate_]: 7.66999e-06
        [merge_forward]: 4.36002e-06
        [cell_reuse_recompute_pass]: 1.62999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.69e-05
        [j_node_and_user_rematch]: 1.276e-05
        [meta_fg_expand]: 2.71e-06
        [replace_old_param]: 9.89001e-06
        [inline_without_move]: 7.75e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 9.66e-06
        [cse]: 2.027e-05
        [replace_applicator]: 7.93999e-06
[py_interpret_to_execute_after_opt_a]: 9.87001e-06
[rewriter_after_opt_a]: 0.00029367
[convert_after_rewriter]: 1.122e-05
[order_py_execute_after_rewriter]: 7.1e-06
[mutable_eliminate]: 0.0006057
[jit_opt_b]: 6.767e-05, [1]
    [Cycle 1]: 6.084e-05, [2]
        [frontend_op_eliminate]: 2.537e-05
        [inline_after_opt_a]: 2.409e-05
[cconv]: 2.425e-05
[loop_unroll]: 0.00044265
[jit_opt_after_cconv]: 0.00018582, [1]
    [Cycle 1]: 0.00017933, [11]
        [c_1]: 4.363e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 7.51001e-06
        [updatestate_assign_eliminate]: 4.26001e-06
        [updatestate_loads_eliminate]: 3.93999e-06
        [cse]: 2.617e-05
        [call_graph_tuple_transform]: 2.152e-05
        [tuple_list_get_item_eliminator]: 8.28001e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 8.57e-06
[remove_dup_value]: 1.943e-05
[partial_unused_args_eliminate]: 2.34999e-06
[environ_conv]: 1.023e-05
[add_recomputation]: 6.463e-05
[cse_after_recomputation]: 2.852e-05, [1]
    [Cycle 1]: 2.286e-05, [1]
        [cse]: 1.714e-05
[auto_monad_reorder]: 2.409e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 4.35e-06
[opt_after_jit_grad]: 0.00049427
[symbol_engine_optimizer]: 0.00010203, [1]
    [Cycle 1]: 9.587e-05, [6]
        [build]: 1.058e-05
        [elim_shapecalc]: 1.257e-05
        [elim_not_effective]: 2e-05
        [opt_reshape]: 9.32999e-06
        [fold_const_symbol]: 1.415e-05
        [renormalize]: 3.89991e-07
[validate]: 5.028e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.0101626
[execute]: 5.25999e-06
Sums
    bootstrap                            :     0.000268s :  0.73%
    type_inference                       :     0.021221s : 57.81%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000155s :  0.42%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000116s :  0.32%
    rewriter_before_opt_a                :     0.000099s :  0.27%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.19%
    jit_opt_a.loop_unroll                :     0.000044s :  0.12%
    jit_opt_a.a_1                        :     0.000851s :  2.32%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  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 :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000958s :  2.61%
    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      :     0.000030s :  0.08%
    jit_opt_a.cse                        :     0.000057s :  0.16%
    jit_opt_a.replace_applicator         :     0.000024s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000294s :  0.80%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000606s :  1.65%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000024s :  0.07%
    loop_unroll                          :     0.000443s :  1.21%
    jit_opt_after_cconv.c_1              :     0.000044s :  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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000065s :  0.18%
    cse_after_recomputation.cse          :     0.000017s :  0.05%
    auto_monad_reorder                   :     0.000024s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000494s :  1.35%
    symbol_engine_optimizer.build        :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    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                             :     0.000050s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010163s : 27.69%
    execute                              :     0.000005s :  0.01%


Time group info:
------[substitution.]   0.000255    47
 3.41% :     0.000009s :      2: substitution.depend_value_elim
 1.08% :     0.000003s :      4: substitution.elim_not_effective
 0.96% :     0.000002s :      4: substitution.fold_const_symbol
 2.46% :     0.000006s :      5: substitution.graph_param_transform
74.85% :     0.000191s :      4: substitution.inline
 1.79% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.93% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.52% :     0.000004s :      2: substitution.replace_old_param
 4.71% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.12% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.17% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021156     2
93.22% :     0.019722s :      1: type_inference.infer
 6.78% :     0.001434s :      1: type_inference.specialize
------[replace.]   0.000087     6
67.00% :     0.000059s :      4: replace.inline
33.00% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     6
94.54% :     0.000188s :      4: match.inline
 5.46% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1021
 1.35% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.24% :     0.000002s :     15: predicate.addn_check_dump
 1.35% :     0.000002s :     15: predicate.addn_zero_filter
 2.01% :     0.000003s :     15: predicate.arithmetic_simplify
 1.46% :     0.000002s :     15: predicate.cast_eliminate
 0.49% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     15: predicate.compare_switch_simplify
 1.31% :     0.000002s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.92% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.54% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.34% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     15: predicate.environ_get_eliminate
 1.22% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.01% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 5.20% :     0.000009s :     31: predicate.inline
 1.06% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :     10: predicate.less_batch_normalization
 1.36% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     22: predicate.load_eliminater
 1.30% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.48% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.90% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     15: predicate.merge_addn
 1.11% :     0.000002s :     15: predicate.minmaximum_grad
 1.44% :     0.000002s :      5: predicate.mutable_eliminate
 0.47% :     0.000001s :      5: predicate.opt_reshape
 2.33% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.71% :     0.000003s :     15: predicate.reduce_eliminate
 1.55% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.75% :     0.000003s :     27: predicate.replace_applicator
 0.59% :     0.000001s :     10: predicate.replace_old_param
 0.37% :     0.000001s :      5: predicate.reset_defer_inline
 1.28% :     0.000002s :     15: predicate.reshape_eliminate
 1.30% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.90% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.11% :     0.000002s :     10: predicate.special_op_eliminate
 1.20% :     0.000002s :     10: predicate.specialize_transform
 1.60% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.27% :     0.000004s :     21: predicate.switch_defer_inline
 1.91% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.92% :     0.000012s :     68: predicate.switch_simplify
 1.18% :     0.000002s :     15: predicate.tile_eliminate
 1.24% :     0.000002s :     15: predicate.transpose_eliminate
 1.47% :     0.000002s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.96% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.83% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.24% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000776    13
42.74% :     0.000332s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.17% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.09% :     0.000404s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042958    76
 0.16% :     0.000068s :      1: add_recomputation
 0.37% :     0.000160s :      1: auto_monad
 0.06% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000282s :      1: bootstrap
 0.06% :     0.000027s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.06% :     0.000024s :      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
13.86% :     0.005952s :      1: jit_opt_a
 0.44% :     0.000189s :      1: jit_opt_after_cconv
 0.16% :     0.000071s :      1: jit_opt_b
 1.05% :     0.000451s :      1: loop_unroll
 1.43% :     0.000614s :      1: mutable_eliminate
 2.67% :     0.001148s :     26: opt.transform.jit_opt_a
 0.18% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.17% :     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% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000120s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.30% :     0.000560s :      1: renormalize.infer
 0.91% :     0.000390s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.69% :     0.000298s :      1: rewriter_after_opt_a
 0.24% :     0.000103s :      1: rewriter_before_opt_a
 0.24% :     0.000105s :      1: symbol_engine_optimizer
23.68% :     0.010173s :      1: task_emit
49.43% :     0.021233s :      1: type_inference
 0.17% :     0.000073s :      1: validate


TotalTime = 0.0135503, [30]
[bootstrap]: 0.00027122
[type_inference]: 0.00466401
[event_method]: 2.073e-05
[auto_monad]: 6.389e-05
[graph_reusing]: 6.06998e-06
[pre_auto_parallel]: 2.23998e-06
[py_interpret_to_execute]: 3.651e-05
[rewriter_before_opt_a]: 8.525e-05
[expand_dump_flag]: 2.66e-06
[jit_opt_a]: 0.005734, [2]
    [Cycle 1]: 0.00203229, [27]
        [switch_simplify]: 6.302e-05
        [loop_unroll]: 3.717e-05
        [a_1]: 0.00066533
        [with_stream_mark]: 2.28e-05
        [recompute_prepare]: 8.485e-05
        [updatestate_depend_eliminate]: 5.55001e-06
        [updatestate_assign_eliminate]: 3.86999e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 2.73e-06
        [specialize_transform]: 8.55001e-06
        [updatestate_useless_node_eliminater]: 6.78e-06
        [accelerated_algorithm]: 6.53e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 6.44001e-06
        [merge_forward]: 4.45999e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.679e-05
        [j_node_and_user_rematch]: 1.115e-05
        [meta_fg_expand]: 2.70002e-06
        [replace_old_param]: 1.172e-05
        [inline_without_move]: 5.96e-06
        [renormalize]: 0.00078304
        [add_forward_monad_depend]: 7.4e-06
        [auto_monad_grad]: 2.78e-06
        [auto_monad_eliminator]: 1.749e-05
        [cse]: 3.074e-05
        [replace_applicator]: 1.888e-05
    [Cycle 2]: 0.00042142, [27]
        [switch_simplify]: 8.05e-06
        [loop_unroll]: 6.21998e-06
        [a_1]: 0.00012263
        [with_stream_mark]: 1.241e-05
        [recompute_prepare]: 6.36e-06
        [updatestate_depend_eliminate]: 3.77002e-06
        [updatestate_assign_eliminate]: 3.32002e-06
        [updatestate_loads_eliminate]: 2.70002e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 6.35002e-06
        [updatestate_useless_node_eliminater]: 6.16e-06
        [accelerated_algorithm]: 6.74999e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 5.92001e-06
        [merge_forward]: 3.71999e-06
        [cell_reuse_recompute_pass]: 2.07001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.708e-05
        [j_node_and_user_rematch]: 1.044e-05
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 1.104e-05
        [inline_without_move]: 6.44999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 1.47001e-06
        [auto_monad_eliminator]: 8.87999e-06
        [cse]: 2.154e-05
        [replace_applicator]: 8.2e-06
[py_interpret_to_execute_after_opt_a]: 1.573e-05
[rewriter_after_opt_a]: 5.033e-05
[convert_after_rewriter]: 7.77e-06
[order_py_execute_after_rewriter]: 6.06998e-06
[mutable_eliminate]: 0.00074056
[jit_opt_b]: 6.406e-05, [1]
    [Cycle 1]: 5.459e-05, [2]
        [frontend_op_eliminate]: 2.116e-05
        [inline_after_opt_a]: 2.069e-05
[cconv]: 2.707e-05
[loop_unroll]: 0.00053641
[jit_opt_after_cconv]: 0.00019078, [1]
    [Cycle 1]: 0.00018235, [11]
        [c_1]: 2.799e-05
        [parameter_eliminate]: 4.57998e-06
        [updatestate_depend_eliminate]: 8.07998e-06
        [updatestate_assign_eliminate]: 3.75998e-06
        [updatestate_loads_eliminate]: 3.32002e-06
        [cse]: 3.416e-05
        [call_graph_tuple_transform]: 2.534e-05
        [tuple_list_get_item_eliminator]: 7.25e-06
        [none_parameter_eliminate]: 1.73997e-06
        [renormalize]: 1.11002e-06
        [switch_simplify]: 7.48e-06
[remove_dup_value]: 1.644e-05
[partial_unused_args_eliminate]: 2.58e-06
[environ_conv]: 6.17001e-06
[add_recomputation]: 5.521e-05
[cse_after_recomputation]: 2.757e-05, [1]
    [Cycle 1]: 2.137e-05, [1]
        [cse]: 1.46e-05
[auto_monad_reorder]: 2.021e-05
[get_jit_bprop_graph]: 2.21e-06
[rewriter_after_jit_bprop_graph]: 7.57998e-06
[opt_after_jit_grad]: 0.00057992
[symbol_engine_optimizer]: 7.978e-05, [1]
    [Cycle 1]: 7.322e-05, [6]
        [build]: 3.97e-06
        [elim_shapecalc]: 9.42001e-06
        [elim_not_effective]: 1.486e-05
        [opt_reshape]: 6.33e-06
        [fold_const_symbol]: 9.94001e-06
        [renormalize]: 7.59988e-07
[validate]: 4.446e-05
Sums
    bootstrap                            :     0.000271s :  2.82%
    type_inference                       :     0.004664s : 48.49%
    event_method                         :     0.000021s :  0.22%
    auto_monad                           :     0.000064s :  0.66%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.38%
    rewriter_before_opt_a                :     0.000085s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000071s :  0.74%
    jit_opt_a.loop_unroll                :     0.000043s :  0.45%
    jit_opt_a.a_1                        :     0.000788s :  8.19%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000091s :  0.95%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.10%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000015s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.13%
    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_        :     0.000012s :  0.13%
    jit_opt_a.merge_forward              :     0.000008s :  0.09%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000023s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000783s :  8.14%
    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                        :     0.000052s :  0.54%
    jit_opt_a.replace_applicator         :     0.000027s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.16%
    rewriter_after_opt_a                 :     0.000050s :  0.52%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000741s :  7.70%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.22%
    cconv                                :     0.000027s :  0.28%
    loop_unroll                          :     0.000536s :  5.58%
    jit_opt_after_cconv.c_1              :     0.000028s :  0.29%
    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 :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000034s :  0.36%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.26%
    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      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.08%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000055s :  0.57%
    cse_after_recomputation.cse          :     0.000015s :  0.15%
    auto_monad_reorder                   :     0.000020s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000580s :  6.03%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  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.]   0.000231    28
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000002s :      2: substitution.fold_const_symbol
 2.80% :     0.000006s :      4: substitution.graph_param_transform
81.22% :     0.000187s :      4: substitution.inline
 1.71% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.58% :     0.000006s :      4: substitution.replace_old_param
 8.01% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004606     2
80.51% :     0.003708s :      1: type_inference.infer
19.49% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000095     8
55.13% :     0.000052s :      4: replace.inline
44.87% :     0.000043s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     8
91.88% :     0.000184s :      4: match.inline
 8.12% :     0.000016s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000149   912
 1.86% :     0.000003s :     13: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.03% :     0.000002s :     13: predicate.addn_check_dump
 1.36% :     0.000002s :     13: predicate.addn_zero_filter
 2.03% :     0.000003s :     13: predicate.arithmetic_simplify
 1.27% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.02% :     0.000002s :     13: predicate.compare_switch_simplify
 1.09% :     0.000002s :     13: predicate.depend_value_elim
 1.09% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.07% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.58% :     0.000001s :      4: predicate.elim_not_effective
 0.57% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.01% :     0.000002s :     13: predicate.environ_get_eliminate
 1.03% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 1.07% :     0.000002s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 5.53% :     0.000008s :     29: predicate.inline
 1.06% :     0.000002s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :      8: predicate.less_batch_normalization
 1.57% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     21: predicate.load_eliminater
 1.48% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.88% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000002s :     13: predicate.merge_addn
 1.14% :     0.000002s :     13: predicate.minmaximum_grad
 2.24% :     0.000003s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.57% :     0.000004s :     21: predicate.partial_eliminate
 1.28% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     13: predicate.reduce_eliminate
 1.52% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.18% :     0.000003s :     25: predicate.replace_applicator
 0.57% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000001s :      4: predicate.reset_defer_inline
 1.14% :     0.000002s :     13: predicate.reshape_eliminate
 1.08% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     13: predicate.same_eliminate
 0.73% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.13% :     0.000002s :      8: predicate.special_op_eliminate
 1.12% :     0.000002s :      8: predicate.specialize_transform
 1.26% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.56% :     0.000004s :     21: predicate.switch_defer_inline
 2.02% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.50% :     0.000011s :     67: predicate.switch_simplify
 1.14% :     0.000002s :     13: predicate.tile_eliminate
 1.28% :     0.000002s :     13: predicate.transpose_eliminate
 1.43% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.76% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000595    12
46.26% :     0.000275s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.67% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.07% :     0.000286s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015584    72
 0.37% :     0.000058s :      1: add_recomputation
 0.43% :     0.000067s :      1: auto_monad
 0.15% :     0.000023s :      1: auto_monad_reorder
 1.82% :     0.000284s :      1: bootstrap
 0.19% :     0.000030s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.19% :     0.000030s :      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
36.82% :     0.005738s :      1: jit_opt_a
 1.25% :     0.000194s :      1: jit_opt_after_cconv
 0.43% :     0.000067s :      1: jit_opt_b
 3.51% :     0.000547s :      1: loop_unroll
 4.88% :     0.000760s :      1: mutable_eliminate
 7.27% :     0.001132s :     26: opt.transform.jit_opt_a
 0.41% :     0.000064s :      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
 0.14% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.33% :     0.000052s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.79% :     0.000590s :      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% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000039s :      1: py_interpret_to_execute
 0.12% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.67% :     0.000416s :      1: renormalize.infer
 2.30% :     0.000358s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.34% :     0.000054s :      1: rewriter_after_opt_a
 0.57% :     0.000088s :      1: rewriter_before_opt_a
 0.53% :     0.000082s :      1: symbol_engine_optimizer
30.00% :     0.004675s :      1: type_inference


TotalTime = 0.0132446, [30]
[bootstrap]: 0.00028036
[type_inference]: 0.00472592
[event_method]: 2.131e-05
[auto_monad]: 6.562e-05
[graph_reusing]: 5.76e-06
[pre_auto_parallel]: 2.10002e-06
[py_interpret_to_execute]: 3.642e-05
[rewriter_before_opt_a]: 8.782e-05
[expand_dump_flag]: 2.77002e-06
[jit_opt_a]: 0.00571518, [2]
    [Cycle 1]: 0.00187074, [27]
        [switch_simplify]: 5.528e-05
        [loop_unroll]: 3.876e-05
        [a_1]: 0.00066686
        [with_stream_mark]: 1.756e-05
        [recompute_prepare]: 8.63001e-06
        [updatestate_depend_eliminate]: 3.76999e-06
        [updatestate_assign_eliminate]: 3.34001e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 7.49002e-06
        [updatestate_useless_node_eliminater]: 6.64001e-06
        [accelerated_algorithm]: 7.11001e-06
        [meta_shard_fg_expand]: 1.92999e-06
        [get_grad_eliminate_]: 6.54999e-06
        [merge_forward]: 4.37e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.373e-05
        [j_node_and_user_rematch]: 9.81e-06
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 1.193e-05
        [inline_without_move]: 6.63e-06
        [renormalize]: 0.0007593
        [add_forward_monad_depend]: 5.10999e-06
        [auto_monad_grad]: 2.63998e-06
        [auto_monad_eliminator]: 1.437e-05
        [cse]: 2.38e-05
        [replace_applicator]: 1.234e-05
    [Cycle 2]: 0.00039401, [27]
        [switch_simplify]: 6.49001e-06
        [loop_unroll]: 5.66003e-06
        [a_1]: 0.00011531
        [with_stream_mark]: 3.294e-05
        [recompute_prepare]: 7.83001e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 7.07002e-06
        [updatestate_useless_node_eliminater]: 6.10002e-06
        [accelerated_algorithm]: 5.97999e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 5.43002e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.409e-05
        [j_node_and_user_rematch]: 8.95999e-06
        [meta_fg_expand]: 1.86003e-06
        [replace_old_param]: 1.029e-05
        [inline_without_move]: 5.64e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 6.88998e-06
        [cse]: 1.546e-05
        [replace_applicator]: 5.89e-06
[py_interpret_to_execute_after_opt_a]: 1.885e-05
[rewriter_after_opt_a]: 3.914e-05
[convert_after_rewriter]: 7.03e-06
[order_py_execute_after_rewriter]: 5.94999e-06
[mutable_eliminate]: 0.00060063
[jit_opt_b]: 5.581e-05, [1]
    [Cycle 1]: 4.841e-05, [2]
        [frontend_op_eliminate]: 1.837e-05
        [inline_after_opt_a]: 1.871e-05
[cconv]: 2.81e-05
[loop_unroll]: 0.00046223
[jit_opt_after_cconv]: 0.00016477, [1]
    [Cycle 1]: 0.00015776, [11]
        [c_1]: 2.483e-05
        [parameter_eliminate]: 4.99e-06
        [updatestate_depend_eliminate]: 6.44999e-06
        [updatestate_assign_eliminate]: 3.23998e-06
        [updatestate_loads_eliminate]: 2.63998e-06
        [cse]: 2.728e-05
        [call_graph_tuple_transform]: 2.192e-05
        [tuple_list_get_item_eliminator]: 6.44999e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 6.52001e-06
[remove_dup_value]: 1.398e-05
[partial_unused_args_eliminate]: 2.36998e-06
[environ_conv]: 6.37001e-06
[add_recomputation]: 5.158e-05
[cse_after_recomputation]: 2.429e-05, [1]
    [Cycle 1]: 1.882e-05, [1]
        [cse]: 1.283e-05
[auto_monad_reorder]: 1.747e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 5.07e-06
[opt_after_jit_grad]: 0.00051325
[symbol_engine_optimizer]: 7.855e-05, [1]
    [Cycle 1]: 7.205e-05, [6]
        [build]: 3.45e-06
        [elim_shapecalc]: 9.18002e-06
        [elim_not_effective]: 1.479e-05
        [opt_reshape]: 6.34001e-06
        [fold_const_symbol]: 9.39e-06
        [renormalize]: 7.89994e-07
[validate]: 4.313e-05
Sums
    bootstrap                            :     0.000280s :  3.04%
    type_inference                       :     0.004726s : 51.25%
    event_method                         :     0.000021s :  0.23%
    auto_monad                           :     0.000066s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.39%
    rewriter_before_opt_a                :     0.000088s :  0.95%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000062s :  0.67%
    jit_opt_a.loop_unroll                :     0.000044s :  0.48%
    jit_opt_a.a_1                        :     0.000782s :  8.48%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.55%
    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 :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000015s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.14%
    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_        :     0.000012s :  0.13%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    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    :     0.000019s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000022s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000759s :  8.24%
    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                        :     0.000039s :  0.43%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.20%
    rewriter_after_opt_a                 :     0.000039s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000601s :  6.51%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.20%
    cconv                                :     0.000028s :  0.30%
    loop_unroll                          :     0.000462s :  5.01%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  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 :     0.000003s :  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 :  0.07%
    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 :  0.07%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000052s :  0.56%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000017s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000513s :  5.57%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    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                             :     0.000043s :  0.47%


Time group info:
------[substitution.]   0.000232    28
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.41% :     0.000006s :      4: substitution.graph_param_transform
83.68% :     0.000194s :      4: substitution.inline
 1.23% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.32% :     0.000005s :      4: substitution.replace_old_param
 6.96% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004667     2
80.20% :     0.003743s :      1: type_inference.infer
19.80% :     0.000924s :      1: type_inference.specialize
------[replace.]   0.000089     8
56.40% :     0.000050s :      4: replace.inline
43.60% :     0.000039s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     8
93.18% :     0.000191s :      4: match.inline
 6.82% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   912
 1.28% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     13: predicate.addn_check_dump
 1.18% :     0.000002s :     13: predicate.addn_zero_filter
 2.21% :     0.000003s :     13: predicate.arithmetic_simplify
 1.23% :     0.000002s :     13: predicate.cast_eliminate
 0.51% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.14% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.13% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      4: predicate.elim_not_effective
 0.46% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.35% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.07% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     13: predicate.environ_get_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.95% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.06% :     0.000007s :     29: predicate.inline
 1.01% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.07% :     0.000002s :      8: predicate.less_batch_normalization
 1.87% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.94% :     0.000003s :     21: predicate.load_eliminater
 1.42% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.52% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     13: predicate.merge_addn
 1.09% :     0.000002s :     13: predicate.minmaximum_grad
 1.68% :     0.000002s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.43% :     0.000003s :     21: predicate.partial_eliminate
 1.13% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     13: predicate.reduce_eliminate
 1.63% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.00% :     0.000003s :     25: predicate.replace_applicator
 0.79% :     0.000001s :      8: predicate.replace_old_param
 0.51% :     0.000001s :      4: predicate.reset_defer_inline
 1.26% :     0.000002s :     13: predicate.reshape_eliminate
 1.16% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     13: predicate.same_eliminate
 0.62% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      8: predicate.special_op_eliminate
 1.17% :     0.000002s :      8: predicate.specialize_transform
 1.40% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.64% :     0.000004s :     21: predicate.switch_defer_inline
 2.15% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.56% :     0.000011s :     67: predicate.switch_simplify
 1.14% :     0.000002s :     13: predicate.tile_eliminate
 1.26% :     0.000002s :     13: predicate.transpose_eliminate
 1.35% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.61% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 2.20% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     13: predicate.value_based_eliminate
 0.35% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000611    12
46.83% :     0.000286s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.64% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.54% :     0.000284s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015117    72
 0.36% :     0.000055s :      1: add_recomputation
 0.46% :     0.000069s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 1.94% :     0.000293s :      1: bootstrap
 0.20% :     0.000031s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000026s :      1: cse_after_recomputation
 0.06% :     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
37.83% :     0.005719s :      1: jit_opt_a
 1.11% :     0.000168s :      1: jit_opt_after_cconv
 0.39% :     0.000058s :      1: jit_opt_b
 3.11% :     0.000470s :      1: loop_unroll
 4.03% :     0.000610s :      1: mutable_eliminate
 6.82% :     0.001031s :     26: opt.transform.jit_opt_a
 0.37% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000031s :      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
 0.18% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.46% :     0.000522s :      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% :     0.000039s :      1: py_interpret_to_execute
 0.14% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 2.79% :     0.000422s :      1: renormalize.infer
 2.18% :     0.000330s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000042s :      1: rewriter_after_opt_a
 0.60% :     0.000091s :      1: rewriter_before_opt_a
 0.54% :     0.000081s :      1: symbol_engine_optimizer
31.33% :     0.004736s :      1: type_inference


TotalTime = 0.0416761, [33]
[bootstrap]: 0.00028165
[type_inference]: 0.0219108
[event_method]: 2.054e-05
[auto_monad]: 0.00015425
[graph_reusing]: 6.68e-06
[pre_auto_parallel]: 3.55e-06
[py_interpret_to_execute]: 0.00018592
[rewriter_before_opt_a]: 9.909e-05
[expand_dump_flag]: 3.21999e-06
[jit_opt_a]: 0.00616162, [2]
    [Cycle 1]: 0.00238445, [27]
        [switch_simplify]: 6.386e-05
        [loop_unroll]: 3.62e-05
        [a_1]: 0.0008027
        [with_stream_mark]: 1.811e-05
        [recompute_prepare]: 1.299e-05
        [updatestate_depend_eliminate]: 6.63e-06
        [updatestate_assign_eliminate]: 6.28e-06
        [updatestate_loads_eliminate]: 5.51e-06
        [parameter_eliminate]: 2.41e-06
        [specialize_transform]: 1.027e-05
        [updatestate_useless_node_eliminater]: 1.328e-05
        [accelerated_algorithm]: 9.51998e-06
        [meta_shard_fg_expand]: 3.33e-06
        [get_grad_eliminate_]: 9.47001e-06
        [merge_forward]: 5.61e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.111e-05
        [j_node_and_user_rematch]: 1.559e-05
        [meta_fg_expand]: 3.86001e-06
        [replace_old_param]: 1.231e-05
        [inline_without_move]: 9.20999e-06
        [renormalize]: 0.00104629
        [add_forward_monad_depend]: 5.44e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 2.017e-05
        [cse]: 3.854e-05
        [replace_applicator]: 1.516e-05
    [Cycle 2]: 0.00048417, [27]
        [switch_simplify]: 9.22999e-06
        [loop_unroll]: 8.69e-06
        [a_1]: 0.0001781
        [with_stream_mark]: 1.092e-05
        [recompute_prepare]: 8.28999e-06
        [updatestate_depend_eliminate]: 4.89003e-06
        [updatestate_assign_eliminate]: 4.62998e-06
        [updatestate_loads_eliminate]: 4.13001e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 8.67e-06
        [updatestate_useless_node_eliminater]: 1.196e-05
        [accelerated_algorithm]: 8.90001e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 8.26002e-06
        [merge_forward]: 4.84998e-06
        [cell_reuse_recompute_pass]: 2.01e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.599e-05
        [j_node_and_user_rematch]: 1.23e-05
        [meta_fg_expand]: 3.28e-06
        [replace_old_param]: 9.79e-06
        [inline_without_move]: 7.36999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 9.40025e-07
        [auto_monad_grad]: 1.32999e-06
        [auto_monad_eliminator]: 9.83998e-06
        [cse]: 2.035e-05
        [replace_applicator]: 7.61999e-06
[py_interpret_to_execute_after_opt_a]: 1.099e-05
[rewriter_after_opt_a]: 0.00029478
[convert_after_rewriter]: 1.03e-05
[order_py_execute_after_rewriter]: 8.32998e-06
[mutable_eliminate]: 0.0005628
[jit_opt_b]: 6.895e-05, [1]
    [Cycle 1]: 6.215e-05, [2]
        [frontend_op_eliminate]: 2.591e-05
        [inline_after_opt_a]: 2.42e-05
[cconv]: 2.602e-05
[loop_unroll]: 0.00046298
[jit_opt_after_cconv]: 0.00020779, [1]
    [Cycle 1]: 0.00020066, [11]
        [c_1]: 4.429e-05
        [parameter_eliminate]: 2.75002e-06
        [updatestate_depend_eliminate]: 8.05e-06
        [updatestate_assign_eliminate]: 4.33001e-06
        [updatestate_loads_eliminate]: 3.95e-06
        [cse]: 4.04e-05
        [call_graph_tuple_transform]: 2.299e-05
        [tuple_list_get_item_eliminator]: 8.66002e-06
        [none_parameter_eliminate]: 1.80001e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 9.28002e-06
[remove_dup_value]: 1.981e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 1.046e-05
[add_recomputation]: 6.66e-05
[cse_after_recomputation]: 3.015e-05, [1]
    [Cycle 1]: 2.384e-05, [1]
        [cse]: 1.796e-05
[auto_monad_reorder]: 2.57e-05
[get_jit_bprop_graph]: 1.48002e-06
[rewriter_after_jit_bprop_graph]: 5.46e-06
[opt_after_jit_grad]: 0.00051621
[symbol_engine_optimizer]: 0.00010469, [1]
    [Cycle 1]: 9.734e-05, [6]
        [build]: 9.97999e-06
        [elim_shapecalc]: 1.195e-05
        [elim_not_effective]: 2.123e-05
        [opt_reshape]: 9.15001e-06
        [fold_const_symbol]: 1.414e-05
        [renormalize]: 6.39993e-07
[validate]: 5.704e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.0101229
[execute]: 7.81001e-06
Sums
    bootstrap                            :     0.000282s :  0.75%
    type_inference                       :     0.021911s : 58.08%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000154s :  0.41%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000186s :  0.49%
    rewriter_before_opt_a                :     0.000099s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000981s :  2.60%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000018s :  0.05%
    jit_opt_a.merge_forward              :     0.000010s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001046s :  2.77%
    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      :     0.000030s :  0.08%
    jit_opt_a.cse                        :     0.000059s :  0.16%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000295s :  0.78%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000563s :  1.49%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000463s :  1.23%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.12%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000040s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000067s :  0.18%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000516s :  1.37%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000057s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010123s : 26.83%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000329    47
 2.57% :     0.000008s :      2: substitution.depend_value_elim
 0.92% :     0.000003s :      4: substitution.elim_not_effective
 0.63% :     0.000002s :      4: substitution.fold_const_symbol
 2.01% :     0.000007s :      5: substitution.graph_param_transform
79.38% :     0.000261s :      4: substitution.inline
 1.42% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.13% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.11% :     0.000004s :      2: substitution.replace_old_param
 4.18% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 2.84% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.81% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021843     2
93.52% :     0.020428s :      1: type_inference.infer
 6.48% :     0.001415s :      1: type_inference.specialize
------[replace.]   0.000100     6
65.13% :     0.000065s :      4: replace.inline
34.87% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000271     6
95.33% :     0.000258s :      4: match.inline
 4.67% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1021
 1.50% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.12% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.17% :     0.000002s :     15: predicate.addn_check_dump
 1.38% :     0.000002s :     15: predicate.addn_zero_filter
 2.10% :     0.000004s :     15: predicate.arithmetic_simplify
 1.36% :     0.000002s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     15: predicate.compare_switch_simplify
 1.58% :     0.000003s :     15: predicate.depend_value_elim
 1.18% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.91% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      5: predicate.elim_not_effective
 0.63% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.34% :     0.000002s :     10: predicate.get_grad_eliminate
 0.30% :     0.000001s :      5: predicate.graph_param_transform
 4.72% :     0.000008s :     31: predicate.inline
 1.09% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.50% :     0.000003s :     10: predicate.less_batch_normalization
 1.41% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.94% :     0.000003s :     22: predicate.load_eliminater
 1.23% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.44% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.04% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.23% :     0.000002s :      5: predicate.mutable_eliminate
 0.47% :     0.000001s :      5: predicate.opt_reshape
 2.38% :     0.000004s :     22: predicate.partial_eliminate
 1.43% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     15: predicate.reduce_eliminate
 1.65% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.72% :     0.000003s :     27: predicate.replace_applicator
 0.64% :     0.000001s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.26% :     0.000002s :     15: predicate.reshape_eliminate
 1.30% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.99% :     0.000002s :     10: predicate.special_op_eliminate
 1.14% :     0.000002s :     10: predicate.specialize_transform
 1.41% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.17% :     0.000004s :     21: predicate.switch_defer_inline
 2.04% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.42% :     0.000011s :     68: predicate.switch_simplify
 1.19% :     0.000002s :     15: predicate.tile_eliminate
 1.37% :     0.000002s :     15: predicate.transpose_eliminate
 1.50% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.79% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.49% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000892    13
42.43% :     0.000379s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.15% :     0.000108s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.43% :     0.000405s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044211    76
 0.16% :     0.000069s :      1: add_recomputation
 0.36% :     0.000159s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.67% :     0.000296s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.06% :     0.000025s :      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
13.94% :     0.006165s :      1: jit_opt_a
 0.48% :     0.000211s :      1: jit_opt_after_cconv
 0.16% :     0.000072s :      1: jit_opt_b
 1.07% :     0.000471s :      1: loop_unroll
 1.29% :     0.000572s :      1: mutable_eliminate
 2.93% :     0.001297s :     26: opt.transform.jit_opt_a
 0.18% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000044s :      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
 0.12% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000527s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.43% :     0.000191s :      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
 1.34% :     0.000593s :      1: renormalize.infer
 1.01% :     0.000444s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.68% :     0.000299s :      1: rewriter_after_opt_a
 0.23% :     0.000103s :      1: rewriter_before_opt_a
 0.24% :     0.000108s :      1: symbol_engine_optimizer
22.93% :     0.010137s :      1: task_emit
49.59% :     0.021926s :      1: type_inference
 0.19% :     0.000083s :      1: validate


TotalTime = 0.0165484, [30]
[bootstrap]: 0.00026834
[type_inference]: 0.00467351
[event_method]: 2.32e-05
[auto_monad]: 6.749e-05
[graph_reusing]: 5.52999e-06
[pre_auto_parallel]: 2.36998e-06
[py_interpret_to_execute]: 3.722e-05
[rewriter_before_opt_a]: 8.831e-05
[expand_dump_flag]: 3.21999e-06
[jit_opt_a]: 0.00926857, [2]
    [Cycle 1]: 0.0018024, [27]
        [switch_simplify]: 6.228e-05
        [loop_unroll]: 3.596e-05
        [a_1]: 0.00064519
        [with_stream_mark]: 1.999e-05
        [recompute_prepare]: 8.60999e-06
        [updatestate_depend_eliminate]: 4.03999e-06
        [updatestate_assign_eliminate]: 3.53e-06
        [updatestate_loads_eliminate]: 2.81e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 7.35e-06
        [updatestate_useless_node_eliminater]: 6.68e-06
        [accelerated_algorithm]: 6.13002e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 5.94999e-06
        [merge_forward]: 3.91001e-06
        [cell_reuse_recompute_pass]: 1.01002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.51e-05
        [j_node_and_user_rematch]: 9.89001e-06
        [meta_fg_expand]: 2.45002e-06
        [replace_old_param]: 1.115e-05
        [inline_without_move]: 6.02999e-06
        [renormalize]: 0.00070141
        [add_forward_monad_depend]: 5.64e-06
        [auto_monad_grad]: 2.16e-06
        [auto_monad_eliminator]: 1.521e-05
        [cse]: 3.313e-05
        [replace_applicator]: 1.27e-05
    [Cycle 2]: 0.00037198, [27]
        [switch_simplify]: 6.73e-06
        [loop_unroll]: 5.72999e-06
        [a_1]: 0.00012373
        [with_stream_mark]: 9.59e-06
        [recompute_prepare]: 6.36e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 5.63002e-06
        [updatestate_useless_node_eliminater]: 5.97001e-06
        [accelerated_algorithm]: 6.28e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 5.38002e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.362e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 9.46e-06
        [inline_without_move]: 5.48002e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 6.39999e-06
        [cse]: 1.406e-05
        [replace_applicator]: 5.87999e-06
[py_interpret_to_execute_after_opt_a]: 9.75002e-06
[rewriter_after_opt_a]: 3.525e-05
[convert_after_rewriter]: 7.2e-06
[order_py_execute_after_rewriter]: 5.10999e-06
[mutable_eliminate]: 0.00051561
[jit_opt_b]: 5.432e-05, [1]
    [Cycle 1]: 4.739e-05, [2]
        [frontend_op_eliminate]: 1.932e-05
        [inline_after_opt_a]: 1.699e-05
[cconv]: 2.303e-05
[loop_unroll]: 0.00045078
[jit_opt_after_cconv]: 0.000154, [1]
    [Cycle 1]: 0.00014764, [11]
        [c_1]: 2.523e-05
        [parameter_eliminate]: 2.21e-06
        [updatestate_depend_eliminate]: 5.35999e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.024e-05
        [call_graph_tuple_transform]: 2.037e-05
        [tuple_list_get_item_eliminator]: 6.69001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 3.59985e-07
        [switch_simplify]: 6.64999e-06
[remove_dup_value]: 1.518e-05
[partial_unused_args_eliminate]: 2.29999e-06
[environ_conv]: 6.18998e-06
[add_recomputation]: 4.749e-05
[cse_after_recomputation]: 2.356e-05, [1]
    [Cycle 1]: 1.803e-05, [1]
        [cse]: 1.209e-05
[auto_monad_reorder]: 1.723e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 4.23001e-06
[opt_after_jit_grad]: 0.00046719
[symbol_engine_optimizer]: 7.819e-05, [1]
    [Cycle 1]: 7.215e-05, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 9.14e-06
        [elim_not_effective]: 1.408e-05
        [opt_reshape]: 6.61999e-06
        [fold_const_symbol]: 1.011e-05
        [renormalize]: 4.00003e-07
[validate]: 3.553e-05
Sums
    bootstrap                            :     0.000268s :  3.02%
    type_inference                       :     0.004674s : 52.57%
    event_method                         :     0.000023s :  0.26%
    auto_monad                           :     0.000067s :  0.76%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.42%
    rewriter_before_opt_a                :     0.000088s :  0.99%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000069s :  0.78%
    jit_opt_a.loop_unroll                :     0.000042s :  0.47%
    jit_opt_a.a_1                        :     0.000769s :  8.65%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.33%
    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 :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.14%
    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_        :     0.000011s :  0.13%
    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 :     0.000029s :  0.32%
    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          :     0.000021s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000702s :  7.89%
    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      :     0.000022s :  0.24%
    jit_opt_a.cse                        :     0.000047s :  0.53%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000035s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000516s :  5.80%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000023s :  0.26%
    loop_unroll                          :     0.000451s :  5.07%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.28%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.23%
    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  :     0.000007s :  0.07%
    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 :  0.53%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000017s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000467s :  5.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    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 :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.40%


Time group info:
------[substitution.]   0.000222    28
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000006s :      4: substitution.graph_param_transform
82.35% :     0.000183s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.49% :     0.000006s :      4: substitution.replace_old_param
 7.69% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004612     2
80.16% :     0.003697s :      1: type_inference.infer
19.84% :     0.000915s :      1: type_inference.specialize
------[replace.]   0.000090     8
57.10% :     0.000051s :      4: replace.inline
42.90% :     0.000039s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000195     8
92.23% :     0.000180s :      4: match.inline
 7.77% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   912
 1.20% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     13: predicate.addn_check_dump
 1.17% :     0.000002s :     13: predicate.addn_zero_filter
 1.85% :     0.000003s :     13: predicate.arithmetic_simplify
 1.28% :     0.000002s :     13: predicate.cast_eliminate
 0.45% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     13: predicate.compare_switch_simplify
 1.21% :     0.000002s :     13: predicate.depend_value_elim
 1.14% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.17% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.57% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     13: predicate.environ_get_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      4: predicate.fold_const_symbol
 0.97% :     0.000001s :      8: predicate.get_grad_eliminate
 0.30% :     0.000000s :      4: predicate.graph_param_transform
 5.52% :     0.000008s :     29: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      8: predicate.less_batch_normalization
 1.56% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.37% :     0.000003s :     21: predicate.load_eliminater
 1.25% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.27% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     13: predicate.merge_addn
 1.14% :     0.000002s :     13: predicate.minmaximum_grad
 1.14% :     0.000002s :      4: predicate.mutable_eliminate
 0.51% :     0.000001s :      4: predicate.opt_reshape
 2.62% :     0.000004s :     21: predicate.partial_eliminate
 1.17% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     13: predicate.reduce_eliminate
 1.63% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.92% :     0.000003s :     25: predicate.replace_applicator
 0.71% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.19% :     0.000002s :      8: predicate.special_op_eliminate
 1.00% :     0.000001s :      8: predicate.specialize_transform
 1.42% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     21: predicate.switch_defer_inline
 2.22% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.70% :     0.000011s :     67: predicate.switch_simplify
 1.22% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.63% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     13: predicate.value_based_eliminate
 0.31% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000580    12
46.75% :     0.000271s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.31% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.94% :     0.000278s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018339    72
 0.28% :     0.000050s :      1: add_recomputation
 0.39% :     0.000071s :      1: auto_monad
 0.11% :     0.000020s :      1: auto_monad_reorder
 1.52% :     0.000279s :      1: bootstrap
 0.14% :     0.000026s :      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% :     0.000027s :      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
50.55% :     0.009271s :      1: jit_opt_a
 0.85% :     0.000157s :      1: jit_opt_after_cconv
 0.31% :     0.000057s :      1: jit_opt_b
 2.50% :     0.000458s :      1: loop_unroll
 2.85% :     0.000523s :      1: mutable_eliminate
 5.53% :     0.001014s :     26: opt.transform.jit_opt_a
 0.30% :     0.000056s :      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% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.59% :     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
 0.22% :     0.000040s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.02% :     0.000371s :      1: renormalize.infer
 1.75% :     0.000321s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000038s :      1: rewriter_after_opt_a
 0.50% :     0.000091s :      1: rewriter_before_opt_a
 0.44% :     0.000081s :      1: symbol_engine_optimizer
25.55% :     0.004685s :      1: type_inference


TotalTime = 0.0137458, [30]
[bootstrap]: 0.00028761
[type_inference]: 0.00484606
[event_method]: 2.316e-05
[auto_monad]: 6.693e-05
[graph_reusing]: 5.92999e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 4.026e-05
[rewriter_before_opt_a]: 9.129e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00607263, [2]
    [Cycle 1]: 0.00203215, [27]
        [switch_simplify]: 6.803e-05
        [loop_unroll]: 4.096e-05
        [a_1]: 0.00075124
        [with_stream_mark]: 1.995e-05
        [recompute_prepare]: 9.04e-06
        [updatestate_depend_eliminate]: 3.91999e-06
        [updatestate_assign_eliminate]: 3.70998e-06
        [updatestate_loads_eliminate]: 3.32002e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 8.05e-06
        [updatestate_useless_node_eliminater]: 6.43e-06
        [accelerated_algorithm]: 6.99001e-06
        [meta_shard_fg_expand]: 2.69999e-06
        [get_grad_eliminate_]: 6.46e-06
        [merge_forward]: 4.33001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.633e-05
        [j_node_and_user_rematch]: 1.058e-05
        [meta_fg_expand]: 2.96001e-06
        [replace_old_param]: 1.209e-05
        [inline_without_move]: 6.44001e-06
        [renormalize]: 0.00078353
        [add_forward_monad_depend]: 6.31998e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.502e-05
        [cse]: 3.342e-05
        [replace_applicator]: 1.454e-05
    [Cycle 2]: 0.00039588, [27]
        [switch_simplify]: 7.30998e-06
        [loop_unroll]: 6.44999e-06
        [a_1]: 0.00012306
        [with_stream_mark]: 1.018e-05
        [recompute_prepare]: 7.08e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.28002e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 6.44999e-06
        [updatestate_useless_node_eliminater]: 5.97999e-06
        [accelerated_algorithm]: 6.41998e-06
        [meta_shard_fg_expand]: 1.31998e-06
        [get_grad_eliminate_]: 5.64998e-06
        [merge_forward]: 3.11999e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.355e-05
        [j_node_and_user_rematch]: 9.31998e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 9.69999e-06
        [inline_without_move]: 6.01e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.43002e-06
        [auto_monad_grad]: 1.19e-06
        [auto_monad_eliminator]: 7.23e-06
        [cse]: 1.563e-05
        [replace_applicator]: 7.23999e-06
[py_interpret_to_execute_after_opt_a]: 1.254e-05
[rewriter_after_opt_a]: 4.062e-05
[convert_after_rewriter]: 7.16999e-06
[order_py_execute_after_rewriter]: 9.36998e-06
[mutable_eliminate]: 0.00056454
[jit_opt_b]: 5.973e-05, [1]
    [Cycle 1]: 5.236e-05, [2]
        [frontend_op_eliminate]: 2.053e-05
        [inline_after_opt_a]: 1.921e-05
[cconv]: 2.447e-05
[loop_unroll]: 0.00045816
[jit_opt_after_cconv]: 0.0001658, [1]
    [Cycle 1]: 0.00015902, [11]
        [c_1]: 2.838e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.77999e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [cse]: 2.105e-05
        [call_graph_tuple_transform]: 2.23e-05
        [tuple_list_get_item_eliminator]: 6.88e-06
        [none_parameter_eliminate]: 1.75001e-06
        [renormalize]: 2.9002e-07
        [switch_simplify]: 6.93e-06
[remove_dup_value]: 1.52e-05
[partial_unused_args_eliminate]: 2.77002e-06
[environ_conv]: 7.39002e-06
[add_recomputation]: 5.078e-05
[cse_after_recomputation]: 2.558e-05, [1]
    [Cycle 1]: 1.983e-05, [1]
        [cse]: 1.341e-05
[auto_monad_reorder]: 1.972e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 5.89e-06
[opt_after_jit_grad]: 0.00053768
[symbol_engine_optimizer]: 8.53e-05, [1]
    [Cycle 1]: 7.848e-05, [6]
        [build]: 4.84e-06
        [elim_shapecalc]: 1.017e-05
        [elim_not_effective]: 1.507e-05
        [opt_reshape]: 6.91001e-06
        [fold_const_symbol]: 1.024e-05
        [renormalize]: 7.2e-07
[validate]: 3.984e-05
Sums
    bootstrap                            :     0.000288s :  3.03%
    type_inference                       :     0.004846s : 51.13%
    event_method                         :     0.000023s :  0.24%
    auto_monad                           :     0.000067s :  0.71%
    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                :     0.000091s :  0.96%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000075s :  0.79%
    jit_opt_a.loop_unroll                :     0.000047s :  0.50%
    jit_opt_a.a_1                        :     0.000874s :  9.23%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.17%
    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 :     0.000006s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000014s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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_        :     0.000012s :  0.13%
    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 :     0.000030s :  0.32%
    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          :     0.000022s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000784s :  8.27%
    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 :  0.23%
    jit_opt_a.cse                        :     0.000049s :  0.52%
    jit_opt_a.replace_applicator         :     0.000022s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.13%
    rewriter_after_opt_a                 :     0.000041s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000009s :  0.10%
    mutable_eliminate                    :     0.000565s :  5.96%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.20%
    cconv                                :     0.000024s :  0.26%
    loop_unroll                          :     0.000458s :  4.83%
    jit_opt_after_cconv.c_1              :     0.000028s :  0.30%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.24%
    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      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.07%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.08%
    add_recomputation                    :     0.000051s :  0.54%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000020s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000538s :  5.67%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.11%
    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 :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.42%


Time group info:
------[substitution.]   0.000234    28
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 2.68% :     0.000006s :      4: substitution.graph_param_transform
82.33% :     0.000192s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.99% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000005s :      4: substitution.replace_old_param
 8.03% :     0.000019s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004779     2
80.06% :     0.003826s :      1: type_inference.infer
19.94% :     0.000953s :      1: type_inference.specialize
------[replace.]   0.000094     8
56.17% :     0.000053s :      4: replace.inline
43.83% :     0.000041s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000206     8
91.94% :     0.000189s :      4: match.inline
 8.06% :     0.000017s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000149   912
 1.30% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.10% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     13: predicate.addn_check_dump
 1.29% :     0.000002s :     13: predicate.addn_zero_filter
 2.14% :     0.000003s :     13: predicate.arithmetic_simplify
 1.20% :     0.000002s :     13: predicate.cast_eliminate
 0.58% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     13: predicate.compare_switch_simplify
 1.13% :     0.000002s :     13: predicate.depend_value_elim
 1.10% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.09% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      4: predicate.elim_not_effective
 0.64% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.09% :     0.000002s :     13: predicate.environ_get_eliminate
 1.03% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.94% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.46% :     0.000008s :     29: predicate.inline
 0.90% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :      8: predicate.less_batch_normalization
 1.76% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.05% :     0.000003s :     21: predicate.load_eliminater
 1.41% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.39% :     0.000007s :     42: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     13: predicate.merge_addn
 1.21% :     0.000002s :     13: predicate.minmaximum_grad
 1.45% :     0.000002s :      4: predicate.mutable_eliminate
 0.52% :     0.000001s :      4: predicate.opt_reshape
 2.48% :     0.000004s :     21: predicate.partial_eliminate
 1.26% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.74% :     0.000003s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.70% :     0.000001s :      8: predicate.replace_old_param
 0.32% :     0.000000s :      4: predicate.reset_defer_inline
 1.24% :     0.000002s :     13: predicate.reshape_eliminate
 1.17% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     13: predicate.same_eliminate
 0.66% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.13% :     0.000002s :      8: predicate.special_op_eliminate
 0.93% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.52% :     0.000004s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.92% :     0.000012s :     67: predicate.switch_simplify
 1.23% :     0.000002s :     13: predicate.tile_eliminate
 1.19% :     0.000002s :     13: predicate.transpose_eliminate
 1.50% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000617    12
48.22% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.68% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.11% :     0.000284s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015755    72
 0.34% :     0.000054s :      1: add_recomputation
 0.45% :     0.000070s :      1: auto_monad
 0.14% :     0.000022s :      1: auto_monad_reorder
 1.90% :     0.000300s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000028s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.17% :     0.000027s :      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
38.57% :     0.006076s :      1: jit_opt_a
 1.07% :     0.000169s :      1: jit_opt_after_cconv
 0.40% :     0.000063s :      1: jit_opt_b
 2.96% :     0.000466s :      1: loop_unroll
 3.64% :     0.000573s :      1: mutable_eliminate
 7.25% :     0.001142s :     26: opt.transform.jit_opt_a
 0.39% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000033s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.47% :     0.000547s :      1: opt_after_jit_grad
 0.07% :     0.000012s :      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% :     0.000043s :      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
 2.70% :     0.000425s :      1: renormalize.infer
 2.22% :     0.000350s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000044s :      1: rewriter_after_opt_a
 0.60% :     0.000094s :      1: rewriter_before_opt_a
 0.56% :     0.000088s :      1: symbol_engine_optimizer
30.85% :     0.004860s :      1: type_inference


TotalTime = 0.0437327, [33]
[bootstrap]: 0.00025885
[type_inference]: 0.0220478
[event_method]: 2.076e-05
[auto_monad]: 0.00016066
[graph_reusing]: 6.65998e-06
[pre_auto_parallel]: 2.96999e-06
[py_interpret_to_execute]: 0.00013687
[rewriter_before_opt_a]: 9.941e-05
[expand_dump_flag]: 3.79002e-06
[jit_opt_a]: 0.00677991, [2]
    [Cycle 1]: 0.00253029, [27]
        [switch_simplify]: 6.713e-05
        [loop_unroll]: 3.607e-05
        [a_1]: 0.00073602
        [with_stream_mark]: 1.943e-05
        [recompute_prepare]: 1.116e-05
        [updatestate_depend_eliminate]: 6.33002e-06
        [updatestate_assign_eliminate]: 6.68e-06
        [updatestate_loads_eliminate]: 5.32001e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 9.36002e-06
        [updatestate_useless_node_eliminater]: 1.175e-05
        [accelerated_algorithm]: 9.15999e-06
        [meta_shard_fg_expand]: 3.34001e-06
        [get_grad_eliminate_]: 8.52998e-06
        [merge_forward]: 5.61e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.041e-05
        [j_node_and_user_rematch]: 1.418e-05
        [meta_fg_expand]: 3.57997e-06
        [replace_old_param]: 1.244e-05
        [inline_without_move]: 8.38999e-06
        [renormalize]: 0.00125431
        [add_forward_monad_depend]: 6.44001e-06
        [auto_monad_grad]: 3.45e-06
        [auto_monad_eliminator]: 2.279e-05
        [cse]: 4.025e-05
        [replace_applicator]: 1.763e-05
    [Cycle 2]: 0.00048728, [27]
        [switch_simplify]: 1.013e-05
        [loop_unroll]: 8.52998e-06
        [a_1]: 0.00017697
        [with_stream_mark]: 1.312e-05
        [recompute_prepare]: 8.15e-06
        [updatestate_depend_eliminate]: 4.95999e-06
        [updatestate_assign_eliminate]: 4.72e-06
        [updatestate_loads_eliminate]: 4.20999e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 8.17e-06
        [updatestate_useless_node_eliminater]: 1.116e-05
        [accelerated_algorithm]: 8.15999e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 7.97e-06
        [merge_forward]: 4.92e-06
        [cell_reuse_recompute_pass]: 1.92999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.734e-05
        [j_node_and_user_rematch]: 1.319e-05
        [meta_fg_expand]: 3.28e-06
        [replace_old_param]: 1.091e-05
        [inline_without_move]: 7.63999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.00001e-06
        [auto_monad_grad]: 1.01002e-06
        [auto_monad_eliminator]: 1.043e-05
        [cse]: 2.165e-05
        [replace_applicator]: 7.61999e-06
[py_interpret_to_execute_after_opt_a]: 1.567e-05
[rewriter_after_opt_a]: 0.00031144
[convert_after_rewriter]: 1.243e-05
[order_py_execute_after_rewriter]: 8.40999e-06
[mutable_eliminate]: 0.00067671
[jit_opt_b]: 7.449e-05, [1]
    [Cycle 1]: 6.534e-05, [2]
        [frontend_op_eliminate]: 2.63e-05
        [inline_after_opt_a]: 2.654e-05
[cconv]: 3.06e-05
[loop_unroll]: 0.00048359
[jit_opt_after_cconv]: 0.00020576, [1]
    [Cycle 1]: 0.00019884, [11]
        [c_1]: 4.811e-05
        [parameter_eliminate]: 3.02002e-06
        [updatestate_depend_eliminate]: 9.07999e-06
        [updatestate_assign_eliminate]: 4.64002e-06
        [updatestate_loads_eliminate]: 4.33999e-06
        [cse]: 3.401e-05
        [call_graph_tuple_transform]: 2.331e-05
        [tuple_list_get_item_eliminator]: 8.77e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 9.00999e-06
[remove_dup_value]: 2.116e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 1.147e-05
[add_recomputation]: 6.929e-05
[cse_after_recomputation]: 2.994e-05, [1]
    [Cycle 1]: 2.399e-05, [1]
        [cse]: 1.795e-05
[auto_monad_reorder]: 2.635e-05
[get_jit_bprop_graph]: 1.52999e-06
[rewriter_after_jit_bprop_graph]: 4.92e-06
[opt_after_jit_grad]: 0.00052566
[symbol_engine_optimizer]: 0.00010518, [1]
    [Cycle 1]: 9.872e-05, [6]
        [build]: 1.36e-05
        [elim_shapecalc]: 1.251e-05
        [elim_not_effective]: 1.925e-05
        [opt_reshape]: 1.001e-05
        [fold_const_symbol]: 1.377e-05
        [renormalize]: 4.00003e-07
[validate]: 5.943e-05
[backend_pass]: 1.27e-06
[task_emit]: 0.011293
[execute]: 1.061e-05
Sums
    bootstrap                            :     0.000259s :  0.66%
    type_inference                       :     0.022048s : 56.09%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000161s :  0.41%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000137s :  0.35%
    rewriter_before_opt_a                :     0.000099s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.20%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000913s :  2.32%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    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 :     0.000038s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001254s :  3.19%
    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      :     0.000033s :  0.08%
    jit_opt_a.cse                        :     0.000062s :  0.16%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.04%
    rewriter_after_opt_a                 :     0.000311s :  0.79%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000677s :  1.72%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.07%
    cconv                                :     0.000031s :  0.08%
    loop_unroll                          :     0.000484s :  1.23%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000034s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000069s :  0.18%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000526s :  1.34%
    symbol_engine_optimizer.build        :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000059s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011293s : 28.73%
    execute                              :     0.000011s :  0.03%


Time group info:
------[substitution.]   0.000295    47
 3.45% :     0.000010s :      2: substitution.depend_value_elim
 0.96% :     0.000003s :      4: substitution.elim_not_effective
 0.70% :     0.000002s :      4: substitution.fold_const_symbol
 2.38% :     0.000007s :      5: substitution.graph_param_transform
75.32% :     0.000222s :      4: substitution.inline
 1.73% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.81% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.72% :     0.000005s :      2: substitution.replace_old_param
 4.61% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.28% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 3.04% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021974     2
92.95% :     0.020425s :      1: type_inference.infer
 7.05% :     0.001549s :      1: type_inference.specialize
------[replace.]   0.000093     6
65.97% :     0.000062s :      4: replace.inline
34.03% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000232     6
94.59% :     0.000219s :      4: match.inline
 5.41% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000179  1021
 1.17% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.37% :     0.000002s :     15: predicate.addn_zero_filter
 2.12% :     0.000004s :     15: predicate.arithmetic_simplify
 1.33% :     0.000002s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.38% :     0.000002s :     15: predicate.compare_switch_simplify
 1.29% :     0.000002s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.97% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      5: predicate.elim_not_effective
 0.61% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     15: predicate.environ_get_eliminate
 1.06% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.30% :     0.000001s :      5: predicate.fold_const_symbol
 1.11% :     0.000002s :     10: predicate.get_grad_eliminate
 0.44% :     0.000001s :      5: predicate.graph_param_transform
 5.10% :     0.000009s :     31: predicate.inline
 1.04% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.75% :     0.000003s :     10: predicate.less_batch_normalization
 1.37% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     22: predicate.load_eliminater
 1.35% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.33% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.06% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     15: predicate.merge_addn
 1.17% :     0.000002s :     15: predicate.minmaximum_grad
 1.45% :     0.000003s :      5: predicate.mutable_eliminate
 0.51% :     0.000001s :      5: predicate.opt_reshape
 2.37% :     0.000004s :     22: predicate.partial_eliminate
 1.32% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     15: predicate.reduce_eliminate
 1.51% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.64% :     0.000003s :     27: predicate.replace_applicator
 0.76% :     0.000001s :     10: predicate.replace_old_param
 0.37% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000002s :     15: predicate.reshape_eliminate
 1.19% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.00% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     15: predicate.same_eliminate
 0.54% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.98% :     0.000002s :     10: predicate.special_op_eliminate
 1.06% :     0.000002s :     10: predicate.specialize_transform
 1.44% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.08% :     0.000004s :     21: predicate.switch_defer_inline
 1.89% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.72% :     0.000012s :     68: predicate.switch_simplify
 1.41% :     0.000003s :     15: predicate.tile_eliminate
 1.42% :     0.000003s :     15: predicate.transpose_eliminate
 1.78% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000861    13
41.27% :     0.000355s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.61% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
53.12% :     0.000457s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046408    76
 0.16% :     0.000072s :      1: add_recomputation
 0.36% :     0.000166s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.59% :     0.000272s :      1: bootstrap
 0.07% :     0.000033s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000025s :      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
14.62% :     0.006784s :      1: jit_opt_a
 0.45% :     0.000210s :      1: jit_opt_after_cconv
 0.17% :     0.000077s :      1: jit_opt_b
 1.06% :     0.000492s :      1: loop_unroll
 1.48% :     0.000687s :      1: mutable_eliminate
 2.64% :     0.001227s :     26: opt.transform.jit_opt_a
 0.18% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000045s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000535s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000141s :      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
 1.49% :     0.000693s :      1: renormalize.infer
 1.19% :     0.000552s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.68% :     0.000316s :      1: rewriter_after_opt_a
 0.22% :     0.000104s :      1: rewriter_before_opt_a
 0.23% :     0.000108s :      1: symbol_engine_optimizer
24.38% :     0.011314s :      1: task_emit
47.54% :     0.022062s :      1: type_inference
 0.19% :     0.000086s :      1: validate


TotalTime = 0.0127813, [30]
[bootstrap]: 0.00028148
[type_inference]: 0.00474748
[event_method]: 2.137e-05
[auto_monad]: 6.127e-05
[graph_reusing]: 5.98998e-06
[pre_auto_parallel]: 2.32001e-06
[py_interpret_to_execute]: 4.176e-05
[rewriter_before_opt_a]: 9.245e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00540572, [2]
    [Cycle 1]: 0.00177009, [27]
        [switch_simplify]: 5.373e-05
        [loop_unroll]: 3.609e-05
        [a_1]: 0.00069199
        [with_stream_mark]: 1.354e-05
        [recompute_prepare]: 8.22e-06
        [updatestate_depend_eliminate]: 3.35998e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 6.97002e-06
        [updatestate_useless_node_eliminater]: 6.20002e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 5.89e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 9.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.262e-05
        [j_node_and_user_rematch]: 9.85002e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 1.031e-05
        [inline_without_move]: 5.95002e-06
        [renormalize]: 0.00066912
        [add_forward_monad_depend]: 4.57e-06
        [auto_monad_grad]: 1.92999e-06
        [auto_monad_eliminator]: 1.103e-05
        [cse]: 2.304e-05
        [replace_applicator]: 1.317e-05
    [Cycle 2]: 0.00036515, [27]
        [switch_simplify]: 6.89001e-06
        [loop_unroll]: 5.88998e-06
        [a_1]: 0.00012255
        [with_stream_mark]: 1.01e-05
        [recompute_prepare]: 6.06e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 6.22001e-06
        [updatestate_useless_node_eliminater]: 5.84999e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 5.68997e-06
        [merge_forward]: 3.21999e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.293e-05
        [j_node_and_user_rematch]: 9.15001e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 8.95001e-06
        [inline_without_move]: 5.51998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 7.49977e-07
        [auto_monad_eliminator]: 5.87999e-06
        [cse]: 1.305e-05
        [replace_applicator]: 5.61e-06
[py_interpret_to_execute_after_opt_a]: 1.157e-05
[rewriter_after_opt_a]: 3.022e-05
[convert_after_rewriter]: 6.31998e-06
[order_py_execute_after_rewriter]: 5.35001e-06
[mutable_eliminate]: 0.00055529
[jit_opt_b]: 5.362e-05, [1]
    [Cycle 1]: 4.762e-05, [2]
        [frontend_op_eliminate]: 1.882e-05
        [inline_after_opt_a]: 1.747e-05
[cconv]: 2.021e-05
[loop_unroll]: 0.00043222
[jit_opt_after_cconv]: 0.00014477, [1]
    [Cycle 1]: 0.00013907, [11]
        [c_1]: 2.529e-05
        [parameter_eliminate]: 2.42001e-06
        [updatestate_depend_eliminate]: 4.99e-06
        [updatestate_assign_eliminate]: 2.80002e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [cse]: 1.872e-05
        [call_graph_tuple_transform]: 1.876e-05
        [tuple_list_get_item_eliminator]: 6.33e-06
        [none_parameter_eliminate]: 1.19e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 6.16998e-06
[remove_dup_value]: 1.424e-05
[partial_unused_args_eliminate]: 1.72001e-06
[environ_conv]: 5.47999e-06
[add_recomputation]: 4.056e-05
[cse_after_recomputation]: 2.315e-05, [1]
    [Cycle 1]: 1.805e-05, [1]
        [cse]: 1.216e-05
[auto_monad_reorder]: 1.66e-05
[get_jit_bprop_graph]: 1.78997e-06
[rewriter_after_jit_bprop_graph]: 5.87001e-06
[opt_after_jit_grad]: 0.00047956
[symbol_engine_optimizer]: 7.737e-05, [1]
    [Cycle 1]: 7.119e-05, [6]
        [build]: 3.33e-06
        [elim_shapecalc]: 8.32e-06
        [elim_not_effective]: 1.321e-05
        [opt_reshape]: 6.69999e-06
        [fold_const_symbol]: 9.67999e-06
        [renormalize]: 5.60016e-07
[validate]: 3.506e-05
Sums
    bootstrap                            :     0.000281s :  3.14%
    type_inference                       :     0.004747s : 52.98%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000061s :  0.68%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000042s :  0.47%
    rewriter_before_opt_a                :     0.000092s :  1.03%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.68%
    jit_opt_a.loop_unroll                :     0.000042s :  0.47%
    jit_opt_a.a_1                        :     0.000815s :  9.09%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.26%
    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        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  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 :  0.29%
    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 :  0.13%
    jit_opt_a.renormalize                :     0.000669s :  7.47%
    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 :  0.19%
    jit_opt_a.cse                        :     0.000036s :  0.40%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000030s :  0.34%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000555s :  6.20%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000020s :  0.23%
    loop_unroll                          :     0.000432s :  4.82%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.28%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  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 :  0.07%
    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  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000041s :  0.45%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000017s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000480s :  5.35%
    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  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.39%


Time group info:
------[substitution.]   0.000268    28
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 1.65% :     0.000004s :      4: substitution.graph_param_transform
87.57% :     0.000235s :      4: substitution.inline
 1.23% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.53% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.42% :     0.000004s :      4: substitution.replace_old_param
 5.40% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004672     2
80.36% :     0.003754s :      1: type_inference.infer
19.64% :     0.000918s :      1: type_inference.specialize
------[replace.]   0.000090     8
56.48% :     0.000051s :      4: replace.inline
43.52% :     0.000039s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     8
94.91% :     0.000232s :      4: match.inline
 5.09% :     0.000012s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.28% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.13% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     13: predicate.addn_check_dump
 1.28% :     0.000002s :     13: predicate.addn_zero_filter
 2.00% :     0.000003s :     13: predicate.arithmetic_simplify
 1.43% :     0.000002s :     13: predicate.cast_eliminate
 0.49% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     13: predicate.compare_switch_simplify
 1.25% :     0.000002s :     13: predicate.depend_value_elim
 1.15% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      4: predicate.elim_not_effective
 0.46% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.25% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     13: predicate.environ_get_eliminate
 1.07% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.30% :     0.000000s :      4: predicate.fold_const_symbol
 1.07% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.31% :     0.000007s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.16% :     0.000002s :      8: predicate.less_batch_normalization
 1.69% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     21: predicate.load_eliminater
 1.31% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.23% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     13: predicate.merge_addn
 1.14% :     0.000002s :     13: predicate.minmaximum_grad
 1.33% :     0.000002s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.50% :     0.000003s :     21: predicate.partial_eliminate
 1.29% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     13: predicate.reduce_eliminate
 1.57% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.65% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.26% :     0.000002s :     13: predicate.reshape_eliminate
 1.21% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     13: predicate.same_eliminate
 0.67% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.39% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.57% :     0.000004s :     21: predicate.switch_defer_inline
 2.17% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.31% :     0.000010s :     67: predicate.switch_simplify
 1.31% :     0.000002s :     13: predicate.tile_eliminate
 1.37% :     0.000002s :     13: predicate.transpose_eliminate
 1.64% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.81% :     0.000002s :     21: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000580    12
47.41% :     0.000275s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.95% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.64% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014579    72
 0.30% :     0.000044s :      1: add_recomputation
 0.44% :     0.000065s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.01% :     0.000293s :      1: bootstrap
 0.16% :     0.000023s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.17% :     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
37.10% :     0.005409s :      1: jit_opt_a
 1.01% :     0.000148s :      1: jit_opt_after_cconv
 0.39% :     0.000056s :      1: jit_opt_b
 3.02% :     0.000440s :      1: loop_unroll
 3.86% :     0.000563s :      1: mutable_eliminate
 7.20% :     0.001050s :     26: opt.transform.jit_opt_a
 0.36% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000030s :      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
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.34% :     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% :     0.000004s :      1: pre_auto_parallel
 0.31% :     0.000045s :      1: py_interpret_to_execute
 0.10% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.52% :     0.000367s :      1: renormalize.infer
 2.03% :     0.000295s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000033s :      1: rewriter_after_opt_a
 0.65% :     0.000095s :      1: rewriter_before_opt_a
 0.55% :     0.000080s :      1: symbol_engine_optimizer
32.64% :     0.004759s :      1: type_inference


TotalTime = 0.0132866, [30]
[bootstrap]: 0.00032384
[type_inference]: 0.00493706
[event_method]: 2.165e-05
[auto_monad]: 6.764e-05
[graph_reusing]: 6.65998e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 4.145e-05
[rewriter_before_opt_a]: 8.876e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.00561694, [2]
    [Cycle 1]: 0.00176177, [27]
        [switch_simplify]: 6.209e-05
        [loop_unroll]: 3.583e-05
        [a_1]: 0.00062621
        [with_stream_mark]: 1.914e-05
        [recompute_prepare]: 8.54002e-06
        [updatestate_depend_eliminate]: 4.11001e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 3.05998e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 7.21999e-06
        [updatestate_useless_node_eliminater]: 6.49001e-06
        [accelerated_algorithm]: 6.48e-06
        [meta_shard_fg_expand]: 2.36998e-06
        [get_grad_eliminate_]: 6.06998e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.601e-05
        [j_node_and_user_rematch]: 9.56e-06
        [meta_fg_expand]: 2.63998e-06
        [replace_old_param]: 1.144e-05
        [inline_without_move]: 6.09999e-06
        [renormalize]: 0.00067838
        [add_forward_monad_depend]: 6.59999e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.508e-05
        [cse]: 3.064e-05
        [replace_applicator]: 1.482e-05
    [Cycle 2]: 0.00036346, [27]
        [switch_simplify]: 7.30003e-06
        [loop_unroll]: 5.97999e-06
        [a_1]: 0.00011615
        [with_stream_mark]: 1.154e-05
        [recompute_prepare]: 5.99999e-06
        [updatestate_depend_eliminate]: 3.5e-06
        [updatestate_assign_eliminate]: 2.60997e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 5.71003e-06
        [updatestate_useless_node_eliminater]: 5.77001e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 1.46002e-06
        [get_grad_eliminate_]: 5.31002e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.376e-05
        [j_node_and_user_rematch]: 8.83001e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 9.42001e-06
        [inline_without_move]: 5.64e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 6.12001e-06
        [cse]: 1.286e-05
        [replace_applicator]: 6.17999e-06
[py_interpret_to_execute_after_opt_a]: 1.251e-05
[rewriter_after_opt_a]: 3.556e-05
[convert_after_rewriter]: 7.8e-06
[order_py_execute_after_rewriter]: 5.92001e-06
[mutable_eliminate]: 0.00056078
[jit_opt_b]: 5.456e-05, [1]
    [Cycle 1]: 4.705e-05, [2]
        [frontend_op_eliminate]: 1.822e-05
        [inline_after_opt_a]: 1.749e-05
[cconv]: 2.367e-05
[loop_unroll]: 0.0004404
[jit_opt_after_cconv]: 0.00015503, [1]
    [Cycle 1]: 0.00014841, [11]
        [c_1]: 2.502e-05
        [parameter_eliminate]: 2.64001e-06
        [updatestate_depend_eliminate]: 6.21e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [cse]: 2.456e-05
        [call_graph_tuple_transform]: 2.056e-05
        [tuple_list_get_item_eliminator]: 6.37001e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 6.31e-06
[remove_dup_value]: 1.457e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.94e-06
[add_recomputation]: 4.95e-05
[cse_after_recomputation]: 2.365e-05, [1]
    [Cycle 1]: 1.822e-05, [1]
        [cse]: 1.216e-05
[auto_monad_reorder]: 1.751e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 4.95001e-06
[opt_after_jit_grad]: 0.00047308
[symbol_engine_optimizer]: 8.053e-05, [1]
    [Cycle 1]: 7.418e-05, [6]
        [build]: 3.41999e-06
        [elim_shapecalc]: 9.09e-06
        [elim_not_effective]: 1.496e-05
        [opt_reshape]: 6.86999e-06
        [fold_const_symbol]: 1.002e-05
        [renormalize]: 4.60015e-07
[validate]: 4.106e-05
Sums
    bootstrap                            :     0.000324s :  3.51%
    type_inference                       :     0.004937s : 53.51%
    event_method                         :     0.000022s :  0.23%
    auto_monad                           :     0.000068s :  0.73%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.45%
    rewriter_before_opt_a                :     0.000089s :  0.96%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.75%
    jit_opt_a.loop_unroll                :     0.000042s :  0.45%
    jit_opt_a.a_1                        :     0.000742s :  8.05%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000015s :  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 :  0.06%
    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 :     0.000012s :  0.13%
    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              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000021s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000678s :  7.35%
    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 :  0.23%
    jit_opt_a.cse                        :     0.000043s :  0.47%
    jit_opt_a.replace_applicator         :     0.000021s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000036s :  0.39%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000561s :  6.08%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000024s :  0.26%
    loop_unroll                          :     0.000440s :  4.77%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.27%
    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              :     0.000025s :  0.27%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  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 :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000049s :  0.54%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000473s :  5.13%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    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 :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.45%


Time group info:
------[substitution.]   0.000209    28
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.76% :     0.000006s :      4: substitution.graph_param_transform
81.06% :     0.000170s :      4: substitution.inline
 1.32% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.58% :     0.000005s :      4: substitution.replace_old_param
 8.34% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004867     2
79.71% :     0.003879s :      1: type_inference.infer
20.29% :     0.000987s :      1: type_inference.specialize
------[replace.]   0.000084     8
55.44% :     0.000047s :      4: replace.inline
44.56% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     8
91.57% :     0.000166s :      4: match.inline
 8.43% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   912
 1.41% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.15% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.07% :     0.000002s :     13: predicate.addn_check_dump
 1.25% :     0.000002s :     13: predicate.addn_zero_filter
 1.71% :     0.000002s :     13: predicate.arithmetic_simplify
 1.36% :     0.000002s :     13: predicate.cast_eliminate
 0.50% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.10% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.70% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      4: predicate.elim_not_effective
 0.56% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.07% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.13% :     0.000002s :     13: predicate.environ_get_eliminate
 1.05% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      4: predicate.fold_const_symbol
 0.92% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.53% :     0.000008s :     29: predicate.inline
 0.89% :     0.000001s :      8: predicate.inline_without_move
 0.45% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      8: predicate.less_batch_normalization
 1.53% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     21: predicate.load_eliminater
 1.29% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.07% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 1.61% :     0.000002s :      4: predicate.mutable_eliminate
 0.60% :     0.000001s :      4: predicate.opt_reshape
 2.76% :     0.000004s :     21: predicate.partial_eliminate
 1.24% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     13: predicate.reduce_eliminate
 1.77% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.92% :     0.000003s :     25: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.38% :     0.000001s :      4: predicate.reset_defer_inline
 1.17% :     0.000002s :     13: predicate.reshape_eliminate
 1.29% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     13: predicate.same_eliminate
 0.62% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.40% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.52% :     0.000004s :     21: predicate.switch_defer_inline
 2.37% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.97% :     0.000011s :     67: predicate.switch_simplify
 1.19% :     0.000002s :     13: predicate.tile_eliminate
 1.36% :     0.000002s :     13: predicate.transpose_eliminate
 1.38% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.41% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.97% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     13: predicate.value_based_eliminate
 0.41% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000649    12
45.58% :     0.000296s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.61% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.81% :     0.000324s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015029    72
 0.35% :     0.000052s :      1: add_recomputation
 0.47% :     0.000071s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 2.24% :     0.000336s :      1: bootstrap
 0.18% :     0.000027s :      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
 0.18% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
37.40% :     0.005620s :      1: jit_opt_a
 1.05% :     0.000158s :      1: jit_opt_after_cconv
 0.38% :     0.000057s :      1: jit_opt_b
 2.98% :     0.000448s :      1: loop_unroll
 3.79% :     0.000570s :      1: mutable_eliminate
 6.58% :     0.000989s :     26: opt.transform.jit_opt_a
 0.37% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.20% :     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
 0.30% :     0.000044s :      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
 2.36% :     0.000355s :      1: renormalize.infer
 2.10% :     0.000316s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000039s :      1: rewriter_after_opt_a
 0.61% :     0.000092s :      1: rewriter_before_opt_a
 0.55% :     0.000083s :      1: symbol_engine_optimizer
32.94% :     0.004951s :      1: type_inference


TotalTime = 0.0433139, [33]
[bootstrap]: 0.00033121
[type_inference]: 0.0234062
[event_method]: 2.033e-05
[auto_monad]: 0.0001482
[graph_reusing]: 6.54001e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 0.00012966
[rewriter_before_opt_a]: 9.606e-05
[expand_dump_flag]: 4.33999e-06
[jit_opt_a]: 0.00621663, [2]
    [Cycle 1]: 0.0023923, [27]
        [switch_simplify]: 6.458e-05
        [loop_unroll]: 3.788e-05
        [a_1]: 0.00072302
        [with_stream_mark]: 2.192e-05
        [recompute_prepare]: 1.133e-05
        [updatestate_depend_eliminate]: 6.34001e-06
        [updatestate_assign_eliminate]: 5.18002e-06
        [updatestate_loads_eliminate]: 4.94998e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 9.31998e-06
        [updatestate_useless_node_eliminater]: 1.129e-05
        [accelerated_algorithm]: 8.79e-06
        [meta_shard_fg_expand]: 3.04999e-06
        [get_grad_eliminate_]: 8.15e-06
        [merge_forward]: 5.66e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.051e-05
        [j_node_and_user_rematch]: 1.385e-05
        [meta_fg_expand]: 3.68e-06
        [replace_old_param]: 1.181e-05
        [inline_without_move]: 8.07e-06
        [renormalize]: 0.00113616
        [add_forward_monad_depend]: 5.51e-06
        [auto_monad_grad]: 2.76e-06
        [auto_monad_eliminator]: 2.174e-05
        [cse]: 4.063e-05
        [replace_applicator]: 1.693e-05
    [Cycle 2]: 0.00046677, [27]
        [switch_simplify]: 9.18002e-06
        [loop_unroll]: 8.23001e-06
        [a_1]: 0.00016745
        [with_stream_mark]: 1.202e-05
        [recompute_prepare]: 8.38001e-06
        [updatestate_depend_eliminate]: 5.17e-06
        [updatestate_assign_eliminate]: 4.74e-06
        [updatestate_loads_eliminate]: 3.83001e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 8.16002e-06
        [updatestate_useless_node_eliminater]: 1.098e-05
        [accelerated_algorithm]: 7.96001e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 7.56001e-06
        [merge_forward]: 4.65999e-06
        [cell_reuse_recompute_pass]: 2.11e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.613e-05
        [j_node_and_user_rematch]: 1.279e-05
        [meta_fg_expand]: 2.87002e-06
        [replace_old_param]: 1.005e-05
        [inline_without_move]: 7.53e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 1.50999e-06
        [auto_monad_eliminator]: 9.38002e-06
        [cse]: 1.934e-05
        [replace_applicator]: 8.12e-06
[py_interpret_to_execute_after_opt_a]: 1.354e-05
[rewriter_after_opt_a]: 0.00028564
[convert_after_rewriter]: 1.084e-05
[order_py_execute_after_rewriter]: 7.33e-06
[mutable_eliminate]: 0.00059784
[jit_opt_b]: 6.865e-05, [1]
    [Cycle 1]: 6.14e-05, [2]
        [frontend_op_eliminate]: 2.503e-05
        [inline_after_opt_a]: 2.447e-05
[cconv]: 2.627e-05
[loop_unroll]: 0.00045318
[jit_opt_after_cconv]: 0.00019037, [1]
    [Cycle 1]: 0.00018385, [11]
        [c_1]: 4.453e-05
        [parameter_eliminate]: 2.46e-06
        [updatestate_depend_eliminate]: 7.97998e-06
        [updatestate_assign_eliminate]: 4.3e-06
        [updatestate_loads_eliminate]: 3.78999e-06
        [cse]: 2.688e-05
        [call_graph_tuple_transform]: 2.305e-05
        [tuple_list_get_item_eliminator]: 8.60999e-06
        [none_parameter_eliminate]: 1.46002e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 8.97999e-06
[remove_dup_value]: 1.912e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 1.043e-05
[add_recomputation]: 6.859e-05
[cse_after_recomputation]: 3.081e-05, [1]
    [Cycle 1]: 2.493e-05, [1]
        [cse]: 1.843e-05
[auto_monad_reorder]: 2.65e-05
[get_jit_bprop_graph]: 1.77001e-06
[rewriter_after_jit_bprop_graph]: 5.02999e-06
[opt_after_jit_grad]: 0.00054991
[symbol_engine_optimizer]: 0.00010591, [1]
    [Cycle 1]: 9.942e-05, [6]
        [build]: 1.257e-05
        [elim_shapecalc]: 1.133e-05
        [elim_not_effective]: 2.288e-05
        [opt_reshape]: 9.09e-06
        [fold_const_symbol]: 1.374e-05
        [renormalize]: 3.89991e-07
[validate]: 5.554e-05
[backend_pass]: 1.18001e-06
[task_emit]: 0.0101797
[execute]: 8.40999e-06
Sums
    bootstrap                            :     0.000331s :  0.84%
    type_inference                       :     0.023406s : 59.56%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000148s :  0.38%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000130s :  0.33%
    rewriter_before_opt_a                :     0.000096s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.19%
    jit_opt_a.loop_unroll                :     0.000046s :  0.12%
    jit_opt_a.a_1                        :     0.000890s :  2.27%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  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    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001136s :  2.89%
    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      :     0.000031s :  0.08%
    jit_opt_a.cse                        :     0.000060s :  0.15%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000286s :  0.73%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000598s :  1.52%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000453s :  1.15%
    jit_opt_after_cconv.c_1              :     0.000045s :  0.11%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  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 :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000069s :  0.17%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000550s :  1.40%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000056s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010180s : 25.90%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000277    47
 3.09% :     0.000009s :      2: substitution.depend_value_elim
 1.01% :     0.000003s :      4: substitution.elim_not_effective
 0.78% :     0.000002s :      4: substitution.fold_const_symbol
 2.35% :     0.000007s :      5: substitution.graph_param_transform
75.99% :     0.000211s :      4: substitution.inline
 1.68% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.54% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.64% :     0.000005s :      2: substitution.replace_old_param
 4.57% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.44% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 2.91% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023336     2
93.83% :     0.021896s :      1: type_inference.infer
 6.17% :     0.001441s :      1: type_inference.specialize
------[replace.]   0.000099     6
68.61% :     0.000068s :      4: replace.inline
31.39% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.72% :     0.000208s :      4: match.inline
 5.28% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1021
 1.51% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.25% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     15: predicate.addn_check_dump
 1.40% :     0.000002s :     15: predicate.addn_zero_filter
 2.12% :     0.000004s :     15: predicate.arithmetic_simplify
 1.35% :     0.000002s :     15: predicate.cast_eliminate
 0.53% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     15: predicate.compare_switch_simplify
 1.43% :     0.000002s :     15: predicate.depend_value_elim
 1.18% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      5: predicate.elim_not_effective
 0.58% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.28% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      5: predicate.fold_const_symbol
 1.12% :     0.000002s :     10: predicate.get_grad_eliminate
 0.38% :     0.000001s :      5: predicate.graph_param_transform
 5.24% :     0.000009s :     31: predicate.inline
 0.95% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :     10: predicate.less_batch_normalization
 1.64% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     22: predicate.load_eliminater
 1.30% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.45% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.03% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     15: predicate.merge_addn
 1.21% :     0.000002s :     15: predicate.minmaximum_grad
 1.42% :     0.000002s :      5: predicate.mutable_eliminate
 0.56% :     0.000001s :      5: predicate.opt_reshape
 2.38% :     0.000004s :     22: predicate.partial_eliminate
 1.22% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.63% :     0.000003s :     15: predicate.reduce_eliminate
 1.40% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.76% :     0.000003s :     27: predicate.replace_applicator
 0.72% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.27% :     0.000002s :     15: predicate.reshape_eliminate
 1.26% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.06% :     0.000002s :     10: predicate.special_op_eliminate
 1.10% :     0.000002s :     10: predicate.specialize_transform
 1.45% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.25% :     0.000004s :     21: predicate.switch_defer_inline
 1.95% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.73% :     0.000012s :     68: predicate.switch_simplify
 1.17% :     0.000002s :     15: predicate.tile_eliminate
 1.29% :     0.000002s :     15: predicate.transpose_eliminate
 1.55% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000824    13
45.89% :     0.000378s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.07% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.03% :     0.000396s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045841    76
 0.16% :     0.000072s :      1: add_recomputation
 0.33% :     0.000153s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.76% :     0.000350s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.05% :     0.000025s :      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
13.57% :     0.006220s :      1: jit_opt_a
 0.42% :     0.000193s :      1: jit_opt_after_cconv
 0.16% :     0.000072s :      1: jit_opt_b
 1.01% :     0.000461s :      1: loop_unroll
 1.32% :     0.000607s :      1: mutable_eliminate
 2.61% :     0.001197s :     26: opt.transform.jit_opt_a
 0.18% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      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
 0.09% :     0.000040s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000054s :      4: opt.transform.symbol_engine_opt
 1.22% :     0.000559s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000135s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.54% :     0.000707s :      1: renormalize.infer
 0.92% :     0.000420s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000290s :      1: rewriter_after_opt_a
 0.22% :     0.000100s :      1: rewriter_before_opt_a
 0.24% :     0.000109s :      1: symbol_engine_optimizer
22.25% :     0.010198s :      1: task_emit
51.09% :     0.023422s :      1: type_inference
 0.17% :     0.000080s :      1: validate


TotalTime = 0.0133057, [30]
[bootstrap]: 0.00028348
[type_inference]: 0.00492811
[event_method]: 2.193e-05
[auto_monad]: 6.473e-05
[graph_reusing]: 6.12001e-06
[pre_auto_parallel]: 3.45e-06
[py_interpret_to_execute]: 3.905e-05
[rewriter_before_opt_a]: 9.142e-05
[expand_dump_flag]: 2.83998e-06
[jit_opt_a]: 0.00569094, [2]
    [Cycle 1]: 0.00181627, [27]
        [switch_simplify]: 6.266e-05
        [loop_unroll]: 3.745e-05
        [a_1]: 0.0006731
        [with_stream_mark]: 1.677e-05
        [recompute_prepare]: 9.60001e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 3.61999e-06
        [updatestate_loads_eliminate]: 3.04001e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 7.13e-06
        [updatestate_useless_node_eliminater]: 6.28998e-06
        [accelerated_algorithm]: 6.36e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 6.04999e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.604e-05
        [j_node_and_user_rematch]: 1.048e-05
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 1.227e-05
        [inline_without_move]: 6.17999e-06
        [renormalize]: 0.00068023
        [add_forward_monad_depend]: 5.07999e-06
        [auto_monad_grad]: 2.56998e-06
        [auto_monad_eliminator]: 1.478e-05
        [cse]: 3.521e-05
        [replace_applicator]: 1.405e-05
    [Cycle 2]: 0.00036653, [27]
        [switch_simplify]: 6.88998e-06
        [loop_unroll]: 6.26e-06
        [a_1]: 0.00011734
        [with_stream_mark]: 9.79e-06
        [recompute_prepare]: 6.38e-06
        [updatestate_depend_eliminate]: 3.82998e-06
        [updatestate_assign_eliminate]: 2.38002e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 6.49999e-06
        [updatestate_useless_node_eliminater]: 5.90002e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 1.52999e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.322e-05
        [j_node_and_user_rematch]: 8.48999e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 9.29998e-06
        [inline_without_move]: 5.56e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 6.36e-06
        [cse]: 1.337e-05
        [replace_applicator]: 6.49001e-06
[py_interpret_to_execute_after_opt_a]: 9.44e-06
[rewriter_after_opt_a]: 3.504e-05
[convert_after_rewriter]: 7.48e-06
[order_py_execute_after_rewriter]: 5.19998e-06
[mutable_eliminate]: 0.00052654
[jit_opt_b]: 5.503e-05, [1]
    [Cycle 1]: 4.843e-05, [2]
        [frontend_op_eliminate]: 1.854e-05
        [inline_after_opt_a]: 1.857e-05
[cconv]: 2.488e-05
[loop_unroll]: 0.00042416
[jit_opt_after_cconv]: 0.00015778, [1]
    [Cycle 1]: 0.00015157, [11]
        [c_1]: 2.532e-05
        [parameter_eliminate]: 3.11999e-06
        [updatestate_depend_eliminate]: 6.61999e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [cse]: 2.479e-05
        [call_graph_tuple_transform]: 2.12e-05
        [tuple_list_get_item_eliminator]: 7.01001e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 6.19999e-06
[remove_dup_value]: 1.367e-05
[partial_unused_args_eliminate]: 2.79999e-06
[environ_conv]: 6.06e-06
[add_recomputation]: 4.966e-05
[cse_after_recomputation]: 2.402e-05, [1]
    [Cycle 1]: 1.852e-05, [1]
        [cse]: 1.284e-05
[auto_monad_reorder]: 1.797e-05
[get_jit_bprop_graph]: 2.00002e-06
[rewriter_after_jit_bprop_graph]: 4.38999e-06
[opt_after_jit_grad]: 0.00052277
[symbol_engine_optimizer]: 7.963e-05, [1]
    [Cycle 1]: 7.327e-05, [6]
        [build]: 3.73001e-06
        [elim_shapecalc]: 9.99001e-06
        [elim_not_effective]: 1.358e-05
        [opt_reshape]: 6.44001e-06
        [fold_const_symbol]: 1.014e-05
        [renormalize]: 4.10015e-07
[validate]: 3.723e-05
Sums
    bootstrap                            :     0.000283s :  3.07%
    type_inference                       :     0.004928s : 53.42%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000065s :  0.70%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000039s :  0.42%
    rewriter_before_opt_a                :     0.000091s :  0.99%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000070s :  0.75%
    jit_opt_a.loop_unroll                :     0.000044s :  0.47%
    jit_opt_a.a_1                        :     0.000790s :  8.57%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.09%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000014s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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_        :     0.000012s :  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 :     0.000029s :  0.32%
    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          :     0.000022s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000680s :  7.37%
    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      :     0.000021s :  0.23%
    jit_opt_a.cse                        :     0.000049s :  0.53%
    jit_opt_a.replace_applicator         :     0.000021s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000035s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000527s :  5.71%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.20%
    cconv                                :     0.000025s :  0.27%
    loop_unroll                          :     0.000424s :  4.60%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.27%
    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              :     0.000025s :  0.27%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.08%
    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 :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000050s :  0.54%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000523s :  5.67%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.11%
    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 :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.40%


Time group info:
------[substitution.]   0.000248    28
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.38% :     0.000006s :      4: substitution.graph_param_transform
71.77% :     0.000178s :      4: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000006s :      4: substitution.replace_old_param
19.06% :     0.000047s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004862     2
79.61% :     0.003871s :      1: type_inference.infer
20.39% :     0.000992s :      1: type_inference.specialize
------[replace.]   0.000090     8
56.29% :     0.000050s :      4: replace.inline
43.71% :     0.000039s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     8
79.77% :     0.000175s :      4: match.inline
20.23% :     0.000044s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000144   912
 1.19% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.88% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     13: predicate.addn_check_dump
 1.30% :     0.000002s :     13: predicate.addn_zero_filter
 1.90% :     0.000003s :     13: predicate.arithmetic_simplify
 1.29% :     0.000002s :     13: predicate.cast_eliminate
 0.38% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.13% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      4: predicate.elim_not_effective
 0.50% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     13: predicate.environ_get_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.30% :     0.000000s :      4: predicate.fold_const_symbol
 1.19% :     0.000002s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 5.48% :     0.000008s :     29: predicate.inline
 0.96% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :      8: predicate.less_batch_normalization
 1.63% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.91% :     0.000003s :     21: predicate.load_eliminater
 1.23% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.31% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.77% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000002s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.39% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.47% :     0.000004s :     21: predicate.partial_eliminate
 1.09% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     13: predicate.reduce_eliminate
 1.61% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.83% :     0.000003s :     25: predicate.replace_applicator
 0.76% :     0.000001s :      8: predicate.replace_old_param
 0.31% :     0.000000s :      4: predicate.reset_defer_inline
 1.47% :     0.000002s :     13: predicate.reshape_eliminate
 1.16% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     13: predicate.same_eliminate
 0.54% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.09% :     0.000002s :      8: predicate.special_op_eliminate
 1.05% :     0.000002s :      8: predicate.specialize_transform
 1.39% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.79% :     0.000004s :     21: predicate.switch_defer_inline
 2.24% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.89% :     0.000011s :     67: predicate.switch_simplify
 1.16% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.37% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.60% :     0.000007s :     25: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.80% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000003s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000580    12
46.58% :     0.000270s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.00% :     0.000029s :      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.]   0.015103    72
 0.35% :     0.000052s :      1: add_recomputation
 0.45% :     0.000068s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 1.97% :     0.000297s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.17% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.70% :     0.005694s :      1: jit_opt_a
 1.07% :     0.000161s :      1: jit_opt_after_cconv
 0.38% :     0.000058s :      1: jit_opt_b
 2.86% :     0.000432s :      1: loop_unroll
 3.55% :     0.000536s :      1: mutable_eliminate
 6.89% :     0.001040s :     26: opt.transform.jit_opt_a
 0.37% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000030s :      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% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.52% :     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
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.28% :     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
 2.53% :     0.000382s :      1: renormalize.infer
 1.93% :     0.000291s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000038s :      1: rewriter_after_opt_a
 0.62% :     0.000094s :      1: rewriter_before_opt_a
 0.55% :     0.000082s :      1: symbol_engine_optimizer
32.70% :     0.004939s :      1: type_inference


TotalTime = 0.0125319, [30]
[bootstrap]: 0.0002705
[type_inference]: 0.00467533
[event_method]: 2.1e-05
[auto_monad]: 4.637e-05
[graph_reusing]: 3.76999e-06
[pre_auto_parallel]: 1.77999e-06
[py_interpret_to_execute]: 5.877e-05
[rewriter_before_opt_a]: 8.423e-05
[expand_dump_flag]: 2.41e-06
[jit_opt_a]: 0.00521769, [2]
    [Cycle 1]: 0.00163025, [27]
        [switch_simplify]: 5.242e-05
        [loop_unroll]: 3.638e-05
        [a_1]: 0.00059062
        [with_stream_mark]: 1.419e-05
        [recompute_prepare]: 8.62e-06
        [updatestate_depend_eliminate]: 3.31999e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 7.77e-06
        [updatestate_useless_node_eliminater]: 6.23e-06
        [accelerated_algorithm]: 6.27001e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 6.38e-06
        [merge_forward]: 2.87002e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 1.019e-05
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 9.81998e-06
        [inline_without_move]: 6.09001e-06
        [renormalize]: 0.00062672
        [add_forward_monad_depend]: 4.63999e-06
        [auto_monad_grad]: 1.36998e-06
        [auto_monad_eliminator]: 1.195e-05
        [cse]: 2.354e-05
        [replace_applicator]: 1.318e-05
    [Cycle 2]: 0.00035889, [27]
        [switch_simplify]: 6.53e-06
        [loop_unroll]: 5.91e-06
        [a_1]: 0.00011634
        [with_stream_mark]: 9.28002e-06
        [recompute_prepare]: 5.87999e-06
        [updatestate_depend_eliminate]: 3.08998e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 5.96e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 3.06999e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.92999e-06
        [inline_without_move]: 5.87999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 5.96e-06
        [cse]: 1.338e-05
        [replace_applicator]: 6.42001e-06
[py_interpret_to_execute_after_opt_a]: 1.006e-05
[rewriter_after_opt_a]: 2.905e-05
[convert_after_rewriter]: 5.79e-06
[order_py_execute_after_rewriter]: 4.29002e-06
[mutable_eliminate]: 0.00056459
[jit_opt_b]: 7.058e-05, [1]
    [Cycle 1]: 6.399e-05, [2]
        [frontend_op_eliminate]: 1.978e-05
        [inline_after_opt_a]: 3.13e-05
[cconv]: 1.59e-05
[loop_unroll]: 0.00044368
[jit_opt_after_cconv]: 0.00015214, [1]
    [Cycle 1]: 0.00014503, [11]
        [c_1]: 2.542e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 6.12999e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 2.258e-05
        [call_graph_tuple_transform]: 1.839e-05
        [tuple_list_get_item_eliminator]: 6.30002e-06
        [none_parameter_eliminate]: 9.79984e-07
        [renormalize]: 3.30008e-07
        [switch_simplify]: 6.69999e-06
[remove_dup_value]: 8.19998e-06
[partial_unused_args_eliminate]: 1.79998e-06
[environ_conv]: 4.38001e-06
[add_recomputation]: 3.267e-05
[cse_after_recomputation]: 2.441e-05, [1]
    [Cycle 1]: 1.902e-05, [1]
        [cse]: 1.297e-05
[auto_monad_reorder]: 1.279e-05
[get_jit_bprop_graph]: 1.42e-06
[rewriter_after_jit_bprop_graph]: 3.94002e-06
[opt_after_jit_grad]: 0.00046958
[symbol_engine_optimizer]: 7.549e-05, [1]
    [Cycle 1]: 6.878e-05, [6]
        [build]: 3.32002e-06
        [elim_shapecalc]: 9.00001e-06
        [elim_not_effective]: 1.293e-05
        [opt_reshape]: 6.38e-06
        [fold_const_symbol]: 9.22999e-06
        [renormalize]: 3.80009e-07
[validate]: 3.018e-05
Sums
    bootstrap                            :     0.000270s :  3.10%
    type_inference                       :     0.004675s : 53.64%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000046s :  0.53%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000059s :  0.67%
    rewriter_before_opt_a                :     0.000084s :  0.97%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000059s :  0.68%
    jit_opt_a.loop_unroll                :     0.000042s :  0.49%
    jit_opt_a.a_1                        :     0.000707s :  8.11%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.27%
    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 :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.29%
    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          :     0.000019s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000627s :  7.19%
    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      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000037s :  0.42%
    jit_opt_a.replace_applicator         :     0.000020s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000029s :  0.33%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000565s :  6.48%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000031s :  0.36%
    cconv                                :     0.000016s :  0.18%
    loop_unroll                          :     0.000444s :  5.09%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.29%
    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              :     0.000023s :  0.26%
    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 :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.08%
    remove_dup_value                     :     0.000008s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000033s :  0.37%
    cse_after_recomputation.cse          :     0.000013s :  0.15%
    auto_monad_reorder                   :     0.000013s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000470s :  5.39%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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 :  0.00%
    validate                             :     0.000030s :  0.35%


Time group info:
------[substitution.]   0.000176    28
 0.83% :     0.000001s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000004s :      4: substitution.graph_param_transform
82.33% :     0.000144s :      4: substitution.inline
 1.89% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      4: substitution.replace_old_param
 7.67% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004618     2
80.03% :     0.003696s :      1: type_inference.infer
19.97% :     0.000922s :      1: type_inference.specialize
------[replace.]   0.000082     8
54.01% :     0.000044s :      4: replace.inline
45.99% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     8
92.61% :     0.000141s :      4: match.inline
 7.39% :     0.000011s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     13: predicate.addn_check_dump
 1.40% :     0.000002s :     13: predicate.addn_zero_filter
 2.04% :     0.000003s :     13: predicate.arithmetic_simplify
 1.37% :     0.000002s :     13: predicate.cast_eliminate
 0.62% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     13: predicate.compare_switch_simplify
 1.29% :     0.000002s :     13: predicate.depend_value_elim
 1.15% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      4: predicate.elim_not_effective
 0.53% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     13: predicate.environ_get_eliminate
 1.39% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.30% :     0.000000s :      4: predicate.fold_const_symbol
 0.98% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.19% :     0.000007s :     29: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.47% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      8: predicate.less_batch_normalization
 1.75% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.05% :     0.000003s :     21: predicate.load_eliminater
 1.38% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.25% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 1.24% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.52% :     0.000003s :     21: predicate.partial_eliminate
 1.23% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     13: predicate.reduce_eliminate
 1.58% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.06% :     0.000003s :     25: predicate.replace_applicator
 0.68% :     0.000001s :      8: predicate.replace_old_param
 0.36% :     0.000001s :      4: predicate.reset_defer_inline
 1.21% :     0.000002s :     13: predicate.reshape_eliminate
 1.16% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.44% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.67% :     0.000004s :     21: predicate.switch_defer_inline
 2.22% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.47% :     0.000010s :     67: predicate.switch_simplify
 1.21% :     0.000002s :     13: predicate.tile_eliminate
 1.24% :     0.000002s :     13: predicate.transpose_eliminate
 1.58% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.14% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000664    12
53.93% :     0.000358s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.05% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.02% :     0.000279s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014185    72
 0.25% :     0.000036s :      1: add_recomputation
 0.35% :     0.000050s :      1: auto_monad
 0.11% :     0.000015s :      1: auto_monad_reorder
 2.00% :     0.000284s :      1: bootstrap
 0.13% :     0.000019s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000027s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.18% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
36.81% :     0.005221s :      1: jit_opt_a
 1.09% :     0.000155s :      1: jit_opt_after_cconv
 0.52% :     0.000074s :      1: jit_opt_b
 3.18% :     0.000451s :      1: loop_unroll
 4.23% :     0.000600s :      1: mutable_eliminate
 6.64% :     0.000942s :     26: opt.transform.jit_opt_a
 0.38% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000031s :      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
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.37% :     0.000478s :      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
 0.44% :     0.000062s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000011s :      1: remove_dup_value
 2.45% :     0.000347s :      1: renormalize.infer
 1.93% :     0.000273s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000032s :      1: rewriter_after_opt_a
 0.61% :     0.000087s :      1: rewriter_before_opt_a
 0.55% :     0.000078s :      1: symbol_engine_optimizer
33.04% :     0.004687s :      1: type_inference


TotalTime = 0.0485501, [33]
[bootstrap]: 0.00028457
[type_inference]: 0.026669
[event_method]: 1.905e-05
[auto_monad]: 0.00013781
[graph_reusing]: 3.98001e-06
[pre_auto_parallel]: 1.97001e-06
[py_interpret_to_execute]: 0.00012507
[rewriter_before_opt_a]: 0.00012904
[expand_dump_flag]: 2.29001e-06
[jit_opt_a]: 0.00637963, [2]
    [Cycle 1]: 0.00227784, [27]
        [switch_simplify]: 5.591e-05
        [loop_unroll]: 3.662e-05
        [a_1]: 0.00079105
        [with_stream_mark]: 1.49e-05
        [recompute_prepare]: 1.101e-05
        [updatestate_depend_eliminate]: 5.49e-06
        [updatestate_assign_eliminate]: 5.45001e-06
        [updatestate_loads_eliminate]: 4.48999e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 9.30001e-06
        [updatestate_useless_node_eliminater]: 1.169e-05
        [accelerated_algorithm]: 8.3e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 8.28999e-06
        [merge_forward]: 4.34002e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.693e-05
        [j_node_and_user_rematch]: 1.362e-05
        [meta_fg_expand]: 3.21001e-06
        [replace_old_param]: 1.081e-05
        [inline_without_move]: 8.22e-06
        [renormalize]: 0.00101519
        [add_forward_monad_depend]: 4.65001e-06
        [auto_monad_grad]: 1.40999e-06
        [auto_monad_eliminator]: 1.723e-05
        [cse]: 2.74e-05
        [replace_applicator]: 1.585e-05
    [Cycle 2]: 0.00045914, [27]
        [switch_simplify]: 9.12001e-06
        [loop_unroll]: 7.92e-06
        [a_1]: 0.00016463
        [with_stream_mark]: 1.081e-05
        [recompute_prepare]: 7.86001e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 4.59998e-06
        [updatestate_loads_eliminate]: 3.97002e-06
        [parameter_eliminate]: 1.26997e-06
        [specialize_transform]: 8.07e-06
        [updatestate_useless_node_eliminater]: 1.093e-05
        [accelerated_algorithm]: 7.96001e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 7.61001e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.676e-05
        [j_node_and_user_rematch]: 1.234e-05
        [meta_fg_expand]: 2.76e-06
        [replace_old_param]: 1.004e-05
        [inline_without_move]: 7.76001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.02998e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 1.003e-05
        [cse]: 2.017e-05
        [replace_applicator]: 7.74002e-06
[py_interpret_to_execute_after_opt_a]: 1.057e-05
[rewriter_after_opt_a]: 0.00027614
[convert_after_rewriter]: 1.004e-05
[order_py_execute_after_rewriter]: 7.01001e-06
[mutable_eliminate]: 0.00055218
[jit_opt_b]: 6.641e-05, [1]
    [Cycle 1]: 6.017e-05, [2]
        [frontend_op_eliminate]: 2.468e-05
        [inline_after_opt_a]: 2.403e-05
[cconv]: 1.898e-05
[loop_unroll]: 0.00044541
[jit_opt_after_cconv]: 0.00024506, [1]
    [Cycle 1]: 0.0002382, [11]
        [c_1]: 4.429e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 7.77998e-06
        [updatestate_assign_eliminate]: 4.74998e-06
        [updatestate_loads_eliminate]: 3.97002e-06
        [cse]: 2.713e-05
        [call_graph_tuple_transform]: 2.172e-05
        [tuple_list_get_item_eliminator]: 8.33001e-06
        [none_parameter_eliminate]: 1.37e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 9.17001e-06
[remove_dup_value]: 1.419e-05
[partial_unused_args_eliminate]: 1.67999e-06
[environ_conv]: 3.834e-05
[add_recomputation]: 8.452e-05
[cse_after_recomputation]: 3.268e-05, [1]
    [Cycle 1]: 2.657e-05, [1]
        [cse]: 1.983e-05
[auto_monad_reorder]: 2.024e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 5.19998e-06
[opt_after_jit_grad]: 0.00050132
[symbol_engine_optimizer]: 9.838e-05, [1]
    [Cycle 1]: 9.24e-05, [6]
        [build]: 1.18e-05
        [elim_shapecalc]: 1.181e-05
        [elim_not_effective]: 1.819e-05
        [opt_reshape]: 8.53001e-06
        [fold_const_symbol]: 1.306e-05
        [renormalize]: 4.09986e-07
[validate]: 7.749e-05
[backend_pass]: 1.34e-06
[task_emit]: 0.0120317
[execute]: 8.44002e-06
Sums
    bootstrap                            :     0.000285s :  0.64%
    type_inference                       :     0.026669s : 60.34%
    event_method                         :     0.000019s :  0.04%
    auto_monad                           :     0.000138s :  0.31%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000125s :  0.28%
    rewriter_before_opt_a                :     0.000129s :  0.29%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.15%
    jit_opt_a.loop_unroll                :     0.000045s :  0.10%
    jit_opt_a.a_1                        :     0.000956s :  2.16%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001015s :  2.30%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.06%
    jit_opt_a.cse                        :     0.000048s :  0.11%
    jit_opt_a.replace_applicator         :     0.000024s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.02%
    rewriter_after_opt_a                 :     0.000276s :  0.62%
    convert_after_rewriter               :     0.000010s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000552s :  1.25%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.05%
    cconv                                :     0.000019s :  0.04%
    loop_unroll                          :     0.000445s :  1.01%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.10%
    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 :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000014s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000038s :  0.09%
    add_recomputation                    :     0.000085s :  0.19%
    cse_after_recomputation.cse          :     0.000020s :  0.04%
    auto_monad_reorder                   :     0.000020s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000501s :  1.13%
    symbol_engine_optimizer.build        :     0.000012s :  0.03%
    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 :     0.000013s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000077s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012032s : 27.22%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000246    47
 3.23% :     0.000008s :      2: substitution.depend_value_elim
 0.96% :     0.000002s :      4: substitution.elim_not_effective
 0.77% :     0.000002s :      4: substitution.fold_const_symbol
 2.10% :     0.000005s :      5: substitution.graph_param_transform
76.35% :     0.000187s :      4: substitution.inline
 1.73% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.92% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.50% :     0.000004s :      2: substitution.replace_old_param
 3.50% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
 3.25% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.67% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026576     2
94.53% :     0.025121s :      1: type_inference.infer
 5.47% :     0.001454s :      1: type_inference.specialize
------[replace.]   0.000087     6
65.34% :     0.000057s :      4: replace.inline
34.66% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
96.11% :     0.000185s :      4: match.inline
 3.89% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1021
 1.31% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.32% :     0.000002s :     15: predicate.addn_zero_filter
 1.90% :     0.000003s :     15: predicate.arithmetic_simplify
 1.34% :     0.000002s :     15: predicate.cast_eliminate
 0.49% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.21% :     0.000002s :     15: predicate.compare_switch_simplify
 1.43% :     0.000002s :     15: predicate.depend_value_elim
 1.16% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.58% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.37% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     15: predicate.environ_get_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 4.80% :     0.000008s :     31: predicate.inline
 1.18% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :     10: predicate.less_batch_normalization
 1.59% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.96% :     0.000003s :     22: predicate.load_eliminater
 1.33% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.54% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.24% :     0.000002s :     15: predicate.merge_addn
 1.15% :     0.000002s :     15: predicate.minmaximum_grad
 1.57% :     0.000003s :      5: predicate.mutable_eliminate
 0.52% :     0.000001s :      5: predicate.opt_reshape
 2.35% :     0.000004s :     22: predicate.partial_eliminate
 1.28% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.68% :     0.000003s :     15: predicate.reduce_eliminate
 1.51% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.77% :     0.000003s :     27: predicate.replace_applicator
 0.60% :     0.000001s :     10: predicate.replace_old_param
 0.41% :     0.000001s :      5: predicate.reset_defer_inline
 1.29% :     0.000002s :     15: predicate.reshape_eliminate
 1.25% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.01% :     0.000002s :     10: predicate.special_op_eliminate
 1.04% :     0.000002s :     10: predicate.specialize_transform
 1.50% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.56% :     0.000004s :     21: predicate.switch_defer_inline
 2.05% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.70% :     0.000011s :     68: predicate.switch_simplify
 1.25% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.60% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.86% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.30% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000003s :     15: predicate.value_based_eliminate
 0.47% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.99% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000775    13
43.76% :     0.000339s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.99% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.25% :     0.000397s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050996    76
 0.17% :     0.000089s :      1: add_recomputation
 0.28% :     0.000143s :      1: auto_monad
 0.05% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.58% :     0.000298s :      1: bootstrap
 0.04% :     0.000022s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000035s :      1: cse_after_recomputation
 0.08% :     0.000042s :      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
 0.01% :     0.000006s :      1: graph_reusing
12.52% :     0.006382s :      1: jit_opt_a
 0.49% :     0.000248s :      1: jit_opt_after_cconv
 0.14% :     0.000069s :      1: jit_opt_b
 0.89% :     0.000453s :      1: loop_unroll
 1.10% :     0.000560s :      1: mutable_eliminate
 2.45% :     0.001250s :     26: opt.transform.jit_opt_a
 0.16% :     0.000080s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000042s :      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
 0.06% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000048s :      4: opt.transform.symbol_engine_opt
 1.00% :     0.000510s :      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
 0.25% :     0.000130s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 1.21% :     0.000617s :      1: renormalize.infer
 0.77% :     0.000391s :      1: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.55% :     0.000281s :      1: rewriter_after_opt_a
 0.26% :     0.000134s :      1: rewriter_before_opt_a
 0.20% :     0.000101s :      1: symbol_engine_optimizer
23.62% :     0.012048s :      1: task_emit
52.38% :     0.026711s :      1: type_inference
 0.20% :     0.000101s :      1: validate


TotalTime = 0.0133892, [30]
[bootstrap]: 0.00031919
[type_inference]: 0.00520387
[event_method]: 2.115e-05
[auto_monad]: 6.621e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 3.906e-05
[rewriter_before_opt_a]: 9.06e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.0054706, [2]
    [Cycle 1]: 0.00179581, [27]
        [switch_simplify]: 6.281e-05
        [loop_unroll]: 3.551e-05
        [a_1]: 0.00067094
        [with_stream_mark]: 1.827e-05
        [recompute_prepare]: 8.07998e-06
        [updatestate_depend_eliminate]: 4.02002e-06
        [updatestate_assign_eliminate]: 3.56999e-06
        [updatestate_loads_eliminate]: 3.2e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 6.90998e-06
        [updatestate_useless_node_eliminater]: 5.96998e-06
        [accelerated_algorithm]: 6.41998e-06
        [meta_shard_fg_expand]: 2.15002e-06
        [get_grad_eliminate_]: 6.34999e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 1.14003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.515e-05
        [j_node_and_user_rematch]: 9.51e-06
        [meta_fg_expand]: 2.57001e-06
        [replace_old_param]: 1.089e-05
        [inline_without_move]: 6.16e-06
        [renormalize]: 0.00066611
        [add_forward_monad_depend]: 5.64e-06
        [auto_monad_grad]: 2.26998e-06
        [auto_monad_eliminator]: 1.498e-05
        [cse]: 3.678e-05
        [replace_applicator]: 1.31e-05
    [Cycle 2]: 0.00036631, [27]
        [switch_simplify]: 7.03e-06
        [loop_unroll]: 5.97999e-06
        [a_1]: 0.00011398
        [with_stream_mark]: 1.045e-05
        [recompute_prepare]: 6.24999e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.68003e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.78997e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 5.25001e-06
        [merge_forward]: 2.58998e-06
        [cell_reuse_recompute_pass]: 1.74998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.427e-05
        [j_node_and_user_rematch]: 8.83001e-06
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 9.72999e-06
        [inline_without_move]: 6.07999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.35001e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 5.98002e-06
        [cse]: 1.38e-05
        [replace_applicator]: 5.94e-06
[py_interpret_to_execute_after_opt_a]: 9.87001e-06
[rewriter_after_opt_a]: 6.227e-05
[convert_after_rewriter]: 7.95998e-06
[order_py_execute_after_rewriter]: 5.15001e-06
[mutable_eliminate]: 0.0005437
[jit_opt_b]: 5.336e-05, [1]
    [Cycle 1]: 4.664e-05, [2]
        [frontend_op_eliminate]: 1.776e-05
        [inline_after_opt_a]: 1.774e-05
[cconv]: 2.398e-05
[loop_unroll]: 0.00043281
[jit_opt_after_cconv]: 0.00015773, [1]
    [Cycle 1]: 0.00015109, [11]
        [c_1]: 2.502e-05
        [parameter_eliminate]: 2.52001e-06
        [updatestate_depend_eliminate]: 5.74e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [cse]: 2.313e-05
        [call_graph_tuple_transform]: 2.041e-05
        [tuple_list_get_item_eliminator]: 6.38998e-06
        [none_parameter_eliminate]: 1.32999e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 6.27001e-06
[remove_dup_value]: 1.494e-05
[partial_unused_args_eliminate]: 2.33002e-06
[environ_conv]: 6.11e-06
[add_recomputation]: 4.875e-05
[cse_after_recomputation]: 2.428e-05, [1]
    [Cycle 1]: 1.837e-05, [1]
        [cse]: 1.259e-05
[auto_monad_reorder]: 1.807e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 4.42e-06
[opt_after_jit_grad]: 0.00046357
[symbol_engine_optimizer]: 7.634e-05, [1]
    [Cycle 1]: 7.032e-05, [6]
        [build]: 3.51001e-06
        [elim_shapecalc]: 8.89e-06
        [elim_not_effective]: 1.368e-05
        [opt_reshape]: 6.41e-06
        [fold_const_symbol]: 9.17001e-06
        [renormalize]: 4.49974e-07
[validate]: 3.885e-05
Sums
    bootstrap                            :     0.000319s :  3.36%
    type_inference                       :     0.005204s : 54.78%
    event_method                         :     0.000021s :  0.22%
    auto_monad                           :     0.000066s :  0.70%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.41%
    rewriter_before_opt_a                :     0.000091s :  0.95%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000070s :  0.74%
    jit_opt_a.loop_unroll                :     0.000041s :  0.44%
    jit_opt_a.a_1                        :     0.000785s :  8.26%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.30%
    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 :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.12%
    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_        :     0.000012s :  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 :     0.000029s :  0.31%
    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          :     0.000021s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000666s :  7.01%
    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      :     0.000021s :  0.22%
    jit_opt_a.cse                        :     0.000051s :  0.53%
    jit_opt_a.replace_applicator         :     0.000019s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000062s :  0.66%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000544s :  5.72%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.19%
    cconv                                :     0.000024s :  0.25%
    loop_unroll                          :     0.000433s :  4.56%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.26%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000023s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  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      :     0.000001s :  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                    :     0.000049s :  0.51%
    cse_after_recomputation.cse          :     0.000013s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000464s :  4.88%
    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 :  0.14%
    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 :  0.41%


Time group info:
------[substitution.]   0.000206    28
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.95% :     0.000006s :      4: substitution.graph_param_transform
80.53% :     0.000166s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.40% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.63% :     0.000005s :      4: substitution.replace_old_param
 8.47% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005124     2
81.22% :     0.004161s :      1: type_inference.infer
18.78% :     0.000962s :      1: type_inference.specialize
------[replace.]   0.000082     8
56.00% :     0.000046s :      4: replace.inline
44.00% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     8
91.33% :     0.000163s :      4: match.inline
 8.67% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000195   912
 0.86% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.78% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
30.84% :     0.000060s :     13: predicate.addn_check_dump
 0.99% :     0.000002s :     13: predicate.addn_zero_filter
 1.22% :     0.000002s :     13: predicate.arithmetic_simplify
 0.93% :     0.000002s :     13: predicate.cast_eliminate
 0.32% :     0.000001s :      4: predicate.check_bprop_eliminate
 0.76% :     0.000001s :     13: predicate.compare_switch_simplify
 1.05% :     0.000002s :     13: predicate.depend_value_elim
 0.79% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 0.93% :     0.000002s :     13: predicate.dict_get_item_eliminator
 0.86% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.56% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      4: predicate.elim_not_effective
 0.37% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 0.83% :     0.000002s :     13: predicate.environ_add_const_eliminate
 0.81% :     0.000002s :     13: predicate.environ_get_add_eliminate
 0.76% :     0.000001s :     13: predicate.environ_get_depend_swap
 0.82% :     0.000002s :     13: predicate.environ_get_eliminate
 0.79% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      4: predicate.fold_const_symbol
 0.62% :     0.000001s :      8: predicate.get_grad_eliminate
 0.22% :     0.000000s :      4: predicate.graph_param_transform
 3.64% :     0.000007s :     29: predicate.inline
 0.71% :     0.000001s :      8: predicate.inline_without_move
 0.30% :     0.000001s :      8: predicate.j_node_and_user_rematch
 0.89% :     0.000002s :      8: predicate.less_batch_normalization
 1.13% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.33% :     0.000003s :     21: predicate.load_eliminater
 0.91% :     0.000002s :      4: predicate.loop_unroll_after_grad
 2.95% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.27% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 0.76% :     0.000001s :     13: predicate.merge_addn
 0.74% :     0.000001s :     13: predicate.minmaximum_grad
 1.12% :     0.000002s :      4: predicate.mutable_eliminate
 0.31% :     0.000001s :      4: predicate.opt_reshape
 1.80% :     0.000004s :     21: predicate.partial_eliminate
 0.81% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.08% :     0.000002s :     13: predicate.reduce_eliminate
 1.08% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.35% :     0.000003s :     25: predicate.replace_applicator
 0.43% :     0.000001s :      8: predicate.replace_old_param
 0.22% :     0.000000s :      4: predicate.reset_defer_inline
 0.84% :     0.000002s :     13: predicate.reshape_eliminate
 0.80% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.54% :     0.000001s :      4: predicate.row_tensor_eliminate
 0.84% :     0.000002s :     13: predicate.same_eliminate
 0.47% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.73% :     0.000001s :      8: predicate.special_op_eliminate
 0.68% :     0.000001s :      8: predicate.specialize_transform
 0.97% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 0.82% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.30% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 1.85% :     0.000004s :     21: predicate.switch_defer_inline
 1.50% :     0.000003s :     21: predicate.switch_layer_defer_inline
 5.64% :     0.000011s :     67: predicate.switch_simplify
 0.82% :     0.000002s :     13: predicate.tile_eliminate
 0.84% :     0.000002s :     13: predicate.transpose_eliminate
 1.04% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 0.88% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 2.92% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.22% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.10% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.29% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.12% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.13% :     0.000002s :     13: predicate.value_based_eliminate
 0.29% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.41% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000638    12
51.85% :     0.000331s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.58% :     0.000029s :      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.]   0.015158    72
 0.34% :     0.000052s :      1: add_recomputation
 0.46% :     0.000070s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 2.24% :     0.000339s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000008s :      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
36.11% :     0.005473s :      1: jit_opt_a
 1.06% :     0.000161s :      1: jit_opt_after_cconv
 0.37% :     0.000056s :      1: jit_opt_b
 2.91% :     0.000441s :      1: loop_unroll
 3.64% :     0.000552s :      1: mutable_eliminate
 6.80% :     0.001031s :     26: opt.transform.jit_opt_a
 0.36% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000030s :      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
 0.16% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.11% :     0.000471s :      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% :     0.000042s :      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
 2.35% :     0.000357s :      1: renormalize.infer
 1.99% :     0.000302s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.43% :     0.000066s :      1: rewriter_after_opt_a
 0.61% :     0.000093s :      1: rewriter_before_opt_a
 0.52% :     0.000079s :      1: symbol_engine_optimizer
34.41% :     0.005216s :      1: type_inference


TotalTime = 0.0135622, [30]
[bootstrap]: 0.000292
[type_inference]: 0.00503565
[event_method]: 2.239e-05
[auto_monad]: 6.133e-05
[graph_reusing]: 5.53002e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 3.873e-05
[rewriter_before_opt_a]: 8.834e-05
[expand_dump_flag]: 3.43999e-06
[jit_opt_a]: 0.00587143, [2]
    [Cycle 1]: 0.00184881, [27]
        [switch_simplify]: 5.658e-05
        [loop_unroll]: 3.534e-05
        [a_1]: 0.00064099
        [with_stream_mark]: 1.775e-05
        [recompute_prepare]: 8.87e-06
        [updatestate_depend_eliminate]: 3.76999e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.28002e-06
        [parameter_eliminate]: 1.40001e-06
        [specialize_transform]: 7.49002e-06
        [updatestate_useless_node_eliminater]: 6.01e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 3.17997e-06
        [cell_reuse_recompute_pass]: 8.59989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.269e-05
        [j_node_and_user_rematch]: 9.27001e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 1.024e-05
        [inline_without_move]: 5.79e-06
        [renormalize]: 0.00077751
        [add_forward_monad_depend]: 5.25999e-06
        [auto_monad_grad]: 3.08998e-06
        [auto_monad_eliminator]: 1.409e-05
        [cse]: 2.327e-05
        [replace_applicator]: 1.353e-05
    [Cycle 2]: 0.00036295, [27]
        [switch_simplify]: 6.50002e-06
        [loop_unroll]: 5.62001e-06
        [a_1]: 0.00011322
        [with_stream_mark]: 1.114e-05
        [recompute_prepare]: 5.77999e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.12999e-06
        [specialize_transform]: 5.78997e-06
        [updatestate_useless_node_eliminater]: 5.47999e-06
        [accelerated_algorithm]: 5.78002e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.448e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 9.96998e-06
        [inline_without_move]: 5.60001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 6.89999e-06
        [cse]: 1.383e-05
        [replace_applicator]: 6.39999e-06
[py_interpret_to_execute_after_opt_a]: 1.212e-05
[rewriter_after_opt_a]: 3.438e-05
[convert_after_rewriter]: 6.16e-06
[order_py_execute_after_rewriter]: 4.90999e-06
[mutable_eliminate]: 0.00055503
[jit_opt_b]: 5.529e-05, [1]
    [Cycle 1]: 4.882e-05, [2]
        [frontend_op_eliminate]: 1.878e-05
        [inline_after_opt_a]: 1.744e-05
[cconv]: 2.112e-05
[loop_unroll]: 0.0004365
[jit_opt_after_cconv]: 0.00015495, [1]
    [Cycle 1]: 0.00014866, [11]
        [c_1]: 2.474e-05
        [parameter_eliminate]: 2.57001e-06
        [updatestate_depend_eliminate]: 6.84999e-06
        [updatestate_assign_eliminate]: 3.10002e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [cse]: 2.513e-05
        [call_graph_tuple_transform]: 1.922e-05
        [tuple_list_get_item_eliminator]: 6.38e-06
        [none_parameter_eliminate]: 9.80013e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 6.46999e-06
[remove_dup_value]: 1.086e-05
[partial_unused_args_eliminate]: 1.79998e-06
[environ_conv]: 5.81e-06
[add_recomputation]: 3.804e-05
[cse_after_recomputation]: 2.356e-05, [1]
    [Cycle 1]: 1.803e-05, [1]
        [cse]: 1.272e-05
[auto_monad_reorder]: 1.47e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 7.26999e-06
[opt_after_jit_grad]: 0.00047604
[symbol_engine_optimizer]: 7.483e-05, [1]
    [Cycle 1]: 6.862e-05, [6]
        [build]: 2.66e-06
        [elim_shapecalc]: 8.42e-06
        [elim_not_effective]: 1.379e-05
        [opt_reshape]: 6.03002e-06
        [fold_const_symbol]: 8.94998e-06
        [renormalize]: 3.7998e-07
[validate]: 3.535e-05
Sums
    bootstrap                            :     0.000292s :  3.13%
    type_inference                       :     0.005036s : 53.99%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000061s :  0.66%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.42%
    rewriter_before_opt_a                :     0.000088s :  0.95%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.68%
    jit_opt_a.loop_unroll                :     0.000041s :  0.44%
    jit_opt_a.a_1                        :     0.000754s :  8.09%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.16%
    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       :     0.000013s :  0.14%
    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_        :     0.000011s :  0.12%
    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 :  0.29%
    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          :     0.000020s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000778s :  8.34%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.23%
    jit_opt_a.cse                        :     0.000037s :  0.40%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000034s :  0.37%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000555s :  5.95%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000021s :  0.23%
    loop_unroll                          :     0.000436s :  4.68%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.27%
    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              :     0.000025s :  0.27%
    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  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000011s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000038s :  0.41%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000476s :  5.10%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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 :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.38%


Time group info:
------[substitution.]   0.000224    28
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.27% :     0.000005s :      4: substitution.graph_param_transform
83.54% :     0.000187s :      4: substitution.inline
 1.22% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.34% :     0.000005s :      4: substitution.replace_old_param
 7.00% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004966     2
81.31% :     0.004038s :      1: type_inference.infer
18.69% :     0.000928s :      1: type_inference.specialize
------[replace.]   0.000090     8
55.39% :     0.000050s :      4: replace.inline
44.61% :     0.000040s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000198     8
93.02% :     0.000184s :      4: match.inline
 6.98% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.15% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     13: predicate.addn_check_dump
 1.33% :     0.000002s :     13: predicate.addn_zero_filter
 1.97% :     0.000003s :     13: predicate.arithmetic_simplify
 1.29% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.50% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     13: predicate.environ_get_eliminate
 1.08% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.97% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.41% :     0.000007s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.46% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      8: predicate.less_batch_normalization
 1.58% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.16% :     0.000003s :     21: predicate.load_eliminater
 1.42% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.11% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     13: predicate.merge_addn
 1.10% :     0.000001s :     13: predicate.minmaximum_grad
 1.58% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.51% :     0.000003s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 2.02% :     0.000003s :     13: predicate.reduce_eliminate
 1.71% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.96% :     0.000003s :     25: predicate.replace_applicator
 0.65% :     0.000001s :      8: predicate.replace_old_param
 0.40% :     0.000001s :      4: predicate.reset_defer_inline
 1.22% :     0.000002s :     13: predicate.reshape_eliminate
 1.20% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     13: predicate.same_eliminate
 0.62% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      8: predicate.special_op_eliminate
 1.07% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.35% :     0.000010s :     67: predicate.switch_simplify
 1.23% :     0.000002s :     13: predicate.tile_eliminate
 1.18% :     0.000002s :     13: predicate.transpose_eliminate
 1.53% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.61% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000608    12
48.26% :     0.000294s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.40% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.34% :     0.000282s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015407    72
 0.27% :     0.000041s :      1: add_recomputation
 0.42% :     0.000065s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.01% :     0.000310s :      1: bootstrap
 0.15% :     0.000024s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.17% :     0.000026s :      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
38.13% :     0.005874s :      1: jit_opt_a
 1.02% :     0.000158s :      1: jit_opt_after_cconv
 0.38% :     0.000058s :      1: jit_opt_b
 2.89% :     0.000445s :      1: loop_unroll
 3.66% :     0.000564s :      1: mutable_eliminate
 6.45% :     0.000993s :     26: opt.transform.jit_opt_a
 0.35% :     0.000054s :      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
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.14% :     0.000484s :      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% :     0.000005s :      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
 0.09% :     0.000013s :      1: remove_dup_value
 2.65% :     0.000409s :      1: renormalize.infer
 2.34% :     0.000361s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000037s :      1: rewriter_after_opt_a
 0.59% :     0.000091s :      1: rewriter_before_opt_a
 0.50% :     0.000077s :      1: symbol_engine_optimizer
32.77% :     0.005050s :      1: type_inference


TotalTime = 0.0443965, [33]
[bootstrap]: 0.0003306
[type_inference]: 0.0231024
[event_method]: 1.941e-05
[auto_monad]: 0.00020065
[graph_reusing]: 6.04999e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 0.0001797
[rewriter_before_opt_a]: 9.896e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00602772, [2]
    [Cycle 1]: 0.00225375, [27]
        [switch_simplify]: 6.306e-05
        [loop_unroll]: 3.632e-05
        [a_1]: 0.00076684
        [with_stream_mark]: 1.819e-05
        [recompute_prepare]: 1.01e-05
        [updatestate_depend_eliminate]: 6.17999e-06
        [updatestate_assign_eliminate]: 5.81998e-06
        [updatestate_loads_eliminate]: 4.75001e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 9.12001e-06
        [updatestate_useless_node_eliminater]: 1.135e-05
        [accelerated_algorithm]: 8.36002e-06
        [meta_shard_fg_expand]: 3.10998e-06
        [get_grad_eliminate_]: 8.27998e-06
        [merge_forward]: 5.34e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.927e-05
        [j_node_and_user_rematch]: 1.36e-05
        [meta_fg_expand]: 3.52997e-06
        [replace_old_param]: 1.127e-05
        [inline_without_move]: 8.13001e-06
        [renormalize]: 0.00100737
        [add_forward_monad_depend]: 3.88999e-06
        [auto_monad_grad]: 1.19e-06
        [auto_monad_eliminator]: 1.577e-05
        [cse]: 2.727e-05
        [replace_applicator]: 1.382e-05
    [Cycle 2]: 0.00045756, [27]
        [switch_simplify]: 9.32001e-06
        [loop_unroll]: 8.38001e-06
        [a_1]: 0.00016783
        [with_stream_mark]: 1.003e-05
        [recompute_prepare]: 7.61001e-06
        [updatestate_depend_eliminate]: 4.79e-06
        [updatestate_assign_eliminate]: 4.70999e-06
        [updatestate_loads_eliminate]: 3.96001e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 7.88001e-06
        [updatestate_useless_node_eliminater]: 1.051e-05
        [accelerated_algorithm]: 7.48e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 7.46001e-06
        [merge_forward]: 3.97e-06
        [cell_reuse_recompute_pass]: 2.34001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.62e-05
        [j_node_and_user_rematch]: 1.207e-05
        [meta_fg_expand]: 2.73998e-06
        [replace_old_param]: 9.61e-06
        [inline_without_move]: 7.33999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.02e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 9.72001e-06
        [cse]: 1.903e-05
        [replace_applicator]: 7.52998e-06
[py_interpret_to_execute_after_opt_a]: 9.81e-06
[rewriter_after_opt_a]: 0.00027086
[convert_after_rewriter]: 1.014e-05
[order_py_execute_after_rewriter]: 6.11e-06
[mutable_eliminate]: 0.00050684
[jit_opt_b]: 6.72e-05, [1]
    [Cycle 1]: 6.074e-05, [2]
        [frontend_op_eliminate]: 2.366e-05
        [inline_after_opt_a]: 2.457e-05
[cconv]: 1.77e-05
[loop_unroll]: 0.0004521
[jit_opt_after_cconv]: 0.00020291, [1]
    [Cycle 1]: 0.00019629, [11]
        [c_1]: 4.506e-05
        [parameter_eliminate]: 2.17001e-06
        [updatestate_depend_eliminate]: 7.63999e-06
        [updatestate_assign_eliminate]: 4.27e-06
        [updatestate_loads_eliminate]: 4.48999e-06
        [cse]: 2.604e-05
        [call_graph_tuple_transform]: 1.99e-05
        [tuple_list_get_item_eliminator]: 8.64e-06
        [none_parameter_eliminate]: 9.00007e-07
        [renormalize]: 4.50003e-07
        [switch_simplify]: 8.25999e-06
[remove_dup_value]: 1.335e-05
[partial_unused_args_eliminate]: 1.38002e-06
[environ_conv]: 1.023e-05
[add_recomputation]: 4.841e-05
[cse_after_recomputation]: 2.928e-05, [1]
    [Cycle 1]: 2.336e-05, [1]
        [cse]: 1.753e-05
[auto_monad_reorder]: 1.918e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 3.76999e-06
[opt_after_jit_grad]: 0.00049091
[symbol_engine_optimizer]: 9.954e-05, [1]
    [Cycle 1]: 9.344e-05, [6]
        [build]: 1.093e-05
        [elim_shapecalc]: 1.147e-05
        [elim_not_effective]: 1.662e-05
        [opt_reshape]: 9.48002e-06
        [fold_const_symbol]: 1.465e-05
        [renormalize]: 4.89992e-07
[validate]: 4.853e-05
[backend_pass]: 7.30011e-07
[task_emit]: 0.0118897
[execute]: 7.59002e-06
Sums
    bootstrap                            :     0.000331s :  0.82%
    type_inference                       :     0.023102s : 57.13%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000201s :  0.50%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000180s :  0.44%
    rewriter_before_opt_a                :     0.000099s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.18%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000935s :  2.31%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    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 :     0.000035s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001007s :  2.49%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.06%
    jit_opt_a.cse                        :     0.000046s :  0.11%
    jit_opt_a.replace_applicator         :     0.000021s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.02%
    rewriter_after_opt_a                 :     0.000271s :  0.67%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000507s :  1.25%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000018s :  0.04%
    loop_unroll                          :     0.000452s :  1.12%
    jit_opt_after_cconv.c_1              :     0.000045s :  0.11%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000013s :  0.03%
    partial_unused_args_eliminate        :     0.000001s :  0.00%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000048s :  0.12%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000019s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000491s :  1.21%
    symbol_engine_optimizer.build        :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  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                             :     0.000049s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011890s : 29.40%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000310    47
 2.41% :     0.000007s :      2: substitution.depend_value_elim
 0.66% :     0.000002s :      4: substitution.elim_not_effective
 0.64% :     0.000002s :      4: substitution.fold_const_symbol
 1.48% :     0.000005s :      5: substitution.graph_param_transform
80.51% :     0.000250s :      4: substitution.inline
 1.33% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.44% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.16% :     0.000004s :      2: substitution.replace_old_param
 3.98% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 2.74% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.65% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023038     2
94.07% :     0.021671s :      1: type_inference.infer
 5.93% :     0.001367s :      1: type_inference.specialize
------[replace.]   0.000095     6
66.61% :     0.000063s :      4: replace.inline
33.39% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000258     6
95.67% :     0.000247s :      4: match.inline
 4.33% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1021
 1.28% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.38% :     0.000002s :     15: predicate.addn_zero_filter
 2.01% :     0.000003s :     15: predicate.arithmetic_simplify
 1.41% :     0.000002s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     15: predicate.compare_switch_simplify
 1.45% :     0.000003s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.45% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.43% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      5: predicate.elim_not_effective
 0.57% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.46% :     0.000003s :     15: predicate.environ_get_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 4.86% :     0.000008s :     31: predicate.inline
 1.01% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :     10: predicate.less_batch_normalization
 1.55% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     22: predicate.load_eliminater
 1.32% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.47% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.50% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     15: predicate.merge_addn
 1.15% :     0.000002s :     15: predicate.minmaximum_grad
 1.26% :     0.000002s :      5: predicate.mutable_eliminate
 0.47% :     0.000001s :      5: predicate.opt_reshape
 2.39% :     0.000004s :     22: predicate.partial_eliminate
 1.37% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.75% :     0.000003s :     15: predicate.reduce_eliminate
 1.50% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.60% :     0.000003s :     27: predicate.replace_applicator
 0.65% :     0.000001s :     10: predicate.replace_old_param
 0.31% :     0.000001s :      5: predicate.reset_defer_inline
 1.37% :     0.000002s :     15: predicate.reshape_eliminate
 1.27% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :     10: predicate.special_op_eliminate
 1.11% :     0.000002s :     10: predicate.specialize_transform
 1.54% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.23% :     0.000004s :     21: predicate.switch_defer_inline
 1.91% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.45% :     0.000011s :     68: predicate.switch_simplify
 1.41% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000002s :     15: predicate.transpose_eliminate
 1.54% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.77% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.50% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000003s :     15: predicate.value_based_eliminate
 0.48% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000825    13
48.29% :     0.000398s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.74% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.97% :     0.000387s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046813    76
 0.11% :     0.000051s :      1: add_recomputation
 0.44% :     0.000207s :      1: auto_monad
 0.05% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.74% :     0.000349s :      1: bootstrap
 0.04% :     0.000020s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000012s :      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% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
12.88% :     0.006031s :      1: jit_opt_a
 0.44% :     0.000206s :      1: jit_opt_after_cconv
 0.15% :     0.000070s :      1: jit_opt_b
 0.98% :     0.000461s :      1: loop_unroll
 1.10% :     0.000515s :      1: mutable_eliminate
 2.63% :     0.001231s :     26: opt.transform.jit_opt_a
 0.17% :     0.000078s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.07% :     0.000500s :      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
 0.40% :     0.000187s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000016s :      1: remove_dup_value
 1.30% :     0.000606s :      1: renormalize.infer
 0.84% :     0.000393s :      1: renormalize.specialize
 0.01% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.59% :     0.000275s :      1: rewriter_after_opt_a
 0.22% :     0.000104s :      1: rewriter_before_opt_a
 0.22% :     0.000102s :      1: symbol_engine_optimizer
25.43% :     0.011905s :      1: task_emit
49.38% :     0.023116s :      1: type_inference
 0.14% :     0.000068s :      1: validate


TotalTime = 0.0133001, [30]
[bootstrap]: 0.00028256
[type_inference]: 0.00509446
[event_method]: 2.399e-05
[auto_monad]: 6.372e-05
[graph_reusing]: 5.79999e-06
[pre_auto_parallel]: 2.50002e-06
[py_interpret_to_execute]: 4.077e-05
[rewriter_before_opt_a]: 8.749e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.00539512, [2]
    [Cycle 1]: 0.001737, [27]
        [switch_simplify]: 5.904e-05
        [loop_unroll]: 3.425e-05
        [a_1]: 0.00061459
        [with_stream_mark]: 1.702e-05
        [recompute_prepare]: 8.94e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 3.34001e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 6.98998e-06
        [updatestate_useless_node_eliminater]: 6.36e-06
        [accelerated_algorithm]: 6.58e-06
        [meta_shard_fg_expand]: 2.51998e-06
        [get_grad_eliminate_]: 6.19001e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.574e-05
        [j_node_and_user_rematch]: 9.82999e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 1.107e-05
        [inline_without_move]: 6.19001e-06
        [renormalize]: 0.00066334
        [add_forward_monad_depend]: 6.50002e-06
        [auto_monad_grad]: 2.26998e-06
        [auto_monad_eliminator]: 1.548e-05
        [cse]: 3.411e-05
        [replace_applicator]: 1.418e-05
    [Cycle 2]: 0.00036415, [27]
        [switch_simplify]: 6.84001e-06
        [loop_unroll]: 5.72999e-06
        [a_1]: 0.00011575
        [with_stream_mark]: 1.156e-05
        [recompute_prepare]: 6.11e-06
        [updatestate_depend_eliminate]: 3.29001e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [parameter_eliminate]: 1.11997e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.403e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 1.91998e-06
        [replace_old_param]: 9.34e-06
        [inline_without_move]: 5.40999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 1.37e-06
        [auto_monad_eliminator]: 6.66999e-06
        [cse]: 1.316e-05
        [replace_applicator]: 5.67999e-06
[py_interpret_to_execute_after_opt_a]: 1.039e-05
[rewriter_after_opt_a]: 3.713e-05
[convert_after_rewriter]: 7.28999e-06
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.00061863
[jit_opt_b]: 5.54e-05, [1]
    [Cycle 1]: 4.828e-05, [2]
        [frontend_op_eliminate]: 1.833e-05
        [inline_after_opt_a]: 1.829e-05
[cconv]: 2.74e-05
[loop_unroll]: 0.0004254
[jit_opt_after_cconv]: 0.00015179, [1]
    [Cycle 1]: 0.00014478, [11]
        [c_1]: 2.529e-05
        [parameter_eliminate]: 2.99001e-06
        [updatestate_depend_eliminate]: 5.82999e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [cse]: 2.206e-05
        [call_graph_tuple_transform]: 1.977e-05
        [tuple_list_get_item_eliminator]: 5.89e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.82001e-06
[remove_dup_value]: 1.486e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 6.46e-06
[add_recomputation]: 4.855e-05
[cse_after_recomputation]: 2.364e-05, [1]
    [Cycle 1]: 1.757e-05, [1]
        [cse]: 1.194e-05
[auto_monad_reorder]: 1.692e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 5.07e-06
[opt_after_jit_grad]: 0.00053183
[symbol_engine_optimizer]: 8.299e-05, [1]
    [Cycle 1]: 7.611e-05, [6]
        [build]: 4.92e-06
        [elim_shapecalc]: 9.51e-06
        [elim_not_effective]: 1.46e-05
        [opt_reshape]: 6.64001e-06
        [fold_const_symbol]: 1.018e-05
        [renormalize]: 3.9002e-07
[validate]: 4.708e-05
Sums
    bootstrap                            :     0.000283s :  3.00%
    type_inference                       :     0.005094s : 54.11%
    event_method                         :     0.000024s :  0.25%
    auto_monad                           :     0.000064s :  0.68%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.43%
    rewriter_before_opt_a                :     0.000087s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000066s :  0.70%
    jit_opt_a.loop_unroll                :     0.000040s :  0.42%
    jit_opt_a.a_1                        :     0.000730s :  7.76%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000015s :  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 :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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_        :     0.000012s :  0.13%
    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 :     0.000030s :  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          :     0.000020s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000663s :  7.05%
    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 :  0.24%
    jit_opt_a.cse                        :     0.000047s :  0.50%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000037s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000619s :  6.57%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.19%
    cconv                                :     0.000027s :  0.29%
    loop_unroll                          :     0.000425s :  4.52%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.27%
    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 :  0.04%
    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                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000049s :  0.52%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000017s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000532s :  5.65%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.10%
    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 :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000047s :  0.50%


Time group info:
------[substitution.]   0.000203    28
 1.24% :     0.000003s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000006s :      4: substitution.graph_param_transform
80.64% :     0.000164s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000005s :      4: substitution.replace_old_param
 8.25% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005026     2
78.52% :     0.003946s :      1: type_inference.infer
21.48% :     0.001080s :      1: type_inference.specialize
------[replace.]   0.000081     8
56.10% :     0.000046s :      4: replace.inline
43.90% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     8
91.63% :     0.000161s :      4: match.inline
 8.37% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   912
 1.36% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     13: predicate.addn_check_dump
 1.46% :     0.000002s :     13: predicate.addn_zero_filter
 1.94% :     0.000003s :     13: predicate.arithmetic_simplify
 1.53% :     0.000002s :     13: predicate.cast_eliminate
 0.42% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     13: predicate.compare_switch_simplify
 1.13% :     0.000002s :     13: predicate.depend_value_elim
 1.13% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.11% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      4: predicate.elim_not_effective
 0.72% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.22% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     13: predicate.environ_get_eliminate
 1.08% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000001s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 5.11% :     0.000007s :     29: predicate.inline
 0.88% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      8: predicate.less_batch_normalization
 1.98% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.96% :     0.000003s :     21: predicate.load_eliminater
 1.48% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.89% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     13: predicate.merge_addn
 1.10% :     0.000002s :     13: predicate.minmaximum_grad
 1.60% :     0.000002s :      4: predicate.mutable_eliminate
 0.49% :     0.000001s :      4: predicate.opt_reshape
 2.48% :     0.000004s :     21: predicate.partial_eliminate
 1.16% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     13: predicate.reduce_eliminate
 1.79% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.98% :     0.000003s :     25: predicate.replace_applicator
 0.65% :     0.000001s :      8: predicate.replace_old_param
 0.48% :     0.000001s :      4: predicate.reset_defer_inline
 1.15% :     0.000002s :     13: predicate.reshape_eliminate
 1.13% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     13: predicate.same_eliminate
 0.58% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.07% :     0.000002s :      8: predicate.special_op_eliminate
 0.94% :     0.000001s :      8: predicate.specialize_transform
 1.26% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     21: predicate.switch_defer_inline
 2.08% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.42% :     0.000011s :     67: predicate.switch_simplify
 1.18% :     0.000002s :     13: predicate.tile_eliminate
 1.30% :     0.000002s :     13: predicate.transpose_eliminate
 1.37% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.86% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.87% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000614    12
50.49% :     0.000310s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.02% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.49% :     0.000280s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015002    72
 0.34% :     0.000051s :      1: add_recomputation
 0.45% :     0.000067s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 1.99% :     0.000298s :      1: bootstrap
 0.20% :     0.000030s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.18% :     0.000028s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.08% :     0.000012s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
35.98% :     0.005398s :      1: jit_opt_a
 1.03% :     0.000154s :      1: jit_opt_after_cconv
 0.39% :     0.000058s :      1: jit_opt_b
 2.88% :     0.000433s :      1: loop_unroll
 4.18% :     0.000628s :      1: mutable_eliminate
 6.49% :     0.000973s :     26: opt.transform.jit_opt_a
 0.36% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000030s :      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
 0.19% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.60% :     0.000540s :      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% :     0.000044s :      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
 2.44% :     0.000365s :      1: renormalize.infer
 1.93% :     0.000290s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000040s :      1: rewriter_after_opt_a
 0.60% :     0.000091s :      1: rewriter_before_opt_a
 0.57% :     0.000086s :      1: symbol_engine_optimizer
34.04% :     0.005106s :      1: type_inference


TotalTime = 0.0140565, [30]
[bootstrap]: 0.0002937
[type_inference]: 0.00521198
[event_method]: 2.319e-05
[auto_monad]: 6.73e-05
[graph_reusing]: 6.58998e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 3.995e-05
[rewriter_before_opt_a]: 8.975e-05
[expand_dump_flag]: 3.63999e-06
[jit_opt_a]: 0.00588389, [2]
    [Cycle 1]: 0.00206857, [27]
        [switch_simplify]: 6.677e-05
        [loop_unroll]: 3.541e-05
        [a_1]: 0.00077302
        [with_stream_mark]: 2.492e-05
        [recompute_prepare]: 9.20999e-06
        [updatestate_depend_eliminate]: 4.32003e-06
        [updatestate_assign_eliminate]: 3.43e-06
        [updatestate_loads_eliminate]: 3.18e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 7.15e-06
        [updatestate_useless_node_eliminater]: 5.70001e-06
        [accelerated_algorithm]: 6.34999e-06
        [meta_shard_fg_expand]: 2.24999e-06
        [get_grad_eliminate_]: 5.85002e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.582e-05
        [j_node_and_user_rematch]: 9.92999e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 1.217e-05
        [inline_without_move]: 6.14001e-06
        [renormalize]: 0.00080155
        [add_forward_monad_depend]: 5.60001e-06
        [auto_monad_grad]: 2.94001e-06
        [auto_monad_eliminator]: 1.682e-05
        [cse]: 3.347e-05
        [replace_applicator]: 1.353e-05
    [Cycle 2]: 0.00036169, [27]
        [switch_simplify]: 6.59999e-06
        [loop_unroll]: 5.57001e-06
        [a_1]: 0.00011456
        [with_stream_mark]: 1.034e-05
        [recompute_prepare]: 5.79999e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 6.16998e-06
        [updatestate_useless_node_eliminater]: 5.70001e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.31998e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.389e-05
        [j_node_and_user_rematch]: 8.69e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 1.04e-05
        [inline_without_move]: 5.37001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 6.38998e-06
        [cse]: 1.345e-05
        [replace_applicator]: 5.70001e-06
[py_interpret_to_execute_after_opt_a]: 1.343e-05
[rewriter_after_opt_a]: 3.795e-05
[convert_after_rewriter]: 7.45998e-06
[order_py_execute_after_rewriter]: 5.30001e-06
[mutable_eliminate]: 0.00063805
[jit_opt_b]: 6.064e-05, [1]
    [Cycle 1]: 5.337e-05, [2]
        [frontend_op_eliminate]: 2.219e-05
        [inline_after_opt_a]: 1.833e-05
[cconv]: 2.798e-05
[loop_unroll]: 0.0004975
[jit_opt_after_cconv]: 0.00016405, [1]
    [Cycle 1]: 0.00015641, [11]
        [c_1]: 2.533e-05
        [parameter_eliminate]: 2.99999e-06
        [updatestate_depend_eliminate]: 7.26999e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [cse]: 2.79e-05
        [call_graph_tuple_transform]: 2.172e-05
        [tuple_list_get_item_eliminator]: 6.08002e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 6.23002e-06
[remove_dup_value]: 1.492e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 5.92001e-06
[add_recomputation]: 5.183e-05
[cse_after_recomputation]: 2.354e-05, [1]
    [Cycle 1]: 1.812e-05, [1]
        [cse]: 1.21e-05
[auto_monad_reorder]: 1.801e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 6.59999e-06
[opt_after_jit_grad]: 0.00049804
[symbol_engine_optimizer]: 0.00013458, [1]
    [Cycle 1]: 0.00012809, [6]
        [build]: 4.47e-06
        [elim_shapecalc]: 8.59002e-06
        [elim_not_effective]: 1.343e-05
        [opt_reshape]: 6.02001e-06
        [fold_const_symbol]: 6.642e-05
        [renormalize]: 4.80009e-07
[validate]: 4.336e-05
Sums
    bootstrap                            :     0.000294s :  2.94%
    type_inference                       :     0.005212s : 52.17%
    event_method                         :     0.000023s :  0.23%
    auto_monad                           :     0.000067s :  0.67%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.40%
    rewriter_before_opt_a                :     0.000090s :  0.90%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000073s :  0.73%
    jit_opt_a.loop_unroll                :     0.000041s :  0.41%
    jit_opt_a.a_1                        :     0.000888s :  8.88%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.35%
    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       :     0.000013s :  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       :     0.000004s :  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  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    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                :     0.000802s :  8.02%
    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      :     0.000023s :  0.23%
    jit_opt_a.cse                        :     0.000047s :  0.47%
    jit_opt_a.replace_applicator         :     0.000019s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.13%
    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                    :     0.000638s :  6.39%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.18%
    cconv                                :     0.000028s :  0.28%
    loop_unroll                          :     0.000498s :  4.98%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.25%
    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              :     0.000028s :  0.28%
    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 :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000052s :  0.52%
    cse_after_recomputation.cse          :     0.000012s :  0.12%
    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                   :     0.000498s :  4.98%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    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 :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000066s :  0.66%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.43%


Time group info:
------[substitution.]   0.000255    28
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.49% :     0.000006s :      4: substitution.graph_param_transform
83.68% :     0.000214s :      4: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.69% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.66% :     0.000007s :      4: substitution.replace_old_param
 6.96% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005137     2
81.71% :     0.004197s :      1: type_inference.infer
18.29% :     0.000939s :      1: type_inference.specialize
------[replace.]   0.000096     8
53.74% :     0.000052s :      4: replace.inline
46.26% :     0.000045s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     8
93.06% :     0.000211s :      4: match.inline
 6.94% :     0.000016s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   912
 1.23% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.10% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     13: predicate.addn_check_dump
 1.52% :     0.000002s :     13: predicate.addn_zero_filter
 1.86% :     0.000003s :     13: predicate.arithmetic_simplify
 1.41% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     13: predicate.compare_switch_simplify
 1.08% :     0.000002s :     13: predicate.depend_value_elim
 1.10% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.72% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      4: predicate.elim_not_effective
 0.48% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     13: predicate.environ_get_eliminate
 1.29% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.88% :     0.000001s :      8: predicate.get_grad_eliminate
 0.37% :     0.000001s :      4: predicate.graph_param_transform
 5.57% :     0.000008s :     29: predicate.inline
 0.89% :     0.000001s :      8: predicate.inline_without_move
 0.39% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      8: predicate.less_batch_normalization
 1.50% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     21: predicate.load_eliminater
 1.54% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.10% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     13: predicate.merge_addn
 1.08% :     0.000002s :     13: predicate.minmaximum_grad
 1.80% :     0.000003s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.47% :     0.000004s :     21: predicate.partial_eliminate
 1.12% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     13: predicate.reduce_eliminate
 1.60% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.89% :     0.000003s :     25: predicate.replace_applicator
 0.79% :     0.000001s :      8: predicate.replace_old_param
 0.31% :     0.000000s :      4: predicate.reset_defer_inline
 1.17% :     0.000002s :     13: predicate.reshape_eliminate
 1.15% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     13: predicate.same_eliminate
 0.58% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      8: predicate.special_op_eliminate
 0.92% :     0.000001s :      8: predicate.specialize_transform
 1.27% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.69% :     0.000004s :     21: predicate.switch_defer_inline
 2.07% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.63% :     0.000011s :     67: predicate.switch_simplify
 1.27% :     0.000002s :     13: predicate.tile_eliminate
 1.15% :     0.000002s :     13: predicate.transpose_eliminate
 1.36% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.23% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.95% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     13: predicate.value_based_eliminate
 0.33% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000651    12
47.30% :     0.000308s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.44% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.26% :     0.000295s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016127    72
 0.34% :     0.000055s :      1: add_recomputation
 0.44% :     0.000071s :      1: auto_monad
 0.13% :     0.000021s :      1: auto_monad_reorder
 1.91% :     0.000308s :      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% :     0.000008s :      1: environ_conv
 0.17% :     0.000027s :      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
36.51% :     0.005888s :      1: jit_opt_a
 1.04% :     0.000167s :      1: jit_opt_after_cconv
 0.39% :     0.000064s :      1: jit_opt_b
 3.15% :     0.000508s :      1: loop_unroll
 4.02% :     0.000648s :      1: mutable_eliminate
 7.05% :     0.001138s :     26: opt.transform.jit_opt_a
 0.35% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000034s :      4: opt.transform.jit_opt_b
 0.11% :     0.000017s :      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
 0.57% :     0.000091s :      4: opt.transform.symbol_engine_opt
 3.14% :     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% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000043s :      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
 2.76% :     0.000445s :      1: renormalize.infer
 2.16% :     0.000348s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000041s :      1: rewriter_after_opt_a
 0.57% :     0.000093s :      1: rewriter_before_opt_a
 0.85% :     0.000138s :      1: symbol_engine_optimizer
32.43% :     0.005230s :      1: type_inference


TotalTime = 0.041432, [33]
[bootstrap]: 0.00029343
[type_inference]: 0.021747
[event_method]: 2.009e-05
[auto_monad]: 0.00013901
[graph_reusing]: 5.37999e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 0.00012678
[rewriter_before_opt_a]: 9.322e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00587302, [2]
    [Cycle 1]: 0.00219102, [27]
        [switch_simplify]: 5.31e-05
        [loop_unroll]: 3.651e-05
        [a_1]: 0.00067328
        [with_stream_mark]: 1.404e-05
        [recompute_prepare]: 1.026e-05
        [updatestate_depend_eliminate]: 5.34e-06
        [updatestate_assign_eliminate]: 4.4e-06
        [updatestate_loads_eliminate]: 4.35e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 9.21998e-06
        [updatestate_useless_node_eliminater]: 1.135e-05
        [accelerated_algorithm]: 8.48001e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 8.15e-06
        [merge_forward]: 4.28001e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.622e-05
        [j_node_and_user_rematch]: 1.401e-05
        [meta_fg_expand]: 3.45998e-06
        [replace_old_param]: 1.06e-05
        [inline_without_move]: 8.27e-06
        [renormalize]: 0.00105348
        [add_forward_monad_depend]: 5.03002e-06
        [auto_monad_grad]: 2.13998e-06
        [auto_monad_eliminator]: 1.772e-05
        [cse]: 3.121e-05
        [replace_applicator]: 1.533e-05
    [Cycle 2]: 0.00046126, [27]
        [switch_simplify]: 9.51e-06
        [loop_unroll]: 8.01001e-06
        [a_1]: 0.00016516
        [with_stream_mark]: 1.094e-05
        [recompute_prepare]: 8.43999e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 4.55001e-06
        [updatestate_loads_eliminate]: 4.88001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 7.98999e-06
        [updatestate_useless_node_eliminater]: 1.055e-05
        [accelerated_algorithm]: 7.69002e-06
        [meta_shard_fg_expand]: 1.75001e-06
        [get_grad_eliminate_]: 7.48e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.78002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.573e-05
        [j_node_and_user_rematch]: 1.311e-05
        [meta_fg_expand]: 2.99999e-06
        [replace_old_param]: 9.14e-06
        [inline_without_move]: 7.53999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 9.94001e-06
        [cse]: 1.925e-05
        [replace_applicator]: 7.90998e-06
[py_interpret_to_execute_after_opt_a]: 1.144e-05
[rewriter_after_opt_a]: 0.00027889
[convert_after_rewriter]: 1.004e-05
[order_py_execute_after_rewriter]: 7.08e-06
[mutable_eliminate]: 0.00055572
[jit_opt_b]: 6.668e-05, [1]
    [Cycle 1]: 5.985e-05, [2]
        [frontend_op_eliminate]: 2.388e-05
        [inline_after_opt_a]: 2.438e-05
[cconv]: 2.018e-05
[loop_unroll]: 0.00045615
[jit_opt_after_cconv]: 0.00018589, [1]
    [Cycle 1]: 0.00017936, [11]
        [c_1]: 4.436e-05
        [parameter_eliminate]: 1.97999e-06
        [updatestate_depend_eliminate]: 7.48e-06
        [updatestate_assign_eliminate]: 4.4e-06
        [updatestate_loads_eliminate]: 3.80998e-06
        [cse]: 2.55e-05
        [call_graph_tuple_transform]: 2.179e-05
        [tuple_list_get_item_eliminator]: 8.62e-06
        [none_parameter_eliminate]: 1.27999e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 8.85999e-06
[remove_dup_value]: 1.467e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 9.25999e-06
[add_recomputation]: 5.626e-05
[cse_after_recomputation]: 2.897e-05, [1]
    [Cycle 1]: 2.309e-05, [1]
        [cse]: 1.692e-05
[auto_monad_reorder]: 2.258e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 4.05e-06
[opt_after_jit_grad]: 0.00049036
[symbol_engine_optimizer]: 0.00010164, [1]
    [Cycle 1]: 9.534e-05, [6]
        [build]: 9.84001e-06
        [elim_shapecalc]: 1.13e-05
        [elim_not_effective]: 1.937e-05
        [opt_reshape]: 9.51998e-06
        [fold_const_symbol]: 1.443e-05
        [renormalize]: 4.39992e-07
[validate]: 4.911e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.010544
[execute]: 7.9e-06
Sums
    bootstrap                            :     0.000293s :  0.78%
    type_inference                       :     0.021747s : 57.83%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000139s :  0.37%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000127s :  0.34%
    rewriter_before_opt_a                :     0.000093s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000063s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000838s :  2.23%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  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 :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001054s :  2.80%
    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      :     0.000028s :  0.07%
    jit_opt_a.cse                        :     0.000050s :  0.13%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000279s :  0.74%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000556s :  1.48%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000020s :  0.05%
    loop_unroll                          :     0.000456s :  1.21%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.12%
    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              :     0.000025s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000015s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000009s :  0.02%
    add_recomputation                    :     0.000056s :  0.15%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000023s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000490s :  1.30%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000049s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010544s : 28.04%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000240    47
 3.16% :     0.000008s :      2: substitution.depend_value_elim
 1.22% :     0.000003s :      4: substitution.elim_not_effective
 0.92% :     0.000002s :      4: substitution.fold_const_symbol
 2.56% :     0.000006s :      5: substitution.graph_param_transform
75.01% :     0.000180s :      4: substitution.inline
 2.04% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.61% :     0.000006s :      8: substitution.remove_not_recompute_node
 1.22% :     0.000003s :      2: substitution.replace_old_param
 4.20% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
 3.46% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.59% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021685     2
93.28% :     0.020228s :      1: type_inference.infer
 6.72% :     0.001457s :      1: type_inference.specialize
------[replace.]   0.000091     6
65.50% :     0.000060s :      4: replace.inline
34.50% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000186     6
95.17% :     0.000177s :      4: match.inline
 4.83% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1021
 1.46% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.35% :     0.000002s :     15: predicate.addn_zero_filter
 1.98% :     0.000003s :     15: predicate.arithmetic_simplify
 1.43% :     0.000002s :     15: predicate.cast_eliminate
 0.48% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     15: predicate.compare_switch_simplify
 1.39% :     0.000002s :     15: predicate.depend_value_elim
 1.31% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.53% :     0.000001s :      5: predicate.elim_not_effective
 0.55% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      5: predicate.fold_const_symbol
 1.13% :     0.000002s :     10: predicate.get_grad_eliminate
 0.61% :     0.000001s :      5: predicate.graph_param_transform
 4.79% :     0.000008s :     31: predicate.inline
 1.05% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :     10: predicate.less_batch_normalization
 1.67% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     22: predicate.load_eliminater
 1.25% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.43% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.24% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     15: predicate.merge_addn
 1.16% :     0.000002s :     15: predicate.minmaximum_grad
 1.49% :     0.000002s :      5: predicate.mutable_eliminate
 0.52% :     0.000001s :      5: predicate.opt_reshape
 2.34% :     0.000004s :     22: predicate.partial_eliminate
 1.32% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.72% :     0.000003s :     15: predicate.reduce_eliminate
 1.65% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.83% :     0.000003s :     27: predicate.replace_applicator
 0.70% :     0.000001s :     10: predicate.replace_old_param
 0.32% :     0.000001s :      5: predicate.reset_defer_inline
 1.26% :     0.000002s :     15: predicate.reshape_eliminate
 1.23% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     15: predicate.same_eliminate
 0.61% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :     10: predicate.special_op_eliminate
 1.05% :     0.000002s :     10: predicate.specialize_transform
 1.40% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.21% :     0.000004s :     21: predicate.switch_defer_inline
 2.17% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.35% :     0.000011s :     68: predicate.switch_simplify
 1.28% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.57% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.33% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000790    13
43.22% :     0.000342s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.73% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.05% :     0.000411s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043799    76
 0.14% :     0.000059s :      1: add_recomputation
 0.33% :     0.000144s :      1: auto_monad
 0.06% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.70% :     0.000306s :      1: bootstrap
 0.05% :     0.000023s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.05% :     0.000024s :      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
13.42% :     0.005877s :      1: jit_opt_a
 0.43% :     0.000189s :      1: jit_opt_after_cconv
 0.16% :     0.000070s :      1: jit_opt_b
 1.06% :     0.000463s :      1: loop_unroll
 1.29% :     0.000564s :      1: mutable_eliminate
 2.57% :     0.001127s :     26: opt.transform.jit_opt_a
 0.18% :     0.000080s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      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
 1.14% :     0.000498s :      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
 0.30% :     0.000131s :      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
 1.43% :     0.000626s :      1: renormalize.infer
 0.96% :     0.000419s :      1: renormalize.specialize
 0.01% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.65% :     0.000284s :      1: rewriter_after_opt_a
 0.22% :     0.000097s :      1: rewriter_before_opt_a
 0.24% :     0.000104s :      1: symbol_engine_optimizer
24.11% :     0.010559s :      1: task_emit
49.67% :     0.021757s :      1: type_inference
 0.16% :     0.000070s :      1: validate


TotalTime = 0.0149075, [30]
[bootstrap]: 0.00029244
[type_inference]: 0.00520631
[event_method]: 2.348e-05
[auto_monad]: 6.103e-05
[graph_reusing]: 5.69999e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 4.447e-05
[rewriter_before_opt_a]: 8.808e-05
[expand_dump_flag]: 3.01999e-06
[jit_opt_a]: 0.00662911, [2]
    [Cycle 1]: 0.00193451, [27]
        [switch_simplify]: 6.206e-05
        [loop_unroll]: 3.528e-05
        [a_1]: 0.00064886
        [with_stream_mark]: 1.831e-05
        [recompute_prepare]: 7.73001e-06
        [updatestate_depend_eliminate]: 4.13001e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 6.88998e-06
        [updatestate_useless_node_eliminater]: 5.94999e-06
        [accelerated_algorithm]: 6.46e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 6.02999e-06
        [merge_forward]: 3.06999e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.239e-05
        [j_node_and_user_rematch]: 1.045e-05
        [meta_fg_expand]: 2.52001e-06
        [replace_old_param]: 1.177e-05
        [inline_without_move]: 6.47001e-06
        [renormalize]: 0.00083119
        [add_forward_monad_depend]: 6.23e-06
        [auto_monad_grad]: 1.35001e-06
        [auto_monad_eliminator]: 1.32e-05
        [cse]: 2.536e-05
        [replace_applicator]: 1.596e-05
    [Cycle 2]: 0.00038445, [27]
        [switch_simplify]: 6.86999e-06
        [loop_unroll]: 6.04001e-06
        [a_1]: 0.00012004
        [with_stream_mark]: 1.053e-05
        [recompute_prepare]: 6.43998e-06
        [updatestate_depend_eliminate]: 3.2e-06
        [updatestate_assign_eliminate]: 2.65002e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.57001e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.72001e-06
        [accelerated_algorithm]: 6.89001e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 5.53002e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 2.15002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.455e-05
        [j_node_and_user_rematch]: 9.15001e-06
        [meta_fg_expand]: 2.88e-06
        [replace_old_param]: 1.048e-05
        [inline_without_move]: 5.74999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.29003e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 6.82002e-06
        [cse]: 1.659e-05
        [replace_applicator]: 6.49001e-06
[py_interpret_to_execute_after_opt_a]: 1.206e-05
[rewriter_after_opt_a]: 3.335e-05
[convert_after_rewriter]: 7.63999e-06
[order_py_execute_after_rewriter]: 4.52e-06
[mutable_eliminate]: 0.00067782
[jit_opt_b]: 6.078e-05, [1]
    [Cycle 1]: 5.305e-05, [2]
        [frontend_op_eliminate]: 2.192e-05
        [inline_after_opt_a]: 1.86e-05
[cconv]: 2.064e-05
[loop_unroll]: 0.00051267
[jit_opt_after_cconv]: 0.00024056, [1]
    [Cycle 1]: 0.00023232, [11]
        [c_1]: 8.866e-05
        [parameter_eliminate]: 3.76999e-06
        [updatestate_depend_eliminate]: 7.97e-06
        [updatestate_assign_eliminate]: 3.45e-06
        [updatestate_loads_eliminate]: 2.80002e-06
        [cse]: 3.184e-05
        [call_graph_tuple_transform]: 2.223e-05
        [tuple_list_get_item_eliminator]: 6.66e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 4.70027e-07
        [switch_simplify]: 6.93e-06
[remove_dup_value]: 1.138e-05
[partial_unused_args_eliminate]: 1.94e-06
[environ_conv]: 4.92999e-06
[add_recomputation]: 4.127e-05
[cse_after_recomputation]: 2.71e-05, [1]
    [Cycle 1]: 2.011e-05, [1]
        [cse]: 1.34e-05
[auto_monad_reorder]: 1.423e-05
[get_jit_bprop_graph]: 1.49998e-06
[rewriter_after_jit_bprop_graph]: 7.06999e-06
[opt_after_jit_grad]: 0.00056034
[symbol_engine_optimizer]: 8.078e-05, [1]
    [Cycle 1]: 7.357e-05, [6]
        [build]: 3.98001e-06
        [elim_shapecalc]: 9.20999e-06
        [elim_not_effective]: 1.477e-05
        [opt_reshape]: 6.58998e-06
        [fold_const_symbol]: 1.017e-05
        [renormalize]: 4.80009e-07
[validate]: 4.085e-05
Sums
    bootstrap                            :     0.000292s :  2.93%
    type_inference                       :     0.005206s : 52.22%
    event_method                         :     0.000023s :  0.24%
    auto_monad                           :     0.000061s :  0.61%
    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                :     0.000088s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.69%
    jit_opt_a.loop_unroll                :     0.000041s :  0.41%
    jit_opt_a.a_1                        :     0.000769s :  7.71%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.29%
    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       :     0.000013s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.12%
    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 :     0.000027s :  0.27%
    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        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000831s :  8.34%
    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      :     0.000020s :  0.20%
    jit_opt_a.cse                        :     0.000042s :  0.42%
    jit_opt_a.replace_applicator         :     0.000022s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.33%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000678s :  6.80%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.19%
    cconv                                :     0.000021s :  0.21%
    loop_unroll                          :     0.000513s :  5.14%
    jit_opt_after_cconv.c_1              :     0.000089s :  0.89%
    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              :     0.000032s :  0.32%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  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  :     0.000007s :  0.07%
    remove_dup_value                     :     0.000011s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000041s :  0.41%
    cse_after_recomputation.cse          :     0.000013s :  0.13%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000560s :  5.62%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.41%


Time group info:
------[substitution.]   0.000226    28
 0.63% :     0.000001s :      2: substitution.elim_not_effective
 0.44% :     0.000001s :      2: substitution.fold_const_symbol
 2.12% :     0.000005s :      4: substitution.graph_param_transform
83.86% :     0.000190s :      4: substitution.inline
 1.57% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.99% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.49% :     0.000006s :      4: substitution.replace_old_param
 6.91% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005137     2
80.69% :     0.004145s :      1: type_inference.infer
19.31% :     0.000992s :      1: type_inference.specialize
------[replace.]   0.000091     8
56.63% :     0.000051s :      4: replace.inline
43.37% :     0.000039s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     8
93.26% :     0.000187s :      4: match.inline
 6.74% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000145   912
 1.19% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     13: predicate.addn_check_dump
 1.34% :     0.000002s :     13: predicate.addn_zero_filter
 2.10% :     0.000003s :     13: predicate.arithmetic_simplify
 1.86% :     0.000003s :     13: predicate.cast_eliminate
 0.50% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.36% :     0.000002s :     13: predicate.compare_switch_simplify
 1.11% :     0.000002s :     13: predicate.depend_value_elim
 1.15% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.05% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      4: predicate.elim_not_effective
 0.63% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     13: predicate.environ_get_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      4: predicate.fold_const_symbol
 0.99% :     0.000001s :      8: predicate.get_grad_eliminate
 0.24% :     0.000000s :      4: predicate.graph_param_transform
 4.89% :     0.000007s :     29: predicate.inline
 0.84% :     0.000001s :      8: predicate.inline_without_move
 0.39% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      8: predicate.less_batch_normalization
 1.56% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.86% :     0.000003s :     21: predicate.load_eliminater
 1.46% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.81% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.13% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     13: predicate.merge_addn
 1.08% :     0.000002s :     13: predicate.minmaximum_grad
 2.32% :     0.000003s :      4: predicate.mutable_eliminate
 0.61% :     0.000001s :      4: predicate.opt_reshape
 2.24% :     0.000003s :     21: predicate.partial_eliminate
 1.20% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.90% :     0.000003s :     13: predicate.reduce_eliminate
 1.77% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.84% :     0.000003s :     25: predicate.replace_applicator
 0.74% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.23% :     0.000002s :     13: predicate.reshape_eliminate
 1.24% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     13: predicate.same_eliminate
 0.54% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      8: predicate.special_op_eliminate
 1.03% :     0.000001s :      8: predicate.specialize_transform
 1.39% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     21: predicate.switch_defer_inline
 2.02% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.86% :     0.000011s :     67: predicate.switch_simplify
 1.39% :     0.000002s :     13: predicate.tile_eliminate
 1.15% :     0.000002s :     13: predicate.transpose_eliminate
 1.38% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.75% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000640    12
43.61% :     0.000279s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.56% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.83% :     0.000325s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016907    72
 0.26% :     0.000044s :      1: add_recomputation
 0.38% :     0.000065s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.82% :     0.000307s :      1: bootstrap
 0.14% :     0.000023s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000029s :      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
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.23% :     0.006633s :      1: jit_opt_a
 1.45% :     0.000244s :      1: jit_opt_after_cconv
 0.38% :     0.000064s :      1: jit_opt_b
 3.09% :     0.000522s :      1: loop_unroll
 4.08% :     0.000690s :      1: mutable_eliminate
 6.03% :     0.001020s :     26: opt.transform.jit_opt_a
 0.72% :     0.000121s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000034s :      4: opt.transform.jit_opt_b
 0.10% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.37% :     0.000569s :      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% :     0.000048s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 2.73% :     0.000461s :      1: renormalize.infer
 2.14% :     0.000362s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000036s :      1: rewriter_after_opt_a
 0.54% :     0.000091s :      1: rewriter_before_opt_a
 0.50% :     0.000084s :      1: symbol_engine_optimizer
30.91% :     0.005227s :      1: type_inference


TotalTime = 0.0451602, [33]
[bootstrap]: 0.00029748
[type_inference]: 0.0228413
[event_method]: 2.087e-05
[auto_monad]: 0.00015352
[graph_reusing]: 7e-06
[pre_auto_parallel]: 3.23e-06
[py_interpret_to_execute]: 0.00016361
[rewriter_before_opt_a]: 9.777e-05
[expand_dump_flag]: 3.53e-06
[jit_opt_a]: 0.00675406, [2]
    [Cycle 1]: 0.00253344, [27]
        [switch_simplify]: 7.012e-05
        [loop_unroll]: 3.684e-05
        [a_1]: 0.00076563
        [with_stream_mark]: 2.041e-05
        [recompute_prepare]: 1.095e-05
        [updatestate_depend_eliminate]: 6.05002e-06
        [updatestate_assign_eliminate]: 5.64e-06
        [updatestate_loads_eliminate]: 4.58999e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 9.34e-06
        [updatestate_useless_node_eliminater]: 1.19e-05
        [accelerated_algorithm]: 8.24002e-06
        [meta_shard_fg_expand]: 2.88e-06
        [get_grad_eliminate_]: 8.04002e-06
        [merge_forward]: 5.16002e-06
        [cell_reuse_recompute_pass]: 1.36998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.048e-05
        [j_node_and_user_rematch]: 1.454e-05
        [meta_fg_expand]: 3.63e-06
        [replace_old_param]: 1.225e-05
        [inline_without_move]: 8.27998e-06
        [renormalize]: 0.00123061
        [add_forward_monad_depend]: 6.36e-06
        [auto_monad_grad]: 3.26001e-06
        [auto_monad_eliminator]: 2.346e-05
        [cse]: 3.8e-05
        [replace_applicator]: 1.836e-05
    [Cycle 2]: 0.00048208, [27]
        [switch_simplify]: 9.24e-06
        [loop_unroll]: 8.54e-06
        [a_1]: 0.00017439
        [with_stream_mark]: 1.291e-05
        [recompute_prepare]: 8.60999e-06
        [updatestate_depend_eliminate]: 5.09e-06
        [updatestate_assign_eliminate]: 5.14e-06
        [updatestate_loads_eliminate]: 4.12e-06
        [parameter_eliminate]: 1.65001e-06
        [specialize_transform]: 8.27998e-06
        [updatestate_useless_node_eliminater]: 1.103e-05
        [accelerated_algorithm]: 7.91001e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 7.66999e-06
        [merge_forward]: 4.70001e-06
        [cell_reuse_recompute_pass]: 2.17001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.644e-05
        [j_node_and_user_rematch]: 1.328e-05
        [meta_fg_expand]: 3.19001e-06
        [replace_old_param]: 1.107e-05
        [inline_without_move]: 7.58999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 1.006e-05
        [cse]: 2.023e-05
        [replace_applicator]: 7.73001e-06
[py_interpret_to_execute_after_opt_a]: 1.33e-05
[rewriter_after_opt_a]: 0.00029662
[convert_after_rewriter]: 1.11e-05
[order_py_execute_after_rewriter]: 7.86001e-06
[mutable_eliminate]: 0.0007332
[jit_opt_b]: 7.168e-05, [1]
    [Cycle 1]: 6.376e-05, [2]
        [frontend_op_eliminate]: 2.627e-05
        [inline_after_opt_a]: 2.557e-05
[cconv]: 2.691e-05
[loop_unroll]: 0.00048873
[jit_opt_after_cconv]: 0.00020265, [1]
    [Cycle 1]: 0.0001964, [11]
        [c_1]: 4.835e-05
        [parameter_eliminate]: 3.38999e-06
        [updatestate_depend_eliminate]: 8.84e-06
        [updatestate_assign_eliminate]: 4.69998e-06
        [updatestate_loads_eliminate]: 4.57998e-06
        [cse]: 3.337e-05
        [call_graph_tuple_transform]: 2.201e-05
        [tuple_list_get_item_eliminator]: 8.43999e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 1.04e-06
        [switch_simplify]: 9.15001e-06
[remove_dup_value]: 1.933e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 1.2e-05
[add_recomputation]: 6.704e-05
[cse_after_recomputation]: 3.03e-05, [1]
    [Cycle 1]: 2.406e-05, [1]
        [cse]: 1.807e-05
[auto_monad_reorder]: 2.56e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 6.65002e-06
[opt_after_jit_grad]: 0.00051534
[symbol_engine_optimizer]: 0.00010621, [1]
    [Cycle 1]: 9.933e-05, [6]
        [build]: 1.438e-05
        [elim_shapecalc]: 1.19e-05
        [elim_not_effective]: 1.946e-05
        [opt_reshape]: 9.02999e-06
        [fold_const_symbol]: 1.415e-05
        [renormalize]: 5.59987e-07
[validate]: 5.597e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.0118874
[execute]: 8.73001e-06
Sums
    bootstrap                            :     0.000297s :  0.73%
    type_inference                       :     0.022841s : 56.03%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000154s :  0.38%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000164s :  0.40%
    rewriter_before_opt_a                :     0.000098s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000079s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000940s :  2.31%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  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 :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001231s :  3.02%
    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      :     0.000034s :  0.08%
    jit_opt_a.cse                        :     0.000058s :  0.14%
    jit_opt_a.replace_applicator         :     0.000026s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000297s :  0.73%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000733s :  1.80%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000027s :  0.07%
    loop_unroll                          :     0.000489s :  1.20%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000067s :  0.16%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000515s :  1.26%
    symbol_engine_optimizer.build        :     0.000014s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000056s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011887s : 29.16%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000328    47
 3.27% :     0.000011s :      2: substitution.depend_value_elim
 1.02% :     0.000003s :      4: substitution.elim_not_effective
 0.71% :     0.000002s :      4: substitution.fold_const_symbol
 2.00% :     0.000007s :      5: substitution.graph_param_transform
76.82% :     0.000252s :      4: substitution.inline
 1.66% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.28% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.75% :     0.000006s :      2: substitution.replace_old_param
 4.15% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.33% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.01% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022772     2
93.09% :     0.021197s :      1: type_inference.infer
 6.91% :     0.001574s :      1: type_inference.specialize
------[replace.]   0.000097     6
68.15% :     0.000066s :      4: replace.inline
31.85% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
95.21% :     0.000248s :      4: match.inline
 4.79% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1021
 1.31% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.07% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.08% :     0.000002s :     15: predicate.addn_check_dump
 1.43% :     0.000002s :     15: predicate.addn_zero_filter
 2.03% :     0.000003s :     15: predicate.arithmetic_simplify
 1.45% :     0.000002s :     15: predicate.cast_eliminate
 0.48% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     15: predicate.compare_switch_simplify
 1.41% :     0.000002s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.66% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     15: predicate.environ_get_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.23% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 5.20% :     0.000009s :     31: predicate.inline
 1.06% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :     10: predicate.less_batch_normalization
 1.60% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.92% :     0.000003s :     22: predicate.load_eliminater
 1.42% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.46% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     15: predicate.merge_addn
 1.27% :     0.000002s :     15: predicate.minmaximum_grad
 1.45% :     0.000002s :      5: predicate.mutable_eliminate
 0.56% :     0.000001s :      5: predicate.opt_reshape
 2.43% :     0.000004s :     22: predicate.partial_eliminate
 1.21% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.74% :     0.000003s :     15: predicate.reduce_eliminate
 1.62% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.66% :     0.000003s :     27: predicate.replace_applicator
 0.75% :     0.000001s :     10: predicate.replace_old_param
 0.27% :     0.000000s :      5: predicate.reset_defer_inline
 1.30% :     0.000002s :     15: predicate.reshape_eliminate
 1.29% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.90% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     15: predicate.same_eliminate
 0.60% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.18% :     0.000002s :     10: predicate.special_op_eliminate
 1.05% :     0.000002s :     10: predicate.specialize_transform
 1.41% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.12% :     0.000004s :     21: predicate.switch_defer_inline
 1.84% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.56% :     0.000011s :     68: predicate.switch_simplify
 1.20% :     0.000002s :     15: predicate.tile_eliminate
 1.24% :     0.000002s :     15: predicate.transpose_eliminate
 1.59% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.95% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000003s :     15: predicate.value_based_eliminate
 0.48% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000829    13
41.06% :     0.000340s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.22% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.72% :     0.000429s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047840    76
 0.15% :     0.000070s :      1: add_recomputation
 0.33% :     0.000159s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000312s :      1: bootstrap
 0.06% :     0.000030s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000025s :      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% :     0.000010s :      1: graph_reusing
14.12% :     0.006757s :      1: jit_opt_a
 0.43% :     0.000206s :      1: jit_opt_after_cconv
 0.16% :     0.000075s :      1: jit_opt_b
 1.04% :     0.000498s :      1: loop_unroll
 1.55% :     0.000743s :      1: mutable_eliminate
 2.63% :     0.001257s :     26: opt.transform.jit_opt_a
 0.18% :     0.000084s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000045s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.10% :     0.000525s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.35% :     0.000168s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.54% :     0.000736s :      1: renormalize.infer
 1.01% :     0.000484s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000302s :      1: rewriter_after_opt_a
 0.21% :     0.000102s :      1: rewriter_before_opt_a
 0.23% :     0.000109s :      1: symbol_engine_optimizer
24.88% :     0.011903s :      1: task_emit
47.77% :     0.022854s :      1: type_inference
 0.17% :     0.000081s :      1: validate


TotalTime = 0.0134825, [30]
[bootstrap]: 0.00032875
[type_inference]: 0.00494156
[event_method]: 2.935e-05
[auto_monad]: 5.579e-05
[graph_reusing]: 4.53999e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 4.046e-05
[rewriter_before_opt_a]: 8.522e-05
[expand_dump_flag]: 2.90002e-06
[jit_opt_a]: 0.00563331, [2]
    [Cycle 1]: 0.00179, [27]
        [switch_simplify]: 5.848e-05
        [loop_unroll]: 3.548e-05
        [a_1]: 0.00061607
        [with_stream_mark]: 1.875e-05
        [recompute_prepare]: 8.17e-06
        [updatestate_depend_eliminate]: 4.27e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 7.06999e-06
        [updatestate_useless_node_eliminater]: 5.84e-06
        [accelerated_algorithm]: 6.24001e-06
        [meta_shard_fg_expand]: 1.59998e-06
        [get_grad_eliminate_]: 5.99999e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.286e-05
        [j_node_and_user_rematch]: 9.66e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 1.081e-05
        [inline_without_move]: 6.34999e-06
        [renormalize]: 0.00073791
        [add_forward_monad_depend]: 5.53002e-06
        [auto_monad_grad]: 1.87001e-06
        [auto_monad_eliminator]: 1.326e-05
        [cse]: 2.429e-05
        [replace_applicator]: 1.363e-05
    [Cycle 2]: 0.00036999, [27]
        [switch_simplify]: 7.05998e-06
        [loop_unroll]: 5.87001e-06
        [a_1]: 0.00011398
        [with_stream_mark]: 1.252e-05
        [recompute_prepare]: 5.51e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.61e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.80001e-06
        [get_grad_eliminate_]: 5.83997e-06
        [merge_forward]: 3.15998e-06
        [cell_reuse_recompute_pass]: 1.63002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.425e-05
        [j_node_and_user_rematch]: 8.33999e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 9.72001e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 9.89996e-07
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 6.68998e-06
        [cse]: 1.585e-05
        [replace_applicator]: 6.19999e-06
[py_interpret_to_execute_after_opt_a]: 1.239e-05
[rewriter_after_opt_a]: 3.465e-05
[convert_after_rewriter]: 6.81001e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00059567
[jit_opt_b]: 5.689e-05, [1]
    [Cycle 1]: 4.935e-05, [2]
        [frontend_op_eliminate]: 1.917e-05
        [inline_after_opt_a]: 1.835e-05
[cconv]: 2.49e-05
[loop_unroll]: 0.00049993
[jit_opt_after_cconv]: 0.00016688, [1]
    [Cycle 1]: 0.00015996, [11]
        [c_1]: 2.686e-05
        [parameter_eliminate]: 2.79999e-06
        [updatestate_depend_eliminate]: 6.93998e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [cse]: 2.779e-05
        [call_graph_tuple_transform]: 2.283e-05
        [tuple_list_get_item_eliminator]: 6.12999e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 6.86001e-06
[remove_dup_value]: 1.457e-05
[partial_unused_args_eliminate]: 2.21998e-06
[environ_conv]: 5.81e-06
[add_recomputation]: 4.795e-05
[cse_after_recomputation]: 2.561e-05, [1]
    [Cycle 1]: 1.979e-05, [1]
        [cse]: 1.347e-05
[auto_monad_reorder]: 1.514e-05
[get_jit_bprop_graph]: 2.21e-06
[rewriter_after_jit_bprop_graph]: 6.84999e-06
[opt_after_jit_grad]: 0.00052767
[symbol_engine_optimizer]: 8.038e-05, [1]
    [Cycle 1]: 7.38e-05, [6]
        [build]: 3.83999e-06
        [elim_shapecalc]: 9.17001e-06
        [elim_not_effective]: 1.474e-05
        [opt_reshape]: 6.68e-06
        [fold_const_symbol]: 9.34998e-06
        [renormalize]: 5.50004e-07
[validate]: 4.517e-05
Sums
    bootstrap                            :     0.000329s :  3.49%
    type_inference                       :     0.004942s : 52.50%
    event_method                         :     0.000029s :  0.31%
    auto_monad                           :     0.000056s :  0.59%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.43%
    rewriter_before_opt_a                :     0.000085s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000066s :  0.70%
    jit_opt_a.loop_unroll                :     0.000041s :  0.44%
    jit_opt_a.a_1                        :     0.000730s :  7.76%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.33%
    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 :     0.000006s :  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       :     0.000013s :  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       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.13%
    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 :  0.29%
    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          :     0.000021s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000738s :  7.84%
    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                        :     0.000040s :  0.43%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    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                    :     0.000596s :  6.33%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.19%
    cconv                                :     0.000025s :  0.26%
    loop_unroll                          :     0.000500s :  5.31%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.29%
    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              :     0.000028s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.07%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.51%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    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                   :     0.000528s :  5.61%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    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  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.48%


Time group info:
------[substitution.]   0.000206    28
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.95% :     0.000006s :      4: substitution.graph_param_transform
81.67% :     0.000168s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.63% :     0.000005s :      4: substitution.replace_old_param
 7.23% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004875     2
79.99% :     0.003899s :      1: type_inference.infer
20.01% :     0.000975s :      1: type_inference.specialize
------[replace.]   0.000081     8
55.60% :     0.000045s :      4: replace.inline
44.40% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     8
92.70% :     0.000166s :      4: match.inline
 7.30% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.18% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     13: predicate.addn_check_dump
 1.23% :     0.000002s :     13: predicate.addn_zero_filter
 2.06% :     0.000003s :     13: predicate.arithmetic_simplify
 1.30% :     0.000002s :     13: predicate.cast_eliminate
 0.46% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     13: predicate.compare_switch_simplify
 1.18% :     0.000002s :     13: predicate.depend_value_elim
 1.11% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      4: predicate.elim_not_effective
 0.53% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.93% :     0.000001s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 5.31% :     0.000007s :     29: predicate.inline
 0.93% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.13% :     0.000002s :      8: predicate.less_batch_normalization
 1.65% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     21: predicate.load_eliminater
 1.89% :     0.000003s :      4: predicate.loop_unroll_after_grad
 4.08% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     13: predicate.merge_addn
 1.10% :     0.000002s :     13: predicate.minmaximum_grad
 1.71% :     0.000002s :      4: predicate.mutable_eliminate
 0.51% :     0.000001s :      4: predicate.opt_reshape
 2.45% :     0.000003s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     13: predicate.reduce_eliminate
 1.73% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.92% :     0.000003s :     25: predicate.replace_applicator
 0.72% :     0.000001s :      8: predicate.replace_old_param
 0.30% :     0.000000s :      4: predicate.reset_defer_inline
 1.26% :     0.000002s :     13: predicate.reshape_eliminate
 1.13% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      8: predicate.special_op_eliminate
 1.00% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.61% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     21: predicate.switch_defer_inline
 2.17% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.69% :     0.000011s :     67: predicate.switch_simplify
 1.13% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.64% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.91% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000629    12
49.09% :     0.000309s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.50% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.41% :     0.000285s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015268    72
 0.33% :     0.000051s :      1: add_recomputation
 0.39% :     0.000059s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.29% :     0.000349s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000028s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.22% :     0.000033s :      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
36.92% :     0.005637s :      1: jit_opt_a
 1.11% :     0.000170s :      1: jit_opt_after_cconv
 0.39% :     0.000059s :      1: jit_opt_b
 3.33% :     0.000509s :      1: loop_unroll
 3.96% :     0.000604s :      1: mutable_eliminate
 6.36% :     0.000970s :     26: opt.transform.jit_opt_a
 0.39% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000031s :      4: opt.transform.jit_opt_b
 0.12% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.51% :     0.000536s :      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% :     0.000043s :      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
 2.68% :     0.000410s :      1: renormalize.infer
 2.10% :     0.000320s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000038s :      1: rewriter_after_opt_a
 0.58% :     0.000088s :      1: rewriter_before_opt_a
 0.54% :     0.000083s :      1: symbol_engine_optimizer
32.46% :     0.004956s :      1: type_inference


TotalTime = 0.0404886, [33]
[bootstrap]: 0.00033527
[type_inference]: 0.0216918
[event_method]: 1.868e-05
[auto_monad]: 0.00013171
[graph_reusing]: 4.38999e-06
[pre_auto_parallel]: 2.00002e-06
[py_interpret_to_execute]: 0.00013048
[rewriter_before_opt_a]: 8.981e-05
[expand_dump_flag]: 2.61e-06
[jit_opt_a]: 0.00585062, [2]
    [Cycle 1]: 0.00226546, [27]
        [switch_simplify]: 5.379e-05
        [loop_unroll]: 3.942e-05
        [a_1]: 0.00066641
        [with_stream_mark]: 1.46e-05
        [recompute_prepare]: 1.003e-05
        [updatestate_depend_eliminate]: 5.60001e-06
        [updatestate_assign_eliminate]: 4.52e-06
        [updatestate_loads_eliminate]: 3.97998e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 9.09e-06
        [updatestate_useless_node_eliminater]: 1.107e-05
        [accelerated_algorithm]: 8.38001e-06
        [meta_shard_fg_expand]: 2.11998e-06
        [get_grad_eliminate_]: 8.35001e-06
        [merge_forward]: 4.38999e-06
        [cell_reuse_recompute_pass]: 8.99978e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.601e-05
        [j_node_and_user_rematch]: 8.724e-05
        [meta_fg_expand]: 3.84002e-06
        [replace_old_param]: 1.262e-05
        [inline_without_move]: 8.74003e-06
        [renormalize]: 0.00104116
        [add_forward_monad_depend]: 4.33001e-06
        [auto_monad_grad]: 1.91e-06
        [auto_monad_eliminator]: 1.811e-05
        [cse]: 3.114e-05
        [replace_applicator]: 1.642e-05
    [Cycle 2]: 0.00050399, [27]
        [switch_simplify]: 1.005e-05
        [loop_unroll]: 9.14e-06
        [a_1]: 0.00018291
        [with_stream_mark]: 1.101e-05
        [recompute_prepare]: 8.67e-06
        [updatestate_depend_eliminate]: 5.76e-06
        [updatestate_assign_eliminate]: 4.94e-06
        [updatestate_loads_eliminate]: 4.42e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 8.89998e-06
        [updatestate_useless_node_eliminater]: 1.135e-05
        [accelerated_algorithm]: 9.02999e-06
        [meta_shard_fg_expand]: 1.87999e-06
        [get_grad_eliminate_]: 8.27998e-06
        [merge_forward]: 4.85999e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.703e-05
        [j_node_and_user_rematch]: 1.33e-05
        [meta_fg_expand]: 2.76e-06
        [replace_old_param]: 1.043e-05
        [inline_without_move]: 8.61002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 1.17999e-06
        [auto_monad_eliminator]: 1.058e-05
        [cse]: 2.256e-05
        [replace_applicator]: 8.95001e-06
[py_interpret_to_execute_after_opt_a]: 1.071e-05
[rewriter_after_opt_a]: 0.00027989
[convert_after_rewriter]: 1.13e-05
[order_py_execute_after_rewriter]: 7.87e-06
[mutable_eliminate]: 0.00054523
[jit_opt_b]: 7.458e-05, [1]
    [Cycle 1]: 6.752e-05, [2]
        [frontend_op_eliminate]: 2.674e-05
        [inline_after_opt_a]: 2.725e-05
[cconv]: 1.963e-05
[loop_unroll]: 0.00047879
[jit_opt_after_cconv]: 0.00018856, [1]
    [Cycle 1]: 0.00018196, [11]
        [c_1]: 4.56e-05
        [parameter_eliminate]: 2.74001e-06
        [updatestate_depend_eliminate]: 7.68001e-06
        [updatestate_assign_eliminate]: 4.12e-06
        [updatestate_loads_eliminate]: 3.86999e-06
        [cse]: 2.617e-05
        [call_graph_tuple_transform]: 2.128e-05
        [tuple_list_get_item_eliminator]: 8.18999e-06
        [none_parameter_eliminate]: 9.50007e-07
        [renormalize]: 4.00003e-07
        [switch_simplify]: 8.76002e-06
[remove_dup_value]: 1.6e-05
[partial_unused_args_eliminate]: 1.87001e-06
[environ_conv]: 9.10001e-06
[add_recomputation]: 5.253e-05
[cse_after_recomputation]: 2.98e-05, [1]
    [Cycle 1]: 2.43e-05, [1]
        [cse]: 1.848e-05
[auto_monad_reorder]: 2.1e-05
[get_jit_bprop_graph]: 1.39998e-06
[rewriter_after_jit_bprop_graph]: 4.25e-06
[opt_after_jit_grad]: 0.0005329
[symbol_engine_optimizer]: 9.728e-05, [1]
    [Cycle 1]: 9.093e-05, [6]
        [build]: 8.51002e-06
        [elim_shapecalc]: 1.158e-05
        [elim_not_effective]: 1.864e-05
        [opt_reshape]: 8.86002e-06
        [fold_const_symbol]: 1.382e-05
        [renormalize]: 4.19997e-07
[validate]: 4.512e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00957871
[execute]: 7.57002e-06
Sums
    bootstrap                            :     0.000335s :  0.91%
    type_inference                       :     0.021692s : 59.01%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000132s :  0.36%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000130s :  0.35%
    rewriter_before_opt_a                :     0.000090s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.17%
    jit_opt_a.loop_unroll                :     0.000049s :  0.13%
    jit_opt_a.a_1                        :     0.000849s :  2.31%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.05%
    jit_opt_a.merge_forward              :     0.000009s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000101s :  0.27%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000017s :  0.05%
    jit_opt_a.renormalize                :     0.001041s :  2.83%
    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      :     0.000029s :  0.08%
    jit_opt_a.cse                        :     0.000054s :  0.15%
    jit_opt_a.replace_applicator         :     0.000025s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000280s :  0.76%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000545s :  1.48%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.07%
    cconv                                :     0.000020s :  0.05%
    loop_unroll                          :     0.000479s :  1.30%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.12%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000016s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000009s :  0.02%
    add_recomputation                    :     0.000053s :  0.14%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000021s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000533s :  1.45%
    symbol_engine_optimizer.build        :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009579s : 26.06%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000232    47
 3.67% :     0.000009s :      2: substitution.depend_value_elim
 1.00% :     0.000002s :      4: substitution.elim_not_effective
 0.83% :     0.000002s :      4: substitution.fold_const_symbol
 2.40% :     0.000006s :      5: substitution.graph_param_transform
74.32% :     0.000172s :      4: substitution.inline
 2.05% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.86% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.57% :     0.000004s :      2: substitution.replace_old_param
 4.10% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
 3.56% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.66% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021632     2
93.78% :     0.020286s :      1: type_inference.infer
 6.22% :     0.001346s :      1: type_inference.specialize
------[replace.]   0.000088     6
64.35% :     0.000057s :      4: replace.inline
35.65% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     6
95.27% :     0.000169s :      4: match.inline
 4.73% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1021
 1.31% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.24% :     0.000002s :     15: predicate.addn_check_dump
 1.36% :     0.000002s :     15: predicate.addn_zero_filter
 2.07% :     0.000004s :     15: predicate.arithmetic_simplify
 1.23% :     0.000002s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     15: predicate.compare_switch_simplify
 1.46% :     0.000002s :     15: predicate.depend_value_elim
 1.18% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.91% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.54% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     15: predicate.environ_get_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.17% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 4.77% :     0.000008s :     31: predicate.inline
 1.21% :     0.000002s :     10: predicate.inline_without_move
 0.46% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :     10: predicate.less_batch_normalization
 1.50% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.90% :     0.000003s :     22: predicate.load_eliminater
 1.42% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.79% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.27% :     0.000002s :     15: predicate.merge_addn
 1.17% :     0.000002s :     15: predicate.minmaximum_grad
 1.51% :     0.000003s :      5: predicate.mutable_eliminate
 0.52% :     0.000001s :      5: predicate.opt_reshape
 2.29% :     0.000004s :     22: predicate.partial_eliminate
 1.22% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.79% :     0.000003s :     15: predicate.reduce_eliminate
 1.48% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.86% :     0.000003s :     27: predicate.replace_applicator
 0.56% :     0.000001s :     10: predicate.replace_old_param
 0.39% :     0.000001s :      5: predicate.reset_defer_inline
 1.39% :     0.000002s :     15: predicate.reshape_eliminate
 1.22% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     15: predicate.same_eliminate
 0.62% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.10% :     0.000002s :     10: predicate.special_op_eliminate
 1.20% :     0.000002s :     10: predicate.specialize_transform
 1.45% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.35% :     0.000004s :     21: predicate.switch_defer_inline
 1.91% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.34% :     0.000011s :     68: predicate.switch_simplify
 1.39% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.59% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.86% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.23% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000003s :     15: predicate.value_based_eliminate
 0.42% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000823    13
46.21% :     0.000380s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.72% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.08% :     0.000404s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042952    76
 0.13% :     0.000055s :      1: add_recomputation
 0.32% :     0.000137s :      1: auto_monad
 0.06% :     0.000024s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.82% :     0.000352s :      1: bootstrap
 0.05% :     0.000023s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.05% :     0.000022s :      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
13.63% :     0.005854s :      1: jit_opt_a
 0.45% :     0.000192s :      1: jit_opt_after_cconv
 0.18% :     0.000078s :      1: jit_opt_b
 1.14% :     0.000488s :      1: loop_unroll
 1.29% :     0.000555s :      1: mutable_eliminate
 2.86% :     0.001227s :     26: opt.transform.jit_opt_a
 0.19% :     0.000080s :      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% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.26% :     0.000541s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.31% :     0.000135s :      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
 1.40% :     0.000601s :      1: renormalize.infer
 1.01% :     0.000433s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.66% :     0.000285s :      1: rewriter_after_opt_a
 0.22% :     0.000094s :      1: rewriter_before_opt_a
 0.23% :     0.000100s :      1: symbol_engine_optimizer
22.34% :     0.009594s :      1: task_emit
50.53% :     0.021703s :      1: type_inference
 0.15% :     0.000066s :      1: validate


TotalTime = 0.0440275, [33]
[bootstrap]: 0.00029063
[type_inference]: 0.0223994
[event_method]: 1.987e-05
[auto_monad]: 0.00015304
[graph_reusing]: 6.58e-06
[pre_auto_parallel]: 2.79001e-06
[py_interpret_to_execute]: 0.00012434
[rewriter_before_opt_a]: 9.813e-05
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.00672355, [2]
    [Cycle 1]: 0.00277734, [27]
        [switch_simplify]: 6.091e-05
        [loop_unroll]: 3.59e-05
        [a_1]: 0.00105533
        [with_stream_mark]: 2.564e-05
        [recompute_prepare]: 1.457e-05
        [updatestate_depend_eliminate]: 6.83e-06
        [updatestate_assign_eliminate]: 5.76e-06
        [updatestate_loads_eliminate]: 5.42999e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 1.009e-05
        [updatestate_useless_node_eliminater]: 1.435e-05
        [accelerated_algorithm]: 8.80999e-06
        [meta_shard_fg_expand]: 3.14001e-06
        [get_grad_eliminate_]: 8.85999e-06
        [merge_forward]: 5.61e-06
        [cell_reuse_recompute_pass]: 2.31998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.133e-05
        [j_node_and_user_rematch]: 1.413e-05
        [meta_fg_expand]: 3.5e-06
        [replace_old_param]: 1.291e-05
        [inline_without_move]: 8.25e-06
        [renormalize]: 0.00116374
        [add_forward_monad_depend]: 6.43e-06
        [auto_monad_grad]: 2.62001e-06
        [auto_monad_eliminator]: 2.301e-05
        [cse]: 4.475e-05
        [replace_applicator]: 2.019e-05
    [Cycle 2]: 0.00049441, [27]
        [switch_simplify]: 9.34998e-06
        [loop_unroll]: 8.08001e-06
        [a_1]: 0.00017547
        [with_stream_mark]: 1.327e-05
        [recompute_prepare]: 8.44998e-06
        [updatestate_depend_eliminate]: 5.74e-06
        [updatestate_assign_eliminate]: 4.82e-06
        [updatestate_loads_eliminate]: 4.47e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 8.17e-06
        [updatestate_useless_node_eliminater]: 1.031e-05
        [accelerated_algorithm]: 8.38001e-06
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 7.66999e-06
        [merge_forward]: 5.04e-06
        [cell_reuse_recompute_pass]: 2.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.782e-05
        [j_node_and_user_rematch]: 1.347e-05
        [meta_fg_expand]: 2.91e-06
        [replace_old_param]: 1.123e-05
        [inline_without_move]: 7.67998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.55999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 1.186e-05
        [cse]: 2.271e-05
        [replace_applicator]: 8.15e-06
[py_interpret_to_execute_after_opt_a]: 1.491e-05
[rewriter_after_opt_a]: 0.00029106
[convert_after_rewriter]: 1.321e-05
[order_py_execute_after_rewriter]: 7.03e-06
[mutable_eliminate]: 0.00070848
[jit_opt_b]: 6.899e-05, [1]
    [Cycle 1]: 6.139e-05, [2]
        [frontend_op_eliminate]: 2.466e-05
        [inline_after_opt_a]: 2.529e-05
[cconv]: 2.806e-05
[loop_unroll]: 0.00050258
[jit_opt_after_cconv]: 0.00021253, [1]
    [Cycle 1]: 0.00020578, [11]
        [c_1]: 4.949e-05
        [parameter_eliminate]: 2.64001e-06
        [updatestate_depend_eliminate]: 1.04e-05
        [updatestate_assign_eliminate]: 4.65001e-06
        [updatestate_loads_eliminate]: 4.66002e-06
        [cse]: 3.527e-05
        [call_graph_tuple_transform]: 2.387e-05
        [tuple_list_get_item_eliminator]: 9.10999e-06
        [none_parameter_eliminate]: 1.74998e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 9.20001e-06
[remove_dup_value]: 1.827e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 1.213e-05
[add_recomputation]: 6.809e-05
[cse_after_recomputation]: 2.97e-05, [1]
    [Cycle 1]: 2.365e-05, [1]
        [cse]: 1.701e-05
[auto_monad_reorder]: 2.618e-05
[get_jit_bprop_graph]: 1.93002e-06
[rewriter_after_jit_bprop_graph]: 6.68e-06
[opt_after_jit_grad]: 0.00051968
[symbol_engine_optimizer]: 0.00010595, [1]
    [Cycle 1]: 9.943e-05, [6]
        [build]: 1.329e-05
        [elim_shapecalc]: 1.232e-05
        [elim_not_effective]: 1.937e-05
        [opt_reshape]: 9.22001e-06
        [fold_const_symbol]: 1.382e-05
        [renormalize]: 8.80013e-07
[validate]: 5.806e-05
[backend_pass]: 1.18001e-06
[task_emit]: 0.0112732
[execute]: 8.79e-06
Sums
    bootstrap                            :     0.000291s :  0.73%
    type_inference                       :     0.022399s : 56.13%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000153s :  0.38%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000124s :  0.31%
    rewriter_before_opt_a                :     0.000098s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000070s :  0.18%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.001231s :  3.08%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000024s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001164s :  2.92%
    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      :     0.000035s :  0.09%
    jit_opt_a.cse                        :     0.000067s :  0.17%
    jit_opt_a.replace_applicator         :     0.000028s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000291s :  0.73%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000708s :  1.78%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000028s :  0.07%
    loop_unroll                          :     0.000503s :  1.26%
    jit_opt_after_cconv.c_1              :     0.000049s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000018s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000068s :  0.17%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000520s :  1.30%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000058s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011273s : 28.25%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000273    47
 4.02% :     0.000011s :      2: substitution.depend_value_elim
 0.99% :     0.000003s :      4: substitution.elim_not_effective
 0.80% :     0.000002s :      4: substitution.fold_const_symbol
 2.68% :     0.000007s :      5: substitution.graph_param_transform
72.32% :     0.000198s :      4: substitution.inline
 1.98% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.80% :     0.000008s :      8: substitution.remove_not_recompute_node
 2.15% :     0.000006s :      2: substitution.replace_old_param
 4.89% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.45% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.92% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022328     2
93.77% :     0.020936s :      1: type_inference.infer
 6.23% :     0.001392s :      1: type_inference.specialize
------[replace.]   0.000090     6
65.39% :     0.000059s :      4: replace.inline
34.61% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000207     6
94.13% :     0.000195s :      4: match.inline
 5.87% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1021
 1.50% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.25% :     0.000002s :     15: predicate.addn_zero_filter
 2.40% :     0.000004s :     15: predicate.arithmetic_simplify
 1.36% :     0.000002s :     15: predicate.cast_eliminate
 0.47% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.33% :     0.000002s :     15: predicate.depend_value_elim
 1.09% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.86% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.29% :     0.000001s :      5: predicate.elim_not_effective
 0.52% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.06% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     15: predicate.environ_get_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.33% :     0.000002s :     10: predicate.get_grad_eliminate
 0.35% :     0.000001s :      5: predicate.graph_param_transform
 5.48% :     0.000010s :     31: predicate.inline
 0.98% :     0.000002s :     10: predicate.inline_without_move
 0.40% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :     10: predicate.less_batch_normalization
 1.39% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.80% :     0.000003s :     22: predicate.load_eliminater
 1.59% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.15% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 1.66% :     0.000003s :      5: predicate.mutable_eliminate
 0.66% :     0.000001s :      5: predicate.opt_reshape
 2.54% :     0.000005s :     22: predicate.partial_eliminate
 1.18% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.58% :     0.000003s :     15: predicate.reduce_eliminate
 1.39% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.67% :     0.000003s :     27: predicate.replace_applicator
 0.79% :     0.000001s :     10: predicate.replace_old_param
 0.26% :     0.000000s :      5: predicate.reset_defer_inline
 1.17% :     0.000002s :     15: predicate.reshape_eliminate
 1.26% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.95% :     0.000002s :     10: predicate.special_op_eliminate
 1.22% :     0.000002s :     10: predicate.specialize_transform
 1.54% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.27% :     0.000004s :     21: predicate.switch_defer_inline
 1.81% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.29% :     0.000011s :     68: predicate.switch_simplify
 1.28% :     0.000002s :     15: predicate.tile_eliminate
 1.34% :     0.000002s :     15: predicate.transpose_eliminate
 1.67% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 2.02% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.76% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 4.15% :     0.000007s :     32: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000824    13
44.96% :     0.000371s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.75% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.29% :     0.000414s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046938    76
 0.15% :     0.000071s :      1: add_recomputation
 0.34% :     0.000158s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000307s :      1: bootstrap
 0.07% :     0.000031s :      1: cconv
 0.04% :     0.000017s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000023s :      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% :     0.000009s :      1: graph_reusing
14.33% :     0.006727s :      1: jit_opt_a
 0.46% :     0.000215s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 1.09% :     0.000513s :      1: loop_unroll
 1.53% :     0.000719s :      1: mutable_eliminate
 3.29% :     0.001546s :     26: opt.transform.jit_opt_a
 0.19% :     0.000088s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.13% :     0.000530s :      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
 0.27% :     0.000129s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.51% :     0.000709s :      1: renormalize.infer
 0.95% :     0.000445s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000296s :      1: rewriter_after_opt_a
 0.22% :     0.000102s :      1: rewriter_before_opt_a
 0.23% :     0.000109s :      1: symbol_engine_optimizer
24.06% :     0.011292s :      1: task_emit
47.75% :     0.022411s :      1: type_inference
 0.18% :     0.000083s :      1: validate


TotalTime = 0.0479405, [33]
[bootstrap]: 0.00031495
[type_inference]: 0.027512
[event_method]: 2.109e-05
[auto_monad]: 0.00016028
[graph_reusing]: 6.71999e-06
[pre_auto_parallel]: 3.40998e-06
[py_interpret_to_execute]: 0.00012358
[rewriter_before_opt_a]: 0.00010046
[expand_dump_flag]: 3.30998e-06
[jit_opt_a]: 0.00619424, [2]
    [Cycle 1]: 0.0022681, [27]
        [switch_simplify]: 5.641e-05
        [loop_unroll]: 3.687e-05
        [a_1]: 0.00067369
        [with_stream_mark]: 1.646e-05
        [recompute_prepare]: 1.008e-05
        [updatestate_depend_eliminate]: 6.01e-06
        [updatestate_assign_eliminate]: 4.70001e-06
        [updatestate_loads_eliminate]: 3.99002e-06
        [parameter_eliminate]: 1.34998e-06
        [specialize_transform]: 8.75001e-06
        [updatestate_useless_node_eliminater]: 1.173e-05
        [accelerated_algorithm]: 8.43001e-06
        [meta_shard_fg_expand]: 2.13998e-06
        [get_grad_eliminate_]: 7.98001e-06
        [merge_forward]: 4.10998e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.71e-05
        [j_node_and_user_rematch]: 1.4e-05
        [meta_fg_expand]: 3.35e-06
        [replace_old_param]: 1.072e-05
        [inline_without_move]: 8.30999e-06
        [renormalize]: 0.00111499
        [add_forward_monad_depend]: 4.99e-06
        [auto_monad_grad]: 1.86e-06
        [auto_monad_eliminator]: 1.836e-05
        [cse]: 3.116e-05
        [replace_applicator]: 1.638e-05
    [Cycle 2]: 0.00046512, [27]
        [switch_simplify]: 8.64e-06
        [loop_unroll]: 8.02e-06
        [a_1]: 0.00016455
        [with_stream_mark]: 1.234e-05
        [recompute_prepare]: 7.98001e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 4.67998e-06
        [updatestate_loads_eliminate]: 4.01001e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 7.91001e-06
        [updatestate_useless_node_eliminater]: 1.054e-05
        [accelerated_algorithm]: 7.98999e-06
        [meta_shard_fg_expand]: 1.86998e-06
        [get_grad_eliminate_]: 7.28e-06
        [merge_forward]: 4.28001e-06
        [cell_reuse_recompute_pass]: 2.54001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.626e-05
        [j_node_and_user_rematch]: 1.261e-05
        [meta_fg_expand]: 2.79999e-06
        [replace_old_param]: 1.051e-05
        [inline_without_move]: 7.52998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 1.98002e-06
        [auto_monad_eliminator]: 9.77999e-06
        [cse]: 2.119e-05
        [replace_applicator]: 7.7e-06
[py_interpret_to_execute_after_opt_a]: 1.244e-05
[rewriter_after_opt_a]: 0.00029644
[convert_after_rewriter]: 1.001e-05
[order_py_execute_after_rewriter]: 6.66999e-06
[mutable_eliminate]: 0.00060694
[jit_opt_b]: 6.772e-05, [1]
    [Cycle 1]: 6.042e-05, [2]
        [frontend_op_eliminate]: 2.342e-05
        [inline_after_opt_a]: 2.492e-05
[cconv]: 2.242e-05
[loop_unroll]: 0.00045492
[jit_opt_after_cconv]: 0.00019132, [1]
    [Cycle 1]: 0.00018455, [11]
        [c_1]: 4.419e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 8.02998e-06
        [updatestate_assign_eliminate]: 4.35e-06
        [updatestate_loads_eliminate]: 3.73999e-06
        [cse]: 2.963e-05
        [call_graph_tuple_transform]: 2.149e-05
        [tuple_list_get_item_eliminator]: 9.02e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 8.47e-06
[remove_dup_value]: 1.556e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 9.49999e-06
[add_recomputation]: 5.56e-05
[cse_after_recomputation]: 2.917e-05, [1]
    [Cycle 1]: 2.395e-05, [1]
        [cse]: 1.806e-05
[auto_monad_reorder]: 2.225e-05
[get_jit_bprop_graph]: 1.73002e-06
[rewriter_after_jit_bprop_graph]: 5.59998e-06
[opt_after_jit_grad]: 0.00050262
[symbol_engine_optimizer]: 0.00010055, [1]
    [Cycle 1]: 9.33e-05, [6]
        [build]: 1.079e-05
        [elim_shapecalc]: 1.207e-05
        [elim_not_effective]: 1.861e-05
        [opt_reshape]: 8.87999e-06
        [fold_const_symbol]: 1.346e-05
        [renormalize]: 5.00004e-07
[validate]: 5.281e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.0107847
[execute]: 9.86998e-06
Sums
    bootstrap                            :     0.000315s :  0.72%
    type_inference                       :     0.027512s : 62.77%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000160s :  0.37%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000124s :  0.28%
    rewriter_before_opt_a                :     0.000100s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.15%
    jit_opt_a.loop_unroll                :     0.000045s :  0.10%
    jit_opt_a.a_1                        :     0.000838s :  1.91%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  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 :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001115s :  2.54%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.06%
    jit_opt_a.cse                        :     0.000052s :  0.12%
    jit_opt_a.replace_applicator         :     0.000024s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000296s :  0.68%
    convert_after_rewriter               :     0.000010s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000607s :  1.38%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000022s :  0.05%
    loop_unroll                          :     0.000455s :  1.04%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.10%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000016s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000009s :  0.02%
    add_recomputation                    :     0.000056s :  0.13%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000022s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000503s :  1.15%
    symbol_engine_optimizer.build        :     0.000011s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000053s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010785s : 24.61%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000249    47
 3.73% :     0.000009s :      2: substitution.depend_value_elim
 1.25% :     0.000003s :      4: substitution.elim_not_effective
 0.79% :     0.000002s :      4: substitution.fold_const_symbol
 2.44% :     0.000006s :      5: substitution.graph_param_transform
74.38% :     0.000185s :      4: substitution.inline
 1.91% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.73% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.68% :     0.000004s :      2: substitution.replace_old_param
 4.12% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
 3.51% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.45% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.027434     2
94.68% :     0.025975s :      1: type_inference.infer
 5.32% :     0.001458s :      1: type_inference.specialize
------[replace.]   0.000087     6
65.91% :     0.000057s :      4: replace.inline
34.09% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     6
95.27% :     0.000182s :      4: match.inline
 4.73% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000167  1021
 1.35% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     15: predicate.addn_check_dump
 1.29% :     0.000002s :     15: predicate.addn_zero_filter
 1.91% :     0.000003s :     15: predicate.arithmetic_simplify
 1.43% :     0.000002s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     15: predicate.compare_switch_simplify
 1.43% :     0.000002s :     15: predicate.depend_value_elim
 1.21% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.90% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.65% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     15: predicate.environ_get_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      5: predicate.fold_const_symbol
 1.06% :     0.000002s :     10: predicate.get_grad_eliminate
 0.31% :     0.000001s :      5: predicate.graph_param_transform
 5.08% :     0.000008s :     31: predicate.inline
 1.07% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :     10: predicate.less_batch_normalization
 1.51% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.08% :     0.000003s :     22: predicate.load_eliminater
 1.44% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.51% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.21% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.39% :     0.000002s :      5: predicate.mutable_eliminate
 0.61% :     0.000001s :      5: predicate.opt_reshape
 2.30% :     0.000004s :     22: predicate.partial_eliminate
 1.20% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.86% :     0.000003s :     15: predicate.reduce_eliminate
 1.51% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.77% :     0.000003s :     27: predicate.replace_applicator
 0.60% :     0.000001s :     10: predicate.replace_old_param
 0.46% :     0.000001s :      5: predicate.reset_defer_inline
 1.38% :     0.000002s :     15: predicate.reshape_eliminate
 1.33% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.94% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.93% :     0.000002s :     10: predicate.special_op_eliminate
 1.17% :     0.000002s :     10: predicate.specialize_transform
 1.47% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.11% :     0.000004s :     21: predicate.switch_defer_inline
 1.96% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.27% :     0.000010s :     68: predicate.switch_simplify
 1.39% :     0.000002s :     15: predicate.tile_eliminate
 1.23% :     0.000002s :     15: predicate.transpose_eliminate
 1.49% :     0.000002s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.45% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.38% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000003s :     15: predicate.value_based_eliminate
 0.47% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000835    13
46.27% :     0.000386s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.72% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.02% :     0.000409s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050369    76
 0.12% :     0.000059s :      1: add_recomputation
 0.33% :     0.000166s :      1: auto_monad
 0.05% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000338s :      1: bootstrap
 0.05% :     0.000025s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.06% :     0.000031s :      1: cse_after_recomputation
 0.02% :     0.000012s :      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% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
12.30% :     0.006197s :      1: jit_opt_a
 0.39% :     0.000194s :      1: jit_opt_after_cconv
 0.14% :     0.000070s :      1: jit_opt_b
 0.92% :     0.000463s :      1: loop_unroll
 1.22% :     0.000617s :      1: mutable_eliminate
 2.25% :     0.001131s :     26: opt.transform.jit_opt_a
 0.16% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000041s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.02% :     0.000512s :      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% :     0.000006s :      1: pre_auto_parallel
 0.25% :     0.000128s :      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
 1.32% :     0.000666s :      1: renormalize.infer
 0.88% :     0.000441s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.60% :     0.000301s :      1: rewriter_after_opt_a
 0.21% :     0.000105s :      1: rewriter_before_opt_a
 0.21% :     0.000103s :      1: symbol_engine_optimizer
21.45% :     0.010803s :      1: task_emit
54.65% :     0.027529s :      1: type_inference
 0.15% :     0.000076s :      1: validate


TotalTime = 0.0413792, [33]
[bootstrap]: 0.00029735
[type_inference]: 0.0218561
[event_method]: 1.884e-05
[auto_monad]: 0.00013293
[graph_reusing]: 4e-06
[pre_auto_parallel]: 2.32001e-06
[py_interpret_to_execute]: 0.00011345
[rewriter_before_opt_a]: 8.947e-05
[expand_dump_flag]: 2.44001e-06
[jit_opt_a]: 0.00572429, [2]
    [Cycle 1]: 0.00199176, [27]
        [switch_simplify]: 5.033e-05
        [loop_unroll]: 3.602e-05
        [a_1]: 0.00063747
        [with_stream_mark]: 1.276e-05
        [recompute_prepare]: 9.99999e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 4.57e-06
        [updatestate_loads_eliminate]: 3.97e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 9.24e-06
        [updatestate_useless_node_eliminater]: 1.098e-05
        [accelerated_algorithm]: 8.54e-06
        [meta_shard_fg_expand]: 2.30002e-06
        [get_grad_eliminate_]: 8.1e-06
        [merge_forward]: 4.35999e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.65e-05
        [j_node_and_user_rematch]: 1.318e-05
        [meta_fg_expand]: 3.21999e-06
        [replace_old_param]: 1.088e-05
        [inline_without_move]: 8.32003e-06
        [renormalize]: 0.00090538
        [add_forward_monad_depend]: 5.16002e-06
        [auto_monad_grad]: 1.81003e-06
        [auto_monad_eliminator]: 1.611e-05
        [cse]: 2.924e-05
        [replace_applicator]: 1.656e-05
    [Cycle 2]: 0.00045486, [27]
        [switch_simplify]: 8.64e-06
        [loop_unroll]: 7.73001e-06
        [a_1]: 0.00016287
        [with_stream_mark]: 1.113e-05
        [recompute_prepare]: 8.24998e-06
        [updatestate_depend_eliminate]: 4.75999e-06
        [updatestate_assign_eliminate]: 4.55001e-06
        [updatestate_loads_eliminate]: 3.91001e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 7.98999e-06
        [updatestate_useless_node_eliminater]: 1.032e-05
        [accelerated_algorithm]: 7.41999e-06
        [meta_shard_fg_expand]: 1.93997e-06
        [get_grad_eliminate_]: 7.23999e-06
        [merge_forward]: 3.86001e-06
        [cell_reuse_recompute_pass]: 1.86e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.545e-05
        [j_node_and_user_rematch]: 1.231e-05
        [meta_fg_expand]: 2.60002e-06
        [replace_old_param]: 1.03e-05
        [inline_without_move]: 7.56001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 9.79984e-07
        [auto_monad_eliminator]: 9.93998e-06
        [cse]: 1.927e-05
        [replace_applicator]: 7.99002e-06
[py_interpret_to_execute_after_opt_a]: 1.069e-05
[rewriter_after_opt_a]: 0.00027212
[convert_after_rewriter]: 1.054e-05
[order_py_execute_after_rewriter]: 7.35e-06
[mutable_eliminate]: 0.00056271
[jit_opt_b]: 6.601e-05, [1]
    [Cycle 1]: 5.979e-05, [2]
        [frontend_op_eliminate]: 2.358e-05
        [inline_after_opt_a]: 2.431e-05
[cconv]: 2.151e-05
[loop_unroll]: 0.00055519
[jit_opt_after_cconv]: 0.00020148, [1]
    [Cycle 1]: 0.00019454, [11]
        [c_1]: 4.845e-05
        [parameter_eliminate]: 2.48998e-06
        [updatestate_depend_eliminate]: 6.91999e-06
        [updatestate_assign_eliminate]: 4.57e-06
        [updatestate_loads_eliminate]: 4.3e-06
        [cse]: 3.088e-05
        [call_graph_tuple_transform]: 2.229e-05
        [tuple_list_get_item_eliminator]: 8.89998e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 9.59e-06
[remove_dup_value]: 1.749e-05
[partial_unused_args_eliminate]: 2.49999e-06
[environ_conv]: 1.049e-05
[add_recomputation]: 6.084e-05
[cse_after_recomputation]: 2.927e-05, [1]
    [Cycle 1]: 2.364e-05, [1]
        [cse]: 1.661e-05
[auto_monad_reorder]: 2.387e-05
[get_jit_bprop_graph]: 1.74e-06
[rewriter_after_jit_bprop_graph]: 5.86e-06
[opt_after_jit_grad]: 0.00052832
[symbol_engine_optimizer]: 0.00010739, [1]
    [Cycle 1]: 0.00010004, [6]
        [build]: 1.321e-05
        [elim_shapecalc]: 1.163e-05
        [elim_not_effective]: 1.96e-05
        [opt_reshape]: 9.13002e-06
        [fold_const_symbol]: 1.421e-05
        [renormalize]: 4.30009e-07
[validate]: 5.327e-05
[backend_pass]: 1.06002e-06
[task_emit]: 0.010365
[execute]: 7.68001e-06
Sums
    bootstrap                            :     0.000297s :  0.79%
    type_inference                       :     0.021856s : 58.31%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000133s :  0.35%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000113s :  0.30%
    rewriter_before_opt_a                :     0.000089s :  0.24%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000059s :  0.16%
    jit_opt_a.loop_unroll                :     0.000044s :  0.12%
    jit_opt_a.a_1                        :     0.000800s :  2.14%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000021s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  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 :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000025s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000905s :  2.42%
    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      :     0.000026s :  0.07%
    jit_opt_a.cse                        :     0.000049s :  0.13%
    jit_opt_a.replace_applicator         :     0.000025s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000272s :  0.73%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000563s :  1.50%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000022s :  0.06%
    loop_unroll                          :     0.000555s :  1.48%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.13%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.03%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000061s :  0.16%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000024s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000528s :  1.41%
    symbol_engine_optimizer.build        :     0.000013s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000053s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010365s : 27.65%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000225    47
 3.71% :     0.000008s :      2: substitution.depend_value_elim
 1.16% :     0.000003s :      4: substitution.elim_not_effective
 0.90% :     0.000002s :      4: substitution.fold_const_symbol
 2.72% :     0.000006s :      5: substitution.graph_param_transform
73.69% :     0.000166s :      4: substitution.inline
 1.91% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.86% :     0.000006s :      8: substitution.remove_not_recompute_node
 1.70% :     0.000004s :      2: substitution.replace_old_param
 3.59% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
 3.92% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.85% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021797     2
93.78% :     0.020441s :      1: type_inference.infer
 6.22% :     0.001355s :      1: type_inference.specialize
------[replace.]   0.000083     6
66.01% :     0.000055s :      4: replace.inline
33.99% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000170     6
95.90% :     0.000163s :      4: match.inline
 4.10% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1021
 1.31% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.25% :     0.000002s :     15: predicate.addn_zero_filter
 1.92% :     0.000003s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.19% :     0.000002s :     15: predicate.compare_switch_simplify
 1.32% :     0.000002s :     15: predicate.depend_value_elim
 1.23% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     15: predicate.environ_get_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      5: predicate.fold_const_symbol
 1.15% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 4.97% :     0.000008s :     31: predicate.inline
 1.07% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :     10: predicate.less_batch_normalization
 1.53% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.09% :     0.000003s :     22: predicate.load_eliminater
 1.61% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.54% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000002s :     15: predicate.merge_addn
 1.26% :     0.000002s :     15: predicate.minmaximum_grad
 1.32% :     0.000002s :      5: predicate.mutable_eliminate
 0.61% :     0.000001s :      5: predicate.opt_reshape
 2.35% :     0.000004s :     22: predicate.partial_eliminate
 1.19% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.62% :     0.000003s :     15: predicate.reduce_eliminate
 1.53% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.75% :     0.000003s :     27: predicate.replace_applicator
 0.58% :     0.000001s :     10: predicate.replace_old_param
 0.32% :     0.000001s :      5: predicate.reset_defer_inline
 1.30% :     0.000002s :     15: predicate.reshape_eliminate
 1.36% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.06% :     0.000002s :     10: predicate.special_op_eliminate
 1.20% :     0.000002s :     10: predicate.specialize_transform
 1.46% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.38% :     0.000004s :     21: predicate.switch_defer_inline
 2.14% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.42% :     0.000011s :     68: predicate.switch_simplify
 1.39% :     0.000002s :     15: predicate.tile_eliminate
 1.25% :     0.000002s :     15: predicate.transpose_eliminate
 1.52% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.96% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.39% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     15: predicate.value_based_eliminate
 0.46% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000781    13
46.50% :     0.000363s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.91% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.58% :     0.000380s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043564    76
 0.15% :     0.000064s :      1: add_recomputation
 0.32% :     0.000138s :      1: auto_monad
 0.06% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.72% :     0.000313s :      1: bootstrap
 0.06% :     0.000024s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.05% :     0.000022s :      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% :     0.000007s :      1: graph_reusing
13.15% :     0.005727s :      1: jit_opt_a
 0.47% :     0.000205s :      1: jit_opt_after_cconv
 0.16% :     0.000069s :      1: jit_opt_b
 1.30% :     0.000564s :      1: loop_unroll
 1.31% :     0.000571s :      1: mutable_eliminate
 2.49% :     0.001086s :     26: opt.transform.jit_opt_a
 0.19% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      4: opt.transform.jit_opt_b
 0.05% :     0.000020s :      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% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.23% :     0.000537s :      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
 0.27% :     0.000117s :      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
 1.21% :     0.000526s :      1: renormalize.infer
 0.86% :     0.000373s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.64% :     0.000277s :      1: rewriter_after_opt_a
 0.21% :     0.000093s :      1: rewriter_before_opt_a
 0.25% :     0.000110s :      1: symbol_engine_optimizer
23.83% :     0.010382s :      1: task_emit
50.20% :     0.021869s :      1: type_inference
 0.17% :     0.000076s :      1: validate


TotalTime = 0.042124, [33]
[bootstrap]: 0.00028987
[type_inference]: 0.023774
[event_method]: 2.094e-05
[auto_monad]: 0.00014887
[graph_reusing]: 7.46001e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 0.00011959
[rewriter_before_opt_a]: 9.707e-05
[expand_dump_flag]: 2.86e-06
[jit_opt_a]: 0.00590603, [2]
    [Cycle 1]: 0.00204679, [27]
        [switch_simplify]: 5.982e-05
        [loop_unroll]: 3.576e-05
        [a_1]: 0.00066877
        [with_stream_mark]: 1.514e-05
        [recompute_prepare]: 1.011e-05
        [updatestate_depend_eliminate]: 5.37001e-06
        [updatestate_assign_eliminate]: 4.63001e-06
        [updatestate_loads_eliminate]: 4.42998e-06
        [parameter_eliminate]: 1.04003e-06
        [specialize_transform]: 8.92999e-06
        [updatestate_useless_node_eliminater]: 1.099e-05
        [accelerated_algorithm]: 8.43001e-06
        [meta_shard_fg_expand]: 1.96998e-06
        [get_grad_eliminate_]: 8.45001e-06
        [merge_forward]: 4.29002e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.61e-05
        [j_node_and_user_rematch]: 1.308e-05
        [meta_fg_expand]: 3.24001e-06
        [replace_old_param]: 1.063e-05
        [inline_without_move]: 8.25e-06
        [renormalize]: 0.00090559
        [add_forward_monad_depend]: 4.62998e-06
        [auto_monad_grad]: 1.63002e-06
        [auto_monad_eliminator]: 1.78e-05
        [cse]: 2.937e-05
        [replace_applicator]: 1.524e-05
    [Cycle 2]: 0.00045963, [27]
        [switch_simplify]: 9.57001e-06
        [loop_unroll]: 7.82998e-06
        [a_1]: 0.00016475
        [with_stream_mark]: 1.081e-05
        [recompute_prepare]: 8.12e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 5.04e-06
        [updatestate_loads_eliminate]: 3.92998e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 8.24998e-06
        [updatestate_useless_node_eliminater]: 1.088e-05
        [accelerated_algorithm]: 7.95e-06
        [meta_shard_fg_expand]: 1.69e-06
        [get_grad_eliminate_]: 7.5e-06
        [merge_forward]: 4.35e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.583e-05
        [j_node_and_user_rematch]: 1.255e-05
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 9.86e-06
        [inline_without_move]: 7.61999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.02e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 1.008e-05
        [cse]: 1.922e-05
        [replace_applicator]: 7.93999e-06
[py_interpret_to_execute_after_opt_a]: 1.112e-05
[rewriter_after_opt_a]: 0.00026856
[convert_after_rewriter]: 9.94999e-06
[order_py_execute_after_rewriter]: 6.16e-06
[mutable_eliminate]: 0.00051076
[jit_opt_b]: 6.576e-05, [1]
    [Cycle 1]: 5.95e-05, [2]
        [frontend_op_eliminate]: 2.335e-05
        [inline_after_opt_a]: 2.461e-05
[cconv]: 1.634e-05
[loop_unroll]: 0.00052831
[jit_opt_after_cconv]: 0.00018219, [1]
    [Cycle 1]: 0.00017595, [11]
        [c_1]: 4.381e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 7.71999e-06
        [updatestate_assign_eliminate]: 4.13001e-06
        [updatestate_loads_eliminate]: 3.8e-06
        [cse]: 2.553e-05
        [call_graph_tuple_transform]: 1.941e-05
        [tuple_list_get_item_eliminator]: 8.21002e-06
        [none_parameter_eliminate]: 9.40025e-07
        [renormalize]: 3.60014e-07
        [switch_simplify]: 7.8e-06
[remove_dup_value]: 1.314e-05
[partial_unused_args_eliminate]: 1.72001e-06
[environ_conv]: 8.19998e-06
[add_recomputation]: 5.199e-05
[cse_after_recomputation]: 3.004e-05, [1]
    [Cycle 1]: 2.444e-05, [1]
        [cse]: 1.81e-05
[auto_monad_reorder]: 1.966e-05
[get_jit_bprop_graph]: 1.55001e-06
[rewriter_after_jit_bprop_graph]: 4.33001e-06
[opt_after_jit_grad]: 0.00048748
[symbol_engine_optimizer]: 9.461e-05, [1]
    [Cycle 1]: 8.85e-05, [6]
        [build]: 8.74e-06
        [elim_shapecalc]: 1.141e-05
        [elim_not_effective]: 1.743e-05
        [opt_reshape]: 8.63001e-06
        [fold_const_symbol]: 1.253e-05
        [renormalize]: 4.7998e-07
[validate]: 4.304e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00919065
[execute]: 5.85002e-06
Sums
    bootstrap                            :     0.000290s :  0.76%
    type_inference                       :     0.023774s : 62.38%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000149s :  0.39%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000120s :  0.31%
    rewriter_before_opt_a                :     0.000097s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.18%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000834s :  2.19%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  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 :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000906s :  2.38%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.07%
    jit_opt_a.cse                        :     0.000049s :  0.13%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000269s :  0.70%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000511s :  1.34%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000016s :  0.04%
    loop_unroll                          :     0.000528s :  1.39%
    jit_opt_after_cconv.c_1              :     0.000044s :  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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000013s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000008s :  0.02%
    add_recomputation                    :     0.000052s :  0.14%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000020s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000487s :  1.28%
    symbol_engine_optimizer.build        :     0.000009s :  0.02%
    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 :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009191s : 24.11%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000241    47
 3.31% :     0.000008s :      2: substitution.depend_value_elim
 0.83% :     0.000002s :      4: substitution.elim_not_effective
 0.66% :     0.000002s :      4: substitution.fold_const_symbol
 1.98% :     0.000005s :      5: substitution.graph_param_transform
77.49% :     0.000187s :      4: substitution.inline
 1.65% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.69% :     0.000006s :      8: substitution.remove_not_recompute_node
 1.24% :     0.000003s :      2: substitution.replace_old_param
 3.49% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
 3.21% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.45% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023707     2
94.10% :     0.022308s :      1: type_inference.infer
 5.90% :     0.001399s :      1: type_inference.specialize
------[replace.]   0.000088     6
66.54% :     0.000058s :      4: replace.inline
33.46% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     6
96.18% :     0.000184s :      4: match.inline
 3.82% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000167  1021
 1.21% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.08% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.24% :     0.000002s :     15: predicate.addn_check_dump
 1.38% :     0.000002s :     15: predicate.addn_zero_filter
 1.93% :     0.000003s :     15: predicate.arithmetic_simplify
 1.29% :     0.000002s :     15: predicate.cast_eliminate
 0.47% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     15: predicate.compare_switch_simplify
 1.47% :     0.000002s :     15: predicate.depend_value_elim
 1.23% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.66% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     15: predicate.environ_get_eliminate
 1.26% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.11% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 5.27% :     0.000009s :     31: predicate.inline
 1.08% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :     10: predicate.less_batch_normalization
 1.41% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     22: predicate.load_eliminater
 1.30% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.45% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.04% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     15: predicate.merge_addn
 1.19% :     0.000002s :     15: predicate.minmaximum_grad
 1.35% :     0.000002s :      5: predicate.mutable_eliminate
 0.57% :     0.000001s :      5: predicate.opt_reshape
 2.33% :     0.000004s :     22: predicate.partial_eliminate
 1.20% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.64% :     0.000003s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.87% :     0.000003s :     27: predicate.replace_applicator
 0.62% :     0.000001s :     10: predicate.replace_old_param
 0.30% :     0.000001s :      5: predicate.reset_defer_inline
 1.30% :     0.000002s :     15: predicate.reshape_eliminate
 1.32% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     15: predicate.same_eliminate
 0.61% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.24% :     0.000002s :     10: predicate.special_op_eliminate
 1.14% :     0.000002s :     10: predicate.specialize_transform
 1.66% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.21% :     0.000004s :     21: predicate.switch_defer_inline
 2.09% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.60% :     0.000011s :     68: predicate.switch_simplify
 1.27% :     0.000002s :     15: predicate.tile_eliminate
 1.35% :     0.000002s :     15: predicate.transpose_eliminate
 1.51% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.19% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000858    13
51.23% :     0.000439s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.53% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.24% :     0.000379s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044335    76
 0.12% :     0.000055s :      1: add_recomputation
 0.35% :     0.000154s :      1: auto_monad
 0.05% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000304s :      1: bootstrap
 0.04% :     0.000019s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.02% :     0.000011s :      1: environ_conv
 0.06% :     0.000025s :      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% :     0.000010s :      1: graph_reusing
13.33% :     0.005909s :      1: jit_opt_a
 0.42% :     0.000185s :      1: jit_opt_after_cconv
 0.15% :     0.000069s :      1: jit_opt_b
 1.21% :     0.000538s :      1: loop_unroll
 1.17% :     0.000519s :      1: mutable_eliminate
 2.54% :     0.001126s :     26: opt.transform.jit_opt_a
 0.17% :     0.000076s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      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
 0.07% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000047s :      4: opt.transform.symbol_engine_opt
 1.12% :     0.000496s :      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
 0.28% :     0.000124s :      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
 1.18% :     0.000525s :      1: renormalize.infer
 0.84% :     0.000374s :      1: renormalize.specialize
 0.01% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.62% :     0.000273s :      1: rewriter_after_opt_a
 0.23% :     0.000101s :      1: rewriter_before_opt_a
 0.22% :     0.000097s :      1: symbol_engine_optimizer
20.76% :     0.009202s :      1: task_emit
53.65% :     0.023785s :      1: type_inference
 0.14% :     0.000062s :      1: validate


TotalTime = 0.0415892, [33]
[bootstrap]: 0.00027533
[type_inference]: 0.0223668
[event_method]: 2.077e-05
[auto_monad]: 0.00015609
[graph_reusing]: 6.34001e-06
[pre_auto_parallel]: 2.73003e-06
[py_interpret_to_execute]: 0.00011628
[rewriter_before_opt_a]: 9.855e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00625014, [2]
    [Cycle 1]: 0.0021428, [27]
        [switch_simplify]: 5.786e-05
        [loop_unroll]: 3.695e-05
        [a_1]: 0.00067669
        [with_stream_mark]: 1.893e-05
        [recompute_prepare]: 1.04e-05
        [updatestate_depend_eliminate]: 6.46999e-06
        [updatestate_assign_eliminate]: 6.12999e-06
        [updatestate_loads_eliminate]: 5.08002e-06
        [parameter_eliminate]: 2.11998e-06
        [specialize_transform]: 9.07001e-06
        [updatestate_useless_node_eliminater]: 1.152e-05
        [accelerated_algorithm]: 8.41002e-06
        [meta_shard_fg_expand]: 2.82002e-06
        [get_grad_eliminate_]: 8.08001e-06
        [merge_forward]: 4.95001e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.895e-05
        [j_node_and_user_rematch]: 1.32e-05
        [meta_fg_expand]: 3.48e-06
        [replace_old_param]: 1.103e-05
        [inline_without_move]: 8.05e-06
        [renormalize]: 0.0009725
        [add_forward_monad_depend]: 5.56e-06
        [auto_monad_grad]: 2.16e-06
        [auto_monad_eliminator]: 1.961e-05
        [cse]: 3.866e-05
        [replace_applicator]: 1.538e-05
    [Cycle 2]: 0.00073292, [27]
        [switch_simplify]: 9.05001e-06
        [loop_unroll]: 7.66001e-06
        [a_1]: 0.00016453
        [with_stream_mark]: 1.003e-05
        [recompute_prepare]: 7.92e-06
        [updatestate_depend_eliminate]: 4.78001e-06
        [updatestate_assign_eliminate]: 4.42e-06
        [updatestate_loads_eliminate]: 3.83999e-06
        [parameter_eliminate]: 1.19003e-06
        [specialize_transform]: 8.1e-06
        [updatestate_useless_node_eliminater]: 1.061e-05
        [accelerated_algorithm]: 7.92998e-06
        [meta_shard_fg_expand]: 1.63002e-06
        [get_grad_eliminate_]: 7.23e-06
        [merge_forward]: 4.38999e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.555e-05
        [j_node_and_user_rematch]: 1.23e-05
        [meta_fg_expand]: 2.68e-06
        [replace_old_param]: 9.64e-06
        [inline_without_move]: 0.00025349
        [renormalize]: 1.50001e-07
        [add_forward_monad_depend]: 3.08e-06
        [auto_monad_grad]: 1.30999e-06
        [auto_monad_eliminator]: 1.655e-05
        [cse]: 2.55e-05
        [replace_applicator]: 1.052e-05
[py_interpret_to_execute_after_opt_a]: 1.237e-05
[rewriter_after_opt_a]: 0.00028276
[convert_after_rewriter]: 1.097e-05
[order_py_execute_after_rewriter]: 7.24001e-06
[mutable_eliminate]: 0.00052399
[jit_opt_b]: 6.705e-05, [1]
    [Cycle 1]: 5.995e-05, [2]
        [frontend_op_eliminate]: 2.315e-05
        [inline_after_opt_a]: 2.447e-05
[cconv]: 2.645e-05
[loop_unroll]: 0.0004337
[jit_opt_after_cconv]: 0.00018571, [1]
    [Cycle 1]: 0.00017914, [11]
        [c_1]: 4.328e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 7.5e-06
        [updatestate_assign_eliminate]: 4.25999e-06
        [updatestate_loads_eliminate]: 3.86001e-06
        [cse]: 2.524e-05
        [call_graph_tuple_transform]: 2.109e-05
        [tuple_list_get_item_eliminator]: 8.43999e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 9.42001e-06
[remove_dup_value]: 1.878e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 9.67999e-06
[add_recomputation]: 6.45e-05
[cse_after_recomputation]: 2.979e-05, [1]
    [Cycle 1]: 2.378e-05, [1]
        [cse]: 1.732e-05
[auto_monad_reorder]: 2.441e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 5.09998e-06
[opt_after_jit_grad]: 0.00048549
[symbol_engine_optimizer]: 9.61e-05, [1]
    [Cycle 1]: 8.98e-05, [6]
        [build]: 8.96002e-06
        [elim_shapecalc]: 1.146e-05
        [elim_not_effective]: 1.744e-05
        [opt_reshape]: 8.57e-06
        [fold_const_symbol]: 1.339e-05
        [renormalize]: 5.00004e-07
[validate]: 5.149e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00974065
[execute]: 8.00999e-06
Sums
    bootstrap                            :     0.000275s :  0.73%
    type_inference                       :     0.022367s : 59.50%
    event_method                         :     0.000021s :  0.06%
    auto_monad                           :     0.000156s :  0.42%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000116s :  0.31%
    rewriter_before_opt_a                :     0.000099s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000067s :  0.18%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000841s :  2.24%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000025s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000262s :  0.70%
    jit_opt_a.renormalize                :     0.000973s :  2.59%
    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      :     0.000036s :  0.10%
    jit_opt_a.cse                        :     0.000064s :  0.17%
    jit_opt_a.replace_applicator         :     0.000026s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000283s :  0.75%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000524s :  1.39%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000434s :  1.15%
    jit_opt_after_cconv.c_1              :     0.000043s :  0.12%
    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              :     0.000025s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000064s :  0.17%
    cse_after_recomputation.cse          :     0.000017s :  0.05%
    auto_monad_reorder                   :     0.000024s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000485s :  1.29%
    symbol_engine_optimizer.build        :     0.000009s :  0.02%
    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 :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000051s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009741s : 25.91%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000256    47
 3.12% :     0.000008s :      2: substitution.depend_value_elim
 1.02% :     0.000003s :      4: substitution.elim_not_effective
 0.79% :     0.000002s :      4: substitution.fold_const_symbol
 2.51% :     0.000006s :      5: substitution.graph_param_transform
75.86% :     0.000194s :      4: substitution.inline
 1.74% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.56% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.49% :     0.000004s :      2: substitution.replace_old_param
 4.64% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.12% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.15% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022300     2
93.32% :     0.020810s :      1: type_inference.infer
 6.68% :     0.001490s :      1: type_inference.specialize
------[replace.]   0.000085     6
67.06% :     0.000057s :      4: replace.inline
32.94% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     6
94.67% :     0.000191s :      4: match.inline
 5.33% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.20% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.19% :     0.000002s :     15: predicate.addn_check_dump
 1.27% :     0.000002s :     15: predicate.addn_zero_filter
 2.09% :     0.000003s :     15: predicate.arithmetic_simplify
 1.28% :     0.000002s :     15: predicate.cast_eliminate
 0.46% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     15: predicate.compare_switch_simplify
 1.49% :     0.000002s :     15: predicate.depend_value_elim
 1.18% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.99% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     15: predicate.environ_get_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.08% :     0.000002s :     10: predicate.get_grad_eliminate
 0.37% :     0.000001s :      5: predicate.graph_param_transform
 5.19% :     0.000009s :     31: predicate.inline
 1.11% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :     10: predicate.less_batch_normalization
 1.52% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.99% :     0.000003s :     22: predicate.load_eliminater
 1.34% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.56% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     15: predicate.merge_addn
 1.23% :     0.000002s :     15: predicate.minmaximum_grad
 1.44% :     0.000002s :      5: predicate.mutable_eliminate
 0.54% :     0.000001s :      5: predicate.opt_reshape
 2.41% :     0.000004s :     22: predicate.partial_eliminate
 1.19% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.69% :     0.000003s :     15: predicate.reduce_eliminate
 1.46% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.86% :     0.000003s :     27: predicate.replace_applicator
 0.73% :     0.000001s :     10: predicate.replace_old_param
 0.40% :     0.000001s :      5: predicate.reset_defer_inline
 1.25% :     0.000002s :     15: predicate.reshape_eliminate
 1.24% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     15: predicate.same_eliminate
 0.62% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.04% :     0.000002s :     10: predicate.special_op_eliminate
 1.08% :     0.000002s :     10: predicate.specialize_transform
 1.52% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.20% :     0.000004s :     21: predicate.switch_defer_inline
 1.95% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.77% :     0.000011s :     68: predicate.switch_simplify
 1.25% :     0.000002s :     15: predicate.tile_eliminate
 1.33% :     0.000002s :     15: predicate.transpose_eliminate
 1.57% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000003s :     15: predicate.value_based_eliminate
 0.42% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000813    13
41.82% :     0.000340s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.42% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.76% :     0.000429s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044125    76
 0.15% :     0.000068s :      1: add_recomputation
 0.37% :     0.000161s :      1: auto_monad
 0.06% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000288s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.06% :     0.000025s :      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
14.17% :     0.006253s :      1: jit_opt_a
 0.43% :     0.000189s :      1: jit_opt_after_cconv
 0.16% :     0.000070s :      1: jit_opt_b
 1.00% :     0.000442s :      1: loop_unroll
 1.21% :     0.000533s :      1: mutable_eliminate
 3.13% :     0.001380s :     26: opt.transform.jit_opt_a
 0.18% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      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% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000047s :      4: opt.transform.symbol_engine_opt
 1.12% :     0.000493s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000121s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.30% :     0.000576s :      1: renormalize.infer
 0.88% :     0.000389s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.65% :     0.000287s :      1: rewriter_after_opt_a
 0.23% :     0.000103s :      1: rewriter_before_opt_a
 0.22% :     0.000099s :      1: symbol_engine_optimizer
22.10% :     0.009752s :      1: task_emit
50.72% :     0.022382s :      1: type_inference
 0.17% :     0.000074s :      1: validate


TotalTime = 0.0134443, [30]
[bootstrap]: 0.0003169
[type_inference]: 0.00464781
[event_method]: 2.33e-05
[auto_monad]: 7.429e-05
[graph_reusing]: 6.29001e-06
[pre_auto_parallel]: 2.96999e-06
[py_interpret_to_execute]: 4.6e-05
[rewriter_before_opt_a]: 9.492e-05
[expand_dump_flag]: 3.51999e-06
[jit_opt_a]: 0.00591327, [2]
    [Cycle 1]: 0.00182886, [27]
        [switch_simplify]: 6.598e-05
        [loop_unroll]: 3.57e-05
        [a_1]: 0.00065134
        [with_stream_mark]: 1.979e-05
        [recompute_prepare]: 8.23999e-06
        [updatestate_depend_eliminate]: 4.63001e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 2.81e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 7.53e-06
        [updatestate_useless_node_eliminater]: 6.51e-06
        [accelerated_algorithm]: 6.41e-06
        [meta_shard_fg_expand]: 2.78e-06
        [get_grad_eliminate_]: 6.07999e-06
        [merge_forward]: 4.17e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.651e-05
        [j_node_and_user_rematch]: 9.97999e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 1.14e-05
        [inline_without_move]: 6.01e-06
        [renormalize]: 0.000706
        [add_forward_monad_depend]: 5.89e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.445e-05
        [cse]: 3.325e-05
        [replace_applicator]: 1.49e-05
    [Cycle 2]: 0.00035715, [27]
        [switch_simplify]: 6.96001e-06
        [loop_unroll]: 5.84e-06
        [a_1]: 0.00011551
        [with_stream_mark]: 9.56998e-06
        [recompute_prepare]: 5.77001e-06
        [updatestate_depend_eliminate]: 3.53999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 5.71e-06
        [updatestate_useless_node_eliminater]: 5.78002e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 5.51e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 1.89999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.393e-05
        [j_node_and_user_rematch]: 8.76002e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 9.14e-06
        [inline_without_move]: 5.47999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 6.26e-06
        [cse]: 1.279e-05
        [replace_applicator]: 5.66998e-06
[py_interpret_to_execute_after_opt_a]: 1.247e-05
[rewriter_after_opt_a]: 3.803e-05
[convert_after_rewriter]: 7.85998e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.00062054
[jit_opt_b]: 5.419e-05, [1]
    [Cycle 1]: 4.701e-05, [2]
        [frontend_op_eliminate]: 1.827e-05
        [inline_after_opt_a]: 1.754e-05
[cconv]: 2.444e-05
[loop_unroll]: 0.00048997
[jit_opt_after_cconv]: 0.0001552, [1]
    [Cycle 1]: 0.00014884, [11]
        [c_1]: 2.662e-05
        [parameter_eliminate]: 2.19001e-06
        [updatestate_depend_eliminate]: 5.12999e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [cse]: 2.338e-05
        [call_graph_tuple_transform]: 2.159e-05
        [tuple_list_get_item_eliminator]: 6.11998e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 6.10016e-07
        [switch_simplify]: 6.29999e-06
[remove_dup_value]: 1.434e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 6.92002e-06
[add_recomputation]: 5.206e-05
[cse_after_recomputation]: 2.425e-05, [1]
    [Cycle 1]: 1.885e-05, [1]
        [cse]: 1.27e-05
[auto_monad_reorder]: 1.849e-05
[get_jit_bprop_graph]: 1.72999e-06
[rewriter_after_jit_bprop_graph]: 6.04999e-06
[opt_after_jit_grad]: 0.00049181
[symbol_engine_optimizer]: 7.749e-05, [1]
    [Cycle 1]: 7.092e-05, [6]
        [build]: 4.16001e-06
        [elim_shapecalc]: 8.43999e-06
        [elim_not_effective]: 1.417e-05
        [opt_reshape]: 6.49001e-06
        [fold_const_symbol]: 9.27999e-06
        [renormalize]: 5.00004e-07
[validate]: 4.137e-05
Sums
    bootstrap                            :     0.000317s :  3.47%
    type_inference                       :     0.004648s : 50.85%
    event_method                         :     0.000023s :  0.25%
    auto_monad                           :     0.000074s :  0.81%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000046s :  0.50%
    rewriter_before_opt_a                :     0.000095s :  1.04%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000073s :  0.80%
    jit_opt_a.loop_unroll                :     0.000042s :  0.45%
    jit_opt_a.a_1                        :     0.000767s :  8.39%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.09%
    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       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.13%
    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 :     0.000030s :  0.33%
    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          :     0.000021s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000706s :  7.72%
    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                        :     0.000046s :  0.50%
    jit_opt_a.replace_applicator         :     0.000021s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.14%
    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                    :     0.000621s :  6.79%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.19%
    cconv                                :     0.000024s :  0.27%
    loop_unroll                          :     0.000490s :  5.36%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.29%
    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 :     0.000003s :  0.03%
    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 :     0.000022s :  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 :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.08%
    add_recomputation                    :     0.000052s :  0.57%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000492s :  5.38%
    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                             :     0.000041s :  0.45%


Time group info:
------[substitution.]   0.000234    28
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.91% :     0.000007s :      4: substitution.graph_param_transform
82.76% :     0.000194s :      4: substitution.inline
 1.51% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000005s :      4: substitution.replace_old_param
 7.07% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004579     2
77.96% :     0.003570s :      1: type_inference.infer
22.04% :     0.001009s :      1: type_inference.specialize
------[replace.]   0.000087     8
57.16% :     0.000049s :      4: replace.inline
42.84% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     8
93.00% :     0.000191s :      4: match.inline
 7.00% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   912
 1.19% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     13: predicate.addn_check_dump
 1.31% :     0.000002s :     13: predicate.addn_zero_filter
 1.96% :     0.000003s :     13: predicate.arithmetic_simplify
 1.47% :     0.000002s :     13: predicate.cast_eliminate
 0.53% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.06% :     0.000002s :     13: predicate.compare_switch_simplify
 1.27% :     0.000002s :     13: predicate.depend_value_elim
 1.14% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      4: predicate.elim_not_effective
 0.55% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.06% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.06% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     13: predicate.environ_get_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 1.06% :     0.000002s :      8: predicate.get_grad_eliminate
 0.38% :     0.000001s :      4: predicate.graph_param_transform
 5.23% :     0.000007s :     29: predicate.inline
 0.88% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      8: predicate.less_batch_normalization
 1.82% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.26% :     0.000003s :     21: predicate.load_eliminater
 1.10% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.21% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 1.36% :     0.000002s :      4: predicate.mutable_eliminate
 0.46% :     0.000001s :      4: predicate.opt_reshape
 2.64% :     0.000004s :     21: predicate.partial_eliminate
 1.15% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     13: predicate.reduce_eliminate
 1.71% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.84% :     0.000003s :     25: predicate.replace_applicator
 0.75% :     0.000001s :      8: predicate.replace_old_param
 0.32% :     0.000000s :      4: predicate.reset_defer_inline
 1.45% :     0.000002s :     13: predicate.reshape_eliminate
 1.23% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     13: predicate.same_eliminate
 0.52% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      8: predicate.special_op_eliminate
 0.91% :     0.000001s :      8: predicate.specialize_transform
 1.38% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     21: predicate.switch_defer_inline
 2.15% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.69% :     0.000011s :     67: predicate.switch_simplify
 1.29% :     0.000002s :     13: predicate.tile_eliminate
 1.40% :     0.000002s :     13: predicate.transpose_eliminate
 1.44% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.28% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.77% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     13: predicate.value_based_eliminate
 0.41% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000629    12
45.01% :     0.000283s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.44% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.55% :     0.000312s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015236    72
 0.36% :     0.000055s :      1: add_recomputation
 0.51% :     0.000078s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 2.15% :     0.000328s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.18% :     0.000028s :      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
38.84% :     0.005917s :      1: jit_opt_a
 1.04% :     0.000158s :      1: jit_opt_after_cconv
 0.37% :     0.000057s :      1: jit_opt_b
 3.26% :     0.000496s :      1: loop_unroll
 4.13% :     0.000629s :      1: mutable_eliminate
 6.68% :     0.001017s :     26: opt.transform.jit_opt_a
 0.38% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000030s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.28% :     0.000500s :      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
 0.32% :     0.000049s :      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
 2.57% :     0.000392s :      1: renormalize.infer
 2.01% :     0.000307s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000041s :      1: rewriter_after_opt_a
 0.64% :     0.000098s :      1: rewriter_before_opt_a
 0.53% :     0.000080s :      1: symbol_engine_optimizer
30.61% :     0.004663s :      1: type_inference


TotalTime = 0.0129123, [30]
[bootstrap]: 0.00026184
[type_inference]: 0.0046768
[event_method]: 2.17e-05
[auto_monad]: 6.604e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.31998e-06
[py_interpret_to_execute]: 4.098e-05
[rewriter_before_opt_a]: 9.165e-05
[expand_dump_flag]: 3.00002e-06
[jit_opt_a]: 0.0055141, [2]
    [Cycle 1]: 0.00182841, [27]
        [switch_simplify]: 6.355e-05
        [loop_unroll]: 3.741e-05
        [a_1]: 0.00065885
        [with_stream_mark]: 1.717e-05
        [recompute_prepare]: 8.47e-06
        [updatestate_depend_eliminate]: 4.08001e-06
        [updatestate_assign_eliminate]: 3.76001e-06
        [updatestate_loads_eliminate]: 3.03998e-06
        [parameter_eliminate]: 2.00002e-06
        [specialize_transform]: 7.54002e-06
        [updatestate_useless_node_eliminater]: 6.38e-06
        [accelerated_algorithm]: 6.44001e-06
        [meta_shard_fg_expand]: 2.68e-06
        [get_grad_eliminate_]: 6.84001e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.588e-05
        [j_node_and_user_rematch]: 9.97001e-06
        [meta_fg_expand]: 2.64999e-06
        [replace_old_param]: 1.141e-05
        [inline_without_move]: 6.38e-06
        [renormalize]: 0.00070245
        [add_forward_monad_depend]: 5.27001e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.587e-05
        [cse]: 3.363e-05
        [replace_applicator]: 1.425e-05
    [Cycle 2]: 0.00040253, [27]
        [switch_simplify]: 7.2e-06
        [loop_unroll]: 6.19001e-06
        [a_1]: 0.00012308
        [with_stream_mark]: 9.87999e-06
        [recompute_prepare]: 6.24999e-06
        [updatestate_depend_eliminate]: 3.43999e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 6.61999e-06
        [updatestate_useless_node_eliminater]: 6.04999e-06
        [accelerated_algorithm]: 6.13998e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 6.28998e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.84998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.493e-05
        [j_node_and_user_rematch]: 9.37999e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 9.37999e-06
        [inline_without_move]: 5.84e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 8.49977e-07
        [auto_monad_eliminator]: 7.16001e-06
        [cse]: 1.615e-05
        [replace_applicator]: 6.49001e-06
[py_interpret_to_execute_after_opt_a]: 1.175e-05
[rewriter_after_opt_a]: 3.695e-05
[convert_after_rewriter]: 6.77002e-06
[order_py_execute_after_rewriter]: 5.67001e-06
[mutable_eliminate]: 0.00058336
[jit_opt_b]: 5.804e-05, [1]
    [Cycle 1]: 5.057e-05, [2]
        [frontend_op_eliminate]: 1.975e-05
        [inline_after_opt_a]: 1.79e-05
[cconv]: 2.714e-05
[loop_unroll]: 0.00047344
[jit_opt_after_cconv]: 0.00015735, [1]
    [Cycle 1]: 0.00015066, [11]
        [c_1]: 2.587e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 6.29999e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [cse]: 2.609e-05
        [call_graph_tuple_transform]: 2.098e-05
        [tuple_list_get_item_eliminator]: 6.11998e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 6.11998e-06
[remove_dup_value]: 1.498e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 6.64001e-06
[add_recomputation]: 5.061e-05
[cse_after_recomputation]: 2.355e-05, [1]
    [Cycle 1]: 1.801e-05, [1]
        [cse]: 1.215e-05
[auto_monad_reorder]: 1.755e-05
[get_jit_bprop_graph]: 1.71e-06
[rewriter_after_jit_bprop_graph]: 5.96e-06
[opt_after_jit_grad]: 0.00047062
[symbol_engine_optimizer]: 7.493e-05, [1]
    [Cycle 1]: 6.904e-05, [6]
        [build]: 3.76001e-06
        [elim_shapecalc]: 8.36002e-06
        [elim_not_effective]: 1.381e-05
        [opt_reshape]: 6.26e-06
        [fold_const_symbol]: 9.15001e-06
        [renormalize]: 3.7998e-07
[validate]: 3.665e-05
Sums
    bootstrap                            :     0.000262s :  2.90%
    type_inference                       :     0.004677s : 51.76%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000066s :  0.73%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.45%
    rewriter_before_opt_a                :     0.000092s :  1.01%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000071s :  0.78%
    jit_opt_a.loop_unroll                :     0.000044s :  0.48%
    jit_opt_a.a_1                        :     0.000782s :  8.65%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.16%
    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 :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000014s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000013s :  0.15%
    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 :  0.34%
    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          :     0.000021s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000703s :  7.78%
    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      :     0.000023s :  0.25%
    jit_opt_a.cse                        :     0.000050s :  0.55%
    jit_opt_a.replace_applicator         :     0.000021s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000037s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000583s :  6.46%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000027s :  0.30%
    loop_unroll                          :     0.000473s :  5.24%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.29%
    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              :     0.000026s :  0.29%
    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                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000007s :  0.07%
    add_recomputation                    :     0.000051s :  0.56%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000471s :  5.21%
    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 :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.41%


Time group info:
------[substitution.]   0.000217    28
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000006s :      4: substitution.graph_param_transform
82.33% :     0.000178s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000005s :      4: substitution.replace_old_param
 7.53% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004612     2
78.11% :     0.003602s :      1: type_inference.infer
21.89% :     0.001009s :      1: type_inference.specialize
------[replace.]   0.000086     8
56.03% :     0.000048s :      4: replace.inline
43.97% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     8
92.47% :     0.000176s :      4: match.inline
 7.53% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000151   912
 1.22% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.93% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.05% :     0.000002s :     13: predicate.addn_check_dump
 1.19% :     0.000002s :     13: predicate.addn_zero_filter
 1.77% :     0.000003s :     13: predicate.arithmetic_simplify
 1.44% :     0.000002s :     13: predicate.cast_eliminate
 0.51% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     13: predicate.compare_switch_simplify
 1.26% :     0.000002s :     13: predicate.depend_value_elim
 1.08% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 3.60% :     0.000005s :     13: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      4: predicate.elim_not_effective
 0.44% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.06% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     13: predicate.environ_get_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      4: predicate.fold_const_symbol
 0.98% :     0.000001s :      8: predicate.get_grad_eliminate
 0.36% :     0.000001s :      4: predicate.graph_param_transform
 5.24% :     0.000008s :     29: predicate.inline
 0.90% :     0.000001s :      8: predicate.inline_without_move
 0.35% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.12% :     0.000002s :      8: predicate.less_batch_normalization
 1.66% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.07% :     0.000003s :     21: predicate.load_eliminater
 1.58% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.99% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     13: predicate.merge_addn
 1.20% :     0.000002s :     13: predicate.minmaximum_grad
 1.54% :     0.000002s :      4: predicate.mutable_eliminate
 0.41% :     0.000001s :      4: predicate.opt_reshape
 2.41% :     0.000004s :     21: predicate.partial_eliminate
 1.22% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.92% :     0.000003s :     13: predicate.reduce_eliminate
 1.59% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.89% :     0.000003s :     25: predicate.replace_applicator
 0.74% :     0.000001s :      8: predicate.replace_old_param
 0.32% :     0.000000s :      4: predicate.reset_defer_inline
 1.16% :     0.000002s :     13: predicate.reshape_eliminate
 1.22% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.58% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     13: predicate.same_eliminate
 0.57% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.26% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.58% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.59% :     0.000004s :     21: predicate.switch_defer_inline
 2.12% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.57% :     0.000011s :     67: predicate.switch_simplify
 1.28% :     0.000002s :     13: predicate.tile_eliminate
 1.09% :     0.000002s :     13: predicate.transpose_eliminate
 1.38% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.87% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000631    12
45.82% :     0.000289s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.58% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.60% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014736    72
 0.37% :     0.000054s :      1: add_recomputation
 0.47% :     0.000070s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 1.87% :     0.000275s :      1: bootstrap
 0.20% :     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
 0.17% :     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
37.44% :     0.005517s :      1: jit_opt_a
 1.09% :     0.000160s :      1: jit_opt_after_cconv
 0.41% :     0.000061s :      1: jit_opt_b
 3.27% :     0.000482s :      1: loop_unroll
 4.02% :     0.000592s :      1: mutable_eliminate
 7.04% :     0.001038s :     26: opt.transform.jit_opt_a
 0.38% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000031s :      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
 0.16% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.25% :     0.000478s :      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% :     0.000044s :      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
 2.45% :     0.000361s :      1: renormalize.infer
 2.26% :     0.000334s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000040s :      1: rewriter_after_opt_a
 0.64% :     0.000095s :      1: rewriter_before_opt_a
 0.53% :     0.000078s :      1: symbol_engine_optimizer
31.82% :     0.004689s :      1: type_inference


TotalTime = 0.0457714, [33]
[bootstrap]: 0.00029592
[type_inference]: 0.0261038
[event_method]: 1.866e-05
[auto_monad]: 0.00014182
[graph_reusing]: 4.68001e-06
[pre_auto_parallel]: 1.83002e-06
[py_interpret_to_execute]: 0.00011083
[rewriter_before_opt_a]: 8.877e-05
[expand_dump_flag]: 3.88999e-06
[jit_opt_a]: 0.00583958, [2]
    [Cycle 1]: 0.00217185, [27]
        [switch_simplify]: 5.449e-05
        [loop_unroll]: 3.622e-05
        [a_1]: 0.00066841
        [with_stream_mark]: 1.72e-05
        [recompute_prepare]: 1.07e-05
        [updatestate_depend_eliminate]: 6.17999e-06
        [updatestate_assign_eliminate]: 5.17999e-06
        [updatestate_loads_eliminate]: 4.67e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 8.94e-06
        [updatestate_useless_node_eliminater]: 1.142e-05
        [accelerated_algorithm]: 8.40001e-06
        [meta_shard_fg_expand]: 3.16999e-06
        [get_grad_eliminate_]: 8.04997e-06
        [merge_forward]: 4.61002e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.786e-05
        [j_node_and_user_rematch]: 1.328e-05
        [meta_fg_expand]: 3.27997e-06
        [replace_old_param]: 1.148e-05
        [inline_without_move]: 8.25999e-06
        [renormalize]: 0.0010239
        [add_forward_monad_depend]: 5.80002e-06
        [auto_monad_grad]: 2.94001e-06
        [auto_monad_eliminator]: 1.892e-05
        [cse]: 2.991e-05
        [replace_applicator]: 1.72e-05
    [Cycle 2]: 0.00047484, [27]
        [switch_simplify]: 9.67001e-06
        [loop_unroll]: 7.96001e-06
        [a_1]: 0.0001658
        [with_stream_mark]: 1.212e-05
        [recompute_prepare]: 8.1e-06
        [updatestate_depend_eliminate]: 4.98001e-06
        [updatestate_assign_eliminate]: 4.38001e-06
        [updatestate_loads_eliminate]: 3.86001e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 8.42e-06
        [updatestate_useless_node_eliminater]: 1.131e-05
        [accelerated_algorithm]: 9.44998e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 7.81001e-06
        [merge_forward]: 4.35e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.616e-05
        [j_node_and_user_rematch]: 1.266e-05
        [meta_fg_expand]: 3.28998e-06
        [replace_old_param]: 1.035e-05
        [inline_without_move]: 8.38999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 8.90024e-07
        [auto_monad_eliminator]: 1.036e-05
        [cse]: 2.208e-05
        [replace_applicator]: 8.59e-06
[py_interpret_to_execute_after_opt_a]: 1.243e-05
[rewriter_after_opt_a]: 0.00028805
[convert_after_rewriter]: 1.221e-05
[order_py_execute_after_rewriter]: 7.75e-06
[mutable_eliminate]: 0.00059492
[jit_opt_b]: 6.927e-05, [1]
    [Cycle 1]: 6.157e-05, [2]
        [frontend_op_eliminate]: 2.414e-05
        [inline_after_opt_a]: 2.439e-05
[cconv]: 2.84e-05
[loop_unroll]: 0.00048878
[jit_opt_after_cconv]: 0.00020044, [1]
    [Cycle 1]: 0.00019337, [11]
        [c_1]: 4.612e-05
        [parameter_eliminate]: 3.18e-06
        [updatestate_depend_eliminate]: 9.91e-06
        [updatestate_assign_eliminate]: 4.45e-06
        [updatestate_loads_eliminate]: 4.29997e-06
        [cse]: 3.227e-05
        [call_graph_tuple_transform]: 2.247e-05
        [tuple_list_get_item_eliminator]: 8.09002e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 8.52998e-06
[remove_dup_value]: 1.927e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 1.099e-05
[add_recomputation]: 7.056e-05
[cse_after_recomputation]: 3.336e-05, [1]
    [Cycle 1]: 2.695e-05, [1]
        [cse]: 1.989e-05
[auto_monad_reorder]: 2.583e-05
[get_jit_bprop_graph]: 2.26e-06
[rewriter_after_jit_bprop_graph]: 6.25002e-06
[opt_after_jit_grad]: 0.00054921
[symbol_engine_optimizer]: 0.00010248, [1]
    [Cycle 1]: 9.642e-05, [6]
        [build]: 1.238e-05
        [elim_shapecalc]: 1.169e-05
        [elim_not_effective]: 1.899e-05
        [opt_reshape]: 8.97e-06
        [fold_const_symbol]: 1.376e-05
        [renormalize]: 4.2998e-07
[validate]: 5.901e-05
[backend_pass]: 1.15999e-06
[task_emit]: 0.0103449
[execute]: 8.22e-06
Sums
    bootstrap                            :     0.000296s :  0.71%
    type_inference                       :     0.026104s : 62.24%
    event_method                         :     0.000019s :  0.04%
    auto_monad                           :     0.000142s :  0.34%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000111s :  0.26%
    rewriter_before_opt_a                :     0.000089s :  0.21%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.15%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000834s :  1.99%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001024s :  2.44%
    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      :     0.000029s :  0.07%
    jit_opt_a.cse                        :     0.000052s :  0.12%
    jit_opt_a.replace_applicator         :     0.000026s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000288s :  0.69%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000595s :  1.42%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000028s :  0.07%
    loop_unroll                          :     0.000489s :  1.17%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  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              :     0.000032s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000071s :  0.17%
    cse_after_recomputation.cse          :     0.000020s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000549s :  1.31%
    symbol_engine_optimizer.build        :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000059s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010345s : 24.67%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000242    47
 3.35% :     0.000008s :      2: substitution.depend_value_elim
 1.14% :     0.000003s :      4: substitution.elim_not_effective
 0.93% :     0.000002s :      4: substitution.fold_const_symbol
 2.77% :     0.000007s :      5: substitution.graph_param_transform
73.18% :     0.000177s :      4: substitution.inline
 1.97% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.68% :     0.000006s :      8: substitution.remove_not_recompute_node
 1.59% :     0.000004s :      2: substitution.replace_old_param
 5.14% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.31% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.95% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026037     2
95.28% :     0.024809s :      1: type_inference.infer
 4.72% :     0.001228s :      1: type_inference.specialize
------[replace.]   0.000089     6
66.37% :     0.000059s :      4: replace.inline
33.63% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000186     6
93.98% :     0.000174s :      4: match.inline
 6.02% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1021
 1.22% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.29% :     0.000002s :     15: predicate.addn_zero_filter
 2.04% :     0.000003s :     15: predicate.arithmetic_simplify
 1.29% :     0.000002s :     15: predicate.cast_eliminate
 0.44% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.33% :     0.000002s :     15: predicate.depend_value_elim
 1.21% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.00% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.57% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     15: predicate.environ_get_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.16% :     0.000002s :     10: predicate.get_grad_eliminate
 0.39% :     0.000001s :      5: predicate.graph_param_transform
 5.33% :     0.000009s :     31: predicate.inline
 1.15% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.73% :     0.000003s :     10: predicate.less_batch_normalization
 1.52% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.85% :     0.000003s :     22: predicate.load_eliminater
 1.50% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.47% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.08% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     15: predicate.merge_addn
 1.13% :     0.000002s :     15: predicate.minmaximum_grad
 1.41% :     0.000002s :      5: predicate.mutable_eliminate
 0.66% :     0.000001s :      5: predicate.opt_reshape
 2.34% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.89% :     0.000003s :     15: predicate.reduce_eliminate
 1.46% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.85% :     0.000003s :     27: predicate.replace_applicator
 0.81% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.40% :     0.000002s :     15: predicate.reshape_eliminate
 1.22% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     15: predicate.same_eliminate
 0.55% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.02% :     0.000002s :     10: predicate.special_op_eliminate
 1.09% :     0.000002s :     10: predicate.specialize_transform
 1.67% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.16% :     0.000004s :     21: predicate.switch_defer_inline
 1.99% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.34% :     0.000011s :     68: predicate.switch_simplify
 1.27% :     0.000002s :     15: predicate.tile_eliminate
 1.29% :     0.000002s :     15: predicate.transpose_eliminate
 1.53% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.27% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     15: predicate.value_based_eliminate
 0.46% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000835    13
51.21% :     0.000428s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.05% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.74% :     0.000365s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048117    76
 0.15% :     0.000074s :      1: add_recomputation
 0.31% :     0.000147s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000313s :      1: bootstrap
 0.07% :     0.000031s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000036s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.05% :     0.000022s :      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% :     0.000007s :      1: graph_reusing
12.14% :     0.005843s :      1: jit_opt_a
 0.42% :     0.000203s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 1.03% :     0.000498s :      1: loop_unroll
 1.26% :     0.000605s :      1: mutable_eliminate
 2.35% :     0.001131s :     26: opt.transform.jit_opt_a
 0.17% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.16% :     0.000559s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000115s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.29% :     0.000623s :      1: renormalize.infer
 0.82% :     0.000393s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000293s :      1: rewriter_after_opt_a
 0.19% :     0.000093s :      1: rewriter_before_opt_a
 0.22% :     0.000105s :      1: symbol_engine_optimizer
21.53% :     0.010358s :      1: task_emit
54.27% :     0.026116s :      1: type_inference
 0.18% :     0.000086s :      1: validate


TotalTime = 0.0113273, [30]
[bootstrap]: 0.00025386
[type_inference]: 0.00412521
[event_method]: 1.883e-05
[auto_monad]: 4.358e-05
[graph_reusing]: 4.03999e-06
[pre_auto_parallel]: 2.04e-06
[py_interpret_to_execute]: 3.227e-05
[rewriter_before_opt_a]: 7.876e-05
[expand_dump_flag]: 2.63998e-06
[jit_opt_a]: 0.00488974, [2]
    [Cycle 1]: 0.00155634, [27]
        [switch_simplify]: 4.849e-05
        [loop_unroll]: 3.441e-05
        [a_1]: 0.00058058
        [with_stream_mark]: 1.114e-05
        [recompute_prepare]: 7.73001e-06
        [updatestate_depend_eliminate]: 3.23998e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 7.02002e-06
        [updatestate_useless_node_eliminater]: 5.86e-06
        [accelerated_algorithm]: 6.31e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 5.87999e-06
        [merge_forward]: 2.67001e-06
        [cell_reuse_recompute_pass]: 7.80012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.206e-05
        [j_node_and_user_rematch]: 9.07001e-06
        [meta_fg_expand]: 1.86998e-06
        [replace_old_param]: 9.58002e-06
        [inline_without_move]: 5.71e-06
        [renormalize]: 0.00059142
        [add_forward_monad_depend]: 4.31002e-06
        [auto_monad_grad]: 1.15001e-06
        [auto_monad_eliminator]: 9.27999e-06
        [cse]: 2.102e-05
        [replace_applicator]: 1.182e-05
    [Cycle 2]: 0.00035786, [27]
        [switch_simplify]: 6.69999e-06
        [loop_unroll]: 5.78002e-06
        [a_1]: 0.00011207
        [with_stream_mark]: 9.02999e-06
        [recompute_prepare]: 5.94999e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 5.62999e-06
        [updatestate_useless_node_eliminater]: 5.94e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 2.63003e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.35e-05
        [j_node_and_user_rematch]: 8.49002e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 9.27001e-06
        [inline_without_move]: 5.50001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.73002e-06
        [cse]: 1.427e-05
        [replace_applicator]: 5.82001e-06
[py_interpret_to_execute_after_opt_a]: 9.10999e-06
[rewriter_after_opt_a]: 2.808e-05
[convert_after_rewriter]: 6.27001e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.0004534
[jit_opt_b]: 5.28e-05, [1]
    [Cycle 1]: 4.653e-05, [2]
        [frontend_op_eliminate]: 1.803e-05
        [inline_after_opt_a]: 1.698e-05
[cconv]: 1.615e-05
[loop_unroll]: 0.00039768
[jit_opt_after_cconv]: 0.00014018, [1]
    [Cycle 1]: 0.00013432, [11]
        [c_1]: 2.382e-05
        [parameter_eliminate]: 2.13002e-06
        [updatestate_depend_eliminate]: 4.48001e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [cse]: 1.934e-05
        [call_graph_tuple_transform]: 1.773e-05
        [tuple_list_get_item_eliminator]: 6.38003e-06
        [none_parameter_eliminate]: 1.02998e-06
        [renormalize]: 3.10014e-07
        [switch_simplify]: 6.03998e-06
[remove_dup_value]: 8.67e-06
[partial_unused_args_eliminate]: 1.64e-06
[environ_conv]: 4.52e-06
[add_recomputation]: 3.477e-05
[cse_after_recomputation]: 2.307e-05, [1]
    [Cycle 1]: 1.808e-05, [1]
        [cse]: 1.259e-05
[auto_monad_reorder]: 1.337e-05
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 4.07e-06
[opt_after_jit_grad]: 0.00043894
[symbol_engine_optimizer]: 7.172e-05, [1]
    [Cycle 1]: 6.64e-05, [6]
        [build]: 2.12999e-06
        [elim_shapecalc]: 8.06001e-06
        [elim_not_effective]: 1.338e-05
        [opt_reshape]: 6.28e-06
        [fold_const_symbol]: 9.05999e-06
        [renormalize]: 6.10016e-07
[validate]: 2.749e-05
Sums
    bootstrap                            :     0.000254s :  3.24%
    type_inference                       :     0.004125s : 52.72%
    event_method                         :     0.000019s :  0.24%
    auto_monad                           :     0.000044s :  0.56%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.41%
    rewriter_before_opt_a                :     0.000079s :  1.01%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000055s :  0.71%
    jit_opt_a.loop_unroll                :     0.000040s :  0.51%
    jit_opt_a.a_1                        :     0.000693s :  8.85%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000014s :  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        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.15%
    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_        :     0.000011s :  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 :  0.33%
    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          :     0.000019s :  0.24%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000591s :  7.56%
    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 :  0.19%
    jit_opt_a.cse                        :     0.000035s :  0.45%
    jit_opt_a.replace_applicator         :     0.000018s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000028s :  0.36%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000453s :  5.79%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.22%
    cconv                                :     0.000016s :  0.21%
    loop_unroll                          :     0.000398s :  5.08%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.30%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  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 :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    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 :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000009s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000035s :  0.44%
    cse_after_recomputation.cse          :     0.000013s :  0.16%
    auto_monad_reorder                   :     0.000013s :  0.17%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000439s :  5.61%
    symbol_engine_optimizer.build        :     0.000002s :  0.03%
    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  :     0.000001s :  0.01%
    validate                             :     0.000027s :  0.35%


Time group info:
------[substitution.]   0.000183    28
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.23% :     0.000004s :      4: substitution.graph_param_transform
83.19% :     0.000153s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000004s :      4: substitution.replace_old_param
 7.09% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004080     2
80.11% :     0.003269s :      1: type_inference.infer
19.89% :     0.000811s :      1: type_inference.specialize
------[replace.]   0.000077     8
55.47% :     0.000042s :      4: replace.inline
44.53% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     8
93.09% :     0.000150s :      4: match.inline
 6.91% :     0.000011s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.91% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     13: predicate.addn_check_dump
 1.44% :     0.000002s :     13: predicate.addn_zero_filter
 1.96% :     0.000003s :     13: predicate.arithmetic_simplify
 1.45% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     13: predicate.compare_switch_simplify
 1.25% :     0.000002s :     13: predicate.depend_value_elim
 1.20% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      4: predicate.elim_not_effective
 0.61% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     13: predicate.environ_get_eliminate
 1.16% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.95% :     0.000001s :      8: predicate.get_grad_eliminate
 0.29% :     0.000000s :      4: predicate.graph_param_transform
 5.20% :     0.000007s :     29: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.45% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      8: predicate.less_batch_normalization
 1.62% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.04% :     0.000003s :     21: predicate.load_eliminater
 1.13% :     0.000001s :      4: predicate.loop_unroll_after_grad
 4.27% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     13: predicate.merge_addn
 1.15% :     0.000001s :     13: predicate.minmaximum_grad
 1.20% :     0.000002s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.56% :     0.000003s :     21: predicate.partial_eliminate
 1.20% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     13: predicate.reduce_eliminate
 1.77% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.61% :     0.000001s :      8: predicate.replace_old_param
 0.30% :     0.000000s :      4: predicate.reset_defer_inline
 1.23% :     0.000002s :     13: predicate.reshape_eliminate
 1.23% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     13: predicate.same_eliminate
 0.67% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     21: predicate.switch_defer_inline
 2.29% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.65% :     0.000010s :     67: predicate.switch_simplify
 1.30% :     0.000002s :     13: predicate.tile_eliminate
 1.27% :     0.000002s :     13: predicate.transpose_eliminate
 1.44% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000002s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000528    12
47.07% :     0.000248s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.85% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.08% :     0.000254s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012915    72
 0.29% :     0.000037s :      1: add_recomputation
 0.36% :     0.000047s :      1: auto_monad
 0.12% :     0.000016s :      1: auto_monad_reorder
 2.05% :     0.000265s :      1: bootstrap
 0.14% :     0.000019s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000025s :      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
 0.05% :     0.000006s :      1: graph_reusing
37.88% :     0.004892s :      1: jit_opt_a
 1.11% :     0.000143s :      1: jit_opt_after_cconv
 0.43% :     0.000055s :      1: jit_opt_b
 3.13% :     0.000404s :      1: loop_unroll
 3.56% :     0.000460s :      1: mutable_eliminate
 7.09% :     0.000916s :     26: opt.transform.jit_opt_a
 0.39% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000029s :      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
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.44% :     0.000445s :      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
 0.09% :     0.000011s :      1: remove_dup_value
 2.48% :     0.000321s :      1: renormalize.infer
 2.05% :     0.000264s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000031s :      1: rewriter_after_opt_a
 0.63% :     0.000081s :      1: rewriter_before_opt_a
 0.57% :     0.000074s :      1: symbol_engine_optimizer
32.00% :     0.004133s :      1: type_inference


TotalTime = 0.0415916, [33]
[bootstrap]: 0.00028052
[type_inference]: 0.0207196
[event_method]: 1.799e-05
[auto_monad]: 0.00012719
[graph_reusing]: 4.05e-06
[pre_auto_parallel]: 2.32001e-06
[py_interpret_to_execute]: 0.00010778
[rewriter_before_opt_a]: 8.821e-05
[expand_dump_flag]: 2.18002e-06
[jit_opt_a]: 0.00549194, [2]
    [Cycle 1]: 0.0019781, [27]
        [switch_simplify]: 5.196e-05
        [loop_unroll]: 3.638e-05
        [a_1]: 0.00063883
        [with_stream_mark]: 1.437e-05
        [recompute_prepare]: 1.025e-05
        [updatestate_depend_eliminate]: 5.56e-06
        [updatestate_assign_eliminate]: 4.55999e-06
        [updatestate_loads_eliminate]: 4.33001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 9.20999e-06
        [updatestate_useless_node_eliminater]: 1.143e-05
        [accelerated_algorithm]: 8.40001e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 8.11002e-06
        [merge_forward]: 4.21001e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.626e-05
        [j_node_and_user_rematch]: 1.313e-05
        [meta_fg_expand]: 3.23e-06
        [replace_old_param]: 1.065e-05
        [inline_without_move]: 8.25e-06
        [renormalize]: 0.00088327
        [add_forward_monad_depend]: 4.57e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 1.745e-05
        [cse]: 2.85e-05
        [replace_applicator]: 1.459e-05
    [Cycle 2]: 0.00044928, [27]
        [switch_simplify]: 8.79e-06
        [loop_unroll]: 7.72002e-06
        [a_1]: 0.00016175
        [with_stream_mark]: 1.002e-05
        [recompute_prepare]: 7.84002e-06
        [updatestate_depend_eliminate]: 4.85001e-06
        [updatestate_assign_eliminate]: 4.37e-06
        [updatestate_loads_eliminate]: 3.75e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 7.71999e-06
        [updatestate_useless_node_eliminater]: 1.073e-05
        [accelerated_algorithm]: 7.87e-06
        [meta_shard_fg_expand]: 1.59998e-06
        [get_grad_eliminate_]: 7.31001e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.524e-05
        [j_node_and_user_rematch]: 1.228e-05
        [meta_fg_expand]: 2.69999e-06
        [replace_old_param]: 9.69999e-06
        [inline_without_move]: 7.35998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 9.79984e-07
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 9.91e-06
        [cse]: 1.848e-05
        [replace_applicator]: 7.73001e-06
[py_interpret_to_execute_after_opt_a]: 9.50001e-06
[rewriter_after_opt_a]: 0.00026296
[convert_after_rewriter]: 1.018e-05
[order_py_execute_after_rewriter]: 6.64001e-06
[mutable_eliminate]: 0.00059187
[jit_opt_b]: 6.915e-05, [1]
    [Cycle 1]: 6.27e-05, [2]
        [frontend_op_eliminate]: 2.568e-05
        [inline_after_opt_a]: 2.546e-05
[cconv]: 1.825e-05
[loop_unroll]: 0.00047731
[jit_opt_after_cconv]: 0.00019781, [1]
    [Cycle 1]: 0.00019112, [11]
        [c_1]: 4.807e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 7.99997e-06
        [updatestate_assign_eliminate]: 4.32e-06
        [updatestate_loads_eliminate]: 3.87002e-06
        [cse]: 2.742e-05
        [call_graph_tuple_transform]: 2.292e-05
        [tuple_list_get_item_eliminator]: 8.81002e-06
        [none_parameter_eliminate]: 1.02e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 9.63997e-06
[remove_dup_value]: 1.422e-05
[partial_unused_args_eliminate]: 1.86998e-06
[environ_conv]: 9.27001e-06
[add_recomputation]: 5.124e-05
[cse_after_recomputation]: 3.142e-05, [1]
    [Cycle 1]: 2.586e-05, [1]
        [cse]: 1.956e-05
[auto_monad_reorder]: 2.213e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 5.19998e-06
[opt_after_jit_grad]: 0.00052012
[symbol_engine_optimizer]: 0.0001057, [1]
    [Cycle 1]: 9.907e-05, [6]
        [build]: 1.099e-05
        [elim_shapecalc]: 1.275e-05
        [elim_not_effective]: 1.991e-05
        [opt_reshape]: 9.49e-06
        [fold_const_symbol]: 1.473e-05
        [renormalize]: 4.19997e-07
[validate]: 5.063e-05
[backend_pass]: 1.21002e-06
[task_emit]: 0.0120756
[execute]: 7.5e-06
Sums
    bootstrap                            :     0.000281s :  0.74%
    type_inference                       :     0.020720s : 54.65%
    event_method                         :     0.000018s :  0.05%
    auto_monad                           :     0.000127s :  0.34%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000108s :  0.28%
    rewriter_before_opt_a                :     0.000088s :  0.23%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000061s :  0.16%
    jit_opt_a.loop_unroll                :     0.000044s :  0.12%
    jit_opt_a.a_1                        :     0.000801s :  2.11%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  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 :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000025s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000883s :  2.33%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.07%
    jit_opt_a.cse                        :     0.000047s :  0.12%
    jit_opt_a.replace_applicator         :     0.000022s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000263s :  0.69%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000592s :  1.56%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.07%
    cconv                                :     0.000018s :  0.05%
    loop_unroll                          :     0.000477s :  1.26%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.13%
    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 :     0.000004s :  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 :     0.000009s :  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  :     0.000010s :  0.03%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000009s :  0.02%
    add_recomputation                    :     0.000051s :  0.14%
    cse_after_recomputation.cse          :     0.000020s :  0.05%
    auto_monad_reorder                   :     0.000022s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000520s :  1.37%
    symbol_engine_optimizer.build        :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000051s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012076s : 31.85%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000222    47
 3.39% :     0.000008s :      2: substitution.depend_value_elim
 1.11% :     0.000002s :      4: substitution.elim_not_effective
 0.79% :     0.000002s :      4: substitution.fold_const_symbol
 2.61% :     0.000006s :      5: substitution.graph_param_transform
74.63% :     0.000165s :      4: substitution.inline
 2.00% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.81% :     0.000006s :      8: substitution.remove_not_recompute_node
 1.52% :     0.000003s :      2: substitution.replace_old_param
 3.85% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
 3.58% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.70% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.020662     2
94.03% :     0.019428s :      1: type_inference.infer
 5.97% :     0.001234s :      1: type_inference.specialize
------[replace.]   0.000082     6
65.83% :     0.000054s :      4: replace.inline
34.17% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000170     6
95.67% :     0.000162s :      4: match.inline
 4.33% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000167  1021
 1.26% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.24% :     0.000002s :     15: predicate.addn_check_dump
 1.27% :     0.000002s :     15: predicate.addn_zero_filter
 2.01% :     0.000003s :     15: predicate.arithmetic_simplify
 1.34% :     0.000002s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     15: predicate.compare_switch_simplify
 1.33% :     0.000002s :     15: predicate.depend_value_elim
 1.16% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.92% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      5: predicate.elim_not_effective
 0.70% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      5: predicate.fold_const_symbol
 1.09% :     0.000002s :     10: predicate.get_grad_eliminate
 0.39% :     0.000001s :      5: predicate.graph_param_transform
 5.33% :     0.000009s :     31: predicate.inline
 1.06% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :     10: predicate.less_batch_normalization
 1.52% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.99% :     0.000003s :     22: predicate.load_eliminater
 1.56% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.47% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.11% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.24% :     0.000002s :     15: predicate.minmaximum_grad
 1.81% :     0.000003s :      5: predicate.mutable_eliminate
 0.62% :     0.000001s :      5: predicate.opt_reshape
 2.31% :     0.000004s :     22: predicate.partial_eliminate
 1.25% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.71% :     0.000003s :     15: predicate.reduce_eliminate
 1.38% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.75% :     0.000003s :     27: predicate.replace_applicator
 0.61% :     0.000001s :     10: predicate.replace_old_param
 0.37% :     0.000001s :      5: predicate.reset_defer_inline
 1.37% :     0.000002s :     15: predicate.reshape_eliminate
 1.22% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.15% :     0.000002s :     10: predicate.special_op_eliminate
 1.04% :     0.000002s :     10: predicate.specialize_transform
 1.48% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.60% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.31% :     0.000004s :     21: predicate.switch_defer_inline
 1.91% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.52% :     0.000011s :     68: predicate.switch_simplify
 1.27% :     0.000002s :     15: predicate.tile_eliminate
 1.20% :     0.000002s :     15: predicate.transpose_eliminate
 1.55% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.91% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.29% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000003s :     15: predicate.value_based_eliminate
 0.50% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.91% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000732    13
45.29% :     0.000332s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.05% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.66% :     0.000364s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043766    76
 0.12% :     0.000054s :      1: add_recomputation
 0.30% :     0.000132s :      1: auto_monad
 0.06% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000293s :      1: bootstrap
 0.05% :     0.000021s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.08% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.05% :     0.000022s :      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
12.56% :     0.005496s :      1: jit_opt_a
 0.46% :     0.000201s :      1: jit_opt_after_cconv
 0.16% :     0.000072s :      1: jit_opt_b
 1.11% :     0.000486s :      1: loop_unroll
 1.37% :     0.000601s :      1: mutable_eliminate
 2.48% :     0.001085s :     26: opt.transform.jit_opt_a
 0.20% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000044s :      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
 0.08% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.21% :     0.000529s :      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
 0.26% :     0.000112s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 1.17% :     0.000513s :      1: renormalize.infer
 0.83% :     0.000363s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000268s :      1: rewriter_after_opt_a
 0.21% :     0.000093s :      1: rewriter_before_opt_a
 0.25% :     0.000109s :      1: symbol_engine_optimizer
27.62% :     0.012090s :      1: task_emit
47.37% :     0.020731s :      1: type_inference
 0.17% :     0.000073s :      1: validate


TotalTime = 0.0116789, [30]
[bootstrap]: 0.00026713
[type_inference]: 0.00453922
[event_method]: 1.981e-05
[auto_monad]: 4.8e-05
[graph_reusing]: 4.31002e-06
[pre_auto_parallel]: 1.99e-06
[py_interpret_to_execute]: 3.322e-05
[rewriter_before_opt_a]: 8.166e-05
[expand_dump_flag]: 2.29999e-06
[jit_opt_a]: 0.00472061, [2]
    [Cycle 1]: 0.00148481, [27]
        [switch_simplify]: 5.067e-05
        [loop_unroll]: 3.441e-05
        [a_1]: 0.00056753
        [with_stream_mark]: 1.196e-05
        [recompute_prepare]: 7.45998e-06
        [updatestate_depend_eliminate]: 3.31001e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 6.34999e-06
        [updatestate_useless_node_eliminater]: 5.65001e-06
        [accelerated_algorithm]: 6.14999e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 2.78998e-06
        [cell_reuse_recompute_pass]: 8.59989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.164e-05
        [j_node_and_user_rematch]: 9.52001e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 1.023e-05
        [inline_without_move]: 5.87999e-06
        [renormalize]: 0.00052518
        [add_forward_monad_depend]: 4.85999e-06
        [auto_monad_grad]: 1.51002e-06
        [auto_monad_eliminator]: 1.184e-05
        [cse]: 2.121e-05
        [replace_applicator]: 1.269e-05
    [Cycle 2]: 0.00035556, [27]
        [switch_simplify]: 6.64001e-06
        [loop_unroll]: 5.64e-06
        [a_1]: 0.00011035
        [with_stream_mark]: 8.3e-06
        [recompute_prepare]: 5.62999e-06
        [updatestate_depend_eliminate]: 3.14999e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 5.31002e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.58998e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.322e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 9.59e-06
        [inline_without_move]: 5.56e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 6.47001e-06
        [cse]: 1.39e-05
        [replace_applicator]: 5.69e-06
[py_interpret_to_execute_after_opt_a]: 8.89e-06
[rewriter_after_opt_a]: 2.901e-05
[convert_after_rewriter]: 6.09999e-06
[order_py_execute_after_rewriter]: 4.98001e-06
[mutable_eliminate]: 0.00049592
[jit_opt_b]: 5.273e-05, [1]
    [Cycle 1]: 4.652e-05, [2]
        [frontend_op_eliminate]: 1.784e-05
        [inline_after_opt_a]: 1.701e-05
[cconv]: 1.869e-05
[loop_unroll]: 0.00040885
[jit_opt_after_cconv]: 0.00014629, [1]
    [Cycle 1]: 0.00014001, [11]
        [c_1]: 2.567e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 5.47999e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [cse]: 1.915e-05
        [call_graph_tuple_transform]: 1.867e-05
        [tuple_list_get_item_eliminator]: 6.34999e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 5.82999e-06
[remove_dup_value]: 9.05001e-06
[partial_unused_args_eliminate]: 1.70001e-06
[environ_conv]: 4.22e-06
[add_recomputation]: 3.343e-05
[cse_after_recomputation]: 2.26e-05, [1]
    [Cycle 1]: 1.759e-05, [1]
        [cse]: 1.206e-05
[auto_monad_reorder]: 1.294e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.43999e-06
[opt_after_jit_grad]: 0.00045083
[symbol_engine_optimizer]: 7.259e-05, [1]
    [Cycle 1]: 6.678e-05, [6]
        [build]: 2.67001e-06
        [elim_shapecalc]: 8.37e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 5.96e-06
        [fold_const_symbol]: 8.94e-06
        [renormalize]: 5.39992e-07
[validate]: 2.826e-05
Sums
    bootstrap                            :     0.000267s :  3.23%
    type_inference                       :     0.004539s : 54.96%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000048s :  0.58%
    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                :     0.000082s :  0.99%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000057s :  0.69%
    jit_opt_a.loop_unroll                :     0.000040s :  0.48%
    jit_opt_a.a_1                        :     0.000678s :  8.21%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.25%
    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        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    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       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  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 :     0.000025s :  0.30%
    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          :     0.000020s :  0.24%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000525s :  6.36%
    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      :     0.000018s :  0.22%
    jit_opt_a.cse                        :     0.000035s :  0.43%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000029s :  0.35%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000496s :  6.00%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000019s :  0.23%
    loop_unroll                          :     0.000409s :  4.95%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.31%
    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 :     0.000003s :  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 :  0.08%
    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 :  0.07%
    remove_dup_value                     :     0.000009s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000033s :  0.40%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000013s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000451s :  5.46%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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                             :     0.000028s :  0.34%


Time group info:
------[substitution.]   0.000174    28
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000004s :      4: substitution.graph_param_transform
81.64% :     0.000142s :      4: substitution.inline
 1.59% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.40% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.69% :     0.000005s :      4: substitution.replace_old_param
 7.60% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004490     2
80.85% :     0.003630s :      1: type_inference.infer
19.15% :     0.000860s :      1: type_inference.specialize
------[replace.]   0.000077     8
56.43% :     0.000043s :      4: replace.inline
43.57% :     0.000033s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     8
92.48% :     0.000139s :      4: match.inline
 7.52% :     0.000011s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   912
 1.36% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     13: predicate.addn_check_dump
 1.30% :     0.000002s :     13: predicate.addn_zero_filter
 1.80% :     0.000002s :     13: predicate.arithmetic_simplify
 1.45% :     0.000002s :     13: predicate.cast_eliminate
 0.39% :     0.000000s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     13: predicate.compare_switch_simplify
 1.15% :     0.000001s :     13: predicate.depend_value_elim
 1.17% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      4: predicate.elim_not_effective
 0.57% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.94% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.39% :     0.000007s :     29: predicate.inline
 0.93% :     0.000001s :      8: predicate.inline_without_move
 0.45% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      8: predicate.less_batch_normalization
 1.78% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     21: predicate.load_eliminater
 1.26% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.29% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     13: predicate.merge_addn
 1.14% :     0.000001s :     13: predicate.minmaximum_grad
 1.45% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.61% :     0.000003s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.00% :     0.000003s :     25: predicate.replace_applicator
 0.70% :     0.000001s :      8: predicate.replace_old_param
 0.41% :     0.000001s :      4: predicate.reset_defer_inline
 1.21% :     0.000002s :     13: predicate.reshape_eliminate
 1.22% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     13: predicate.same_eliminate
 0.64% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      8: predicate.special_op_eliminate
 1.05% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     21: predicate.switch_defer_inline
 2.18% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.68% :     0.000010s :     67: predicate.switch_simplify
 1.20% :     0.000002s :     13: predicate.tile_eliminate
 1.23% :     0.000002s :     13: predicate.transpose_eliminate
 1.44% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000002s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
47.32% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.49% :     0.000260s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013194    72
 0.27% :     0.000036s :      1: add_recomputation
 0.39% :     0.000051s :      1: auto_monad
 0.12% :     0.000015s :      1: auto_monad_reorder
 2.11% :     0.000279s :      1: bootstrap
 0.16% :     0.000021s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.18% :     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
35.80% :     0.004723s :      1: jit_opt_a
 1.13% :     0.000149s :      1: jit_opt_after_cconv
 0.42% :     0.000055s :      1: jit_opt_b
 3.15% :     0.000416s :      1: loop_unroll
 3.82% :     0.000504s :      1: mutable_eliminate
 6.85% :     0.000903s :     26: opt.transform.jit_opt_a
 0.40% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000029s :      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
 0.25% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.48% :     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
 0.27% :     0.000036s :      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
 2.05% :     0.000271s :      1: renormalize.infer
 1.88% :     0.000248s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000032s :      1: rewriter_after_opt_a
 0.64% :     0.000084s :      1: rewriter_before_opt_a
 0.57% :     0.000075s :      1: symbol_engine_optimizer
34.48% :     0.004549s :      1: type_inference


TotalTime = 0.0113385, [30]
[bootstrap]: 0.0002517
[type_inference]: 0.00425392
[event_method]: 1.95e-05
[auto_monad]: 5.441e-05
[graph_reusing]: 4.65001e-06
[pre_auto_parallel]: 2.29999e-06
[py_interpret_to_execute]: 3.394e-05
[rewriter_before_opt_a]: 8.312e-05
[expand_dump_flag]: 2.34999e-06
[jit_opt_a]: 0.00471521, [2]
    [Cycle 1]: 0.00150852, [27]
        [switch_simplify]: 5.094e-05
        [loop_unroll]: 3.405e-05
        [a_1]: 0.00058981
        [with_stream_mark]: 1.228e-05
        [recompute_prepare]: 7.71999e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 6.54001e-06
        [updatestate_useless_node_eliminater]: 5.69999e-06
        [accelerated_algorithm]: 5.94999e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.78002e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.177e-05
        [j_node_and_user_rematch]: 9.51998e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 9.74999e-06
        [inline_without_move]: 6.18998e-06
        [renormalize]: 0.00052487
        [add_forward_monad_depend]: 5.03002e-06
        [auto_monad_grad]: 1.84998e-06
        [auto_monad_eliminator]: 1.187e-05
        [cse]: 1.947e-05
        [replace_applicator]: 1.312e-05
    [Cycle 2]: 0.00035335, [27]
        [switch_simplify]: 6.78e-06
        [loop_unroll]: 5.59e-06
        [a_1]: 0.00011198
        [with_stream_mark]: 7.93001e-06
        [recompute_prepare]: 5.49998e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 9.00001e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 9.18002e-06
        [inline_without_move]: 5.41002e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.66e-06
        [cse]: 1.303e-05
        [replace_applicator]: 5.67001e-06
[py_interpret_to_execute_after_opt_a]: 8.92e-06
[rewriter_after_opt_a]: 2.836e-05
[convert_after_rewriter]: 6.21e-06
[order_py_execute_after_rewriter]: 4.80001e-06
[mutable_eliminate]: 0.00046226
[jit_opt_b]: 5.325e-05, [1]
    [Cycle 1]: 4.701e-05, [2]
        [frontend_op_eliminate]: 1.842e-05
        [inline_after_opt_a]: 1.69e-05
[cconv]: 1.773e-05
[loop_unroll]: 0.00040487
[jit_opt_after_cconv]: 0.00014631, [1]
    [Cycle 1]: 0.00014026, [11]
        [c_1]: 2.453e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.84999e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.041e-05
        [call_graph_tuple_transform]: 1.877e-05
        [tuple_list_get_item_eliminator]: 6.02999e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 6.07999e-06
[remove_dup_value]: 1.024e-05
[partial_unused_args_eliminate]: 1.87999e-06
[environ_conv]: 4.47e-06
[add_recomputation]: 3.751e-05
[cse_after_recomputation]: 2.269e-05, [1]
    [Cycle 1]: 1.763e-05, [1]
        [cse]: 1.214e-05
[auto_monad_reorder]: 1.428e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 4.65999e-06
[opt_after_jit_grad]: 0.00044022
[symbol_engine_optimizer]: 7.253e-05, [1]
    [Cycle 1]: 6.662e-05, [6]
        [build]: 2.58e-06
        [elim_shapecalc]: 8.29998e-06
        [elim_not_effective]: 1.315e-05
        [opt_reshape]: 6.24001e-06
        [fold_const_symbol]: 9.14998e-06
        [renormalize]: 3.50003e-07
[validate]: 2.984e-05
Sums
    bootstrap                            :     0.000252s :  3.17%
    type_inference                       :     0.004254s : 53.52%
    event_method                         :     0.000020s :  0.25%
    auto_monad                           :     0.000054s :  0.68%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.43%
    rewriter_before_opt_a                :     0.000083s :  1.05%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000058s :  0.73%
    jit_opt_a.loop_unroll                :     0.000040s :  0.50%
    jit_opt_a.a_1                        :     0.000702s :  8.83%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.25%
    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 :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    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      :     0.000011s :  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 :     0.000025s :  0.31%
    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          :     0.000019s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.15%
    jit_opt_a.renormalize                :     0.000525s :  6.60%
    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 :  0.22%
    jit_opt_a.cse                        :     0.000033s :  0.41%
    jit_opt_a.replace_applicator         :     0.000019s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000028s :  0.36%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000462s :  5.82%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000018s :  0.22%
    loop_unroll                          :     0.000405s :  5.09%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.31%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  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 :  0.08%
    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 :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000038s :  0.47%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    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                   :     0.000440s :  5.54%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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                             :     0.000030s :  0.38%


Time group info:
------[substitution.]   0.000175    28
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.70% :     0.000005s :      4: substitution.graph_param_transform
81.41% :     0.000143s :      4: substitution.inline
 1.84% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000004s :      4: substitution.replace_old_param
 7.71% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004202     2
79.70% :     0.003349s :      1: type_inference.infer
20.30% :     0.000853s :      1: type_inference.specialize
------[replace.]   0.000078     8
57.06% :     0.000044s :      4: replace.inline
42.94% :     0.000033s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     8
92.31% :     0.000140s :      4: match.inline
 7.69% :     0.000012s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   912
 1.25% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.06% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     13: predicate.addn_check_dump
 1.28% :     0.000002s :     13: predicate.addn_zero_filter
 1.82% :     0.000002s :     13: predicate.arithmetic_simplify
 1.39% :     0.000002s :     13: predicate.cast_eliminate
 0.60% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     13: predicate.compare_switch_simplify
 1.20% :     0.000002s :     13: predicate.depend_value_elim
 1.18% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.55% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     13: predicate.environ_get_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.30% :     0.000007s :     29: predicate.inline
 0.99% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      8: predicate.less_batch_normalization
 1.64% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     21: predicate.load_eliminater
 1.38% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     13: predicate.merge_addn
 1.16% :     0.000002s :     13: predicate.minmaximum_grad
 1.35% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.67% :     0.000004s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.78% :     0.000001s :      8: predicate.replace_old_param
 0.36% :     0.000000s :      4: predicate.reset_defer_inline
 1.19% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.29% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     21: predicate.switch_defer_inline
 2.31% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.88% :     0.000010s :     67: predicate.switch_simplify
 1.23% :     0.000002s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.38% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.73% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000002s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000534    12
46.86% :     0.000250s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.00% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.13% :     0.000257s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012879    72
 0.31% :     0.000040s :      1: add_recomputation
 0.45% :     0.000058s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.04% :     0.000263s :      1: bootstrap
 0.16% :     0.000020s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.18% :     0.000023s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000003s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
36.63% :     0.004718s :      1: jit_opt_a
 1.16% :     0.000149s :      1: jit_opt_after_cconv
 0.43% :     0.000056s :      1: jit_opt_b
 3.20% :     0.000412s :      1: loop_unroll
 3.65% :     0.000470s :      1: mutable_eliminate
 7.20% :     0.000927s :     26: opt.transform.jit_opt_a
 0.41% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.48% :     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% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000037s :      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
 2.10% :     0.000270s :      1: renormalize.infer
 1.93% :     0.000248s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000031s :      1: rewriter_after_opt_a
 0.67% :     0.000086s :      1: rewriter_before_opt_a
 0.58% :     0.000075s :      1: symbol_engine_optimizer
33.11% :     0.004264s :      1: type_inference


TotalTime = 0.0112152, [30]
[bootstrap]: 0.00024909
[type_inference]: 0.00416274
[event_method]: 2.012e-05
[auto_monad]: 5.002e-05
[graph_reusing]: 4.18001e-06
[pre_auto_parallel]: 1.87999e-06
[py_interpret_to_execute]: 3.137e-05
[rewriter_before_opt_a]: 8.102e-05
[expand_dump_flag]: 2.27999e-06
[jit_opt_a]: 0.00467404, [2]
    [Cycle 1]: 0.00148055, [27]
        [switch_simplify]: 4.95e-05
        [loop_unroll]: 3.472e-05
        [a_1]: 0.00057088
        [with_stream_mark]: 1.196e-05
        [recompute_prepare]: 8.03001e-06
        [updatestate_depend_eliminate]: 3.4e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 6.89001e-06
        [updatestate_useless_node_eliminater]: 6.21998e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 5.76998e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 8.79983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.189e-05
        [j_node_and_user_rematch]: 9.47001e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 1.001e-05
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.0005167
        [add_forward_monad_depend]: 4.25e-06
        [auto_monad_grad]: 1.73002e-06
        [auto_monad_eliminator]: 1.161e-05
        [cse]: 1.939e-05
        [replace_applicator]: 1.306e-05
    [Cycle 2]: 0.00035506, [27]
        [switch_simplify]: 6.44999e-06
        [loop_unroll]: 5.62999e-06
        [a_1]: 0.0001163
        [with_stream_mark]: 8.21002e-06
        [recompute_prepare]: 5.71e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.60001e-06
        [updatestate_useless_node_eliminater]: 5.51e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 5.13002e-06
        [merge_forward]: 2.73998e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.472e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 9.66e-06
        [inline_without_move]: 5.62999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 6.21e-06
        [cse]: 1.275e-05
        [replace_applicator]: 5.67001e-06
[py_interpret_to_execute_after_opt_a]: 9.38002e-06
[rewriter_after_opt_a]: 2.89e-05
[convert_after_rewriter]: 6.53e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.0004658
[jit_opt_b]: 5.21e-05, [1]
    [Cycle 1]: 4.592e-05, [2]
        [frontend_op_eliminate]: 1.78e-05
        [inline_after_opt_a]: 1.668e-05
[cconv]: 1.798e-05
[loop_unroll]: 0.00040955
[jit_opt_after_cconv]: 0.00014281, [1]
    [Cycle 1]: 0.00013687, [11]
        [c_1]: 2.401e-05
        [parameter_eliminate]: 2.68998e-06
        [updatestate_depend_eliminate]: 5.58002e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [cse]: 1.861e-05
        [call_graph_tuple_transform]: 1.878e-05
        [tuple_list_get_item_eliminator]: 5.92999e-06
        [none_parameter_eliminate]: 9.29984e-07
        [renormalize]: 3.59985e-07
        [switch_simplify]: 5.87001e-06
[remove_dup_value]: 9.51998e-06
[partial_unused_args_eliminate]: 1.66e-06
[environ_conv]: 4.75001e-06
[add_recomputation]: 4.76e-05
[cse_after_recomputation]: 2.42e-05, [1]
    [Cycle 1]: 1.906e-05, [1]
        [cse]: 1.291e-05
[auto_monad_reorder]: 1.487e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 4.72998e-06
[opt_after_jit_grad]: 0.00044296
[symbol_engine_optimizer]: 7.381e-05, [1]
    [Cycle 1]: 6.763e-05, [6]
        [build]: 2.69999e-06
        [elim_shapecalc]: 8.32998e-06
        [elim_not_effective]: 1.339e-05
        [opt_reshape]: 6.58e-06
        [fold_const_symbol]: 9.17999e-06
        [renormalize]: 4.89992e-07
[validate]: 2.894e-05
Sums
    bootstrap                            :     0.000249s :  3.18%
    type_inference                       :     0.004163s : 53.06%
    event_method                         :     0.000020s :  0.26%
    auto_monad                           :     0.000050s :  0.64%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.40%
    rewriter_before_opt_a                :     0.000081s :  1.03%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000056s :  0.71%
    jit_opt_a.loop_unroll                :     0.000040s :  0.51%
    jit_opt_a.a_1                        :     0.000687s :  8.76%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.26%
    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 :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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_        :     0.000011s :  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 :     0.000027s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000020s :  0.25%
    jit_opt_a.inline_without_move        :     0.000011s :  0.15%
    jit_opt_a.renormalize                :     0.000517s :  6.59%
    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      :     0.000018s :  0.23%
    jit_opt_a.cse                        :     0.000032s :  0.41%
    jit_opt_a.replace_applicator         :     0.000019s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000029s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000466s :  5.94%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000018s :  0.23%
    loop_unroll                          :     0.000410s :  5.22%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.31%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  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      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000010s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000048s :  0.61%
    cse_after_recomputation.cse          :     0.000013s :  0.16%
    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                   :     0.000443s :  5.65%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000029s :  0.37%


Time group info:
------[substitution.]   0.000175    28
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000005s :      4: substitution.graph_param_transform
81.51% :     0.000143s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.48% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.52% :     0.000004s :      4: substitution.replace_old_param
 7.47% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004115     2
80.02% :     0.003292s :      1: type_inference.infer
19.98% :     0.000822s :      1: type_inference.specialize
------[replace.]   0.000074     8
55.77% :     0.000041s :      4: replace.inline
44.23% :     0.000033s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     8
92.58% :     0.000140s :      4: match.inline
 7.42% :     0.000011s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   912
 1.30% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     13: predicate.addn_check_dump
 1.25% :     0.000002s :     13: predicate.addn_zero_filter
 1.84% :     0.000002s :     13: predicate.arithmetic_simplify
 1.27% :     0.000002s :     13: predicate.cast_eliminate
 0.53% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     13: predicate.compare_switch_simplify
 1.25% :     0.000002s :     13: predicate.depend_value_elim
 1.17% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.48% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      4: predicate.elim_not_effective
 0.51% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     13: predicate.environ_get_eliminate
 1.21% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 1.01% :     0.000001s :      8: predicate.get_grad_eliminate
 0.29% :     0.000000s :      4: predicate.graph_param_transform
 5.21% :     0.000007s :     29: predicate.inline
 1.04% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      8: predicate.less_batch_normalization
 1.61% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     21: predicate.load_eliminater
 1.12% :     0.000001s :      4: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     13: predicate.merge_addn
 1.24% :     0.000002s :     13: predicate.minmaximum_grad
 1.24% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.72% :     0.000004s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     13: predicate.reduce_eliminate
 1.69% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.05% :     0.000003s :     25: predicate.replace_applicator
 0.76% :     0.000001s :      8: predicate.replace_old_param
 0.37% :     0.000000s :      4: predicate.reset_defer_inline
 1.22% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     13: predicate.same_eliminate
 0.65% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.37% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.66% :     0.000003s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.85% :     0.000010s :     67: predicate.switch_simplify
 1.17% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.41% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000526    12
46.99% :     0.000247s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.11% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.90% :     0.000252s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012732    72
 0.40% :     0.000051s :      1: add_recomputation
 0.42% :     0.000053s :      1: auto_monad
 0.14% :     0.000017s :      1: auto_monad_reorder
 2.04% :     0.000259s :      1: bootstrap
 0.16% :     0.000021s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.21% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.18% :     0.000023s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000003s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
36.73% :     0.004677s :      1: jit_opt_a
 1.14% :     0.000146s :      1: jit_opt_after_cconv
 0.43% :     0.000055s :      1: jit_opt_b
 3.28% :     0.000417s :      1: loop_unroll
 3.72% :     0.000474s :      1: mutable_eliminate
 7.19% :     0.000915s :     26: opt.transform.jit_opt_a
 0.40% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.54% :     0.000451s :      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% :     0.000034s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000012s :      1: remove_dup_value
 2.09% :     0.000266s :      1: renormalize.infer
 1.92% :     0.000245s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000032s :      1: rewriter_after_opt_a
 0.66% :     0.000084s :      1: rewriter_before_opt_a
 0.60% :     0.000076s :      1: symbol_engine_optimizer
32.78% :     0.004173s :      1: type_inference


TotalTime = 0.0413494, [33]
[bootstrap]: 0.00025816
[type_inference]: 0.020487
[event_method]: 2.015e-05
[auto_monad]: 0.00014596
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 2.71999e-06
[py_interpret_to_execute]: 0.00011676
[rewriter_before_opt_a]: 9.832e-05
[expand_dump_flag]: 3.53999e-06
[jit_opt_a]: 0.00576682, [2]
    [Cycle 1]: 0.00211497, [27]
        [switch_simplify]: 5.965e-05
        [loop_unroll]: 3.652e-05
        [a_1]: 0.00068031
        [with_stream_mark]: 1.885e-05
        [recompute_prepare]: 1.001e-05
        [updatestate_depend_eliminate]: 5.90002e-06
        [updatestate_assign_eliminate]: 5.65001e-06
        [updatestate_loads_eliminate]: 4.93001e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 9.10999e-06
        [updatestate_useless_node_eliminater]: 1.126e-05
        [accelerated_algorithm]: 8.58001e-06
        [meta_shard_fg_expand]: 2.88e-06
        [get_grad_eliminate_]: 7.93999e-06
        [merge_forward]: 5.05001e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.943e-05
        [j_node_and_user_rematch]: 1.337e-05
        [meta_fg_expand]: 3.68999e-06
        [replace_old_param]: 1.08e-05
        [inline_without_move]: 7.85e-06
        [renormalize]: 0.00094337
        [add_forward_monad_depend]: 5.39e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.989e-05
        [cse]: 3.709e-05
        [replace_applicator]: 1.486e-05
    [Cycle 2]: 0.00045275, [27]
        [switch_simplify]: 8.55001e-06
        [loop_unroll]: 8.1e-06
        [a_1]: 0.00016406
        [with_stream_mark]: 1.022e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 4.99e-06
        [updatestate_assign_eliminate]: 4.25e-06
        [updatestate_loads_eliminate]: 3.83001e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 7.63001e-06
        [updatestate_useless_node_eliminater]: 1.056e-05
        [accelerated_algorithm]: 7.61001e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 7.28999e-06
        [merge_forward]: 3.96001e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.563e-05
        [j_node_and_user_rematch]: 1.212e-05
        [meta_fg_expand]: 2.81999e-06
        [replace_old_param]: 9.67999e-06
        [inline_without_move]: 7.38999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 9.69999e-06
        [cse]: 1.946e-05
        [replace_applicator]: 7.64002e-06
[py_interpret_to_execute_after_opt_a]: 1.033e-05
[rewriter_after_opt_a]: 0.00029777
[convert_after_rewriter]: 1.152e-05
[order_py_execute_after_rewriter]: 7.36999e-06
[mutable_eliminate]: 0.00051303
[jit_opt_b]: 6.586e-05, [1]
    [Cycle 1]: 5.949e-05, [2]
        [frontend_op_eliminate]: 2.346e-05
        [inline_after_opt_a]: 2.436e-05
[cconv]: 2.571e-05
[loop_unroll]: 0.00044609
[jit_opt_after_cconv]: 0.00018812, [1]
    [Cycle 1]: 0.00018182, [11]
        [c_1]: 4.388e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 7.97998e-06
        [updatestate_assign_eliminate]: 4.37e-06
        [updatestate_loads_eliminate]: 3.76001e-06
        [cse]: 2.683e-05
        [call_graph_tuple_transform]: 2.205e-05
        [tuple_list_get_item_eliminator]: 8.13001e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 8.46002e-06
[remove_dup_value]: 2.059e-05
[partial_unused_args_eliminate]: 2.33002e-06
[environ_conv]: 1.117e-05
[add_recomputation]: 6.269e-05
[cse_after_recomputation]: 2.932e-05, [1]
    [Cycle 1]: 2.392e-05, [1]
        [cse]: 1.803e-05
[auto_monad_reorder]: 2.533e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 5.04e-06
[opt_after_jit_grad]: 0.00049001
[symbol_engine_optimizer]: 9.989e-05, [1]
    [Cycle 1]: 9.358e-05, [6]
        [build]: 1.143e-05
        [elim_shapecalc]: 1.174e-05
        [elim_not_effective]: 1.85e-05
        [opt_reshape]: 8.74e-06
        [fold_const_symbol]: 1.323e-05
        [renormalize]: 4.80009e-07
[validate]: 5.251e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.0118598
[execute]: 8.03999e-06
Sums
    bootstrap                            :     0.000258s :  0.69%
    type_inference                       :     0.020487s : 54.58%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000146s :  0.39%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000117s :  0.31%
    rewriter_before_opt_a                :     0.000098s :  0.26%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.18%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000844s :  2.25%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000025s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.05%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.000943s :  2.51%
    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      :     0.000030s :  0.08%
    jit_opt_a.cse                        :     0.000057s :  0.15%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000298s :  0.79%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000513s :  1.37%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000446s :  1.19%
    jit_opt_after_cconv.c_1              :     0.000044s :  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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  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                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000063s :  0.17%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000025s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000490s :  1.31%
    symbol_engine_optimizer.build        :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000053s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011860s : 31.59%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000251    47
 3.19% :     0.000008s :      2: substitution.depend_value_elim
 1.08% :     0.000003s :      4: substitution.elim_not_effective
 0.78% :     0.000002s :      4: substitution.fold_const_symbol
 2.76% :     0.000007s :      5: substitution.graph_param_transform
74.96% :     0.000188s :      4: substitution.inline
 1.71% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.85% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.33% :     0.000003s :      2: substitution.replace_old_param
 4.98% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 2.99% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.36% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.020422     2
93.67% :     0.019129s :      1: type_inference.infer
 6.33% :     0.001293s :      1: type_inference.specialize
------[replace.]   0.000087     6
66.64% :     0.000058s :      4: replace.inline
33.36% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
94.18% :     0.000186s :      4: match.inline
 5.82% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1021
 1.26% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.01% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.32% :     0.000002s :     15: predicate.addn_zero_filter
 1.94% :     0.000003s :     15: predicate.arithmetic_simplify
 1.32% :     0.000002s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.53% :     0.000003s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.91% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      5: predicate.fold_const_symbol
 1.11% :     0.000002s :     10: predicate.get_grad_eliminate
 0.39% :     0.000001s :      5: predicate.graph_param_transform
 5.50% :     0.000009s :     31: predicate.inline
 1.04% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :     10: predicate.less_batch_normalization
 1.44% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.92% :     0.000003s :     22: predicate.load_eliminater
 1.24% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.51% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     15: predicate.merge_addn
 1.30% :     0.000002s :     15: predicate.minmaximum_grad
 1.62% :     0.000003s :      5: predicate.mutable_eliminate
 0.52% :     0.000001s :      5: predicate.opt_reshape
 2.39% :     0.000004s :     22: predicate.partial_eliminate
 1.24% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.69% :     0.000003s :     15: predicate.reduce_eliminate
 1.40% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.82% :     0.000003s :     27: predicate.replace_applicator
 0.75% :     0.000001s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.26% :     0.000002s :     15: predicate.reshape_eliminate
 1.26% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.01% :     0.000002s :     10: predicate.special_op_eliminate
 1.09% :     0.000002s :     10: predicate.specialize_transform
 1.48% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.26% :     0.000004s :     21: predicate.switch_defer_inline
 1.93% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.68% :     0.000011s :     68: predicate.switch_simplify
 1.30% :     0.000002s :     15: predicate.tile_eliminate
 1.34% :     0.000002s :     15: predicate.transpose_eliminate
 1.64% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.94% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.28% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000003s :     15: predicate.value_based_eliminate
 0.46% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000738    13
44.36% :     0.000327s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.70% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.94% :     0.000369s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043606    76
 0.15% :     0.000066s :      1: add_recomputation
 0.35% :     0.000151s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.62% :     0.000270s :      1: bootstrap
 0.06% :     0.000028s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.06% :     0.000025s :      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
13.23% :     0.005770s :      1: jit_opt_a
 0.44% :     0.000191s :      1: jit_opt_after_cconv
 0.16% :     0.000069s :      1: jit_opt_b
 1.04% :     0.000454s :      1: loop_unroll
 1.20% :     0.000522s :      1: mutable_eliminate
 2.60% :     0.001136s :     26: opt.transform.jit_opt_a
 0.18% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000048s :      4: opt.transform.symbol_engine_opt
 1.14% :     0.000498s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000120s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000023s :      1: remove_dup_value
 1.26% :     0.000548s :      1: renormalize.infer
 0.89% :     0.000387s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.69% :     0.000302s :      1: rewriter_after_opt_a
 0.23% :     0.000102s :      1: rewriter_before_opt_a
 0.24% :     0.000103s :      1: symbol_engine_optimizer
27.22% :     0.011870s :      1: task_emit
47.02% :     0.020502s :      1: type_inference
 0.17% :     0.000075s :      1: validate


TotalTime = 0.0125089, [30]
[bootstrap]: 0.00025023
[type_inference]: 0.00447066
[event_method]: 2.011e-05
[auto_monad]: 6.277e-05
[graph_reusing]: 5.54998e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 3.456e-05
[rewriter_before_opt_a]: 8.69e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.00556982, [2]
    [Cycle 1]: 0.00179073, [27]
        [switch_simplify]: 5.789e-05
        [loop_unroll]: 3.454e-05
        [a_1]: 0.00061418
        [with_stream_mark]: 1.767e-05
        [recompute_prepare]: 8.08001e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.99999e-06
        [specialize_transform]: 6.84999e-06
        [updatestate_useless_node_eliminater]: 5.69999e-06
        [accelerated_algorithm]: 6.03002e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 3.65998e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 9.61e-06
        [meta_fg_expand]: 2.35002e-06
        [replace_old_param]: 1.05e-05
        [inline_without_move]: 5.91998e-06
        [renormalize]: 0.00073992
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 2.11e-06
        [auto_monad_eliminator]: 1.461e-05
        [cse]: 2.985e-05
        [replace_applicator]: 1.315e-05
    [Cycle 2]: 0.00036447, [27]
        [switch_simplify]: 6.69001e-06
        [loop_unroll]: 6.14001e-06
        [a_1]: 0.00011713
        [with_stream_mark]: 9.51998e-06
        [recompute_prepare]: 1.048e-05
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 6.36998e-06
        [updatestate_useless_node_eliminater]: 5.66998e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 5.20999e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.29e-05
        [j_node_and_user_rematch]: 8.70999e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 9.52001e-06
        [inline_without_move]: 5.96e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 6.14001e-06
        [cse]: 1.303e-05
        [replace_applicator]: 6.09001e-06
[py_interpret_to_execute_after_opt_a]: 1.065e-05
[rewriter_after_opt_a]: 3.282e-05
[convert_after_rewriter]: 6.78e-06
[order_py_execute_after_rewriter]: 5.69999e-06
[mutable_eliminate]: 0.00047691
[jit_opt_b]: 5.338e-05, [1]
    [Cycle 1]: 4.731e-05, [2]
        [frontend_op_eliminate]: 1.894e-05
        [inline_after_opt_a]: 1.699e-05
[cconv]: 2.438e-05
[loop_unroll]: 0.0004114
[jit_opt_after_cconv]: 0.00014732, [1]
    [Cycle 1]: 0.00014099, [11]
        [c_1]: 2.491e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 5.76e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [cse]: 1.946e-05
        [call_graph_tuple_transform]: 2.033e-05
        [tuple_list_get_item_eliminator]: 6.03002e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 3.09985e-07
        [switch_simplify]: 6.06e-06
[remove_dup_value]: 1.36e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.88002e-06
[add_recomputation]: 4.68e-05
[cse_after_recomputation]: 2.261e-05, [1]
    [Cycle 1]: 1.74e-05, [1]
        [cse]: 1.194e-05
[auto_monad_reorder]: 1.79e-05
[get_jit_bprop_graph]: 1.51002e-06
[rewriter_after_jit_bprop_graph]: 4.97e-06
[opt_after_jit_grad]: 0.00045404
[symbol_engine_optimizer]: 7.448e-05, [1]
    [Cycle 1]: 6.84e-05, [6]
        [build]: 3.31999e-06
        [elim_shapecalc]: 8.3e-06
        [elim_not_effective]: 1.345e-05
        [opt_reshape]: 6.04999e-06
        [fold_const_symbol]: 9.56e-06
        [renormalize]: 5.40022e-07
[validate]: 3.443e-05
Sums
    bootstrap                            :     0.000250s :  2.93%
    type_inference                       :     0.004471s : 52.32%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000063s :  0.73%
    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                :     0.000087s :  1.02%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.76%
    jit_opt_a.loop_unroll                :     0.000041s :  0.48%
    jit_opt_a.a_1                        :     0.000731s :  8.56%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.22%
    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       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.13%
    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_        :     0.000011s :  0.13%
    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    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000740s :  8.66%
    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      :     0.000021s :  0.24%
    jit_opt_a.cse                        :     0.000043s :  0.50%
    jit_opt_a.replace_applicator         :     0.000019s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000477s :  5.58%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000024s :  0.29%
    loop_unroll                          :     0.000411s :  4.81%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.29%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  0.23%
    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 :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000047s :  0.55%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000454s :  5.31%
    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 :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000034s :  0.40%


Time group info:
------[substitution.]   0.000211    28
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000006s :      4: substitution.graph_param_transform
81.96% :     0.000173s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000005s :      4: substitution.replace_old_param
 7.88% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004408     2
80.25% :     0.003537s :      1: type_inference.infer
19.75% :     0.000871s :      1: type_inference.specialize
------[replace.]   0.000079     8
56.39% :     0.000045s :      4: replace.inline
43.61% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     8
92.03% :     0.000170s :      4: match.inline
 7.97% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   912
 1.25% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.03% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     13: predicate.addn_check_dump
 1.29% :     0.000002s :     13: predicate.addn_zero_filter
 2.13% :     0.000003s :     13: predicate.arithmetic_simplify
 1.33% :     0.000002s :     13: predicate.cast_eliminate
 0.49% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     13: predicate.compare_switch_simplify
 1.17% :     0.000002s :     13: predicate.depend_value_elim
 1.11% :     0.000001s :     13: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.57% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     13: predicate.environ_get_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 0.92% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.35% :     0.000007s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.14% :     0.000002s :      8: predicate.less_batch_normalization
 1.59% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     21: predicate.load_eliminater
 1.28% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.12% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     13: predicate.merge_addn
 1.17% :     0.000002s :     13: predicate.minmaximum_grad
 1.18% :     0.000002s :      4: predicate.mutable_eliminate
 0.46% :     0.000001s :      4: predicate.opt_reshape
 2.62% :     0.000004s :     21: predicate.partial_eliminate
 1.22% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     13: predicate.reduce_eliminate
 1.64% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.96% :     0.000003s :     25: predicate.replace_applicator
 0.78% :     0.000001s :      8: predicate.replace_old_param
 0.41% :     0.000001s :      4: predicate.reset_defer_inline
 1.21% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     13: predicate.same_eliminate
 0.66% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.37% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.73% :     0.000004s :     21: predicate.switch_defer_inline
 2.16% :     0.000003s :     21: predicate.switch_layer_defer_inline
 8.00% :     0.000011s :     67: predicate.switch_simplify
 1.19% :     0.000002s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.51% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.95% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     13: predicate.value_based_eliminate
 0.36% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000561    12
46.87% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.89% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.25% :     0.000265s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014300    72
 0.35% :     0.000050s :      1: add_recomputation
 0.46% :     0.000066s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 1.83% :     0.000262s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.17% :     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
38.97% :     0.005573s :      1: jit_opt_a
 1.05% :     0.000150s :      1: jit_opt_after_cconv
 0.39% :     0.000056s :      1: jit_opt_b
 2.93% :     0.000419s :      1: loop_unroll
 3.39% :     0.000485s :      1: mutable_eliminate
 6.81% :     0.000973s :     26: opt.transform.jit_opt_a
 0.38% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000030s :      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
 0.16% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.23% :     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% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000038s :      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
 3.13% :     0.000448s :      1: renormalize.infer
 1.99% :     0.000285s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000036s :      1: rewriter_after_opt_a
 0.63% :     0.000090s :      1: rewriter_before_opt_a
 0.54% :     0.000077s :      1: symbol_engine_optimizer
31.35% :     0.004483s :      1: type_inference


TotalTime = 0.0116972, [30]
[bootstrap]: 0.00024719
[type_inference]: 0.00423863
[event_method]: 1.989e-05
[auto_monad]: 5.711e-05
[graph_reusing]: 4.97e-06
[pre_auto_parallel]: 2.20002e-06
[py_interpret_to_execute]: 3.19e-05
[rewriter_before_opt_a]: 8.286e-05
[expand_dump_flag]: 2.84001e-06
[jit_opt_a]: 0.00499238, [2]
    [Cycle 1]: 0.00159994, [27]
        [switch_simplify]: 5.468e-05
        [loop_unroll]: 3.49e-05
        [a_1]: 0.0005967
        [with_stream_mark]: 1.525e-05
        [recompute_prepare]: 8.75001e-06
        [updatestate_depend_eliminate]: 4.03001e-06
        [updatestate_assign_eliminate]: 3.58e-06
        [updatestate_loads_eliminate]: 2.70002e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 7.06999e-06
        [updatestate_useless_node_eliminater]: 5.87001e-06
        [accelerated_algorithm]: 6.19999e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.74999e-06
        [merge_forward]: 3.09999e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.491e-05
        [j_node_and_user_rematch]: 9.89001e-06
        [meta_fg_expand]: 2.61999e-06
        [replace_old_param]: 1.074e-05
        [inline_without_move]: 5.94999e-06
        [renormalize]: 0.00057829
        [add_forward_monad_depend]: 4.58999e-06
        [auto_monad_grad]: 1.76e-06
        [auto_monad_eliminator]: 1.32e-05
        [cse]: 2.777e-05
        [replace_applicator]: 1.246e-05
    [Cycle 2]: 0.00035363, [27]
        [switch_simplify]: 6.41998e-06
        [loop_unroll]: 5.56e-06
        [a_1]: 0.00011161
        [with_stream_mark]: 8.2e-06
        [recompute_prepare]: 5.46002e-06
        [updatestate_depend_eliminate]: 3.07002e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.46998e-06
        [updatestate_useless_node_eliminater]: 1.057e-05
        [accelerated_algorithm]: 5.53002e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 1.29003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.301e-05
        [j_node_and_user_rematch]: 8.62e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 1.034e-05
        [inline_without_move]: 5.89999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.82999e-06
        [cse]: 1.346e-05
        [replace_applicator]: 6.16e-06
[py_interpret_to_execute_after_opt_a]: 8.98002e-06
[rewriter_after_opt_a]: 3.137e-05
[convert_after_rewriter]: 6.81001e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00051571
[jit_opt_b]: 5.283e-05, [1]
    [Cycle 1]: 4.618e-05, [2]
        [frontend_op_eliminate]: 1.794e-05
        [inline_after_opt_a]: 1.669e-05
[cconv]: 2.3e-05
[loop_unroll]: 0.00040774
[jit_opt_after_cconv]: 0.00014614, [1]
    [Cycle 1]: 0.00014004, [11]
        [c_1]: 2.473e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 5.59998e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 1.947e-05
        [call_graph_tuple_transform]: 1.88e-05
        [tuple_list_get_item_eliminator]: 6.02999e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.92999e-06
[remove_dup_value]: 1.35e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.30999e-06
[add_recomputation]: 4.359e-05
[cse_after_recomputation]: 2.34e-05, [1]
    [Cycle 1]: 1.8e-05, [1]
        [cse]: 1.203e-05
[auto_monad_reorder]: 1.671e-05
[get_jit_bprop_graph]: 1.47999e-06
[rewriter_after_jit_bprop_graph]: 5.47999e-06
[opt_after_jit_grad]: 0.00045517
[symbol_engine_optimizer]: 7.327e-05, [1]
    [Cycle 1]: 6.722e-05, [6]
        [build]: 2.79999e-06
        [elim_shapecalc]: 8.10999e-06
        [elim_not_effective]: 1.353e-05
        [opt_reshape]: 6.26e-06
        [fold_const_symbol]: 9.00001e-06
        [renormalize]: 5.00004e-07
[validate]: 3.251e-05
Sums
    bootstrap                            :     0.000247s :  3.04%
    type_inference                       :     0.004239s : 52.20%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000057s :  0.70%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.39%
    rewriter_before_opt_a                :     0.000083s :  1.02%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.75%
    jit_opt_a.loop_unroll                :     0.000040s :  0.50%
    jit_opt_a.a_1                        :     0.000708s :  8.72%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.09%
    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        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.20%
    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_        :     0.000011s :  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 :     0.000028s :  0.34%
    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          :     0.000021s :  0.26%
    jit_opt_a.inline_without_move        :     0.000012s :  0.15%
    jit_opt_a.renormalize                :     0.000578s :  7.12%
    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      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000041s :  0.51%
    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 :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000516s :  6.35%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000023s :  0.28%
    loop_unroll                          :     0.000408s :  5.02%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.30%
    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              :     0.000019s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  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                     :     0.000013s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000044s :  0.54%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000017s :  0.21%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000455s :  5.61%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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 :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000033s :  0.40%


Time group info:
------[substitution.]   0.000198    28
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000005s :      4: substitution.graph_param_transform
81.73% :     0.000162s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.43% :     0.000005s :      4: substitution.replace_old_param
 7.91% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004188     2
79.90% :     0.003347s :      1: type_inference.infer
20.10% :     0.000842s :      1: type_inference.specialize
------[replace.]   0.000080     8
57.03% :     0.000045s :      4: replace.inline
42.97% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000173     8
92.05% :     0.000160s :      4: match.inline
 7.95% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   912
 1.21% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.01% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     13: predicate.addn_check_dump
 1.26% :     0.000002s :     13: predicate.addn_zero_filter
 1.82% :     0.000002s :     13: predicate.arithmetic_simplify
 1.53% :     0.000002s :     13: predicate.cast_eliminate
 0.40% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     13: predicate.compare_switch_simplify
 1.21% :     0.000002s :     13: predicate.depend_value_elim
 1.16% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.52% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     13: predicate.environ_get_eliminate
 1.13% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 0.91% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.54% :     0.000007s :     29: predicate.inline
 0.93% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      8: predicate.less_batch_normalization
 1.76% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     21: predicate.load_eliminater
 1.35% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     42: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     13: predicate.merge_addn
 1.18% :     0.000002s :     13: predicate.minmaximum_grad
 1.53% :     0.000002s :      4: predicate.mutable_eliminate
 0.46% :     0.000001s :      4: predicate.opt_reshape
 2.65% :     0.000003s :     21: predicate.partial_eliminate
 1.24% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.07% :     0.000003s :     25: predicate.replace_applicator
 0.64% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000000s :      4: predicate.reset_defer_inline
 1.21% :     0.000002s :     13: predicate.reshape_eliminate
 1.22% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     21: predicate.switch_defer_inline
 2.21% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.73% :     0.000010s :     67: predicate.switch_simplify
 1.20% :     0.000002s :     13: predicate.tile_eliminate
 1.28% :     0.000002s :     13: predicate.transpose_eliminate
 1.41% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000002s :     21: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     13: predicate.value_based_eliminate
 0.43% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000542    12
47.40% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.24% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.36% :     0.000256s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013304    72
 0.35% :     0.000046s :      1: add_recomputation
 0.45% :     0.000060s :      1: auto_monad
 0.14% :     0.000019s :      1: auto_monad_reorder
 1.94% :     0.000258s :      1: bootstrap
 0.19% :     0.000025s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.18% :     0.000023s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
37.55% :     0.004995s :      1: jit_opt_a
 1.12% :     0.000149s :      1: jit_opt_after_cconv
 0.42% :     0.000055s :      1: jit_opt_b
 3.12% :     0.000415s :      1: loop_unroll
 3.93% :     0.000523s :      1: mutable_eliminate
 7.12% :     0.000948s :     26: opt.transform.jit_opt_a
 0.39% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.48% :     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% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000035s :      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
 2.30% :     0.000306s :      1: renormalize.infer
 2.00% :     0.000265s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000034s :      1: rewriter_after_opt_a
 0.64% :     0.000086s :      1: rewriter_before_opt_a
 0.57% :     0.000076s :      1: symbol_engine_optimizer
31.94% :     0.004249s :      1: type_inference


TotalTime = 0.0486754, [33]
[bootstrap]: 0.00026858
[type_inference]: 0.0292008
[event_method]: 1.917e-05
[auto_monad]: 0.00014603
[graph_reusing]: 5.76e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 0.00010984
[rewriter_before_opt_a]: 9.441e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.0059531, [2]
    [Cycle 1]: 0.00208175, [27]
        [switch_simplify]: 5.614e-05
        [loop_unroll]: 3.576e-05
        [a_1]: 0.0006719
        [with_stream_mark]: 1.652e-05
        [recompute_prepare]: 1.026e-05
        [updatestate_depend_eliminate]: 5.97001e-06
        [updatestate_assign_eliminate]: 5.42999e-06
        [updatestate_loads_eliminate]: 4.63001e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 9.14e-06
        [updatestate_useless_node_eliminater]: 1.15e-05
        [accelerated_algorithm]: 8.75999e-06
        [meta_shard_fg_expand]: 2.63998e-06
        [get_grad_eliminate_]: 8.15e-06
        [merge_forward]: 5.40999e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.845e-05
        [j_node_and_user_rematch]: 1.374e-05
        [meta_fg_expand]: 4.03001e-06
        [replace_old_param]: 1.144e-05
        [inline_without_move]: 8.64e-06
        [renormalize]: 0.00092469
        [add_forward_monad_depend]: 5.05999e-06
        [auto_monad_grad]: 2.14e-06
        [auto_monad_eliminator]: 1.933e-05
        [cse]: 3.408e-05
        [replace_applicator]: 1.453e-05
    [Cycle 2]: 0.00045234, [27]
        [switch_simplify]: 8.85001e-06
        [loop_unroll]: 8.17e-06
        [a_1]: 0.00016308
        [with_stream_mark]: 9.37001e-06
        [recompute_prepare]: 8.15999e-06
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 4.38999e-06
        [updatestate_loads_eliminate]: 3.68e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 8.17e-06
        [updatestate_useless_node_eliminater]: 1.07e-05
        [accelerated_algorithm]: 7.66001e-06
        [meta_shard_fg_expand]: 1.64998e-06
        [get_grad_eliminate_]: 7.29001e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.535e-05
        [j_node_and_user_rematch]: 1.238e-05
        [meta_fg_expand]: 2.73e-06
        [replace_old_param]: 9.65002e-06
        [inline_without_move]: 7.68001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 9.86998e-06
        [cse]: 1.9e-05
        [replace_applicator]: 7.82998e-06
[py_interpret_to_execute_after_opt_a]: 1.036e-05
[rewriter_after_opt_a]: 0.00027791
[convert_after_rewriter]: 1.043e-05
[order_py_execute_after_rewriter]: 7.72998e-06
[mutable_eliminate]: 0.0005243
[jit_opt_b]: 6.711e-05, [1]
    [Cycle 1]: 6.057e-05, [2]
        [frontend_op_eliminate]: 2.402e-05
        [inline_after_opt_a]: 2.41e-05
[cconv]: 2.473e-05
[loop_unroll]: 0.00043872
[jit_opt_after_cconv]: 0.00018393, [1]
    [Cycle 1]: 0.00017773, [11]
        [c_1]: 4.217e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 7.87e-06
        [updatestate_assign_eliminate]: 4.42003e-06
        [updatestate_loads_eliminate]: 3.85e-06
        [cse]: 2.427e-05
        [call_graph_tuple_transform]: 2.25e-05
        [tuple_list_get_item_eliminator]: 8.42998e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 8.77e-06
[remove_dup_value]: 1.848e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 9.82999e-06
[add_recomputation]: 6.093e-05
[cse_after_recomputation]: 3.102e-05, [1]
    [Cycle 1]: 2.525e-05, [1]
        [cse]: 1.882e-05
[auto_monad_reorder]: 2.557e-05
[get_jit_bprop_graph]: 1.85001e-06
[rewriter_after_jit_bprop_graph]: 4.33999e-06
[opt_after_jit_grad]: 0.00049216
[symbol_engine_optimizer]: 9.83e-05, [1]
    [Cycle 1]: 9.153e-05, [6]
        [build]: 9.56e-06
        [elim_shapecalc]: 1.156e-05
        [elim_not_effective]: 1.876e-05
        [opt_reshape]: 8.50999e-06
        [fold_const_symbol]: 1.346e-05
        [renormalize]: 4.50003e-07
[validate]: 4.973e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.010318
[execute]: 7.57002e-06
Sums
    bootstrap                            :     0.000269s :  0.60%
    type_inference                       :     0.029201s : 65.41%
    event_method                         :     0.000019s :  0.04%
    auto_monad                           :     0.000146s :  0.33%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000110s :  0.25%
    rewriter_before_opt_a                :     0.000094s :  0.21%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.15%
    jit_opt_a.loop_unroll                :     0.000044s :  0.10%
    jit_opt_a.a_1                        :     0.000835s :  1.87%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.03%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000925s :  2.07%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.07%
    jit_opt_a.cse                        :     0.000053s :  0.12%
    jit_opt_a.replace_applicator         :     0.000022s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.02%
    rewriter_after_opt_a                 :     0.000278s :  0.62%
    convert_after_rewriter               :     0.000010s :  0.02%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000524s :  1.17%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.05%
    cconv                                :     0.000025s :  0.06%
    loop_unroll                          :     0.000439s :  0.98%
    jit_opt_after_cconv.c_1              :     0.000042s :  0.09%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000018s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.02%
    add_recomputation                    :     0.000061s :  0.14%
    cse_after_recomputation.cse          :     0.000019s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000492s :  1.10%
    symbol_engine_optimizer.build        :     0.000010s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000050s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010318s : 23.11%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000249    47
 3.15% :     0.000008s :      2: substitution.depend_value_elim
 1.12% :     0.000003s :      4: substitution.elim_not_effective
 0.83% :     0.000002s :      4: substitution.fold_const_symbol
 2.80% :     0.000007s :      5: substitution.graph_param_transform
74.76% :     0.000186s :      4: substitution.inline
 1.79% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.78% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.52% :     0.000004s :      2: substitution.replace_old_param
 4.74% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.25% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.24% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.029137     2
95.25% :     0.027752s :      1: type_inference.infer
 4.75% :     0.001385s :      1: type_inference.specialize
------[replace.]   0.000087     6
66.44% :     0.000058s :      4: replace.inline
33.56% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000194     6
94.52% :     0.000183s :      4: match.inline
 5.48% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000164  1021
 1.32% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     15: predicate.addn_check_dump
 1.34% :     0.000002s :     15: predicate.addn_zero_filter
 1.98% :     0.000003s :     15: predicate.arithmetic_simplify
 1.38% :     0.000002s :     15: predicate.cast_eliminate
 0.57% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     15: predicate.compare_switch_simplify
 1.35% :     0.000002s :     15: predicate.depend_value_elim
 1.23% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     15: predicate.environ_get_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.29% :     0.000000s :      5: predicate.graph_param_transform
 5.11% :     0.000008s :     31: predicate.inline
 1.10% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :     10: predicate.less_batch_normalization
 1.52% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     22: predicate.load_eliminater
 1.35% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.56% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.43% :     0.000002s :      5: predicate.mutable_eliminate
 0.51% :     0.000001s :      5: predicate.opt_reshape
 2.35% :     0.000004s :     22: predicate.partial_eliminate
 1.21% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.81% :     0.000003s :     15: predicate.reduce_eliminate
 1.42% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.65% :     0.000003s :     27: predicate.replace_applicator
 0.77% :     0.000001s :     10: predicate.replace_old_param
 0.32% :     0.000001s :      5: predicate.reset_defer_inline
 1.21% :     0.000002s :     15: predicate.reshape_eliminate
 1.37% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.99% :     0.000002s :     10: predicate.special_op_eliminate
 1.14% :     0.000002s :     10: predicate.specialize_transform
 1.50% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.24% :     0.000004s :     21: predicate.switch_defer_inline
 1.93% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.62% :     0.000011s :     68: predicate.switch_simplify
 1.29% :     0.000002s :     15: predicate.tile_eliminate
 1.32% :     0.000002s :     15: predicate.transpose_eliminate
 1.60% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.61% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 2.01% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.36% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000782    13
45.11% :     0.000353s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.18% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.71% :     0.000388s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050907    76
 0.13% :     0.000064s :      1: add_recomputation
 0.30% :     0.000151s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.56% :     0.000283s :      1: bootstrap
 0.05% :     0.000027s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.02% :     0.000012s :      1: environ_conv
 0.05% :     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% :     0.000009s :      1: graph_reusing
11.70% :     0.005956s :      1: jit_opt_a
 0.37% :     0.000187s :      1: jit_opt_after_cconv
 0.14% :     0.000070s :      1: jit_opt_b
 0.88% :     0.000447s :      1: loop_unroll
 1.05% :     0.000533s :      1: mutable_eliminate
 2.21% :     0.001126s :     26: opt.transform.jit_opt_a
 0.15% :     0.000078s :      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% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000049s :      4: opt.transform.symbol_engine_opt
 0.98% :     0.000501s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000114s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.04% :     0.000530s :      1: renormalize.infer
 0.76% :     0.000386s :      1: renormalize.specialize
 0.01% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.56% :     0.000283s :      1: rewriter_after_opt_a
 0.19% :     0.000099s :      1: rewriter_before_opt_a
 0.20% :     0.000101s :      1: symbol_engine_optimizer
20.29% :     0.010330s :      1: task_emit
57.39% :     0.029213s :      1: type_inference
 0.14% :     0.000071s :      1: validate


TotalTime = 0.0114505, [30]
[bootstrap]: 0.00023974
[type_inference]: 0.0042877
[event_method]: 1.969e-05
[auto_monad]: 4.315e-05
[graph_reusing]: 4.60999e-06
[pre_auto_parallel]: 1.69e-06
[py_interpret_to_execute]: 3.052e-05
[rewriter_before_opt_a]: 8.01e-05
[expand_dump_flag]: 1.92999e-06
[jit_opt_a]: 0.00475927, [2]
    [Cycle 1]: 0.00149656, [27]
        [switch_simplify]: 5.014e-05
        [loop_unroll]: 3.459e-05
        [a_1]: 0.00056204
        [with_stream_mark]: 1.173e-05
        [recompute_prepare]: 8.03999e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 7.11001e-06
        [updatestate_useless_node_eliminater]: 6.12001e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 8.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.184e-05
        [j_node_and_user_rematch]: 9.51e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 9.59e-06
        [inline_without_move]: 5.79e-06
        [renormalize]: 0.00054119
        [add_forward_monad_depend]: 4.18999e-06
        [auto_monad_grad]: 1.45001e-06
        [auto_monad_eliminator]: 1.094e-05
        [cse]: 2.189e-05
        [replace_applicator]: 1.208e-05
    [Cycle 2]: 0.00035902, [27]
        [switch_simplify]: 6.61999e-06
        [loop_unroll]: 5.60001e-06
        [a_1]: 0.00011569
        [with_stream_mark]: 8.38999e-06
        [recompute_prepare]: 5.64e-06
        [updatestate_depend_eliminate]: 3.01001e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 5.70001e-06
        [updatestate_useless_node_eliminater]: 5.49998e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 5.13002e-06
        [merge_forward]: 2.75002e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.278e-05
        [j_node_and_user_rematch]: 8.40001e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 9.87999e-06
        [inline_without_move]: 5.66e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.04998e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 6.05002e-06
        [cse]: 1.407e-05
        [replace_applicator]: 5.79999e-06
[py_interpret_to_execute_after_opt_a]: 9.00999e-06
[rewriter_after_opt_a]: 2.742e-05
[convert_after_rewriter]: 5.84e-06
[order_py_execute_after_rewriter]: 4.27998e-06
[mutable_eliminate]: 0.00048513
[jit_opt_b]: 5.288e-05, [1]
    [Cycle 1]: 4.621e-05, [2]
        [frontend_op_eliminate]: 1.806e-05
        [inline_after_opt_a]: 1.659e-05
[cconv]: 1.769e-05
[loop_unroll]: 0.00040026
[jit_opt_after_cconv]: 0.00019967, [1]
    [Cycle 1]: 0.00019345, [11]
        [c_1]: 2.474e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 1.952e-05
        [call_graph_tuple_transform]: 1.865e-05
        [tuple_list_get_item_eliminator]: 5.972e-05
        [none_parameter_eliminate]: 1.31998e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 6.39999e-06
[remove_dup_value]: 9.67001e-06
[partial_unused_args_eliminate]: 1.81e-06
[environ_conv]: 4.47e-06
[add_recomputation]: 3.636e-05
[cse_after_recomputation]: 2.375e-05, [1]
    [Cycle 1]: 1.828e-05, [1]
        [cse]: 1.236e-05
[auto_monad_reorder]: 1.344e-05
[get_jit_bprop_graph]: 1.37999e-06
[rewriter_after_jit_bprop_graph]: 4.82998e-06
[opt_after_jit_grad]: 0.00043912
[symbol_engine_optimizer]: 7.12e-05, [1]
    [Cycle 1]: 6.533e-05, [6]
        [build]: 2.53e-06
        [elim_shapecalc]: 8.41002e-06
        [elim_not_effective]: 1.228e-05
        [opt_reshape]: 6.14001e-06
        [fold_const_symbol]: 8.94e-06
        [renormalize]: 4.00003e-07
[validate]: 2.808e-05
Sums
    bootstrap                            :     0.000240s :  2.99%
    type_inference                       :     0.004288s : 53.55%
    event_method                         :     0.000020s :  0.25%
    auto_monad                           :     0.000043s :  0.54%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.38%
    rewriter_before_opt_a                :     0.000080s :  1.00%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.71%
    jit_opt_a.loop_unroll                :     0.000040s :  0.50%
    jit_opt_a.a_1                        :     0.000678s :  8.46%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000014s :  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 :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.15%
    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_        :     0.000011s :  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 :  0.31%
    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          :     0.000019s :  0.24%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000541s :  6.76%
    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      :     0.000017s :  0.21%
    jit_opt_a.cse                        :     0.000036s :  0.45%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000027s :  0.34%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000485s :  6.06%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000018s :  0.22%
    loop_unroll                          :     0.000400s :  5.00%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.31%
    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 :     0.000003s :  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 :     0.000060s :  0.75%
    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 :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000036s :  0.45%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    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                   :     0.000439s :  5.48%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    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 :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.35%


Time group info:
------[substitution.]   0.000167    28
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.63% :     0.000004s :      4: substitution.graph_param_transform
81.72% :     0.000136s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.45% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.53% :     0.000004s :      4: substitution.replace_old_param
 7.38% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004240     2
80.29% :     0.003404s :      1: type_inference.infer
19.71% :     0.000836s :      1: type_inference.specialize
------[replace.]   0.000076     8
55.47% :     0.000042s :      4: replace.inline
44.53% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000144     8
92.89% :     0.000133s :      4: match.inline
 7.11% :     0.000010s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   912
 1.25% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     13: predicate.addn_check_dump
 1.42% :     0.000002s :     13: predicate.addn_zero_filter
 1.74% :     0.000002s :     13: predicate.arithmetic_simplify
 1.36% :     0.000002s :     13: predicate.cast_eliminate
 0.49% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     13: predicate.compare_switch_simplify
 1.55% :     0.000002s :     13: predicate.depend_value_elim
 1.16% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.85% :     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
 1.46% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     13: predicate.environ_get_eliminate
 1.19% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.93% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.23% :     0.000007s :     29: predicate.inline
 0.96% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      8: predicate.less_batch_normalization
 1.64% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.16% :     0.000003s :     21: predicate.load_eliminater
 1.21% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.24% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     13: predicate.merge_addn
 1.16% :     0.000002s :     13: predicate.minmaximum_grad
 1.38% :     0.000002s :      4: predicate.mutable_eliminate
 0.50% :     0.000001s :      4: predicate.opt_reshape
 2.51% :     0.000003s :     21: predicate.partial_eliminate
 1.14% :     0.000001s :     13: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     13: predicate.reduce_eliminate
 1.60% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.94% :     0.000003s :     25: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.36% :     0.000000s :      4: predicate.reset_defer_inline
 1.23% :     0.000002s :     13: predicate.reshape_eliminate
 1.21% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     13: predicate.same_eliminate
 0.68% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.39% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     21: predicate.switch_defer_inline
 2.20% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.84% :     0.000010s :     67: predicate.switch_simplify
 1.19% :     0.000002s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.48% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     13: predicate.value_based_eliminate
 0.43% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000524    12
46.18% :     0.000242s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.24% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.59% :     0.000255s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013037    72
 0.30% :     0.000039s :      1: add_recomputation
 0.36% :     0.000046s :      1: auto_monad
 0.12% :     0.000016s :      1: auto_monad_reorder
 1.92% :     0.000251s :      1: bootstrap
 0.15% :     0.000020s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.20% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.18% :     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
36.52% :     0.004762s :      1: jit_opt_a
 1.55% :     0.000203s :      1: jit_opt_after_cconv
 0.42% :     0.000055s :      1: jit_opt_b
 3.13% :     0.000408s :      1: loop_unroll
 3.78% :     0.000493s :      1: mutable_eliminate
 6.93% :     0.000904s :     26: opt.transform.jit_opt_a
 0.82% :     0.000106s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000029s :      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
 0.17% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.42% :     0.000446s :      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
 0.26% :     0.000033s :      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
 2.21% :     0.000289s :      1: renormalize.infer
 1.89% :     0.000247s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000030s :      1: rewriter_after_opt_a
 0.64% :     0.000083s :      1: rewriter_before_opt_a
 0.57% :     0.000074s :      1: symbol_engine_optimizer
32.97% :     0.004299s :      1: type_inference


TotalTime = 0.0111756, [30]
[bootstrap]: 0.00024391
[type_inference]: 0.00414619
[event_method]: 1.807e-05
[auto_monad]: 4.279e-05
[graph_reusing]: 3.95e-06
[pre_auto_parallel]: 2.06e-06
[py_interpret_to_execute]: 3.211e-05
[rewriter_before_opt_a]: 7.964e-05
[expand_dump_flag]: 1.52001e-06
[jit_opt_a]: 0.00469667, [2]
    [Cycle 1]: 0.00148896, [27]
        [switch_simplify]: 4.913e-05
        [loop_unroll]: 3.547e-05
        [a_1]: 0.0005888
        [with_stream_mark]: 1.184e-05
        [recompute_prepare]: 8.02e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 6.94999e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 6.53003e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 6.23e-06
        [merge_forward]: 2.82002e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 9.20999e-06
        [meta_fg_expand]: 1.94e-06
        [replace_old_param]: 9.56e-06
        [inline_without_move]: 5.85002e-06
        [renormalize]: 0.00051759
        [add_forward_monad_depend]: 4.35e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 1.086e-05
        [cse]: 1.845e-05
        [replace_applicator]: 1.179e-05
    [Cycle 2]: 0.00034317, [27]
        [switch_simplify]: 6.71999e-06
        [loop_unroll]: 5.78002e-06
        [a_1]: 0.00011158
        [with_stream_mark]: 7.79002e-06
        [recompute_prepare]: 5.51e-06
        [updatestate_depend_eliminate]: 2.88998e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.16002e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 5.47999e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.59998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.262e-05
        [j_node_and_user_rematch]: 8.28999e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 8.82e-06
        [inline_without_move]: 5.37999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 5.82001e-06
        [cse]: 1.237e-05
        [replace_applicator]: 5.61e-06
[py_interpret_to_execute_after_opt_a]: 8.74998e-06
[rewriter_after_opt_a]: 2.617e-05
[convert_after_rewriter]: 5.49e-06
[order_py_execute_after_rewriter]: 4.36002e-06
[mutable_eliminate]: 0.00046056
[jit_opt_b]: 5.441e-05, [1]
    [Cycle 1]: 4.788e-05, [2]
        [frontend_op_eliminate]: 1.873e-05
        [inline_after_opt_a]: 1.743e-05
[cconv]: 1.533e-05
[loop_unroll]: 0.00040232
[jit_opt_after_cconv]: 0.00014431, [1]
    [Cycle 1]: 0.00013811, [11]
        [c_1]: 2.43e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 5.46e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.28998e-06
        [cse]: 1.944e-05
        [call_graph_tuple_transform]: 1.877e-05
        [tuple_list_get_item_eliminator]: 6.14001e-06
        [none_parameter_eliminate]: 8.09989e-07
        [renormalize]: 3.50003e-07
        [switch_simplify]: 6.10002e-06
[remove_dup_value]: 8.18001e-06
[partial_unused_args_eliminate]: 1.54e-06
[environ_conv]: 4.52e-06
[add_recomputation]: 3.462e-05
[cse_after_recomputation]: 2.318e-05, [1]
    [Cycle 1]: 1.8e-05, [1]
        [cse]: 1.235e-05
[auto_monad_reorder]: 1.236e-05
[get_jit_bprop_graph]: 1.39998e-06
[rewriter_after_jit_bprop_graph]: 4.17e-06
[opt_after_jit_grad]: 0.00045554
[symbol_engine_optimizer]: 7.186e-05, [1]
    [Cycle 1]: 6.599e-05, [6]
        [build]: 2.43e-06
        [elim_shapecalc]: 8.70001e-06
        [elim_not_effective]: 1.271e-05
        [opt_reshape]: 6.17999e-06
        [fold_const_symbol]: 8.90999e-06
        [renormalize]: 5.09986e-07
[validate]: 2.692e-05
Sums
    bootstrap                            :     0.000244s :  3.13%
    type_inference                       :     0.004146s : 53.22%
    event_method                         :     0.000018s :  0.23%
    auto_monad                           :     0.000043s :  0.55%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.41%
    rewriter_before_opt_a                :     0.000080s :  1.02%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.72%
    jit_opt_a.loop_unroll                :     0.000041s :  0.53%
    jit_opt_a.a_1                        :     0.000700s :  8.99%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000014s :  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        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  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 :  0.15%
    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 :     0.000024s :  0.31%
    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          :     0.000018s :  0.24%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000518s :  6.64%
    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 :  0.21%
    jit_opt_a.cse                        :     0.000031s :  0.40%
    jit_opt_a.replace_applicator         :     0.000017s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000026s :  0.34%
    convert_after_rewriter               :     0.000005s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000461s :  5.91%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.22%
    cconv                                :     0.000015s :  0.20%
    loop_unroll                          :     0.000402s :  5.16%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.31%
    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 :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  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      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000008s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000035s :  0.44%
    cse_after_recomputation.cse          :     0.000012s :  0.16%
    auto_monad_reorder                   :     0.000012s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000456s :  5.85%
    symbol_engine_optimizer.build        :     0.000002s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    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                             :     0.000027s :  0.35%


Time group info:
------[substitution.]   0.000163    28
 0.80% :     0.000001s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.33% :     0.000004s :      4: substitution.graph_param_transform
82.66% :     0.000135s :      4: substitution.inline
 1.65% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.42% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.43% :     0.000004s :      4: substitution.replace_old_param
 7.12% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004101     2
78.43% :     0.003217s :      1: type_inference.infer
21.57% :     0.000885s :      1: type_inference.specialize
------[replace.]   0.000077     8
54.67% :     0.000042s :      4: replace.inline
45.33% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000142     8
93.18% :     0.000132s :      4: match.inline
 6.82% :     0.000010s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   912
 1.37% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     13: predicate.addn_check_dump
 1.26% :     0.000002s :     13: predicate.addn_zero_filter
 1.80% :     0.000002s :     13: predicate.arithmetic_simplify
 1.36% :     0.000002s :     13: predicate.cast_eliminate
 0.51% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     13: predicate.compare_switch_simplify
 1.18% :     0.000002s :     13: predicate.depend_value_elim
 1.16% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      4: predicate.elim_not_effective
 0.58% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.26% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.48% :     0.000002s :     13: predicate.environ_get_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 1.00% :     0.000001s :      8: predicate.get_grad_eliminate
 0.40% :     0.000001s :      4: predicate.graph_param_transform
 5.22% :     0.000007s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.45% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      8: predicate.less_batch_normalization
 1.72% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.10% :     0.000003s :     21: predicate.load_eliminater
 1.37% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.37% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     13: predicate.merge_addn
 1.22% :     0.000002s :     13: predicate.minmaximum_grad
 1.57% :     0.000002s :      4: predicate.mutable_eliminate
 0.48% :     0.000001s :      4: predicate.opt_reshape
 2.49% :     0.000003s :     21: predicate.partial_eliminate
 1.17% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     13: predicate.reduce_eliminate
 1.65% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.93% :     0.000003s :     25: predicate.replace_applicator
 0.65% :     0.000001s :      8: predicate.replace_old_param
 0.36% :     0.000000s :      4: predicate.reset_defer_inline
 1.22% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     13: predicate.same_eliminate
 0.64% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     21: predicate.switch_defer_inline
 2.23% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.63% :     0.000010s :     67: predicate.switch_simplify
 1.16% :     0.000002s :     13: predicate.tile_eliminate
 1.25% :     0.000002s :     13: predicate.transpose_eliminate
 1.49% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.94% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000527    12
46.04% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.02% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.94% :     0.000258s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012709    72
 0.29% :     0.000037s :      1: add_recomputation
 0.36% :     0.000046s :      1: auto_monad
 0.12% :     0.000015s :      1: auto_monad_reorder
 1.99% :     0.000253s :      1: bootstrap
 0.14% :     0.000018s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.20% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.17% :     0.000021s :      1: event_method
 0.03% :     0.000003s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
36.97% :     0.004699s :      1: jit_opt_a
 1.16% :     0.000147s :      1: jit_opt_after_cconv
 0.45% :     0.000057s :      1: jit_opt_b
 3.23% :     0.000410s :      1: loop_unroll
 3.69% :     0.000468s :      1: mutable_eliminate
 7.27% :     0.000924s :     26: opt.transform.jit_opt_a
 0.41% :     0.000052s :      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
 0.11% :     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
 3.65% :     0.000464s :      1: opt_after_jit_grad
 0.05% :     0.000006s :      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% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000011s :      1: remove_dup_value
 2.14% :     0.000272s :      1: renormalize.infer
 1.89% :     0.000240s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000029s :      1: rewriter_after_opt_a
 0.65% :     0.000082s :      1: rewriter_before_opt_a
 0.59% :     0.000074s :      1: symbol_engine_optimizer
32.70% :     0.004156s :      1: type_inference


TotalTime = 0.011517, [30]
[bootstrap]: 0.00024319
[type_inference]: 0.00419313
[event_method]: 1.934e-05
[auto_monad]: 5.024e-05
[graph_reusing]: 4.27998e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 3.105e-05
[rewriter_before_opt_a]: 8.023e-05
[expand_dump_flag]: 2.64999e-06
[jit_opt_a]: 0.00490576, [2]
    [Cycle 1]: 0.00159347, [27]
        [switch_simplify]: 5.143e-05
        [loop_unroll]: 3.43e-05
        [a_1]: 0.00059638
        [with_stream_mark]: 1.479e-05
        [recompute_prepare]: 8.07e-06
        [updatestate_depend_eliminate]: 3.30003e-06
        [updatestate_assign_eliminate]: 3.31001e-06
        [updatestate_loads_eliminate]: 2.77002e-06
        [parameter_eliminate]: 1.44998e-06
        [specialize_transform]: 6.58e-06
        [updatestate_useless_node_eliminater]: 5.77001e-06
        [accelerated_algorithm]: 6.14999e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.75001e-06
        [merge_forward]: 3.23e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.257e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 1.048e-05
        [inline_without_move]: 5.94999e-06
        [renormalize]: 0.00059212
        [add_forward_monad_depend]: 5.19e-06
        [auto_monad_grad]: 1.81e-06
        [auto_monad_eliminator]: 1.292e-05
        [cse]: 2.013e-05
        [replace_applicator]: 1.202e-05
    [Cycle 2]: 0.0003551, [27]
        [switch_simplify]: 6.26998e-06
        [loop_unroll]: 5.72001e-06
        [a_1]: 0.00011252
        [with_stream_mark]: 8.17e-06
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.28002e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.98002e-06
        [updatestate_useless_node_eliminater]: 5.86e-06
        [accelerated_algorithm]: 7.21999e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 5.20999e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.33e-05
        [j_node_and_user_rematch]: 9.22001e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 9.05001e-06
        [inline_without_move]: 5.70001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 6.27001e-06
        [cse]: 1.349e-05
        [replace_applicator]: 5.68002e-06
[py_interpret_to_execute_after_opt_a]: 9.37999e-06
[rewriter_after_opt_a]: 3.079e-05
[convert_after_rewriter]: 6.19999e-06
[order_py_execute_after_rewriter]: 5.40999e-06
[mutable_eliminate]: 0.00048168
[jit_opt_b]: 5.335e-05, [1]
    [Cycle 1]: 4.693e-05, [2]
        [frontend_op_eliminate]: 1.833e-05
        [inline_after_opt_a]: 1.705e-05
[cconv]: 2.104e-05
[loop_unroll]: 0.00040786
[jit_opt_after_cconv]: 0.00014481, [1]
    [Cycle 1]: 0.00013845, [11]
        [c_1]: 2.441e-05
        [parameter_eliminate]: 2.28998e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 1.9e-05
        [call_graph_tuple_transform]: 1.928e-05
        [tuple_list_get_item_eliminator]: 6.17001e-06
        [none_parameter_eliminate]: 1.04e-06
        [renormalize]: 3.29979e-07
        [switch_simplify]: 6.12001e-06
[remove_dup_value]: 1.345e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 5.03002e-06
[add_recomputation]: 4.177e-05
[cse_after_recomputation]: 2.293e-05, [1]
    [Cycle 1]: 1.72e-05, [1]
        [cse]: 1.162e-05
[auto_monad_reorder]: 1.661e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 4.45e-06
[opt_after_jit_grad]: 0.00046378
[symbol_engine_optimizer]: 7.29e-05, [1]
    [Cycle 1]: 6.725e-05, [6]
        [build]: 2.64999e-06
        [elim_shapecalc]: 8.59e-06
        [elim_not_effective]: 1.334e-05
        [opt_reshape]: 5.97001e-06
        [fold_const_symbol]: 9.41998e-06
        [renormalize]: 6.60017e-07
[validate]: 3.28e-05
Sums
    bootstrap                            :     0.000243s :  3.03%
    type_inference                       :     0.004193s : 52.27%
    event_method                         :     0.000019s :  0.24%
    auto_monad                           :     0.000050s :  0.63%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.39%
    rewriter_before_opt_a                :     0.000080s :  1.00%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000058s :  0.72%
    jit_opt_a.loop_unroll                :     0.000040s :  0.50%
    jit_opt_a.a_1                        :     0.000709s :  8.84%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.17%
    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        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.17%
    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 :  0.32%
    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          :     0.000020s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.15%
    jit_opt_a.renormalize                :     0.000592s :  7.38%
    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      :     0.000019s :  0.24%
    jit_opt_a.cse                        :     0.000034s :  0.42%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000031s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000482s :  6.00%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000021s :  0.26%
    loop_unroll                          :     0.000408s :  5.08%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.30%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  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      :     0.000000s :  0.00%
    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 :  0.06%
    add_recomputation                    :     0.000042s :  0.52%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000017s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000464s :  5.78%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    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  :     0.000001s :  0.01%
    validate                             :     0.000033s :  0.41%


Time group info:
------[substitution.]   0.000197    28
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.47% :     0.000005s :      4: substitution.graph_param_transform
82.19% :     0.000162s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000004s :      4: substitution.replace_old_param
 7.54% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004145     2
80.04% :     0.003318s :      1: type_inference.infer
19.96% :     0.000827s :      1: type_inference.specialize
------[replace.]   0.000079     8
56.66% :     0.000045s :      4: replace.inline
43.34% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     8
92.48% :     0.000159s :      4: match.inline
 7.52% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   912
 1.26% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     13: predicate.addn_check_dump
 1.40% :     0.000002s :     13: predicate.addn_zero_filter
 1.82% :     0.000002s :     13: predicate.arithmetic_simplify
 1.38% :     0.000002s :     13: predicate.cast_eliminate
 0.51% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.20% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      4: predicate.elim_not_effective
 0.58% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     13: predicate.environ_get_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      4: predicate.fold_const_symbol
 0.92% :     0.000001s :      8: predicate.get_grad_eliminate
 0.37% :     0.000000s :      4: predicate.graph_param_transform
 5.40% :     0.000007s :     29: predicate.inline
 0.92% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      8: predicate.less_batch_normalization
 1.83% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     21: predicate.load_eliminater
 1.41% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     42: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.29% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.62% :     0.000003s :     21: predicate.partial_eliminate
 1.16% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     13: predicate.reduce_eliminate
 1.68% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.79% :     0.000001s :      8: predicate.replace_old_param
 0.38% :     0.000001s :      4: predicate.reset_defer_inline
 1.24% :     0.000002s :     13: predicate.reshape_eliminate
 1.22% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     13: predicate.same_eliminate
 0.58% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      8: predicate.special_op_eliminate
 1.03% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     21: predicate.switch_defer_inline
 2.21% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.42% :     0.000010s :     67: predicate.switch_simplify
 1.22% :     0.000002s :     13: predicate.tile_eliminate
 1.31% :     0.000002s :     13: predicate.transpose_eliminate
 1.56% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.14% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000531    12
45.80% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.78% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.42% :     0.000257s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013130    72
 0.34% :     0.000045s :      1: add_recomputation
 0.41% :     0.000054s :      1: auto_monad
 0.15% :     0.000019s :      1: auto_monad_reorder
 1.92% :     0.000252s :      1: bootstrap
 0.18% :     0.000024s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.17% :     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
37.38% :     0.004908s :      1: jit_opt_a
 1.12% :     0.000148s :      1: jit_opt_after_cconv
 0.43% :     0.000056s :      1: jit_opt_b
 3.17% :     0.000416s :      1: loop_unroll
 3.73% :     0.000490s :      1: mutable_eliminate
 7.14% :     0.000938s :     26: opt.transform.jit_opt_a
 0.40% :     0.000053s :      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
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.59% :     0.000472s :      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
 0.26% :     0.000034s :      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
 2.43% :     0.000320s :      1: renormalize.infer
 2.02% :     0.000265s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000034s :      1: rewriter_after_opt_a
 0.63% :     0.000083s :      1: rewriter_before_opt_a
 0.57% :     0.000075s :      1: symbol_engine_optimizer
32.02% :     0.004204s :      1: type_inference


TotalTime = 0.0409611, [33]
[bootstrap]: 0.00027354
[type_inference]: 0.021252
[event_method]: 1.89e-05
[auto_monad]: 0.0001431
[graph_reusing]: 5.93998e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 0.00011429
[rewriter_before_opt_a]: 9.074e-05
[expand_dump_flag]: 3.6e-06
[jit_opt_a]: 0.0058295, [2]
    [Cycle 1]: 0.00223887, [27]
        [switch_simplify]: 5.4e-05
        [loop_unroll]: 3.684e-05
        [a_1]: 0.00069782
        [with_stream_mark]: 1.572e-05
        [recompute_prepare]: 1.056e-05
        [updatestate_depend_eliminate]: 5.62001e-06
        [updatestate_assign_eliminate]: 5.03002e-06
        [updatestate_loads_eliminate]: 5.22999e-06
        [parameter_eliminate]: 1.34998e-06
        [specialize_transform]: 9.31e-06
        [updatestate_useless_node_eliminater]: 1.104e-05
        [accelerated_algorithm]: 8.52998e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 8.11002e-06
        [merge_forward]: 4.51002e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.718e-05
        [j_node_and_user_rematch]: 1.328e-05
        [meta_fg_expand]: 3.32002e-06
        [replace_old_param]: 1.021e-05
        [inline_without_move]: 7.82998e-06
        [renormalize]: 0.00107211
        [add_forward_monad_depend]: 4.93001e-06
        [auto_monad_grad]: 1.91e-06
        [auto_monad_eliminator]: 1.72e-05
        [cse]: 2.67e-05
        [replace_applicator]: 1.532e-05
    [Cycle 2]: 0.000463, [27]
        [switch_simplify]: 9.35001e-06
        [loop_unroll]: 8.16002e-06
        [a_1]: 0.00016925
        [with_stream_mark]: 1.065e-05
        [recompute_prepare]: 7.87998e-06
        [updatestate_depend_eliminate]: 4.90001e-06
        [updatestate_assign_eliminate]: 4.47e-06
        [updatestate_loads_eliminate]: 3.98001e-06
        [parameter_eliminate]: 1.12999e-06
        [specialize_transform]: 8.18999e-06
        [updatestate_useless_node_eliminater]: 1.049e-05
        [accelerated_algorithm]: 8e-06
        [meta_shard_fg_expand]: 1.67999e-06
        [get_grad_eliminate_]: 7.71999e-06
        [merge_forward]: 4.22e-06
        [cell_reuse_recompute_pass]: 1.60001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.585e-05
        [j_node_and_user_rematch]: 1.248e-05
        [meta_fg_expand]: 2.68e-06
        [replace_old_param]: 9.47999e-06
        [inline_without_move]: 7.43e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 9.86998e-06
        [cse]: 1.961e-05
        [replace_applicator]: 7.74002e-06
[py_interpret_to_execute_after_opt_a]: 9.92001e-06
[rewriter_after_opt_a]: 0.00027136
[convert_after_rewriter]: 1.043e-05
[order_py_execute_after_rewriter]: 6.74999e-06
[mutable_eliminate]: 0.00051889
[jit_opt_b]: 6.529e-05, [1]
    [Cycle 1]: 5.881e-05, [2]
        [frontend_op_eliminate]: 2.304e-05
        [inline_after_opt_a]: 2.413e-05
[cconv]: 2.162e-05
[loop_unroll]: 0.00044494
[jit_opt_after_cconv]: 0.00018494, [1]
    [Cycle 1]: 0.00017835, [11]
        [c_1]: 4.276e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 7.59002e-06
        [updatestate_assign_eliminate]: 4.38999e-06
        [updatestate_loads_eliminate]: 4.12e-06
        [cse]: 2.576e-05
        [call_graph_tuple_transform]: 2.114e-05
        [tuple_list_get_item_eliminator]: 8e-06
        [none_parameter_eliminate]: 1.51998e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 8.74e-06
[remove_dup_value]: 1.647e-05
[partial_unused_args_eliminate]: 1.81003e-06
[environ_conv]: 9.83002e-06
[add_recomputation]: 5.44e-05
[cse_after_recomputation]: 2.788e-05, [1]
    [Cycle 1]: 2.245e-05, [1]
        [cse]: 1.674e-05
[auto_monad_reorder]: 2.255e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 4.32e-06
[opt_after_jit_grad]: 0.00053636
[symbol_engine_optimizer]: 9.846e-05, [1]
    [Cycle 1]: 9.184e-05, [6]
        [build]: 9.34e-06
        [elim_shapecalc]: 1.14e-05
        [elim_not_effective]: 1.876e-05
        [opt_reshape]: 8.63001e-06
        [fold_const_symbol]: 1.36e-05
        [renormalize]: 5.00004e-07
[validate]: 4.818e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.0106521
[execute]: 9.97999e-06
Sums
    bootstrap                            :     0.000274s :  0.73%
    type_inference                       :     0.021252s : 57.10%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000143s :  0.38%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000114s :  0.31%
    rewriter_before_opt_a                :     0.000091s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000063s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000867s :  2.33%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.05%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001072s :  2.88%
    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      :     0.000027s :  0.07%
    jit_opt_a.cse                        :     0.000046s :  0.12%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000271s :  0.73%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000519s :  1.39%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000022s :  0.06%
    loop_unroll                          :     0.000445s :  1.20%
    jit_opt_after_cconv.c_1              :     0.000043s :  0.11%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000016s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000054s :  0.15%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000023s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000536s :  1.44%
    symbol_engine_optimizer.build        :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010652s : 28.62%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000247    47
 3.23% :     0.000008s :      2: substitution.depend_value_elim
 1.10% :     0.000003s :      4: substitution.elim_not_effective
 0.85% :     0.000002s :      4: substitution.fold_const_symbol
 2.28% :     0.000006s :      5: substitution.graph_param_transform
75.73% :     0.000187s :      4: substitution.inline
 1.75% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.67% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.26% :     0.000003s :      2: substitution.replace_old_param
 4.82% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.06% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.25% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021191     2
93.70% :     0.019855s :      1: type_inference.infer
 6.30% :     0.001336s :      1: type_inference.specialize
------[replace.]   0.000085     6
66.61% :     0.000057s :      4: replace.inline
33.39% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000195     6
94.47% :     0.000184s :      4: match.inline
 5.53% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000165  1021
 1.31% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.28% :     0.000002s :     15: predicate.addn_check_dump
 1.42% :     0.000002s :     15: predicate.addn_zero_filter
 2.03% :     0.000003s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     15: predicate.compare_switch_simplify
 1.46% :     0.000002s :     15: predicate.depend_value_elim
 1.21% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.60% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.28% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     15: predicate.environ_get_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 4.90% :     0.000008s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.46% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :     10: predicate.less_batch_normalization
 1.49% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.86% :     0.000003s :     22: predicate.load_eliminater
 1.37% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.58% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     15: predicate.merge_addn
 1.24% :     0.000002s :     15: predicate.minmaximum_grad
 1.45% :     0.000002s :      5: predicate.mutable_eliminate
 0.54% :     0.000001s :      5: predicate.opt_reshape
 2.49% :     0.000004s :     22: predicate.partial_eliminate
 1.28% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.80% :     0.000003s :     15: predicate.reduce_eliminate
 1.48% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.74% :     0.000003s :     27: predicate.replace_applicator
 0.59% :     0.000001s :     10: predicate.replace_old_param
 0.30% :     0.000001s :      5: predicate.reset_defer_inline
 1.24% :     0.000002s :     15: predicate.reshape_eliminate
 1.32% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.12% :     0.000002s :     10: predicate.special_op_eliminate
 1.15% :     0.000002s :     10: predicate.specialize_transform
 1.44% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.18% :     0.000004s :     21: predicate.switch_defer_inline
 1.99% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.79% :     0.000011s :     68: predicate.switch_simplify
 1.42% :     0.000002s :     15: predicate.tile_eliminate
 1.31% :     0.000002s :     15: predicate.transpose_eliminate
 1.54% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.85% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.35% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000802    13
42.90% :     0.000344s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.32% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.78% :     0.000415s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043370    76
 0.13% :     0.000057s :      1: add_recomputation
 0.34% :     0.000148s :      1: auto_monad
 0.06% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000288s :      1: bootstrap
 0.06% :     0.000024s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000030s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.05% :     0.000022s :      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
13.45% :     0.005833s :      1: jit_opt_a
 0.43% :     0.000188s :      1: jit_opt_after_cconv
 0.16% :     0.000068s :      1: jit_opt_b
 1.04% :     0.000453s :      1: loop_unroll
 1.22% :     0.000528s :      1: mutable_eliminate
 2.67% :     0.001156s :     26: opt.transform.jit_opt_a
 0.18% :     0.000077s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000040s :      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
 0.07% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.26% :     0.000545s :      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
 0.27% :     0.000118s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
 1.45% :     0.000629s :      1: renormalize.infer
 1.00% :     0.000435s :      1: renormalize.specialize
 0.01% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.64% :     0.000276s :      1: rewriter_after_opt_a
 0.22% :     0.000094s :      1: rewriter_before_opt_a
 0.23% :     0.000101s :      1: symbol_engine_optimizer
24.60% :     0.010668s :      1: task_emit
49.03% :     0.021264s :      1: type_inference
 0.16% :     0.000069s :      1: validate


TotalTime = 0.0112699, [30]
[bootstrap]: 0.00024358
[type_inference]: 0.00414837
[event_method]: 1.959e-05
[auto_monad]: 5.103e-05
[graph_reusing]: 4.78001e-06
[pre_auto_parallel]: 1.99999e-06
[py_interpret_to_execute]: 3.015e-05
[rewriter_before_opt_a]: 8.17e-05
[expand_dump_flag]: 1.84e-06
[jit_opt_a]: 0.00478898, [2]
    [Cycle 1]: 0.00157696, [27]
        [switch_simplify]: 5.106e-05
        [loop_unroll]: 3.475e-05
        [a_1]: 0.00063781
        [with_stream_mark]: 1.414e-05
        [recompute_prepare]: 7.91001e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 6.83998e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 1.76003e-06
        [get_grad_eliminate_]: 5.76998e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.221e-05
        [j_node_and_user_rematch]: 9.39e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 1.052e-05
        [inline_without_move]: 5.82999e-06
        [renormalize]: 0.00054606
        [add_forward_monad_depend]: 4.94e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 1.122e-05
        [cse]: 1.963e-05
        [replace_applicator]: 1.206e-05
    [Cycle 2]: 0.00035058, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 5.57999e-06
        [a_1]: 0.00011138
        [with_stream_mark]: 8.02e-06
        [recompute_prepare]: 5.52999e-06
        [updatestate_depend_eliminate]: 2.97002e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 5.62999e-06
        [updatestate_useless_node_eliminater]: 5.45001e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 1.21002e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.439e-05
        [j_node_and_user_rematch]: 8.48999e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 9.99999e-06
        [inline_without_move]: 5.89e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.72999e-06
        [cse]: 1.255e-05
        [replace_applicator]: 5.64e-06
[py_interpret_to_execute_after_opt_a]: 8.32e-06
[rewriter_after_opt_a]: 2.729e-05
[convert_after_rewriter]: 5.59e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00045116
[jit_opt_b]: 5.276e-05, [1]
    [Cycle 1]: 4.635e-05, [2]
        [frontend_op_eliminate]: 1.818e-05
        [inline_after_opt_a]: 1.685e-05
[cconv]: 1.926e-05
[loop_unroll]: 0.00040695
[jit_opt_after_cconv]: 0.0001425, [1]
    [Cycle 1]: 0.00013627, [11]
        [c_1]: 2.442e-05
        [parameter_eliminate]: 2.27999e-06
        [updatestate_depend_eliminate]: 5.17e-06
        [updatestate_assign_eliminate]: 2.43002e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [cse]: 1.931e-05
        [call_graph_tuple_transform]: 1.859e-05
        [tuple_list_get_item_eliminator]: 5.86e-06
        [none_parameter_eliminate]: 7.89994e-07
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.76e-06
[remove_dup_value]: 1.103e-05
[partial_unused_args_eliminate]: 1.95001e-06
[environ_conv]: 4.72998e-06
[add_recomputation]: 3.809e-05
[cse_after_recomputation]: 2.156e-05, [1]
    [Cycle 1]: 1.623e-05, [1]
        [cse]: 1.078e-05
[auto_monad_reorder]: 1.447e-05
[get_jit_bprop_graph]: 1.46998e-06
[rewriter_after_jit_bprop_graph]: 4.23999e-06
[opt_after_jit_grad]: 0.00043875
[symbol_engine_optimizer]: 7.173e-05, [1]
    [Cycle 1]: 6.562e-05, [6]
        [build]: 2.63e-06
        [elim_shapecalc]: 8.27e-06
        [elim_not_effective]: 1.293e-05
        [opt_reshape]: 6.12001e-06
        [fold_const_symbol]: 9.05999e-06
        [renormalize]: 4.09986e-07
[validate]: 3.028e-05
Sums
    bootstrap                            :     0.000244s :  3.09%
    type_inference                       :     0.004148s : 52.63%
    event_method                         :     0.000020s :  0.25%
    auto_monad                           :     0.000051s :  0.65%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.38%
    rewriter_before_opt_a                :     0.000082s :  1.04%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.73%
    jit_opt_a.loop_unroll                :     0.000040s :  0.51%
    jit_opt_a.a_1                        :     0.000749s :  9.50%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.28%
    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 :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  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              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  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          :     0.000021s :  0.26%
    jit_opt_a.inline_without_move        :     0.000012s :  0.15%
    jit_opt_a.renormalize                :     0.000546s :  6.93%
    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 :  0.22%
    jit_opt_a.cse                        :     0.000032s :  0.41%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000027s :  0.35%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000451s :  5.72%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000019s :  0.24%
    loop_unroll                          :     0.000407s :  5.16%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.31%
    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 :  0.24%
    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  :     0.000006s :  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                    :     0.000038s :  0.48%
    cse_after_recomputation.cse          :     0.000011s :  0.14%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000439s :  5.57%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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                             :     0.000030s :  0.38%


Time group info:
------[substitution.]   0.000179    28
 0.82% :     0.000001s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.58% :     0.000005s :      4: substitution.graph_param_transform
81.01% :     0.000145s :      4: substitution.inline
 1.68% :     0.000003s :      4: substitution.j_node_and_user_rematch
 3.19% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.50% :     0.000004s :      4: substitution.replace_old_param
 7.60% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004102     2
79.17% :     0.003247s :      1: type_inference.infer
20.83% :     0.000855s :      1: type_inference.specialize
------[replace.]   0.000076     8
54.13% :     0.000041s :      4: replace.inline
45.87% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000154     8
92.45% :     0.000142s :      4: match.inline
 7.55% :     0.000012s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   912
 1.20% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     13: predicate.addn_check_dump
 1.29% :     0.000002s :     13: predicate.addn_zero_filter
 1.87% :     0.000002s :     13: predicate.arithmetic_simplify
 1.41% :     0.000002s :     13: predicate.cast_eliminate
 0.51% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     13: predicate.compare_switch_simplify
 1.12% :     0.000001s :     13: predicate.depend_value_elim
 1.17% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.46% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.57% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     13: predicate.environ_get_eliminate
 1.14% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000001s :      8: predicate.get_grad_eliminate
 0.40% :     0.000001s :      4: predicate.graph_param_transform
 5.19% :     0.000007s :     29: predicate.inline
 0.93% :     0.000001s :      8: predicate.inline_without_move
 0.45% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      8: predicate.less_batch_normalization
 1.68% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.01% :     0.000003s :     21: predicate.load_eliminater
 1.27% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     13: predicate.merge_addn
 1.17% :     0.000002s :     13: predicate.minmaximum_grad
 1.28% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.58% :     0.000003s :     21: predicate.partial_eliminate
 1.14% :     0.000001s :     13: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     13: predicate.reduce_eliminate
 1.73% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.03% :     0.000003s :     25: predicate.replace_applicator
 0.82% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.49% :     0.000002s :     13: predicate.reshape_eliminate
 1.20% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      8: predicate.special_op_eliminate
 0.91% :     0.000001s :      8: predicate.specialize_transform
 1.41% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     21: predicate.switch_defer_inline
 2.28% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.55% :     0.000010s :     67: predicate.switch_simplify
 1.18% :     0.000002s :     13: predicate.tile_eliminate
 1.24% :     0.000002s :     13: predicate.transpose_eliminate
 1.38% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 2.00% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000528    12
45.52% :     0.000240s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.27% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.22% :     0.000260s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012875    72
 0.32% :     0.000041s :      1: add_recomputation
 0.42% :     0.000054s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 1.96% :     0.000252s :      1: bootstrap
 0.17% :     0.000022s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.18% :     0.000023s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
37.22% :     0.004792s :      1: jit_opt_a
 1.13% :     0.000145s :      1: jit_opt_after_cconv
 0.43% :     0.000055s :      1: jit_opt_b
 3.22% :     0.000414s :      1: loop_unroll
 3.56% :     0.000459s :      1: mutable_eliminate
 7.59% :     0.000977s :     26: opt.transform.jit_opt_a
 0.40% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000029s :      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
 0.18% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.46% :     0.000446s :      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% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000013s :      1: remove_dup_value
 2.21% :     0.000284s :      1: renormalize.infer
 1.99% :     0.000256s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000030s :      1: rewriter_after_opt_a
 0.66% :     0.000085s :      1: rewriter_before_opt_a
 0.58% :     0.000074s :      1: symbol_engine_optimizer
32.28% :     0.004156s :      1: type_inference


TotalTime = 0.0114443, [30]
[bootstrap]: 0.00024425
[type_inference]: 0.00412268
[event_method]: 1.955e-05
[auto_monad]: 4.476e-05
[graph_reusing]: 4.12e-06
[pre_auto_parallel]: 1.42e-06
[py_interpret_to_execute]: 3.106e-05
[rewriter_before_opt_a]: 7.871e-05
[expand_dump_flag]: 1.97001e-06
[jit_opt_a]: 0.00484496, [2]
    [Cycle 1]: 0.00160967, [27]
        [switch_simplify]: 5.221e-05
        [loop_unroll]: 3.511e-05
        [a_1]: 0.00059565
        [with_stream_mark]: 1.453e-05
        [recompute_prepare]: 9.12999e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 1.47999e-06
        [specialize_transform]: 6.70998e-06
        [updatestate_useless_node_eliminater]: 5.87001e-06
        [accelerated_algorithm]: 6.14999e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.84999e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 9.40001e-06
        [meta_fg_expand]: 2.37001e-06
        [replace_old_param]: 1.068e-05
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00060017
        [add_forward_monad_depend]: 5.00001e-06
        [auto_monad_grad]: 2.08002e-06
        [auto_monad_eliminator]: 1.306e-05
        [cse]: 2.04e-05
        [replace_applicator]: 1.288e-05
    [Cycle 2]: 0.00035656, [27]
        [switch_simplify]: 6.28e-06
        [loop_unroll]: 5.54998e-06
        [a_1]: 0.00011115
        [with_stream_mark]: 9.10999e-06
        [recompute_prepare]: 5.92999e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.75001e-06
        [updatestate_useless_node_eliminater]: 5.49e-06
        [accelerated_algorithm]: 5.57999e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 5.75001e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 1.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.327e-05
        [j_node_and_user_rematch]: 8.22e-06
        [meta_fg_expand]: 1.79998e-06
        [replace_old_param]: 9.15999e-06
        [inline_without_move]: 6.08002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.03001e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 6.53e-06
        [cse]: 1.397e-05
        [replace_applicator]: 6.24001e-06
[py_interpret_to_execute_after_opt_a]: 1.004e-05
[rewriter_after_opt_a]: 3.014e-05
[convert_after_rewriter]: 6.66999e-06
[order_py_execute_after_rewriter]: 4.90999e-06
[mutable_eliminate]: 0.00052125
[jit_opt_b]: 5.263e-05, [1]
    [Cycle 1]: 4.633e-05, [2]
        [frontend_op_eliminate]: 1.78e-05
        [inline_after_opt_a]: 1.704e-05
[cconv]: 2.004e-05
[loop_unroll]: 0.00045894
[jit_opt_after_cconv]: 0.0001445, [1]
    [Cycle 1]: 0.00013849, [11]
        [c_1]: 2.393e-05
        [parameter_eliminate]: 2.42001e-06
        [updatestate_depend_eliminate]: 5.75001e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 1.926e-05
        [call_graph_tuple_transform]: 2.005e-05
        [tuple_list_get_item_eliminator]: 6.02001e-06
        [none_parameter_eliminate]: 1.26997e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 5.72001e-06
[remove_dup_value]: 1.255e-05
[partial_unused_args_eliminate]: 1.74998e-06
[environ_conv]: 5.18002e-06
[add_recomputation]: 3.892e-05
[cse_after_recomputation]: 2.235e-05, [1]
    [Cycle 1]: 1.735e-05, [1]
        [cse]: 1.178e-05
[auto_monad_reorder]: 1.496e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 5.48002e-06
[opt_after_jit_grad]: 0.00044611
[symbol_engine_optimizer]: 7.327e-05, [1]
    [Cycle 1]: 6.746e-05, [6]
        [build]: 2.91e-06
        [elim_shapecalc]: 8.69e-06
        [elim_not_effective]: 1.287e-05
        [opt_reshape]: 6.07999e-06
        [fold_const_symbol]: 8.99e-06
        [renormalize]: 3.50003e-07
[validate]: 3.325e-05
Sums
    bootstrap                            :     0.000244s :  3.04%
    type_inference                       :     0.004123s : 51.39%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000045s :  0.56%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000001s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.39%
    rewriter_before_opt_a                :     0.000079s :  0.98%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.73%
    jit_opt_a.loop_unroll                :     0.000041s :  0.51%
    jit_opt_a.a_1                        :     0.000707s :  8.81%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.19%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.09%
    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       :     0.000012s :  0.16%
    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_        :     0.000012s :  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 :  0.33%
    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          :     0.000020s :  0.25%
    jit_opt_a.inline_without_move        :     0.000012s :  0.15%
    jit_opt_a.renormalize                :     0.000600s :  7.48%
    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      :     0.000020s :  0.24%
    jit_opt_a.cse                        :     0.000034s :  0.43%
    jit_opt_a.replace_applicator         :     0.000019s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000030s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000521s :  6.50%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000020s :  0.25%
    loop_unroll                          :     0.000459s :  5.72%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.30%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.25%
    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 :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  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                    :     0.000039s :  0.49%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    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                   :     0.000446s :  5.56%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    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 :  0.00%
    validate                             :     0.000033s :  0.41%


Time group info:
------[substitution.]   0.000187    28
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000005s :      4: substitution.graph_param_transform
80.96% :     0.000152s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.54% :     0.000005s :      4: substitution.replace_old_param
 8.14% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004077     2
79.96% :     0.003260s :      1: type_inference.infer
20.04% :     0.000817s :      1: type_inference.specialize
------[replace.]   0.000082     8
55.28% :     0.000045s :      4: replace.inline
44.72% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     8
91.84% :     0.000149s :      4: match.inline
 8.16% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.22% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.25% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.27% :     0.000002s :     13: predicate.addn_check_dump
 1.19% :     0.000002s :     13: predicate.addn_zero_filter
 1.79% :     0.000002s :     13: predicate.arithmetic_simplify
 1.35% :     0.000002s :     13: predicate.cast_eliminate
 0.50% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     13: predicate.compare_switch_simplify
 1.24% :     0.000002s :     13: predicate.depend_value_elim
 1.21% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.48% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     13: predicate.environ_get_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.89% :     0.000001s :      8: predicate.get_grad_eliminate
 0.38% :     0.000001s :      4: predicate.graph_param_transform
 5.15% :     0.000007s :     29: predicate.inline
 0.98% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      8: predicate.less_batch_normalization
 1.60% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     21: predicate.load_eliminater
 1.37% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.64% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.53% :     0.000003s :     21: predicate.partial_eliminate
 1.19% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     13: predicate.reduce_eliminate
 1.72% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.71% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.95% :     0.000003s :     25: predicate.replace_applicator
 0.64% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000000s :      4: predicate.reset_defer_inline
 1.24% :     0.000002s :     13: predicate.reshape_eliminate
 1.18% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.26% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.68% :     0.000004s :     21: predicate.switch_defer_inline
 2.24% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.85% :     0.000010s :     67: predicate.switch_simplify
 1.24% :     0.000002s :     13: predicate.tile_eliminate
 1.29% :     0.000002s :     13: predicate.transpose_eliminate
 1.43% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000538    12
45.52% :     0.000245s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.04% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.44% :     0.000260s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013066    72
 0.32% :     0.000042s :      1: add_recomputation
 0.37% :     0.000048s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 1.95% :     0.000254s :      1: bootstrap
 0.17% :     0.000023s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.18% :     0.000023s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
37.10% :     0.004848s :      1: jit_opt_a
 1.13% :     0.000147s :      1: jit_opt_after_cconv
 0.42% :     0.000055s :      1: jit_opt_b
 3.57% :     0.000467s :      1: loop_unroll
 4.05% :     0.000529s :      1: mutable_eliminate
 7.18% :     0.000938s :     26: opt.transform.jit_opt_a
 0.40% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000029s :      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
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.47% :     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% :     0.000003s :      1: pre_auto_parallel
 0.26% :     0.000034s :      1: py_interpret_to_execute
 0.10% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000015s :      1: remove_dup_value
 2.46% :     0.000322s :      1: renormalize.infer
 2.08% :     0.000271s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000033s :      1: rewriter_after_opt_a
 0.62% :     0.000082s :      1: rewriter_before_opt_a
 0.58% :     0.000076s :      1: symbol_engine_optimizer
31.63% :     0.004133s :      1: type_inference


TotalTime = 0.040307, [33]
[bootstrap]: 0.00027105
[type_inference]: 0.0212319
[event_method]: 1.935e-05
[auto_monad]: 0.00014813
[graph_reusing]: 6.46e-06
[pre_auto_parallel]: 3.20998e-06
[py_interpret_to_execute]: 0.00012171
[rewriter_before_opt_a]: 9.647e-05
[expand_dump_flag]: 3.26999e-06
[jit_opt_a]: 0.00591812, [2]
    [Cycle 1]: 0.00218144, [27]
        [switch_simplify]: 6.279e-05
        [loop_unroll]: 3.828e-05
        [a_1]: 0.00069148
        [with_stream_mark]: 1.86e-05
        [recompute_prepare]: 1.019e-05
        [updatestate_depend_eliminate]: 6.14999e-06
        [updatestate_assign_eliminate]: 5.69e-06
        [updatestate_loads_eliminate]: 4.74e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 8.84998e-06
        [updatestate_useless_node_eliminater]: 1.147e-05
        [accelerated_algorithm]: 8.50001e-06
        [meta_shard_fg_expand]: 2.84001e-06
        [get_grad_eliminate_]: 7.84002e-06
        [merge_forward]: 5.57999e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.902e-05
        [j_node_and_user_rematch]: 1.3e-05
        [meta_fg_expand]: 3.86999e-06
        [replace_old_param]: 1.119e-05
        [inline_without_move]: 8.12e-06
        [renormalize]: 0.00097268
        [add_forward_monad_depend]: 6.38998e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 2.14e-05
        [cse]: 3.779e-05
        [replace_applicator]: 1.509e-05
    [Cycle 2]: 0.00046233, [27]
        [switch_simplify]: 9.22999e-06
        [loop_unroll]: 8.08001e-06
        [a_1]: 0.00016614
        [with_stream_mark]: 1.095e-05
        [recompute_prepare]: 7.9e-06
        [updatestate_depend_eliminate]: 5.02e-06
        [updatestate_assign_eliminate]: 4.63999e-06
        [updatestate_loads_eliminate]: 4.11001e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 8.50999e-06
        [updatestate_useless_node_eliminater]: 1.081e-05
        [accelerated_algorithm]: 7.63001e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 7.21999e-06
        [merge_forward]: 3.95e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.59e-05
        [j_node_and_user_rematch]: 1.217e-05
        [meta_fg_expand]: 2.88e-06
        [replace_old_param]: 9.99001e-06
        [inline_without_move]: 7.76001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 9.92999e-06
        [cse]: 2.039e-05
        [replace_applicator]: 7.82998e-06
[py_interpret_to_execute_after_opt_a]: 1.136e-05
[rewriter_after_opt_a]: 0.00029047
[convert_after_rewriter]: 1.068e-05
[order_py_execute_after_rewriter]: 6.91001e-06
[mutable_eliminate]: 0.00052675
[jit_opt_b]: 6.813e-05, [1]
    [Cycle 1]: 6.137e-05, [2]
        [frontend_op_eliminate]: 2.496e-05
        [inline_after_opt_a]: 2.429e-05
[cconv]: 2.625e-05
[loop_unroll]: 0.00045305
[jit_opt_after_cconv]: 0.00021182, [1]
    [Cycle 1]: 0.00020461, [11]
        [c_1]: 4.767e-05
        [parameter_eliminate]: 1.085e-05
        [updatestate_depend_eliminate]: 8.75999e-06
        [updatestate_assign_eliminate]: 4.60001e-06
        [updatestate_loads_eliminate]: 4.00998e-06
        [cse]: 2.778e-05
        [call_graph_tuple_transform]: 2.458e-05
        [tuple_list_get_item_eliminator]: 9.14e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 8.17998e-06
[remove_dup_value]: 2.102e-05
[partial_unused_args_eliminate]: 2.53003e-06
[environ_conv]: 1.004e-05
[add_recomputation]: 6.442e-05
[cse_after_recomputation]: 3.479e-05, [1]
    [Cycle 1]: 2.841e-05, [1]
        [cse]: 2.174e-05
[auto_monad_reorder]: 2.719e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 5.28002e-06
[opt_after_jit_grad]: 0.00054648
[symbol_engine_optimizer]: 0.00010279, [1]
    [Cycle 1]: 9.588e-05, [6]
        [build]: 1.026e-05
        [elim_shapecalc]: 1.29e-05
        [elim_not_effective]: 1.963e-05
        [opt_reshape]: 9.10999e-06
        [fold_const_symbol]: 1.336e-05
        [renormalize]: 3.89991e-07
[validate]: 5.185e-05
[backend_pass]: 9.90025e-07
[task_emit]: 0.00977827
[execute]: 7.19001e-06
Sums
    bootstrap                            :     0.000271s :  0.75%
    type_inference                       :     0.021232s : 58.37%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000148s :  0.41%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000122s :  0.33%
    rewriter_before_opt_a                :     0.000096s :  0.27%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.20%
    jit_opt_a.loop_unroll                :     0.000046s :  0.13%
    jit_opt_a.a_1                        :     0.000858s :  2.36%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  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 :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000025s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000973s :  2.67%
    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      :     0.000031s :  0.09%
    jit_opt_a.cse                        :     0.000058s :  0.16%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000290s :  0.80%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000527s :  1.45%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000453s :  1.25%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000011s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000028s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.07%
    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      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000021s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000064s :  0.18%
    cse_after_recomputation.cse          :     0.000022s :  0.06%
    auto_monad_reorder                   :     0.000027s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000546s :  1.50%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000052s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009778s : 26.88%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000257    47
 3.07% :     0.000008s :      2: substitution.depend_value_elim
 1.07% :     0.000003s :      4: substitution.elim_not_effective
 0.80% :     0.000002s :      4: substitution.fold_const_symbol
 2.71% :     0.000007s :      5: substitution.graph_param_transform
75.43% :     0.000194s :      4: substitution.inline
 1.70% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.71% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.49% :     0.000004s :      2: substitution.replace_old_param
 4.59% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.10% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.34% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021165     2
93.24% :     0.019734s :      1: type_inference.infer
 6.76% :     0.001431s :      1: type_inference.specialize
------[replace.]   0.000094     6
65.95% :     0.000062s :      4: replace.inline
34.05% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     6
94.69% :     0.000191s :      4: match.inline
 5.31% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1021
 1.18% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     15: predicate.addn_check_dump
 1.34% :     0.000002s :     15: predicate.addn_zero_filter
 2.08% :     0.000004s :     15: predicate.arithmetic_simplify
 1.40% :     0.000002s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     15: predicate.compare_switch_simplify
 1.32% :     0.000002s :     15: predicate.depend_value_elim
 1.23% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.06% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.69% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.22% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.30% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     15: predicate.environ_get_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.02% :     0.000002s :     10: predicate.get_grad_eliminate
 0.32% :     0.000001s :      5: predicate.graph_param_transform
 5.25% :     0.000009s :     31: predicate.inline
 1.06% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :     10: predicate.less_batch_normalization
 1.51% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     22: predicate.load_eliminater
 1.37% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.49% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.89% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 1.47% :     0.000003s :      5: predicate.mutable_eliminate
 0.60% :     0.000001s :      5: predicate.opt_reshape
 2.36% :     0.000004s :     22: predicate.partial_eliminate
 1.25% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.77% :     0.000003s :     15: predicate.reduce_eliminate
 1.47% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.69% :     0.000003s :     27: predicate.replace_applicator
 0.64% :     0.000001s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.17% :     0.000002s :     15: predicate.reshape_eliminate
 1.23% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.88% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     15: predicate.same_eliminate
 0.54% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.13% :     0.000002s :     10: predicate.special_op_eliminate
 1.09% :     0.000002s :     10: predicate.specialize_transform
 1.44% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.14% :     0.000004s :     21: predicate.switch_defer_inline
 1.92% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.50% :     0.000011s :     68: predicate.switch_simplify
 1.36% :     0.000002s :     15: predicate.tile_eliminate
 1.29% :     0.000002s :     15: predicate.transpose_eliminate
 1.65% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 2.00% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.36% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000003s :     15: predicate.value_based_eliminate
 0.52% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000804    13
45.29% :     0.000364s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.95% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.76% :     0.000400s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042633    76
 0.16% :     0.000067s :      1: add_recomputation
 0.36% :     0.000153s :      1: auto_monad
 0.07% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000286s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.09% :     0.000037s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.05% :     0.000023s :      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
13.89% :     0.005922s :      1: jit_opt_a
 0.50% :     0.000215s :      1: jit_opt_after_cconv
 0.17% :     0.000071s :      1: jit_opt_b
 1.11% :     0.000473s :      1: loop_unroll
 1.26% :     0.000536s :      1: mutable_eliminate
 2.71% :     0.001156s :     26: opt.transform.jit_opt_a
 0.20% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      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% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.30% :     0.000556s :      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
 0.30% :     0.000126s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000024s :      1: remove_dup_value
 1.28% :     0.000544s :      1: renormalize.infer
 0.99% :     0.000421s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.69% :     0.000295s :      1: rewriter_after_opt_a
 0.24% :     0.000100s :      1: rewriter_before_opt_a
 0.25% :     0.000106s :      1: symbol_engine_optimizer
22.96% :     0.009791s :      1: task_emit
49.84% :     0.021247s :      1: type_inference
 0.18% :     0.000076s :      1: validate


TotalTime = 0.0164398, [30]
[bootstrap]: 0.00025433
[type_inference]: 0.00434736
[event_method]: 1.992e-05
[auto_monad]: 4.127e-05
[graph_reusing]: 3.81999e-06
[pre_auto_parallel]: 1.64e-06
[py_interpret_to_execute]: 3.063e-05
[rewriter_before_opt_a]: 7.839e-05
[expand_dump_flag]: 1.61002e-06
[jit_opt_a]: 0.00477665, [2]
    [Cycle 1]: 0.00151215, [27]
        [switch_simplify]: 4.987e-05
        [loop_unroll]: 3.506e-05
        [a_1]: 0.00058105
        [with_stream_mark]: 1.295e-05
        [recompute_prepare]: 7.7e-06
        [updatestate_depend_eliminate]: 3.67998e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 6.71999e-06
        [updatestate_useless_node_eliminater]: 5.97999e-06
        [accelerated_algorithm]: 6.29999e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 5.84999e-06
        [merge_forward]: 2.51998e-06
        [cell_reuse_recompute_pass]: 8.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.267e-05
        [j_node_and_user_rematch]: 9.67001e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 9.79999e-06
        [inline_without_move]: 6.06e-06
        [renormalize]: 0.00053048
        [add_forward_monad_depend]: 4.3e-06
        [auto_monad_grad]: 1.76998e-06
        [auto_monad_eliminator]: 1.108e-05
        [cse]: 2.187e-05
        [replace_applicator]: 1.375e-05
    [Cycle 2]: 0.00034347, [27]
        [switch_simplify]: 6.57002e-06
        [loop_unroll]: 5.71998e-06
        [a_1]: 0.00011279
        [with_stream_mark]: 8.67e-06
        [recompute_prepare]: 5.47999e-06
        [updatestate_depend_eliminate]: 3.12002e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 5.37001e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 5.10001e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.385e-05
        [j_node_and_user_rematch]: 8.40001e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 8.55001e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.07998e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 5.67001e-06
        [cse]: 1.161e-05
        [replace_applicator]: 5.68002e-06
[py_interpret_to_execute_after_opt_a]: 9.25001e-06
[rewriter_after_opt_a]: 2.655e-05
[convert_after_rewriter]: 5.80002e-06
[order_py_execute_after_rewriter]: 3.96001e-06
[mutable_eliminate]: 0.00047533
[jit_opt_b]: 5.47e-05, [1]
    [Cycle 1]: 4.832e-05, [2]
        [frontend_op_eliminate]: 1.857e-05
        [inline_after_opt_a]: 1.771e-05
[cconv]: 1.6e-05
[loop_unroll]: 0.00040939
[jit_opt_after_cconv]: 0.0001438, [1]
    [Cycle 1]: 0.00013732, [11]
        [c_1]: 2.385e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 5.52999e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [cse]: 2.023e-05
        [call_graph_tuple_transform]: 1.793e-05
        [tuple_list_get_item_eliminator]: 5.95002e-06
        [none_parameter_eliminate]: 1.14003e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.99999e-06
[remove_dup_value]: 7.51001e-06
[partial_unused_args_eliminate]: 1.45999e-06
[environ_conv]: 4.33001e-06
[add_recomputation]: 3.136e-05
[cse_after_recomputation]: 2.218e-05, [1]
    [Cycle 1]: 1.721e-05, [1]
        [cse]: 1.175e-05
[auto_monad_reorder]: 1.32e-05
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 4.42e-06
[opt_after_jit_grad]: 0.00538054
[symbol_engine_optimizer]: 8.154e-05, [1]
    [Cycle 1]: 7.373e-05, [6]
        [build]: 3.45e-06
        [elim_shapecalc]: 8.47e-06
        [elim_not_effective]: 1.635e-05
        [opt_reshape]: 6.34999e-06
        [fold_const_symbol]: 9.72999e-06
        [renormalize]: 3.69997e-07
[validate]: 3.614e-05
Sums
    bootstrap                            :     0.000254s :  1.96%
    type_inference                       :     0.004347s : 33.51%
    event_method                         :     0.000020s :  0.15%
    auto_monad                           :     0.000041s :  0.32%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.24%
    rewriter_before_opt_a                :     0.000078s :  0.60%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000056s :  0.44%
    jit_opt_a.loop_unroll                :     0.000041s :  0.31%
    jit_opt_a.a_1                        :     0.000694s :  5.35%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.17%
    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        :     0.000002s :  0.01%
    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      :     0.000012s :  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  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.14%
    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        :     0.000012s :  0.09%
    jit_opt_a.renormalize                :     0.000531s :  4.09%
    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      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000033s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.15%
    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                    :     0.000475s :  3.66%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.14%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000409s :  3.16%
    jit_opt_after_cconv.c_1              :     0.000024s :  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 :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  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 :     0.000006s :  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        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000031s :  0.24%
    cse_after_recomputation.cse          :     0.000012s :  0.09%
    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                   :     0.005381s : 41.47%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    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 :     0.000010s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.28%


Time group info:
------[substitution.]   0.000173    28
 1.21% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.20% :     0.000004s :      4: substitution.graph_param_transform
82.81% :     0.000143s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      4: substitution.replace_old_param
 7.04% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004300     2
80.88% :     0.003478s :      1: type_inference.infer
19.12% :     0.000822s :      1: type_inference.specialize
------[replace.]   0.000077     8
56.23% :     0.000043s :      4: replace.inline
43.77% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     8
93.13% :     0.000140s :      4: match.inline
 6.87% :     0.000010s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.23% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.17% :     0.000002s :     13: predicate.addn_check_dump
 1.22% :     0.000002s :     13: predicate.addn_zero_filter
 1.90% :     0.000003s :     13: predicate.arithmetic_simplify
 1.40% :     0.000002s :     13: predicate.cast_eliminate
 0.49% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     13: predicate.compare_switch_simplify
 1.20% :     0.000002s :     13: predicate.depend_value_elim
 1.17% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.23% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      4: predicate.elim_not_effective
 0.59% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     13: predicate.environ_get_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      4: predicate.fold_const_symbol
 0.94% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.43% :     0.000007s :     29: predicate.inline
 0.93% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      8: predicate.less_batch_normalization
 1.67% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.10% :     0.000003s :     21: predicate.load_eliminater
 1.22% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.89% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     13: predicate.merge_addn
 1.19% :     0.000002s :     13: predicate.minmaximum_grad
 1.51% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.49% :     0.000003s :     21: predicate.partial_eliminate
 1.17% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     13: predicate.reduce_eliminate
 1.59% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.00% :     0.000003s :     25: predicate.replace_applicator
 0.64% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.17% :     0.000002s :     13: predicate.reshape_eliminate
 1.20% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     13: predicate.same_eliminate
 0.58% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      8: predicate.special_op_eliminate
 0.95% :     0.000001s :      8: predicate.specialize_transform
 1.38% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     21: predicate.switch_defer_inline
 2.29% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.43% :     0.000010s :     67: predicate.switch_simplify
 1.18% :     0.000002s :     13: predicate.tile_eliminate
 1.28% :     0.000002s :     13: predicate.transpose_eliminate
 1.44% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 2.00% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000533    12
45.64% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.67% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.69% :     0.000259s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017980    72
 0.19% :     0.000034s :      1: add_recomputation
 0.25% :     0.000044s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 1.47% :     0.000265s :      1: bootstrap
 0.10% :     0.000018s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000006s :      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
 0.03% :     0.000006s :      1: graph_reusing
26.58% :     0.004779s :      1: jit_opt_a
 0.82% :     0.000147s :      1: jit_opt_after_cconv
 0.32% :     0.000057s :      1: jit_opt_b
 2.32% :     0.000417s :      1: loop_unroll
 2.69% :     0.000484s :      1: mutable_eliminate
 5.12% :     0.000921s :     26: opt.transform.jit_opt_a
 0.28% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000030s :      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
 0.18% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000037s :      4: opt.transform.symbol_engine_opt
30.00% :     0.005394s :      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% :     0.000033s :      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
 1.53% :     0.000276s :      1: renormalize.infer
 1.38% :     0.000248s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000030s :      1: rewriter_after_opt_a
 0.45% :     0.000081s :      1: rewriter_before_opt_a
 0.47% :     0.000084s :      1: symbol_engine_optimizer
24.24% :     0.004357s :      1: type_inference


TotalTime = 0.0404871, [33]
[bootstrap]: 0.00029402
[type_inference]: 0.0214429
[event_method]: 1.909e-05
[auto_monad]: 0.0001441
[graph_reusing]: 5.22e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 0.00012368
[rewriter_before_opt_a]: 9.327e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00575176, [2]
    [Cycle 1]: 0.00213532, [27]
        [switch_simplify]: 5.493e-05
        [loop_unroll]: 3.541e-05
        [a_1]: 0.00067297
        [with_stream_mark]: 1.589e-05
        [recompute_prepare]: 1.103e-05
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 4.69002e-06
        [updatestate_loads_eliminate]: 4.14997e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 9.00001e-06
        [updatestate_useless_node_eliminater]: 1.198e-05
        [accelerated_algorithm]: 8.70001e-06
        [meta_shard_fg_expand]: 2.28002e-06
        [get_grad_eliminate_]: 8.07e-06
        [merge_forward]: 4.67e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.673e-05
        [j_node_and_user_rematch]: 1.35e-05
        [meta_fg_expand]: 3.63e-06
        [replace_old_param]: 1.126e-05
        [inline_without_move]: 7.85998e-06
        [renormalize]: 0.00099172
        [add_forward_monad_depend]: 5.03002e-06
        [auto_monad_grad]: 1.96e-06
        [auto_monad_eliminator]: 1.857e-05
        [cse]: 2.86e-05
        [replace_applicator]: 1.554e-05
    [Cycle 2]: 0.000457, [27]
        [switch_simplify]: 9.17001e-06
        [loop_unroll]: 8.10999e-06
        [a_1]: 0.00016471
        [with_stream_mark]: 1.071e-05
        [recompute_prepare]: 8.33999e-06
        [updatestate_depend_eliminate]: 4.82998e-06
        [updatestate_assign_eliminate]: 4.75001e-06
        [updatestate_loads_eliminate]: 3.88001e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 8.40001e-06
        [updatestate_useless_node_eliminater]: 1.057e-05
        [accelerated_algorithm]: 7.7e-06
        [meta_shard_fg_expand]: 1.80001e-06
        [get_grad_eliminate_]: 7.32002e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 1.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.576e-05
        [j_node_and_user_rematch]: 1.263e-05
        [meta_fg_expand]: 2.61999e-06
        [replace_old_param]: 9.32001e-06
        [inline_without_move]: 7.26999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 1.011e-05
        [cse]: 1.921e-05
        [replace_applicator]: 7.83001e-06
[py_interpret_to_execute_after_opt_a]: 1.01e-05
[rewriter_after_opt_a]: 0.00027289
[convert_after_rewriter]: 1.08e-05
[order_py_execute_after_rewriter]: 7.68001e-06
[mutable_eliminate]: 0.00059789
[jit_opt_b]: 6.745e-05, [1]
    [Cycle 1]: 6.05e-05, [2]
        [frontend_op_eliminate]: 2.423e-05
        [inline_after_opt_a]: 2.441e-05
[cconv]: 2.365e-05
[loop_unroll]: 0.00046436
[jit_opt_after_cconv]: 0.00019646, [1]
    [Cycle 1]: 0.00018988, [11]
        [c_1]: 4.673e-05
        [parameter_eliminate]: 3.20998e-06
        [updatestate_depend_eliminate]: 8.63001e-06
        [updatestate_assign_eliminate]: 4.57e-06
        [updatestate_loads_eliminate]: 3.9e-06
        [cse]: 2.635e-05
        [call_graph_tuple_transform]: 2.275e-05
        [tuple_list_get_item_eliminator]: 8.58001e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 9.26002e-06
[remove_dup_value]: 1.709e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 9.29e-06
[add_recomputation]: 6.094e-05
[cse_after_recomputation]: 3.103e-05, [1]
    [Cycle 1]: 2.503e-05, [1]
        [cse]: 1.903e-05
[auto_monad_reorder]: 2.459e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.84e-06
[opt_after_jit_grad]: 0.00049974
[symbol_engine_optimizer]: 9.689e-05, [1]
    [Cycle 1]: 9.069e-05, [6]
        [build]: 9.90002e-06
        [elim_shapecalc]: 1.158e-05
        [elim_not_effective]: 1.8e-05
        [opt_reshape]: 9.03002e-06
        [fold_const_symbol]: 1.307e-05
        [renormalize]: 3.9002e-07
[validate]: 4.952e-05
[backend_pass]: 1.16002e-06
[task_emit]: 0.00993246
[execute]: 7.92e-06
Sums
    bootstrap                            :     0.000294s :  0.80%
    type_inference                       :     0.021443s : 58.42%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000144s :  0.39%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000124s :  0.34%
    rewriter_before_opt_a                :     0.000093s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.17%
    jit_opt_a.loop_unroll                :     0.000044s :  0.12%
    jit_opt_a.a_1                        :     0.000838s :  2.28%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  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 :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.000992s :  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      :     0.000029s :  0.08%
    jit_opt_a.cse                        :     0.000048s :  0.13%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000273s :  0.74%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000598s :  1.63%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000024s :  0.06%
    loop_unroll                          :     0.000464s :  1.27%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000009s :  0.03%
    add_recomputation                    :     0.000061s :  0.17%
    cse_after_recomputation.cse          :     0.000019s :  0.05%
    auto_monad_reorder                   :     0.000025s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000500s :  1.36%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000050s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009932s : 27.06%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000242    47
 3.05% :     0.000007s :      2: substitution.depend_value_elim
 0.97% :     0.000002s :      4: substitution.elim_not_effective
 0.79% :     0.000002s :      4: substitution.fold_const_symbol
 2.58% :     0.000006s :      5: substitution.graph_param_transform
76.05% :     0.000184s :      4: substitution.inline
 1.78% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.64% :     0.000006s :      8: substitution.remove_not_recompute_node
 1.42% :     0.000003s :      2: substitution.replace_old_param
 4.05% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
 3.36% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.30% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021380     2
93.52% :     0.019995s :      1: type_inference.infer
 6.48% :     0.001385s :      1: type_inference.specialize
------[replace.]   0.000088     6
65.53% :     0.000058s :      4: replace.inline
34.47% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
95.42% :     0.000181s :      4: match.inline
 4.58% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1021
 1.27% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.25% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.33% :     0.000002s :     15: predicate.addn_check_dump
 1.35% :     0.000002s :     15: predicate.addn_zero_filter
 2.09% :     0.000004s :     15: predicate.arithmetic_simplify
 1.55% :     0.000003s :     15: predicate.cast_eliminate
 0.47% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     15: predicate.compare_switch_simplify
 1.41% :     0.000002s :     15: predicate.depend_value_elim
 1.25% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.65% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     15: predicate.environ_get_eliminate
 1.27% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.08% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 5.06% :     0.000009s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :     10: predicate.less_batch_normalization
 1.47% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.92% :     0.000003s :     22: predicate.load_eliminater
 1.41% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.43% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.63% :     0.000003s :      5: predicate.mutable_eliminate
 0.63% :     0.000001s :      5: predicate.opt_reshape
 2.30% :     0.000004s :     22: predicate.partial_eliminate
 1.21% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.58% :     0.000003s :     15: predicate.reduce_eliminate
 1.69% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.62% :     0.000003s :     27: predicate.replace_applicator
 0.68% :     0.000001s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.33% :     0.000002s :     15: predicate.reshape_eliminate
 1.32% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.08% :     0.000002s :     10: predicate.special_op_eliminate
 1.14% :     0.000002s :     10: predicate.specialize_transform
 1.47% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.20% :     0.000004s :     21: predicate.switch_defer_inline
 2.08% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.41% :     0.000011s :     68: predicate.switch_simplify
 1.40% :     0.000002s :     15: predicate.tile_eliminate
 1.29% :     0.000002s :     15: predicate.transpose_eliminate
 1.50% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.98% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.48% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     15: predicate.value_based_eliminate
 0.46% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000796    13
45.70% :     0.000364s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.26% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.04% :     0.000390s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042793    76
 0.15% :     0.000064s :      1: add_recomputation
 0.35% :     0.000149s :      1: auto_monad
 0.06% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.73% :     0.000310s :      1: bootstrap
 0.06% :     0.000026s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.08% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000012s :      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
13.45% :     0.005755s :      1: jit_opt_a
 0.47% :     0.000200s :      1: jit_opt_after_cconv
 0.16% :     0.000070s :      1: jit_opt_b
 1.11% :     0.000473s :      1: loop_unroll
 1.42% :     0.000608s :      1: mutable_eliminate
 2.63% :     0.001127s :     26: opt.transform.jit_opt_a
 0.20% :     0.000083s :      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% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000048s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000508s :      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
 0.30% :     0.000128s :      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
 1.34% :     0.000575s :      1: renormalize.infer
 0.95% :     0.000408s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.65% :     0.000278s :      1: rewriter_after_opt_a
 0.23% :     0.000097s :      1: rewriter_before_opt_a
 0.23% :     0.000100s :      1: symbol_engine_optimizer
23.24% :     0.009946s :      1: task_emit
50.14% :     0.021456s :      1: type_inference
 0.17% :     0.000071s :      1: validate


TotalTime = 0.0122098, [30]
[bootstrap]: 0.00027119
[type_inference]: 0.00455724
[event_method]: 2.118e-05
[auto_monad]: 6.1e-05
[graph_reusing]: 5.60001e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.361e-05
[rewriter_before_opt_a]: 8.977e-05
[expand_dump_flag]: 3.9e-06
[jit_opt_a]: 0.00513643, [2]
    [Cycle 1]: 0.00162453, [27]
        [switch_simplify]: 5.594e-05
        [loop_unroll]: 3.529e-05
        [a_1]: 0.00059847
        [with_stream_mark]: 1.49e-05
        [recompute_prepare]: 8.02e-06
        [updatestate_depend_eliminate]: 3.76001e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 3.13998e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.79001e-06
        [updatestate_useless_node_eliminater]: 6.01e-06
        [accelerated_algorithm]: 5.91998e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 3.49001e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 9.74999e-06
        [meta_fg_expand]: 2.48002e-06
        [replace_old_param]: 1.042e-05
        [inline_without_move]: 5.99e-06
        [renormalize]: 0.00060354
        [add_forward_monad_depend]: 4.60001e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.302e-05
        [cse]: 2.425e-05
        [replace_applicator]: 1.261e-05
    [Cycle 2]: 0.00036676, [27]
        [switch_simplify]: 6.29001e-06
        [loop_unroll]: 5.80002e-06
        [a_1]: 0.00011202
        [with_stream_mark]: 8.70999e-06
        [recompute_prepare]: 5.90002e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.72999e-06
        [updatestate_useless_node_eliminater]: 1.088e-05
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.307e-05
        [j_node_and_user_rematch]: 8.76002e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 1.175e-05
        [inline_without_move]: 5.49e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 8.90024e-07
        [auto_monad_eliminator]: 6.14999e-06
        [cse]: 2.12e-05
        [replace_applicator]: 6.85998e-06
[py_interpret_to_execute_after_opt_a]: 1.043e-05
[rewriter_after_opt_a]: 3.489e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 5.47001e-06
[mutable_eliminate]: 0.0004783
[jit_opt_b]: 5.286e-05, [1]
    [Cycle 1]: 4.636e-05, [2]
        [frontend_op_eliminate]: 1.832e-05
        [inline_after_opt_a]: 1.651e-05
[cconv]: 2.271e-05
[loop_unroll]: 0.00039908
[jit_opt_after_cconv]: 0.00014466, [1]
    [Cycle 1]: 0.00013859, [11]
        [c_1]: 2.387e-05
        [parameter_eliminate]: 2.52001e-06
        [updatestate_depend_eliminate]: 5.35001e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [cse]: 1.933e-05
        [call_graph_tuple_transform]: 1.91e-05
        [tuple_list_get_item_eliminator]: 6.33e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 6.04999e-06
[remove_dup_value]: 1.334e-05
[partial_unused_args_eliminate]: 2.16998e-06
[environ_conv]: 5.34998e-06
[add_recomputation]: 4.149e-05
[cse_after_recomputation]: 2.262e-05, [1]
    [Cycle 1]: 1.764e-05, [1]
        [cse]: 1.21e-05
[auto_monad_reorder]: 9.021e-05
[get_jit_bprop_graph]: 1.49998e-06
[rewriter_after_jit_bprop_graph]: 4.48001e-06
[opt_after_jit_grad]: 0.00043981
[symbol_engine_optimizer]: 7.511e-05, [1]
    [Cycle 1]: 6.901e-05, [6]
        [build]: 3.30998e-06
        [elim_shapecalc]: 8.99e-06
        [elim_not_effective]: 1.315e-05
        [opt_reshape]: 6.21e-06
        [fold_const_symbol]: 9.47001e-06
        [renormalize]: 3.50003e-07
[validate]: 3.279e-05
Sums
    bootstrap                            :     0.000271s :  3.18%
    type_inference                       :     0.004557s : 53.43%
    event_method                         :     0.000021s :  0.25%
    auto_monad                           :     0.000061s :  0.72%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.39%
    rewriter_before_opt_a                :     0.000090s :  1.05%
    expand_dump_flag                     :     0.000004s :  0.05%
    jit_opt_a.switch_simplify            :     0.000062s :  0.73%
    jit_opt_a.loop_unroll                :     0.000041s :  0.48%
    jit_opt_a.a_1                        :     0.000710s :  8.33%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.28%
    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 :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000017s :  0.20%
    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_        :     0.000011s :  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 :     0.000027s :  0.32%
    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          :     0.000022s :  0.26%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000604s :  7.08%
    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                        :     0.000045s :  0.53%
    jit_opt_a.replace_applicator         :     0.000019s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000035s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000478s :  5.61%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000023s :  0.27%
    loop_unroll                          :     0.000399s :  4.68%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.28%
    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              :     0.000019s :  0.23%
    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 :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000013s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000041s :  0.49%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000090s :  1.06%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000440s :  5.16%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    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 :  0.00%
    validate                             :     0.000033s :  0.38%


Time group info:
------[substitution.]   0.000198    28
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000005s :      4: substitution.graph_param_transform
81.24% :     0.000161s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000004s :      4: substitution.replace_old_param
 8.41% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004504     2
80.24% :     0.003614s :      1: type_inference.infer
19.76% :     0.000890s :      1: type_inference.specialize
------[replace.]   0.000081     8
56.28% :     0.000046s :      4: replace.inline
43.72% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     8
91.51% :     0.000158s :      4: match.inline
 8.49% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     13: predicate.addn_check_dump
 1.25% :     0.000002s :     13: predicate.addn_zero_filter
 2.05% :     0.000003s :     13: predicate.arithmetic_simplify
 1.32% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     13: predicate.compare_switch_simplify
 1.18% :     0.000002s :     13: predicate.depend_value_elim
 1.21% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.56% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     13: predicate.environ_get_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.94% :     0.000001s :      8: predicate.get_grad_eliminate
 0.29% :     0.000000s :      4: predicate.graph_param_transform
 5.27% :     0.000007s :     29: predicate.inline
 0.93% :     0.000001s :      8: predicate.inline_without_move
 0.45% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      8: predicate.less_batch_normalization
 1.62% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.03% :     0.000003s :     21: predicate.load_eliminater
 1.24% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.14% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.04% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     13: predicate.merge_addn
 1.18% :     0.000002s :     13: predicate.minmaximum_grad
 1.39% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.64% :     0.000004s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     13: predicate.reduce_eliminate
 1.70% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.12% :     0.000003s :     25: predicate.replace_applicator
 0.84% :     0.000001s :      8: predicate.replace_old_param
 0.38% :     0.000001s :      4: predicate.reset_defer_inline
 1.19% :     0.000002s :     13: predicate.reshape_eliminate
 1.21% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :      8: predicate.special_op_eliminate
 0.97% :     0.000001s :      8: predicate.specialize_transform
 1.31% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.69% :     0.000010s :     67: predicate.switch_simplify
 1.20% :     0.000002s :     13: predicate.tile_eliminate
 1.23% :     0.000002s :     13: predicate.transpose_eliminate
 1.40% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     13: predicate.value_based_eliminate
 0.44% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000570    12
46.78% :     0.000266s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.82% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.40% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013849    72
 0.32% :     0.000044s :      1: add_recomputation
 0.46% :     0.000064s :      1: auto_monad
 0.67% :     0.000093s :      1: auto_monad_reorder
 2.02% :     0.000280s :      1: bootstrap
 0.18% :     0.000025s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.18% :     0.000025s :      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
37.11% :     0.005140s :      1: jit_opt_a
 1.06% :     0.000147s :      1: jit_opt_after_cconv
 0.40% :     0.000055s :      1: jit_opt_b
 2.93% :     0.000406s :      1: loop_unroll
 3.51% :     0.000485s :      1: mutable_eliminate
 6.88% :     0.000953s :     26: opt.transform.jit_opt_a
 0.38% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000029s :      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
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.23% :     0.000447s :      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% :     0.000036s :      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
 2.24% :     0.000310s :      1: renormalize.infer
 2.07% :     0.000287s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000038s :      1: rewriter_after_opt_a
 0.67% :     0.000093s :      1: rewriter_before_opt_a
 0.56% :     0.000078s :      1: symbol_engine_optimizer
32.96% :     0.004565s :      1: type_inference


TotalTime = 0.0120989, [30]
[bootstrap]: 0.00026921
[type_inference]: 0.00456843
[event_method]: 2.053e-05
[auto_monad]: 6.162e-05
[graph_reusing]: 5.81e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.471e-05
[rewriter_before_opt_a]: 8.741e-05
[expand_dump_flag]: 3.45003e-06
[jit_opt_a]: 0.00507261, [2]
    [Cycle 1]: 0.00167275, [27]
        [switch_simplify]: 5.555e-05
        [loop_unroll]: 3.45e-05
        [a_1]: 0.0005942
        [with_stream_mark]: 1.539e-05
        [recompute_prepare]: 8.41002e-06
        [updatestate_depend_eliminate]: 4.27e-06
        [updatestate_assign_eliminate]: 3.76001e-06
        [updatestate_loads_eliminate]: 2.79999e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 7.3e-06
        [updatestate_useless_node_eliminater]: 5.84e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.501e-05
        [j_node_and_user_rematch]: 9.65002e-06
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 1.064e-05
        [inline_without_move]: 5.75001e-06
        [renormalize]: 0.00064611
        [add_forward_monad_depend]: 4.92999e-06
        [auto_monad_grad]: 1.88002e-06
        [auto_monad_eliminator]: 1.402e-05
        [cse]: 3.117e-05
        [replace_applicator]: 1.23e-05
    [Cycle 2]: 0.00035915, [27]
        [switch_simplify]: 6.73e-06
        [loop_unroll]: 5.84e-06
        [a_1]: 0.0001123
        [with_stream_mark]: 9.80002e-06
        [recompute_prepare]: 5.72999e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 5.95002e-06
        [updatestate_useless_node_eliminater]: 5.56e-06
        [accelerated_algorithm]: 5.54998e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.376e-05
        [j_node_and_user_rematch]: 8.62e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 8.84e-06
        [inline_without_move]: 6.28e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.96e-06
        [cse]: 1.351e-05
        [replace_applicator]: 6.01e-06
[py_interpret_to_execute_after_opt_a]: 8.87999e-06
[rewriter_after_opt_a]: 3.447e-05
[convert_after_rewriter]: 7.31001e-06
[order_py_execute_after_rewriter]: 5.52001e-06
[mutable_eliminate]: 0.0004893
[jit_opt_b]: 5.305e-05, [1]
    [Cycle 1]: 4.621e-05, [2]
        [frontend_op_eliminate]: 1.827e-05
        [inline_after_opt_a]: 1.69e-05
[cconv]: 2.314e-05
[loop_unroll]: 0.00040641
[jit_opt_after_cconv]: 0.00014697, [1]
    [Cycle 1]: 0.00014062, [11]
        [c_1]: 2.414e-05
        [parameter_eliminate]: 2.29999e-06
        [updatestate_depend_eliminate]: 6.06998e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 1.947e-05
        [call_graph_tuple_transform]: 1.995e-05
        [tuple_list_get_item_eliminator]: 5.87001e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.84999e-06
[remove_dup_value]: 1.392e-05
[partial_unused_args_eliminate]: 2.53e-06
[environ_conv]: 5.66e-06
[add_recomputation]: 4.648e-05
[cse_after_recomputation]: 2.342e-05, [1]
    [Cycle 1]: 1.791e-05, [1]
        [cse]: 1.186e-05
[auto_monad_reorder]: 1.753e-05
[get_jit_bprop_graph]: 1.37e-06
[rewriter_after_jit_bprop_graph]: 4.71002e-06
[opt_after_jit_grad]: 0.0004227
[symbol_engine_optimizer]: 7.432e-05, [1]
    [Cycle 1]: 6.849e-05, [6]
        [build]: 3.41999e-06
        [elim_shapecalc]: 8.02998e-06
        [elim_not_effective]: 1.394e-05
        [opt_reshape]: 6.23998e-06
        [fold_const_symbol]: 9.44998e-06
        [renormalize]: 3.69997e-07
[validate]: 3.472e-05
Sums
    bootstrap                            :     0.000269s :  3.16%
    type_inference                       :     0.004568s : 53.69%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000062s :  0.72%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.41%
    rewriter_before_opt_a                :     0.000087s :  1.03%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.73%
    jit_opt_a.loop_unroll                :     0.000040s :  0.47%
    jit_opt_a.a_1                        :     0.000706s :  8.30%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.09%
    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       :     0.000013s :  0.16%
    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_        :     0.000011s :  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 :     0.000029s :  0.34%
    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          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000646s :  7.59%
    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      :     0.000020s :  0.23%
    jit_opt_a.cse                        :     0.000045s :  0.53%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000489s :  5.75%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000023s :  0.27%
    loop_unroll                          :     0.000406s :  4.78%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.28%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  0.23%
    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 :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000046s :  0.55%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.21%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000423s :  4.97%
    symbol_engine_optimizer.build        :     0.000003s :  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  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.41%


Time group info:
------[substitution.]   0.000196    28
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000002s :      2: substitution.fold_const_symbol
 2.88% :     0.000006s :      4: substitution.graph_param_transform
80.53% :     0.000158s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.37% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000005s :      4: substitution.replace_old_param
 8.22% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004511     2
80.44% :     0.003629s :      1: type_inference.infer
19.56% :     0.000882s :      1: type_inference.specialize
------[replace.]   0.000079     8
56.25% :     0.000045s :      4: replace.inline
43.75% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000170     8
91.59% :     0.000155s :      4: match.inline
 8.41% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     13: predicate.addn_check_dump
 1.43% :     0.000002s :     13: predicate.addn_zero_filter
 1.91% :     0.000003s :     13: predicate.arithmetic_simplify
 1.33% :     0.000002s :     13: predicate.cast_eliminate
 0.49% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     13: predicate.compare_switch_simplify
 1.16% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      4: predicate.elim_not_effective
 0.44% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     13: predicate.environ_get_eliminate
 1.19% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 0.89% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.86% :     0.000008s :     29: predicate.inline
 0.88% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      8: predicate.less_batch_normalization
 1.65% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     21: predicate.load_eliminater
 1.61% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.96% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     13: predicate.merge_addn
 1.13% :     0.000002s :     13: predicate.minmaximum_grad
 1.31% :     0.000002s :      4: predicate.mutable_eliminate
 0.41% :     0.000001s :      4: predicate.opt_reshape
 2.55% :     0.000003s :     21: predicate.partial_eliminate
 1.19% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.88% :     0.000003s :     25: predicate.replace_applicator
 0.79% :     0.000001s :      8: predicate.replace_old_param
 0.41% :     0.000001s :      4: predicate.reset_defer_inline
 1.22% :     0.000002s :     13: predicate.reshape_eliminate
 1.15% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     13: predicate.same_eliminate
 0.57% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.39% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.75% :     0.000010s :     67: predicate.switch_simplify
 1.22% :     0.000002s :     13: predicate.tile_eliminate
 1.17% :     0.000002s :     13: predicate.transpose_eliminate
 1.42% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.32% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     13: predicate.value_based_eliminate
 0.41% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000574    12
49.73% :     0.000285s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.07% :     0.000023s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.19% :     0.000265s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013779    72
 0.36% :     0.000049s :      1: add_recomputation
 0.47% :     0.000065s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 2.04% :     0.000281s :      1: bootstrap
 0.19% :     0.000026s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.18% :     0.000024s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
36.84% :     0.005076s :      1: jit_opt_a
 1.09% :     0.000150s :      1: jit_opt_after_cconv
 0.40% :     0.000056s :      1: jit_opt_b
 3.00% :     0.000414s :      1: loop_unroll
 3.61% :     0.000498s :      1: mutable_eliminate
 6.83% :     0.000941s :     26: opt.transform.jit_opt_a
 0.38% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.19% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000430s :      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% :     0.000037s :      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
 2.62% :     0.000361s :      1: renormalize.infer
 2.02% :     0.000278s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000038s :      1: rewriter_after_opt_a
 0.65% :     0.000090s :      1: rewriter_before_opt_a
 0.56% :     0.000077s :      1: symbol_engine_optimizer
33.24% :     0.004580s :      1: type_inference


TotalTime = 0.0111655, [30]
[bootstrap]: 0.00026455
[type_inference]: 0.00418763
[event_method]: 1.891e-05
[auto_monad]: 3.637e-05
[graph_reusing]: 3.23998e-06
[pre_auto_parallel]: 1.60999e-06
[py_interpret_to_execute]: 3.021e-05
[rewriter_before_opt_a]: 7.691e-05
[expand_dump_flag]: 1.63002e-06
[jit_opt_a]: 0.00476448, [2]
    [Cycle 1]: 0.00147687, [27]
        [switch_simplify]: 5.401e-05
        [loop_unroll]: 3.585e-05
        [a_1]: 0.00058132
        [with_stream_mark]: 9.41e-06
        [recompute_prepare]: 7.67002e-06
        [updatestate_depend_eliminate]: 3.2e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 7.13998e-06
        [updatestate_useless_node_eliminater]: 5.94e-06
        [accelerated_algorithm]: 6.33998e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 5.82999e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 7.60017e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.213e-05
        [j_node_and_user_rematch]: 9.66e-06
        [meta_fg_expand]: 1.95001e-06
        [replace_old_param]: 9.69e-06
        [inline_without_move]: 6.17001e-06
        [renormalize]: 0.00050967
        [add_forward_monad_depend]: 3.46001e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 9.06002e-06
        [cse]: 1.875e-05
        [replace_applicator]: 1.073e-05
    [Cycle 2]: 0.0003632, [27]
        [switch_simplify]: 6.39001e-06
        [loop_unroll]: 5.64998e-06
        [a_1]: 0.00011738
        [with_stream_mark]: 7e-06
        [recompute_prepare]: 5.84e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 9.10019e-07
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.59e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 1.19998e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.294e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 9.57999e-06
        [inline_without_move]: 5.81e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.14998e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 6.49999e-06
        [cse]: 1.292e-05
        [replace_applicator]: 5.98998e-06
[py_interpret_to_execute_after_opt_a]: 8.2e-06
[rewriter_after_opt_a]: 2.8e-05
[convert_after_rewriter]: 5.82001e-06
[order_py_execute_after_rewriter]: 4.89998e-06
[mutable_eliminate]: 0.00042654
[jit_opt_b]: 5.09e-05, [1]
    [Cycle 1]: 4.512e-05, [2]
        [frontend_op_eliminate]: 1.748e-05
        [inline_after_opt_a]: 1.67e-05
[cconv]: 1.295e-05
[loop_unroll]: 0.00038492
[jit_opt_after_cconv]: 0.00013456, [1]
    [Cycle 1]: 0.0001293, [11]
        [c_1]: 2.342e-05
        [parameter_eliminate]: 1.60001e-06
        [updatestate_depend_eliminate]: 4.41002e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [cse]: 1.678e-05
        [call_graph_tuple_transform]: 1.703e-05
        [tuple_list_get_item_eliminator]: 5.99999e-06
        [none_parameter_eliminate]: 7.99977e-07
        [renormalize]: 2.70025e-07
        [switch_simplify]: 6.06e-06
[remove_dup_value]: 6.93e-06
[partial_unused_args_eliminate]: 1.26997e-06
[environ_conv]: 3.71001e-06
[add_recomputation]: 2.785e-05
[cse_after_recomputation]: 2.1e-05, [1]
    [Cycle 1]: 1.637e-05, [1]
        [cse]: 1.107e-05
[auto_monad_reorder]: 1.151e-05
[get_jit_bprop_graph]: 1.34e-06
[rewriter_after_jit_bprop_graph]: 4.43001e-06
[opt_after_jit_grad]: 0.00041113
[symbol_engine_optimizer]: 6.722e-05, [1]
    [Cycle 1]: 6.228e-05, [6]
        [build]: 2.04999e-06
        [elim_shapecalc]: 7.76001e-06
        [elim_not_effective]: 1.105e-05
        [opt_reshape]: 5.94e-06
        [fold_const_symbol]: 8.61002e-06
        [renormalize]: 3.39991e-07
[validate]: 2.441e-05
Sums
    bootstrap                            :     0.000265s :  3.43%
    type_inference                       :     0.004188s : 54.30%
    event_method                         :     0.000019s :  0.25%
    auto_monad                           :     0.000036s :  0.47%
    graph_reusing                        :     0.000003s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.39%
    rewriter_before_opt_a                :     0.000077s :  1.00%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.78%
    jit_opt_a.loop_unroll                :     0.000041s :  0.54%
    jit_opt_a.a_1                        :     0.000699s :  9.06%
    jit_opt_a.with_stream_mark           :     0.000016s :  0.21%
    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 :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.17%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.16%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  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 :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.25%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.25%
    jit_opt_a.inline_without_move        :     0.000012s :  0.16%
    jit_opt_a.renormalize                :     0.000510s :  6.61%
    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      :     0.000016s :  0.20%
    jit_opt_a.cse                        :     0.000032s :  0.41%
    jit_opt_a.replace_applicator         :     0.000017s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000028s :  0.36%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000427s :  5.53%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.22%
    cconv                                :     0.000013s :  0.17%
    loop_unroll                          :     0.000385s :  4.99%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.30%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  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              :     0.000017s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    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 :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000007s :  0.09%
    partial_unused_args_eliminate        :     0.000001s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000028s :  0.36%
    cse_after_recomputation.cse          :     0.000011s :  0.14%
    auto_monad_reorder                   :     0.000012s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000411s :  5.33%
    symbol_engine_optimizer.build        :     0.000002s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.14%
    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 :  0.00%
    validate                             :     0.000024s :  0.32%


Time group info:
------[substitution.]   0.000163    28
 0.74% :     0.000001s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.11% :     0.000003s :      4: substitution.graph_param_transform
83.00% :     0.000135s :      4: substitution.inline
 1.63% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.76% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.36% :     0.000004s :      4: substitution.replace_old_param
 6.84% :     0.000011s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004147     2
81.18% :     0.003366s :      1: type_inference.infer
18.82% :     0.000780s :      1: type_inference.specialize
------[replace.]   0.000078     8
56.59% :     0.000044s :      4: replace.inline
43.41% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     8
93.40% :     0.000132s :      4: match.inline
 6.60% :     0.000009s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.45% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.96% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.17% :     0.000002s :     13: predicate.addn_check_dump
 1.29% :     0.000002s :     13: predicate.addn_zero_filter
 1.93% :     0.000003s :     13: predicate.arithmetic_simplify
 1.45% :     0.000002s :     13: predicate.cast_eliminate
 0.42% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     13: predicate.compare_switch_simplify
 1.22% :     0.000002s :     13: predicate.depend_value_elim
 1.18% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      4: predicate.elim_not_effective
 0.49% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     13: predicate.environ_get_eliminate
 1.22% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.95% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.08% :     0.000007s :     29: predicate.inline
 1.02% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      8: predicate.less_batch_normalization
 1.66% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.05% :     0.000003s :     21: predicate.load_eliminater
 1.03% :     0.000001s :      4: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.75% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.24% :     0.000002s :     13: predicate.merge_addn
 1.19% :     0.000002s :     13: predicate.minmaximum_grad
 1.28% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.44% :     0.000003s :     21: predicate.partial_eliminate
 1.24% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     13: predicate.reduce_eliminate
 1.74% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.03% :     0.000003s :     25: predicate.replace_applicator
 0.61% :     0.000001s :      8: predicate.replace_old_param
 0.29% :     0.000000s :      4: predicate.reset_defer_inline
 1.45% :     0.000002s :     13: predicate.reshape_eliminate
 1.24% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     13: predicate.same_eliminate
 0.70% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      8: predicate.special_op_eliminate
 1.09% :     0.000001s :      8: predicate.specialize_transform
 1.38% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.64% :     0.000004s :     21: predicate.switch_defer_inline
 2.30% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.75% :     0.000010s :     67: predicate.switch_simplify
 1.33% :     0.000002s :     13: predicate.tile_eliminate
 1.33% :     0.000002s :     13: predicate.transpose_eliminate
 1.41% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.47% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.96% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.14% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000519    12
48.95% :     0.000254s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.11% :     0.000021s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.94% :     0.000244s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012685    72
 0.24% :     0.000031s :      1: add_recomputation
 0.31% :     0.000039s :      1: auto_monad
 0.11% :     0.000014s :      1: auto_monad_reorder
 2.21% :     0.000280s :      1: bootstrap
 0.12% :     0.000015s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.17% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000005s :      1: graph_reusing
37.58% :     0.004767s :      1: jit_opt_a
 1.08% :     0.000137s :      1: jit_opt_after_cconv
 0.42% :     0.000053s :      1: jit_opt_b
 3.08% :     0.000391s :      1: loop_unroll
 3.41% :     0.000433s :      1: mutable_eliminate
 7.33% :     0.000929s :     26: opt.transform.jit_opt_a
 0.39% :     0.000049s :      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
 0.09% :     0.000011s :      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
 3.29% :     0.000417s :      1: opt_after_jit_grad
 0.06% :     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
 0.26% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000009s :      1: remove_dup_value
 2.03% :     0.000258s :      1: renormalize.infer
 1.94% :     0.000247s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000031s :      1: rewriter_after_opt_a
 0.63% :     0.000079s :      1: rewriter_before_opt_a
 0.55% :     0.000070s :      1: symbol_engine_optimizer
33.08% :     0.004196s :      1: type_inference


TotalTime = 0.0425439, [33]
[bootstrap]: 0.00044131
[type_inference]: 0.0229969
[event_method]: 1.935e-05
[auto_monad]: 0.00015283
[graph_reusing]: 6.90002e-06
[pre_auto_parallel]: 2.97002e-06
[py_interpret_to_execute]: 0.0001314
[rewriter_before_opt_a]: 0.00010074
[expand_dump_flag]: 3.66001e-06
[jit_opt_a]: 0.00638747, [2]
    [Cycle 1]: 0.0024445, [27]
        [switch_simplify]: 6.74e-05
        [loop_unroll]: 3.699e-05
        [a_1]: 0.00070899
        [with_stream_mark]: 2.183e-05
        [recompute_prepare]: 1.115e-05
        [updatestate_depend_eliminate]: 6.00002e-06
        [updatestate_assign_eliminate]: 5.42001e-06
        [updatestate_loads_eliminate]: 4.72998e-06
        [parameter_eliminate]: 2.11e-06
        [specialize_transform]: 9.00999e-06
        [updatestate_useless_node_eliminater]: 1.185e-05
        [accelerated_algorithm]: 9.62999e-06
        [meta_shard_fg_expand]: 3.23e-06
        [get_grad_eliminate_]: 8.3e-06
        [merge_forward]: 5.29e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.966e-05
        [j_node_and_user_rematch]: 1.453e-05
        [meta_fg_expand]: 3.88999e-06
        [replace_old_param]: 1.239e-05
        [inline_without_move]: 8.02998e-06
        [renormalize]: 0.00120555
        [add_forward_monad_depend]: 6.73e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 2.234e-05
        [cse]: 3.914e-05
        [replace_applicator]: 1.759e-05
    [Cycle 2]: 0.00050847, [27]
        [switch_simplify]: 9.37001e-06
        [loop_unroll]: 8.15999e-06
        [a_1]: 0.00016956
        [with_stream_mark]: 1.318e-05
        [recompute_prepare]: 8.28999e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 4.55999e-06
        [updatestate_loads_eliminate]: 3.92002e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 3.486e-05
        [updatestate_useless_node_eliminater]: 1.26e-05
        [accelerated_algorithm]: 8.25999e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 7.88999e-06
        [merge_forward]: 5.03002e-06
        [cell_reuse_recompute_pass]: 2.01e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.632e-05
        [j_node_and_user_rematch]: 1.23e-05
        [meta_fg_expand]: 3.06999e-06
        [replace_old_param]: 1.017e-05
        [inline_without_move]: 7.58001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.69e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 1.161e-05
        [cse]: 2.293e-05
        [replace_applicator]: 7.83001e-06
[py_interpret_to_execute_after_opt_a]: 1.498e-05
[rewriter_after_opt_a]: 0.00029598
[convert_after_rewriter]: 1.145e-05
[order_py_execute_after_rewriter]: 7.21001e-06
[mutable_eliminate]: 0.00063035
[jit_opt_b]: 6.672e-05, [1]
    [Cycle 1]: 5.917e-05, [2]
        [frontend_op_eliminate]: 2.337e-05
        [inline_after_opt_a]: 2.421e-05
[cconv]: 2.768e-05
[loop_unroll]: 0.00046074
[jit_opt_after_cconv]: 0.00019845, [1]
    [Cycle 1]: 0.00019163, [11]
        [c_1]: 4.543e-05
        [parameter_eliminate]: 2.94999e-06
        [updatestate_depend_eliminate]: 9.45001e-06
        [updatestate_assign_eliminate]: 4.45e-06
        [updatestate_loads_eliminate]: 4.12e-06
        [cse]: 2.915e-05
        [call_graph_tuple_transform]: 2.304e-05
        [tuple_list_get_item_eliminator]: 8.52e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 5.49975e-07
        [switch_simplify]: 9.40001e-06
[remove_dup_value]: 1.863e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 1.123e-05
[add_recomputation]: 6.954e-05
[cse_after_recomputation]: 3.064e-05, [1]
    [Cycle 1]: 2.478e-05, [1]
        [cse]: 1.876e-05
[auto_monad_reorder]: 2.679e-05
[get_jit_bprop_graph]: 2.09999e-06
[rewriter_after_jit_bprop_graph]: 2.809e-05
[opt_after_jit_grad]: 0.0005291
[symbol_engine_optimizer]: 0.00010299, [1]
    [Cycle 1]: 9.582e-05, [6]
        [build]: 1.164e-05
        [elim_shapecalc]: 1.194e-05
        [elim_not_effective]: 1.906e-05
        [opt_reshape]: 8.59e-06
        [fold_const_symbol]: 1.387e-05
        [renormalize]: 4.59986e-07
[validate]: 5.735e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.00948303
[execute]: 6.84001e-06
Sums
    bootstrap                            :     0.000441s :  1.15%
    type_inference                       :     0.022997s : 59.78%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000153s :  0.40%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000131s :  0.34%
    rewriter_before_opt_a                :     0.000101s :  0.26%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.20%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000879s :  2.28%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000044s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001206s :  3.13%
    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      :     0.000034s :  0.09%
    jit_opt_a.cse                        :     0.000062s :  0.16%
    jit_opt_a.replace_applicator         :     0.000025s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000296s :  0.77%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000630s :  1.64%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000028s :  0.07%
    loop_unroll                          :     0.000461s :  1.20%
    jit_opt_after_cconv.c_1              :     0.000045s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000029s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000070s :  0.18%
    cse_after_recomputation.cse          :     0.000019s :  0.05%
    auto_monad_reorder                   :     0.000027s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000028s :  0.07%
    opt_after_jit_grad                   :     0.000529s :  1.38%
    symbol_engine_optimizer.build        :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000057s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009483s : 24.65%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000280    47
 2.85% :     0.000008s :      2: substitution.depend_value_elim
 1.00% :     0.000003s :      4: substitution.elim_not_effective
 0.71% :     0.000002s :      4: substitution.fold_const_symbol
 2.41% :     0.000007s :      5: substitution.graph_param_transform
75.93% :     0.000213s :      4: substitution.inline
 1.79% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.55% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.74% :     0.000005s :      2: substitution.replace_old_param
 4.41% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.22% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.41% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022932     2
93.73% :     0.021493s :      1: type_inference.infer
 6.27% :     0.001439s :      1: type_inference.specialize
------[replace.]   0.000092     6
66.49% :     0.000061s :      4: replace.inline
33.51% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
94.91% :     0.000210s :      4: match.inline
 5.09% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1021
 1.29% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     15: predicate.addn_check_dump
 1.36% :     0.000002s :     15: predicate.addn_zero_filter
 1.86% :     0.000003s :     15: predicate.arithmetic_simplify
 1.40% :     0.000002s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.10% :     0.000002s :     15: predicate.compare_switch_simplify
 1.47% :     0.000003s :     15: predicate.depend_value_elim
 1.12% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.62% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.22% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 5.58% :     0.000009s :     31: predicate.inline
 1.02% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :     10: predicate.less_batch_normalization
 1.39% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.88% :     0.000003s :     22: predicate.load_eliminater
 1.40% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.43% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     15: predicate.merge_addn
 1.20% :     0.000002s :     15: predicate.minmaximum_grad
 1.45% :     0.000002s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.36% :     0.000004s :     22: predicate.partial_eliminate
 1.21% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.74% :     0.000003s :     15: predicate.reduce_eliminate
 1.45% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.75% :     0.000003s :     27: predicate.replace_applicator
 0.68% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.37% :     0.000002s :     15: predicate.reshape_eliminate
 1.47% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.89% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.04% :     0.000002s :     10: predicate.special_op_eliminate
 1.12% :     0.000002s :     10: predicate.specialize_transform
 1.54% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.17% :     0.000004s :     21: predicate.switch_defer_inline
 1.89% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.58% :     0.000011s :     68: predicate.switch_simplify
 1.53% :     0.000003s :     15: predicate.tile_eliminate
 1.22% :     0.000002s :     15: predicate.transpose_eliminate
 1.56% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.37% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000003s :     15: predicate.value_based_eliminate
 0.45% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000865    13
41.80% :     0.000361s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.87% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.34% :     0.000453s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045152    76
 0.16% :     0.000073s :      1: add_recomputation
 0.35% :     0.000158s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.01% :     0.000456s :      1: bootstrap
 0.07% :     0.000030s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.05% :     0.000023s :      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
14.15% :     0.006391s :      1: jit_opt_a
 0.45% :     0.000201s :      1: jit_opt_after_cconv
 0.15% :     0.000070s :      1: jit_opt_b
 1.04% :     0.000469s :      1: loop_unroll
 1.42% :     0.000639s :      1: mutable_eliminate
 2.69% :     0.001214s :     26: opt.transform.jit_opt_a
 0.18% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000539s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000136s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.52% :     0.000688s :      1: renormalize.infer
 1.12% :     0.000508s :      1: renormalize.specialize
 0.07% :     0.000031s :      1: rewriter_after_jit_bprop_graph
 0.67% :     0.000301s :      1: rewriter_after_opt_a
 0.23% :     0.000105s :      1: rewriter_before_opt_a
 0.23% :     0.000106s :      1: symbol_engine_optimizer
21.03% :     0.009497s :      1: task_emit
50.96% :     0.023010s :      1: type_inference
 0.18% :     0.000082s :      1: validate


TotalTime = 0.0118572, [30]
[bootstrap]: 0.0002504
[type_inference]: 0.00431139
[event_method]: 2.13e-05
[auto_monad]: 6.278e-05
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 3.478e-05
[rewriter_before_opt_a]: 8.662e-05
[expand_dump_flag]: 3.41001e-06
[jit_opt_a]: 0.00510337, [2]
    [Cycle 1]: 0.00170937, [27]
        [switch_simplify]: 5.77e-05
        [loop_unroll]: 3.475e-05
        [a_1]: 0.00063503
        [with_stream_mark]: 1.605e-05
        [recompute_prepare]: 8.07e-06
        [updatestate_depend_eliminate]: 3.98001e-06
        [updatestate_assign_eliminate]: 3.54002e-06
        [updatestate_loads_eliminate]: 2.90002e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.94999e-06
        [updatestate_useless_node_eliminater]: 5.71e-06
        [accelerated_algorithm]: 5.99999e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 6.04001e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.396e-05
        [j_node_and_user_rematch]: 9.87999e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 1.093e-05
        [inline_without_move]: 5.96998e-06
        [renormalize]: 0.00063688
        [add_forward_monad_depend]: 4.96997e-06
        [auto_monad_grad]: 1.80001e-06
        [auto_monad_eliminator]: 1.417e-05
        [cse]: 3.339e-05
        [replace_applicator]: 1.251e-05
    [Cycle 2]: 0.00036238, [27]
        [switch_simplify]: 6.58e-06
        [loop_unroll]: 5.76e-06
        [a_1]: 0.00011452
        [with_stream_mark]: 8.71002e-06
        [recompute_prepare]: 5.52001e-06
        [updatestate_depend_eliminate]: 2.95998e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 5.66e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 2.44999e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.4e-05
        [j_node_and_user_rematch]: 8.74e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 9.57999e-06
        [inline_without_move]: 5.82001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 6.38e-06
        [cse]: 1.448e-05
        [replace_applicator]: 6.34999e-06
[py_interpret_to_execute_after_opt_a]: 9.75002e-06
[rewriter_after_opt_a]: 3.426e-05
[convert_after_rewriter]: 6.79999e-06
[order_py_execute_after_rewriter]: 5.42999e-06
[mutable_eliminate]: 0.00047604
[jit_opt_b]: 5.463e-05, [1]
    [Cycle 1]: 4.772e-05, [2]
        [frontend_op_eliminate]: 1.872e-05
        [inline_after_opt_a]: 1.732e-05
[cconv]: 2.245e-05
[loop_unroll]: 0.00040309
[jit_opt_after_cconv]: 0.00014528, [1]
    [Cycle 1]: 0.00013894, [11]
        [c_1]: 2.415e-05
        [parameter_eliminate]: 2.35002e-06
        [updatestate_depend_eliminate]: 5.62999e-06
        [updatestate_assign_eliminate]: 2.70997e-06
        [updatestate_loads_eliminate]: 2.25002e-06
        [cse]: 1.995e-05
        [call_graph_tuple_transform]: 1.889e-05
        [tuple_list_get_item_eliminator]: 6.16998e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 5.96e-06
[remove_dup_value]: 1.495e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 5.24e-06
[add_recomputation]: 4.816e-05
[cse_after_recomputation]: 2.215e-05, [1]
    [Cycle 1]: 1.693e-05, [1]
        [cse]: 1.141e-05
[auto_monad_reorder]: 1.741e-05
[get_jit_bprop_graph]: 1.58002e-06
[rewriter_after_jit_bprop_graph]: 4.65999e-06
[opt_after_jit_grad]: 0.00044384
[symbol_engine_optimizer]: 7.424e-05, [1]
    [Cycle 1]: 6.783e-05, [6]
        [build]: 3.26999e-06
        [elim_shapecalc]: 8e-06
        [elim_not_effective]: 1.332e-05
        [opt_reshape]: 6.06e-06
        [fold_const_symbol]: 9.39998e-06
        [renormalize]: 4.09986e-07
[validate]: 3.177e-05
Sums
    bootstrap                            :     0.000250s :  3.03%
    type_inference                       :     0.004311s : 52.11%
    event_method                         :     0.000021s :  0.26%
    auto_monad                           :     0.000063s :  0.76%
    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                :     0.000087s :  1.05%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000064s :  0.78%
    jit_opt_a.loop_unroll                :     0.000041s :  0.49%
    jit_opt_a.a_1                        :     0.000750s :  9.06%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.30%
    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 :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  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_        :     0.000011s :  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 :     0.000028s :  0.34%
    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          :     0.000021s :  0.25%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000637s :  7.70%
    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      :     0.000021s :  0.25%
    jit_opt_a.cse                        :     0.000048s :  0.58%
    jit_opt_a.replace_applicator         :     0.000019s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000034s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000476s :  5.75%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000022s :  0.27%
    loop_unroll                          :     0.000403s :  4.87%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.29%
    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 :  0.03%
    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      :     0.000000s :  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                    :     0.000048s :  0.58%
    cse_after_recomputation.cse          :     0.000011s :  0.14%
    auto_monad_reorder                   :     0.000017s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000444s :  5.36%
    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 :  0.00%
    validate                             :     0.000032s :  0.38%


Time group info:
------[substitution.]   0.000228    28
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.22% :     0.000005s :      4: substitution.graph_param_transform
83.11% :     0.000190s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000005s :      4: substitution.replace_old_param
 7.45% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004257     2
79.18% :     0.003370s :      1: type_inference.infer
20.82% :     0.000886s :      1: type_inference.specialize
------[replace.]   0.000084     8
56.82% :     0.000047s :      4: replace.inline
43.18% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     8
92.59% :     0.000187s :      4: match.inline
 7.41% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   912
 1.21% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.17% :     0.000002s :     13: predicate.addn_check_dump
 1.27% :     0.000002s :     13: predicate.addn_zero_filter
 1.86% :     0.000003s :     13: predicate.arithmetic_simplify
 1.43% :     0.000002s :     13: predicate.cast_eliminate
 0.45% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     13: predicate.compare_switch_simplify
 1.17% :     0.000002s :     13: predicate.depend_value_elim
 1.17% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.52% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     13: predicate.environ_get_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.37% :     0.000007s :     29: predicate.inline
 1.03% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      8: predicate.less_batch_normalization
 1.63% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.02% :     0.000003s :     21: predicate.load_eliminater
 1.21% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.09% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     13: predicate.merge_addn
 1.18% :     0.000002s :     13: predicate.minmaximum_grad
 1.42% :     0.000002s :      4: predicate.mutable_eliminate
 0.46% :     0.000001s :      4: predicate.opt_reshape
 2.64% :     0.000004s :     21: predicate.partial_eliminate
 1.36% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     13: predicate.reduce_eliminate
 1.69% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.89% :     0.000003s :     25: predicate.replace_applicator
 0.75% :     0.000001s :      8: predicate.replace_old_param
 0.37% :     0.000001s :      4: predicate.reset_defer_inline
 1.25% :     0.000002s :     13: predicate.reshape_eliminate
 1.17% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      8: predicate.special_op_eliminate
 1.03% :     0.000001s :      8: predicate.specialize_transform
 1.30% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     21: predicate.switch_defer_inline
 2.26% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.63% :     0.000010s :     67: predicate.switch_simplify
 1.26% :     0.000002s :     13: predicate.tile_eliminate
 1.32% :     0.000002s :     13: predicate.transpose_eliminate
 1.51% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.07% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.85% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000559    12
46.04% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.01% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.94% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013562    72
 0.38% :     0.000051s :      1: add_recomputation
 0.49% :     0.000066s :      1: auto_monad
 0.15% :     0.000020s :      1: auto_monad_reorder
 1.94% :     0.000263s :      1: bootstrap
 0.18% :     0.000025s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.18% :     0.000025s :      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
37.65% :     0.005107s :      1: jit_opt_a
 1.09% :     0.000148s :      1: jit_opt_after_cconv
 0.42% :     0.000057s :      1: jit_opt_b
 3.03% :     0.000411s :      1: loop_unroll
 3.57% :     0.000484s :      1: mutable_eliminate
 7.27% :     0.000986s :     26: opt.transform.jit_opt_a
 0.38% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000030s :      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
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.32% :     0.000451s :      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% :     0.000037s :      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
 2.53% :     0.000343s :      1: renormalize.infer
 2.12% :     0.000287s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000037s :      1: rewriter_after_opt_a
 0.66% :     0.000089s :      1: rewriter_before_opt_a
 0.57% :     0.000077s :      1: symbol_engine_optimizer
31.87% :     0.004323s :      1: type_inference


TotalTime = 0.0118047, [30]
[bootstrap]: 0.00025133
[type_inference]: 0.00443929
[event_method]: 2.061e-05
[auto_monad]: 6.32e-05
[graph_reusing]: 6.05002e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 3.698e-05
[rewriter_before_opt_a]: 8.549e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.0049547, [2]
    [Cycle 1]: 0.00162035, [27]
        [switch_simplify]: 5.528e-05
        [loop_unroll]: 3.467e-05
        [a_1]: 0.00064403
        [with_stream_mark]: 1.599e-05
        [recompute_prepare]: 9.02999e-06
        [updatestate_depend_eliminate]: 3.71001e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.94001e-06
        [parameter_eliminate]: 2.01998e-06
        [specialize_transform]: 6.87002e-06
        [updatestate_useless_node_eliminater]: 5.81998e-06
        [accelerated_algorithm]: 6.32001e-06
        [meta_shard_fg_expand]: 2.81e-06
        [get_grad_eliminate_]: 5.69999e-06
        [merge_forward]: 4.33999e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.456e-05
        [j_node_and_user_rematch]: 9.24998e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.96e-06
        [inline_without_move]: 6.01e-06
        [renormalize]: 0.00055232
        [add_forward_monad_depend]: 4.91002e-06
        [auto_monad_grad]: 2.06e-06
        [auto_monad_eliminator]: 1.44e-05
        [cse]: 2.562e-05
        [replace_applicator]: 1.189e-05
    [Cycle 2]: 0.0003437, [27]
        [switch_simplify]: 6.36e-06
        [loop_unroll]: 5.86998e-06
        [a_1]: 0.00011071
        [with_stream_mark]: 8.26002e-06
        [recompute_prepare]: 5.66e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.08998e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 5.71998e-06
        [updatestate_useless_node_eliminater]: 5.45001e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 1.21002e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.278e-05
        [j_node_and_user_rematch]: 8.59998e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 9.07001e-06
        [inline_without_move]: 5.36998e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 9.70002e-07
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.66e-06
        [cse]: 1.271e-05
        [replace_applicator]: 5.77999e-06
[py_interpret_to_execute_after_opt_a]: 8.76997e-06
[rewriter_after_opt_a]: 2.985e-05
[convert_after_rewriter]: 6.26e-06
[order_py_execute_after_rewriter]: 4.23999e-06
[mutable_eliminate]: 0.0004676
[jit_opt_b]: 5.218e-05, [1]
    [Cycle 1]: 4.623e-05, [2]
        [frontend_op_eliminate]: 1.798e-05
        [inline_after_opt_a]: 1.707e-05
[cconv]: 1.84e-05
[loop_unroll]: 0.00041099
[jit_opt_after_cconv]: 0.00015297, [1]
    [Cycle 1]: 0.00014676, [11]
        [c_1]: 2.505e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 5.56e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 1.948e-05
        [call_graph_tuple_transform]: 1.765e-05
        [tuple_list_get_item_eliminator]: 6.54001e-06
        [none_parameter_eliminate]: 1.09e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 6.78998e-06
[remove_dup_value]: 8.29002e-06
[partial_unused_args_eliminate]: 1.46998e-06
[environ_conv]: 4.4e-06
[add_recomputation]: 3.234e-05
[cse_after_recomputation]: 2.339e-05, [1]
    [Cycle 1]: 1.789e-05, [1]
        [cse]: 1.222e-05
[auto_monad_reorder]: 1.165e-05
[get_jit_bprop_graph]: 1.42e-06
[rewriter_after_jit_bprop_graph]: 4.49998e-06
[opt_after_jit_grad]: 0.00045677
[symbol_engine_optimizer]: 7.261e-05, [1]
    [Cycle 1]: 6.656e-05, [6]
        [build]: 2.91999e-06
        [elim_shapecalc]: 8.79003e-06
        [elim_not_effective]: 1.264e-05
        [opt_reshape]: 5.84e-06
        [fold_const_symbol]: 8.62e-06
        [renormalize]: 4.39992e-07
[validate]: 2.745e-05
Sums
    bootstrap                            :     0.000251s :  3.04%
    type_inference                       :     0.004439s : 53.66%
    event_method                         :     0.000021s :  0.25%
    auto_monad                           :     0.000063s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.45%
    rewriter_before_opt_a                :     0.000085s :  1.03%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.75%
    jit_opt_a.loop_unroll                :     0.000041s :  0.49%
    jit_opt_a.a_1                        :     0.000755s :  9.12%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000015s :  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 :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  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       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.13%
    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 :  0.33%
    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          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000552s :  6.68%
    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      :     0.000020s :  0.24%
    jit_opt_a.cse                        :     0.000038s :  0.46%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000030s :  0.36%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000468s :  5.65%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000018s :  0.22%
    loop_unroll                          :     0.000411s :  4.97%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.30%
    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              :     0.000019s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  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  :     0.000007s :  0.08%
    remove_dup_value                     :     0.000008s :  0.10%
    partial_unused_args_eliminate        :     0.000001s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000032s :  0.39%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000012s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000457s :  5.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    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                             :     0.000027s :  0.33%


Time group info:
------[substitution.]   0.000188    28
 0.68% :     0.000001s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 1.99% :     0.000004s :      4: substitution.graph_param_transform
82.13% :     0.000155s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.31% :     0.000004s :      4: substitution.replace_old_param
 8.53% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004381     2
78.11% :     0.003422s :      1: type_inference.infer
21.89% :     0.000959s :      1: type_inference.specialize
------[replace.]   0.000080     8
56.38% :     0.000045s :      4: replace.inline
43.62% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     8
91.55% :     0.000152s :      4: match.inline
 8.45% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     13: predicate.addn_check_dump
 1.22% :     0.000002s :     13: predicate.addn_zero_filter
 1.87% :     0.000002s :     13: predicate.arithmetic_simplify
 1.50% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.13% :     0.000001s :     13: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.58% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     13: predicate.environ_get_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.55% :     0.000007s :     29: predicate.inline
 0.93% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      8: predicate.less_batch_normalization
 1.62% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     21: predicate.load_eliminater
 1.41% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     42: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     13: predicate.merge_addn
 1.13% :     0.000002s :     13: predicate.minmaximum_grad
 1.49% :     0.000002s :      4: predicate.mutable_eliminate
 0.39% :     0.000001s :      4: predicate.opt_reshape
 2.79% :     0.000004s :     21: predicate.partial_eliminate
 1.20% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     13: predicate.reduce_eliminate
 1.70% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.95% :     0.000003s :     25: predicate.replace_applicator
 0.59% :     0.000001s :      8: predicate.replace_old_param
 0.32% :     0.000000s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.31% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     13: predicate.same_eliminate
 0.64% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      8: predicate.special_op_eliminate
 1.00% :     0.000001s :      8: predicate.specialize_transform
 1.28% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     21: predicate.switch_defer_inline
 2.36% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.88% :     0.000010s :     67: predicate.switch_simplify
 1.14% :     0.000002s :     13: predicate.tile_eliminate
 1.23% :     0.000002s :     13: predicate.transpose_eliminate
 1.37% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.32% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000561    12
45.89% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.20% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.91% :     0.000274s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013430    72
 0.26% :     0.000035s :      1: add_recomputation
 0.50% :     0.000067s :      1: auto_monad
 0.11% :     0.000014s :      1: auto_monad_reorder
 1.95% :     0.000262s :      1: bootstrap
 0.16% :     0.000021s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.18% :     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
36.91% :     0.004957s :      1: jit_opt_a
 1.16% :     0.000156s :      1: jit_opt_after_cconv
 0.41% :     0.000055s :      1: jit_opt_b
 3.12% :     0.000419s :      1: loop_unroll
 3.54% :     0.000475s :      1: mutable_eliminate
 7.33% :     0.000985s :     26: opt.transform.jit_opt_a
 0.39% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000029s :      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% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.46% :     0.000465s :      1: opt_after_jit_grad
 0.05% :     0.000006s :      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% :     0.000040s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000011s :      1: remove_dup_value
 2.17% :     0.000291s :      1: renormalize.infer
 1.90% :     0.000255s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000033s :      1: rewriter_after_opt_a
 0.66% :     0.000088s :      1: rewriter_before_opt_a
 0.56% :     0.000075s :      1: symbol_engine_optimizer
33.13% :     0.004450s :      1: type_inference


TotalTime = 0.0400622, [33]
[bootstrap]: 0.00025822
[type_inference]: 0.0210332
[event_method]: 1.889e-05
[auto_monad]: 0.00013112
[graph_reusing]: 4.13001e-06
[pre_auto_parallel]: 2.22001e-06
[py_interpret_to_execute]: 0.00011802
[rewriter_before_opt_a]: 8.75e-05
[expand_dump_flag]: 2.19001e-06
[jit_opt_a]: 0.00574071, [2]
    [Cycle 1]: 0.00211071, [27]
        [switch_simplify]: 5.729e-05
        [loop_unroll]: 3.852e-05
        [a_1]: 0.00066294
        [with_stream_mark]: 1.582e-05
        [recompute_prepare]: 1.03e-05
        [updatestate_depend_eliminate]: 5.65001e-06
        [updatestate_assign_eliminate]: 4.72e-06
        [updatestate_loads_eliminate]: 4.36002e-06
        [parameter_eliminate]: 1.14003e-06
        [specialize_transform]: 9.10999e-06
        [updatestate_useless_node_eliminater]: 1.104e-05
        [accelerated_algorithm]: 8.62998e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 8.35001e-06
        [merge_forward]: 3.98999e-06
        [cell_reuse_recompute_pass]: 1.04998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.631e-05
        [j_node_and_user_rematch]: 1.294e-05
        [meta_fg_expand]: 3.17002e-06
        [replace_old_param]: 1.053e-05
        [inline_without_move]: 8.12e-06
        [renormalize]: 0.00097704
        [add_forward_monad_depend]: 4.80001e-06
        [auto_monad_grad]: 1.42e-06
        [auto_monad_eliminator]: 1.774e-05
        [cse]: 2.905e-05
        [replace_applicator]: 1.564e-05
    [Cycle 2]: 0.0004553, [27]
        [switch_simplify]: 9.16998e-06
        [loop_unroll]: 8.07e-06
        [a_1]: 0.00016362
        [with_stream_mark]: 1.09e-05
        [recompute_prepare]: 8.18001e-06
        [updatestate_depend_eliminate]: 4.89e-06
        [updatestate_assign_eliminate]: 4.79002e-06
        [updatestate_loads_eliminate]: 3.79002e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 8.2e-06
        [updatestate_useless_node_eliminater]: 1.077e-05
        [accelerated_algorithm]: 7.83999e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 7.22002e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.57e-05
        [j_node_and_user_rematch]: 1.25e-05
        [meta_fg_expand]: 2.74001e-06
        [replace_old_param]: 9.79999e-06
        [inline_without_move]: 7.36999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.06002e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 1.043e-05
        [cse]: 1.928e-05
        [replace_applicator]: 7.55e-06
[py_interpret_to_execute_after_opt_a]: 1.003e-05
[rewriter_after_opt_a]: 0.00028679
[convert_after_rewriter]: 1.036e-05
[order_py_execute_after_rewriter]: 6.48e-06
[mutable_eliminate]: 0.00052611
[jit_opt_b]: 6.716e-05, [1]
    [Cycle 1]: 6.093e-05, [2]
        [frontend_op_eliminate]: 2.479e-05
        [inline_after_opt_a]: 2.436e-05
[cconv]: 1.945e-05
[loop_unroll]: 0.00046755
[jit_opt_after_cconv]: 0.00018485, [1]
    [Cycle 1]: 0.00017796, [11]
        [c_1]: 4.484e-05
        [parameter_eliminate]: 2.74001e-06
        [updatestate_depend_eliminate]: 7.21999e-06
        [updatestate_assign_eliminate]: 4.15999e-06
        [updatestate_loads_eliminate]: 3.88001e-06
        [cse]: 2.573e-05
        [call_graph_tuple_transform]: 2.047e-05
        [tuple_list_get_item_eliminator]: 8.13999e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 8.42e-06
[remove_dup_value]: 1.364e-05
[partial_unused_args_eliminate]: 1.45999e-06
[environ_conv]: 8.97e-06
[add_recomputation]: 5.221e-05
[cse_after_recomputation]: 2.837e-05, [1]
    [Cycle 1]: 2.268e-05, [1]
        [cse]: 1.698e-05
[auto_monad_reorder]: 2.02e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 4.81002e-06
[opt_after_jit_grad]: 0.0004944
[symbol_engine_optimizer]: 9.722e-05, [1]
    [Cycle 1]: 9.1e-05, [6]
        [build]: 1.038e-05
        [elim_shapecalc]: 1.113e-05
        [elim_not_effective]: 1.818e-05
        [opt_reshape]: 8.45999e-06
        [fold_const_symbol]: 1.317e-05
        [renormalize]: 4.89992e-07
[validate]: 4.594e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.0100938
[execute]: 8.75999e-06
Sums
    bootstrap                            :     0.000258s :  0.71%
    type_inference                       :     0.021033s : 57.99%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000131s :  0.36%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000118s :  0.33%
    rewriter_before_opt_a                :     0.000087s :  0.24%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.18%
    jit_opt_a.loop_unroll                :     0.000047s :  0.13%
    jit_opt_a.a_1                        :     0.000827s :  2.28%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  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 :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000025s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.000977s :  2.69%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.08%
    jit_opt_a.cse                        :     0.000048s :  0.13%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000287s :  0.79%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000526s :  1.45%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000019s :  0.05%
    loop_unroll                          :     0.000468s :  1.29%
    jit_opt_after_cconv.c_1              :     0.000045s :  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              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000001s :  0.00%
    environ_conv                         :     0.000009s :  0.02%
    add_recomputation                    :     0.000052s :  0.14%
    cse_after_recomputation.cse          :     0.000017s :  0.05%
    auto_monad_reorder                   :     0.000020s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000494s :  1.36%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000046s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010094s : 27.83%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000232    47
 3.47% :     0.000008s :      2: substitution.depend_value_elim
 1.02% :     0.000002s :      4: substitution.elim_not_effective
 0.79% :     0.000002s :      4: substitution.fold_const_symbol
 2.25% :     0.000005s :      5: substitution.graph_param_transform
75.84% :     0.000176s :      4: substitution.inline
 1.85% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.74% :     0.000006s :      8: substitution.remove_not_recompute_node
 1.40% :     0.000003s :      2: substitution.replace_old_param
 3.98% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
 3.26% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.40% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.020976     2
93.59% :     0.019631s :      1: type_inference.infer
 6.41% :     0.001344s :      1: type_inference.specialize
------[replace.]   0.000090     6
66.12% :     0.000060s :      4: replace.inline
33.88% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
95.50% :     0.000173s :      4: match.inline
 4.50% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1021
 1.31% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     15: predicate.addn_check_dump
 1.34% :     0.000002s :     15: predicate.addn_zero_filter
 2.10% :     0.000003s :     15: predicate.arithmetic_simplify
 1.42% :     0.000002s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.21% :     0.000002s :     15: predicate.compare_switch_simplify
 1.42% :     0.000002s :     15: predicate.depend_value_elim
 1.20% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.93% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.49% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     15: predicate.environ_get_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.30% :     0.000001s :      5: predicate.graph_param_transform
 5.04% :     0.000008s :     31: predicate.inline
 1.11% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :     10: predicate.less_batch_normalization
 1.51% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.92% :     0.000003s :     22: predicate.load_eliminater
 1.58% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.74% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.15% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.57% :     0.000003s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.45% :     0.000004s :     22: predicate.partial_eliminate
 1.21% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     15: predicate.reduce_eliminate
 1.47% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.91% :     0.000003s :     27: predicate.replace_applicator
 0.60% :     0.000001s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.32% :     0.000002s :     15: predicate.reshape_eliminate
 1.24% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.12% :     0.000002s :     10: predicate.special_op_eliminate
 1.15% :     0.000002s :     10: predicate.specialize_transform
 1.56% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.15% :     0.000004s :     21: predicate.switch_defer_inline
 1.98% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.65% :     0.000011s :     68: predicate.switch_simplify
 1.31% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.64% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.47% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.80% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.24% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000772    13
44.06% :     0.000340s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.97% :     0.000393s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042345    76
 0.13% :     0.000055s :      1: add_recomputation
 0.32% :     0.000136s :      1: auto_monad
 0.05% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000277s :      1: bootstrap
 0.05% :     0.000022s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.03% :     0.000014s :      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
13.56% :     0.005744s :      1: jit_opt_a
 0.44% :     0.000188s :      1: jit_opt_after_cconv
 0.17% :     0.000070s :      1: jit_opt_b
 1.13% :     0.000477s :      1: loop_unroll
 1.26% :     0.000534s :      1: mutable_eliminate
 2.65% :     0.001121s :     26: opt.transform.jit_opt_a
 0.18% :     0.000078s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000047s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000503s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000122s :      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
 1.34% :     0.000566s :      1: renormalize.infer
 0.95% :     0.000404s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.69% :     0.000291s :      1: rewriter_after_opt_a
 0.22% :     0.000092s :      1: rewriter_before_opt_a
 0.24% :     0.000100s :      1: symbol_engine_optimizer
23.87% :     0.010109s :      1: task_emit
49.70% :     0.021046s :      1: type_inference
 0.16% :     0.000066s :      1: validate


TotalTime = 0.0115716, [30]
[bootstrap]: 0.00025744
[type_inference]: 0.00439594
[event_method]: 2.08e-05
[auto_monad]: 4.397e-05
[graph_reusing]: 4.18999e-06
[pre_auto_parallel]: 1.97001e-06
[py_interpret_to_execute]: 3.292e-05
[rewriter_before_opt_a]: 8.056e-05
[expand_dump_flag]: 1.76e-06
[jit_opt_a]: 0.00477328, [2]
    [Cycle 1]: 0.00149629, [27]
        [switch_simplify]: 4.98e-05
        [loop_unroll]: 3.511e-05
        [a_1]: 0.00056475
        [with_stream_mark]: 1.171e-05
        [recompute_prepare]: 8.07998e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 6.76e-06
        [updatestate_useless_node_eliminater]: 5.77999e-06
        [accelerated_algorithm]: 6.49001e-06
        [meta_shard_fg_expand]: 1.37999e-06
        [get_grad_eliminate_]: 5.95002e-06
        [merge_forward]: 3.08998e-06
        [cell_reuse_recompute_pass]: 7.99977e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.16e-05
        [j_node_and_user_rematch]: 8.98002e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 9.51e-06
        [inline_without_move]: 5.89999e-06
        [renormalize]: 0.00053919
        [add_forward_monad_depend]: 4.50999e-06
        [auto_monad_grad]: 1.33002e-06
        [auto_monad_eliminator]: 1.039e-05
        [cse]: 2.158e-05
        [replace_applicator]: 1.266e-05
    [Cycle 2]: 0.00035248, [27]
        [switch_simplify]: 6.47001e-06
        [loop_unroll]: 5.54998e-06
        [a_1]: 0.00011202
        [with_stream_mark]: 8.15999e-06
        [recompute_prepare]: 5.72999e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 5.50001e-06
        [updatestate_useless_node_eliminater]: 5.48002e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.297e-05
        [j_node_and_user_rematch]: 8.07e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 9.32001e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 6.11e-06
        [cse]: 1.342e-05
        [replace_applicator]: 5.66e-06
[py_interpret_to_execute_after_opt_a]: 8.74e-06
[rewriter_after_opt_a]: 2.707e-05
[convert_after_rewriter]: 5.84e-06
[order_py_execute_after_rewriter]: 4.09002e-06
[mutable_eliminate]: 0.00050113
[jit_opt_b]: 5.612e-05, [1]
    [Cycle 1]: 4.995e-05, [2]
        [frontend_op_eliminate]: 2.088e-05
        [inline_after_opt_a]: 1.714e-05
[cconv]: 1.554e-05
[loop_unroll]: 0.00041211
[jit_opt_after_cconv]: 0.00014379, [1]
    [Cycle 1]: 0.00013773, [11]
        [c_1]: 2.399e-05
        [parameter_eliminate]: 2.72001e-06
        [updatestate_depend_eliminate]: 5.56e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [cse]: 1.958e-05
        [call_graph_tuple_transform]: 1.826e-05
        [tuple_list_get_item_eliminator]: 6.17001e-06
        [none_parameter_eliminate]: 1.04e-06
        [renormalize]: 3.9002e-07
        [switch_simplify]: 5.96e-06
[remove_dup_value]: 8.46002e-06
[partial_unused_args_eliminate]: 1.72001e-06
[environ_conv]: 4.05e-06
[add_recomputation]: 3.051e-05
[cse_after_recomputation]: 2.215e-05, [1]
    [Cycle 1]: 1.726e-05, [1]
        [cse]: 1.157e-05
[auto_monad_reorder]: 1.207e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.63001e-06
[opt_after_jit_grad]: 0.00045088
[symbol_engine_optimizer]: 7.263e-05, [1]
    [Cycle 1]: 6.677e-05, [6]
        [build]: 2.63e-06
        [elim_shapecalc]: 8.13999e-06
        [elim_not_effective]: 1.257e-05
        [opt_reshape]: 6.47001e-06
        [fold_const_symbol]: 8.83001e-06
        [renormalize]: 6.00005e-07
[validate]: 2.92e-05
Sums
    bootstrap                            :     0.000257s :  3.17%
    type_inference                       :     0.004396s : 54.21%
    event_method                         :     0.000021s :  0.26%
    auto_monad                           :     0.000044s :  0.54%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.41%
    rewriter_before_opt_a                :     0.000081s :  0.99%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.69%
    jit_opt_a.loop_unroll                :     0.000041s :  0.50%
    jit_opt_a.a_1                        :     0.000677s :  8.35%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000014s :  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 :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    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 :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    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 :  0.02%
    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 :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000539s :  6.65%
    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      :     0.000016s :  0.20%
    jit_opt_a.cse                        :     0.000035s :  0.43%
    jit_opt_a.replace_applicator         :     0.000018s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000027s :  0.33%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000501s :  6.18%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.26%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000016s :  0.19%
    loop_unroll                          :     0.000412s :  5.08%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.30%
    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              :     0.000020s :  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 :  0.08%
    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 :  0.07%
    remove_dup_value                     :     0.000008s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000031s :  0.38%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    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                   :     0.000451s :  5.56%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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                             :     0.000029s :  0.36%


Time group info:
------[substitution.]   0.000164    28
 0.83% :     0.000001s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.42% :     0.000004s :      4: substitution.graph_param_transform
82.52% :     0.000135s :      4: substitution.inline
 1.49% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.41% :     0.000004s :      4: substitution.replace_old_param
 7.22% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004346     2
80.84% :     0.003513s :      1: type_inference.infer
19.16% :     0.000832s :      1: type_inference.specialize
------[replace.]   0.000078     8
55.63% :     0.000043s :      4: replace.inline
44.37% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000143     8
93.12% :     0.000133s :      4: match.inline
 6.88% :     0.000010s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     13: predicate.addn_check_dump
 1.20% :     0.000002s :     13: predicate.addn_zero_filter
 1.79% :     0.000002s :     13: predicate.arithmetic_simplify
 1.35% :     0.000002s :     13: predicate.cast_eliminate
 0.47% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     13: predicate.compare_switch_simplify
 1.15% :     0.000002s :     13: predicate.depend_value_elim
 1.21% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      4: predicate.elim_not_effective
 0.56% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     13: predicate.environ_get_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      4: predicate.fold_const_symbol
 1.04% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.08% :     0.000007s :     29: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      8: predicate.less_batch_normalization
 1.60% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.03% :     0.000003s :     21: predicate.load_eliminater
 1.46% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.24% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.63% :     0.000002s :      4: predicate.mutable_eliminate
 0.57% :     0.000001s :      4: predicate.opt_reshape
 2.54% :     0.000003s :     21: predicate.partial_eliminate
 1.26% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     13: predicate.reduce_eliminate
 1.67% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.93% :     0.000003s :     25: predicate.replace_applicator
 0.65% :     0.000001s :      8: predicate.replace_old_param
 0.31% :     0.000000s :      4: predicate.reset_defer_inline
 1.19% :     0.000002s :     13: predicate.reshape_eliminate
 1.21% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     13: predicate.same_eliminate
 0.57% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      8: predicate.special_op_eliminate
 1.02% :     0.000001s :      8: predicate.specialize_transform
 1.31% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     21: predicate.switch_defer_inline
 2.27% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.65% :     0.000010s :     67: predicate.switch_simplify
 1.15% :     0.000002s :     13: predicate.tile_eliminate
 1.29% :     0.000002s :     13: predicate.transpose_eliminate
 1.38% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.73% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.92% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000565    12
48.97% :     0.000277s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.75% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.28% :     0.000262s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013101    72
 0.25% :     0.000033s :      1: add_recomputation
 0.36% :     0.000048s :      1: auto_monad
 0.11% :     0.000015s :      1: auto_monad_reorder
 2.06% :     0.000270s :      1: bootstrap
 0.14% :     0.000018s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.19% :     0.000024s :      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
36.45% :     0.004776s :      1: jit_opt_a
 1.12% :     0.000147s :      1: jit_opt_after_cconv
 0.45% :     0.000059s :      1: jit_opt_b
 3.20% :     0.000419s :      1: loop_unroll
 3.89% :     0.000509s :      1: mutable_eliminate
 6.88% :     0.000901s :     26: opt.transform.jit_opt_a
 0.39% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.25% :     0.000032s :      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
 0.18% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.49% :     0.000458s :      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
 0.27% :     0.000036s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000011s :      1: remove_dup_value
 2.10% :     0.000275s :      1: renormalize.infer
 1.97% :     0.000258s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000030s :      1: rewriter_after_opt_a
 0.64% :     0.000083s :      1: rewriter_before_opt_a
 0.57% :     0.000075s :      1: symbol_engine_optimizer
33.63% :     0.004405s :      1: type_inference


TotalTime = 0.0110559, [30]
[bootstrap]: 0.00024625
[type_inference]: 0.00413397
[event_method]: 2.06e-05
[auto_monad]: 4.183e-05
[graph_reusing]: 3.91999e-06
[pre_auto_parallel]: 1.71e-06
[py_interpret_to_execute]: 3.179e-05
[rewriter_before_opt_a]: 7.934e-05
[expand_dump_flag]: 1.97999e-06
[jit_opt_a]: 0.00470832, [2]
    [Cycle 1]: 0.00147551, [27]
        [switch_simplify]: 4.986e-05
        [loop_unroll]: 3.709e-05
        [a_1]: 0.00055717
        [with_stream_mark]: 1.218e-05
        [recompute_prepare]: 7.61999e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 7.11001e-06
        [updatestate_useless_node_eliminater]: 6.26998e-06
        [accelerated_algorithm]: 6.24999e-06
        [meta_shard_fg_expand]: 1.37999e-06
        [get_grad_eliminate_]: 5.91998e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.138e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 1.045e-05
        [inline_without_move]: 6.33998e-06
        [renormalize]: 0.00053611
        [add_forward_monad_depend]: 3.13e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 8.89e-06
        [cse]: 1.767e-05
        [replace_applicator]: 1.038e-05
    [Cycle 2]: 0.00033892, [27]
        [switch_simplify]: 6.45997e-06
        [loop_unroll]: 5.57001e-06
        [a_1]: 0.00011107
        [with_stream_mark]: 6.88e-06
        [recompute_prepare]: 5.53997e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 7.29982e-07
        [specialize_transform]: 5.72999e-06
        [updatestate_useless_node_eliminater]: 5.49e-06
        [accelerated_algorithm]: 5.86998e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.122e-05
        [j_node_and_user_rematch]: 8.18999e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 8.93002e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 8.70001e-07
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.15999e-06
        [cse]: 1.179e-05
        [replace_applicator]: 5.42999e-06
[py_interpret_to_execute_after_opt_a]: 7.45e-06
[rewriter_after_opt_a]: 2.75e-05
[convert_after_rewriter]: 5.34e-06
[order_py_execute_after_rewriter]: 4.05e-06
[mutable_eliminate]: 0.00041974
[jit_opt_b]: 5.116e-05, [1]
    [Cycle 1]: 4.612e-05, [2]
        [frontend_op_eliminate]: 1.692e-05
        [inline_after_opt_a]: 1.863e-05
[cconv]: 1.055e-05
[loop_unroll]: 0.00038461
[jit_opt_after_cconv]: 0.00013594, [1]
    [Cycle 1]: 0.00012997, [11]
        [c_1]: 2.486e-05
        [parameter_eliminate]: 1.42999e-06
        [updatestate_depend_eliminate]: 4.59002e-06
        [updatestate_assign_eliminate]: 2.38998e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [cse]: 1.673e-05
        [call_graph_tuple_transform]: 1.598e-05
        [tuple_list_get_item_eliminator]: 6.11e-06
        [none_parameter_eliminate]: 6.60017e-07
        [renormalize]: 3.30008e-07
        [switch_simplify]: 5.89999e-06
[remove_dup_value]: 6.15002e-06
[partial_unused_args_eliminate]: 1.25001e-06
[environ_conv]: 3.4e-06
[add_recomputation]: 3.047e-05
[cse_after_recomputation]: 2.251e-05, [1]
    [Cycle 1]: 1.756e-05, [1]
        [cse]: 1.214e-05
[auto_monad_reorder]: 1.103e-05
[get_jit_bprop_graph]: 1.22e-06
[rewriter_after_jit_bprop_graph]: 3.73001e-06
[opt_after_jit_grad]: 0.00042774
[symbol_engine_optimizer]: 6.779e-05, [1]
    [Cycle 1]: 6.253e-05, [6]
        [build]: 2.17001e-06
        [elim_shapecalc]: 7.73999e-06
        [elim_not_effective]: 1.136e-05
        [opt_reshape]: 5.90002e-06
        [fold_const_symbol]: 8.52e-06
        [renormalize]: 3.89991e-07
[validate]: 2.564e-05
Sums
    bootstrap                            :     0.000246s :  3.22%
    type_inference                       :     0.004134s : 54.07%
    event_method                         :     0.000021s :  0.27%
    auto_monad                           :     0.000042s :  0.55%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.42%
    rewriter_before_opt_a                :     0.000079s :  1.04%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000056s :  0.74%
    jit_opt_a.loop_unroll                :     0.000043s :  0.56%
    jit_opt_a.a_1                        :     0.000668s :  8.74%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.25%
    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        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.17%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.16%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    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    :     0.000018s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.25%
    jit_opt_a.inline_without_move        :     0.000012s :  0.15%
    jit_opt_a.renormalize                :     0.000536s :  7.01%
    jit_opt_a.add_forward_monad_depend   :     0.000004s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.18%
    jit_opt_a.cse                        :     0.000029s :  0.39%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.10%
    rewriter_after_opt_a                 :     0.000028s :  0.36%
    convert_after_rewriter               :     0.000005s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000420s :  5.49%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.24%
    cconv                                :     0.000011s :  0.14%
    loop_unroll                          :     0.000385s :  5.03%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.33%
    jit_opt_after_cconv.parameter_eliminate :     0.000001s :  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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.21%
    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 :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000006s :  0.08%
    partial_unused_args_eliminate        :     0.000001s :  0.02%
    environ_conv                         :     0.000003s :  0.04%
    add_recomputation                    :     0.000030s :  0.40%
    cse_after_recomputation.cse          :     0.000012s :  0.16%
    auto_monad_reorder                   :     0.000011s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000428s :  5.59%
    symbol_engine_optimizer.build        :     0.000002s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.15%
    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                             :     0.000026s :  0.34%


Time group info:
------[substitution.]   0.000159    28
 0.78% :     0.000001s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.15% :     0.000003s :      4: substitution.graph_param_transform
82.20% :     0.000130s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.67% :     0.000004s :      4: substitution.replace_old_param
 7.68% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004083     2
80.67% :     0.003294s :      1: type_inference.infer
19.33% :     0.000789s :      1: type_inference.specialize
------[replace.]   0.000073     8
54.62% :     0.000040s :      4: replace.inline
45.38% :     0.000033s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000138     8
92.61% :     0.000128s :      4: match.inline
 7.39% :     0.000010s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.01% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     13: predicate.addn_check_dump
 1.24% :     0.000002s :     13: predicate.addn_zero_filter
 1.86% :     0.000002s :     13: predicate.arithmetic_simplify
 1.27% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.26% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      4: predicate.elim_not_effective
 0.55% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     13: predicate.environ_get_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.98% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.22% :     0.000007s :     29: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      8: predicate.less_batch_normalization
 1.68% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.01% :     0.000003s :     21: predicate.load_eliminater
 1.20% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.54% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     13: predicate.merge_addn
 1.14% :     0.000001s :     13: predicate.minmaximum_grad
 1.35% :     0.000002s :      4: predicate.mutable_eliminate
 0.46% :     0.000001s :      4: predicate.opt_reshape
 2.49% :     0.000003s :     21: predicate.partial_eliminate
 1.24% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     13: predicate.reduce_eliminate
 1.78% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.95% :     0.000003s :     25: predicate.replace_applicator
 0.59% :     0.000001s :      8: predicate.replace_old_param
 0.30% :     0.000000s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.21% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     13: predicate.same_eliminate
 0.57% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      8: predicate.special_op_eliminate
 1.07% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     21: predicate.switch_defer_inline
 2.29% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.90% :     0.000010s :     67: predicate.switch_simplify
 1.17% :     0.000002s :     13: predicate.tile_eliminate
 1.24% :     0.000002s :     13: predicate.transpose_eliminate
 1.48% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.48% :     0.000004s :     25: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 2.02% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     13: predicate.value_based_eliminate
 0.36% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000513    12
47.12% :     0.000242s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.08% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.80% :     0.000245s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012571    72
 0.26% :     0.000033s :      1: add_recomputation
 0.36% :     0.000045s :      1: auto_monad
 0.11% :     0.000013s :      1: auto_monad_reorder
 2.05% :     0.000257s :      1: bootstrap
 0.10% :     0.000013s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.20% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.19% :     0.000024s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000003s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
37.48% :     0.004711s :      1: jit_opt_a
 1.10% :     0.000138s :      1: jit_opt_after_cconv
 0.43% :     0.000054s :      1: jit_opt_b
 3.11% :     0.000391s :      1: loop_unroll
 3.39% :     0.000426s :      1: mutable_eliminate
 7.10% :     0.000893s :     26: opt.transform.jit_opt_a
 0.40% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.24% :     0.000030s :      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
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.45% :     0.000434s :      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
 0.27% :     0.000035s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000008s :      1: remove_dup_value
 2.29% :     0.000287s :      1: renormalize.infer
 1.94% :     0.000244s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000030s :      1: rewriter_after_opt_a
 0.65% :     0.000082s :      1: rewriter_before_opt_a
 0.56% :     0.000070s :      1: symbol_engine_optimizer
32.97% :     0.004145s :      1: type_inference


TotalTime = 0.0401376, [33]
[bootstrap]: 0.00025563
[type_inference]: 0.0212634
[event_method]: 2.007e-05
[auto_monad]: 0.00014618
[graph_reusing]: 6.14001e-06
[pre_auto_parallel]: 2.91999e-06
[py_interpret_to_execute]: 0.00012261
[rewriter_before_opt_a]: 9.728e-05
[expand_dump_flag]: 3.88999e-06
[jit_opt_a]: 0.00572554, [2]
    [Cycle 1]: 0.00209594, [27]
        [switch_simplify]: 6.003e-05
        [loop_unroll]: 3.767e-05
        [a_1]: 0.00066597
        [with_stream_mark]: 1.613e-05
        [recompute_prepare]: 1.071e-05
        [updatestate_depend_eliminate]: 6.16e-06
        [updatestate_assign_eliminate]: 5.59998e-06
        [updatestate_loads_eliminate]: 4.67e-06
        [parameter_eliminate]: 2.22001e-06
        [specialize_transform]: 9.17001e-06
        [updatestate_useless_node_eliminater]: 1.138e-05
        [accelerated_algorithm]: 8.89e-06
        [meta_shard_fg_expand]: 2.76e-06
        [get_grad_eliminate_]: 8.25e-06
        [merge_forward]: 5.42999e-06
        [cell_reuse_recompute_pass]: 1.39998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.89e-05
        [j_node_and_user_rematch]: 1.348e-05
        [meta_fg_expand]: 3.55003e-06
        [replace_old_param]: 1.074e-05
        [inline_without_move]: 8.07e-06
        [renormalize]: 0.00093104
        [add_forward_monad_depend]: 5.11997e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 2.088e-05
        [cse]: 3.662e-05
        [replace_applicator]: 1.562e-05
    [Cycle 2]: 0.00045737, [27]
        [switch_simplify]: 8.99998e-06
        [loop_unroll]: 7.61001e-06
        [a_1]: 0.00016622
        [with_stream_mark]: 1.099e-05
        [recompute_prepare]: 7.80998e-06
        [updatestate_depend_eliminate]: 4.94e-06
        [updatestate_assign_eliminate]: 4.28001e-06
        [updatestate_loads_eliminate]: 3.74002e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 8.13999e-06
        [updatestate_useless_node_eliminater]: 1.04e-05
        [accelerated_algorithm]: 7.92e-06
        [meta_shard_fg_expand]: 1.65001e-06
        [get_grad_eliminate_]: 7.75998e-06
        [merge_forward]: 4.08999e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.559e-05
        [j_node_and_user_rematch]: 1.254e-05
        [meta_fg_expand]: 2.65002e-06
        [replace_old_param]: 9.84001e-06
        [inline_without_move]: 7.53e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 9.61e-06
        [cse]: 1.924e-05
        [replace_applicator]: 7.77002e-06
[py_interpret_to_execute_after_opt_a]: 1.064e-05
[rewriter_after_opt_a]: 0.00027282
[convert_after_rewriter]: 1.089e-05
[order_py_execute_after_rewriter]: 6.83e-06
[mutable_eliminate]: 0.0005107
[jit_opt_b]: 6.659e-05, [1]
    [Cycle 1]: 6e-05, [2]
        [frontend_op_eliminate]: 2.458e-05
        [inline_after_opt_a]: 2.365e-05
[cconv]: 2.418e-05
[loop_unroll]: 0.00046067
[jit_opt_after_cconv]: 0.00018615, [1]
    [Cycle 1]: 0.00017962, [11]
        [c_1]: 4.291e-05
        [parameter_eliminate]: 2.12001e-06
        [updatestate_depend_eliminate]: 7.16001e-06
        [updatestate_assign_eliminate]: 4.22e-06
        [updatestate_loads_eliminate]: 3.83001e-06
        [cse]: 2.563e-05
        [call_graph_tuple_transform]: 2.245e-05
        [tuple_list_get_item_eliminator]: 8.48001e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 9.01998e-06
[remove_dup_value]: 1.875e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 9.20001e-06
[add_recomputation]: 6.353e-05
[cse_after_recomputation]: 3.06e-05, [1]
    [Cycle 1]: 2.439e-05, [1]
        [cse]: 1.829e-05
[auto_monad_reorder]: 2.599e-05
[get_jit_bprop_graph]: 1.59998e-06
[rewriter_after_jit_bprop_graph]: 4.65999e-06
[opt_after_jit_grad]: 0.00048474
[symbol_engine_optimizer]: 9.729e-05, [1]
    [Cycle 1]: 9.09e-05, [6]
        [build]: 8.87999e-06
        [elim_shapecalc]: 1.147e-05
        [elim_not_effective]: 1.75e-05
        [opt_reshape]: 8.94e-06
        [fold_const_symbol]: 1.402e-05
        [renormalize]: 4.19997e-07
[validate]: 4.856e-05
[backend_pass]: 1.05001e-06
[task_emit]: 0.00994106
[execute]: 7.55998e-06
Sums
    bootstrap                            :     0.000256s :  0.70%
    type_inference                       :     0.021263s : 58.50%
    event_method                         :     0.000020s :  0.06%
    auto_monad                           :     0.000146s :  0.40%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000123s :  0.34%
    rewriter_before_opt_a                :     0.000097s :  0.27%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000832s :  2.29%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.000931s :  2.56%
    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      :     0.000030s :  0.08%
    jit_opt_a.cse                        :     0.000056s :  0.15%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000273s :  0.75%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000511s :  1.41%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000024s :  0.07%
    loop_unroll                          :     0.000461s :  1.27%
    jit_opt_after_cconv.c_1              :     0.000043s :  0.12%
    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              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000009s :  0.03%
    add_recomputation                    :     0.000064s :  0.17%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000485s :  1.33%
    symbol_engine_optimizer.build        :     0.000009s :  0.02%
    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 :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000049s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009941s : 27.35%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000240    47
 3.41% :     0.000008s :      2: substitution.depend_value_elim
 1.09% :     0.000003s :      4: substitution.elim_not_effective
 0.86% :     0.000002s :      4: substitution.fold_const_symbol
 2.99% :     0.000007s :      5: substitution.graph_param_transform
74.13% :     0.000178s :      4: substitution.inline
 1.84% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.87% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.48% :     0.000004s :      2: substitution.replace_old_param
 4.91% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.19% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.22% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021201     2
93.34% :     0.019788s :      1: type_inference.infer
 6.66% :     0.001413s :      1: type_inference.specialize
------[replace.]   0.000085     6
65.87% :     0.000056s :      4: replace.inline
34.13% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
94.24% :     0.000174s :      4: match.inline
 5.76% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1021
 1.33% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.30% :     0.000002s :     15: predicate.addn_zero_filter
 2.02% :     0.000003s :     15: predicate.arithmetic_simplify
 1.40% :     0.000002s :     15: predicate.cast_eliminate
 0.53% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.21% :     0.000002s :     15: predicate.compare_switch_simplify
 1.37% :     0.000002s :     15: predicate.depend_value_elim
 1.23% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.58% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     15: predicate.environ_get_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      5: predicate.fold_const_symbol
 1.04% :     0.000002s :     10: predicate.get_grad_eliminate
 0.30% :     0.000001s :      5: predicate.graph_param_transform
 5.13% :     0.000009s :     31: predicate.inline
 1.10% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :     10: predicate.less_batch_normalization
 1.45% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.87% :     0.000003s :     22: predicate.load_eliminater
 1.33% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.65% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     15: predicate.merge_addn
 1.16% :     0.000002s :     15: predicate.minmaximum_grad
 1.66% :     0.000003s :      5: predicate.mutable_eliminate
 0.55% :     0.000001s :      5: predicate.opt_reshape
 2.44% :     0.000004s :     22: predicate.partial_eliminate
 1.19% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.86% :     0.000003s :     15: predicate.reduce_eliminate
 1.46% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.77% :     0.000003s :     27: predicate.replace_applicator
 0.59% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.29% :     0.000002s :     15: predicate.reshape_eliminate
 1.25% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.15% :     0.000002s :     10: predicate.special_op_eliminate
 1.15% :     0.000002s :     10: predicate.specialize_transform
 1.45% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.10% :     0.000003s :     21: predicate.switch_defer_inline
 2.01% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.81% :     0.000011s :     68: predicate.switch_simplify
 1.28% :     0.000002s :     15: predicate.tile_eliminate
 1.31% :     0.000002s :     15: predicate.transpose_eliminate
 1.65% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.46% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.95% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.25% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     15: predicate.value_based_eliminate
 0.42% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000762    13
44.86% :     0.000342s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.68% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.46% :     0.000385s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042382    76
 0.16% :     0.000067s :      1: add_recomputation
 0.36% :     0.000151s :      1: auto_monad
 0.07% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.64% :     0.000270s :      1: bootstrap
 0.06% :     0.000027s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.08% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.06% :     0.000024s :      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% :     0.000009s :      1: graph_reusing
13.52% :     0.005728s :      1: jit_opt_a
 0.45% :     0.000189s :      1: jit_opt_after_cconv
 0.16% :     0.000069s :      1: jit_opt_b
 1.11% :     0.000469s :      1: loop_unroll
 1.23% :     0.000520s :      1: mutable_eliminate
 2.66% :     0.001129s :     26: opt.transform.jit_opt_a
 0.19% :     0.000079s :      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% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000048s :      4: opt.transform.symbol_engine_opt
 1.16% :     0.000494s :      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
 0.30% :     0.000127s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.22% :     0.000518s :      1: renormalize.infer
 0.96% :     0.000405s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.65% :     0.000277s :      1: rewriter_after_opt_a
 0.24% :     0.000102s :      1: rewriter_before_opt_a
 0.24% :     0.000100s :      1: symbol_engine_optimizer
23.48% :     0.009951s :      1: task_emit
50.21% :     0.021278s :      1: type_inference
 0.17% :     0.000070s :      1: validate


TotalTime = 0.0117177, [30]
[bootstrap]: 0.00028323
[type_inference]: 0.00431791
[event_method]: 2.008e-05
[auto_monad]: 5.471e-05
[graph_reusing]: 5.22999e-06
[pre_auto_parallel]: 2.13002e-06
[py_interpret_to_execute]: 3.195e-05
[rewriter_before_opt_a]: 8.249e-05
[expand_dump_flag]: 2.53003e-06
[jit_opt_a]: 0.00494195, [2]
    [Cycle 1]: 0.00157916, [27]
        [switch_simplify]: 5.14e-05
        [loop_unroll]: 3.477e-05
        [a_1]: 0.00059349
        [with_stream_mark]: 1.399e-05
        [recompute_prepare]: 7.73001e-06
        [updatestate_depend_eliminate]: 3.31999e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 6.95998e-06
        [updatestate_useless_node_eliminater]: 5.89999e-06
        [accelerated_algorithm]: 6.24001e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 6.04001e-06
        [merge_forward]: 2.86e-06
        [cell_reuse_recompute_pass]: 9.29984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.183e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 1.092e-05
        [inline_without_move]: 6.14001e-06
        [renormalize]: 0.00058724
        [add_forward_monad_depend]: 4.43999e-06
        [auto_monad_grad]: 1.73997e-06
        [auto_monad_eliminator]: 1.243e-05
        [cse]: 2.187e-05
        [replace_applicator]: 1.252e-05
    [Cycle 2]: 0.000349, [27]
        [switch_simplify]: 6.41998e-06
        [loop_unroll]: 5.56e-06
        [a_1]: 0.00011068
        [with_stream_mark]: 8.77e-06
        [recompute_prepare]: 5.74e-06
        [updatestate_depend_eliminate]: 3.05002e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 5.62001e-06
        [updatestate_useless_node_eliminater]: 5.55001e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.337e-05
        [j_node_and_user_rematch]: 8.48001e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 8.69e-06
        [inline_without_move]: 5.42999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.02e-06
        [auto_monad_grad]: 8.40024e-07
        [auto_monad_eliminator]: 5.89e-06
        [cse]: 1.333e-05
        [replace_applicator]: 5.72001e-06
[py_interpret_to_execute_after_opt_a]: 8.74e-06
[rewriter_after_opt_a]: 3.111e-05
[convert_after_rewriter]: 6.43998e-06
[order_py_execute_after_rewriter]: 4.62e-06
[mutable_eliminate]: 0.0004739
[jit_opt_b]: 5.45e-05, [1]
    [Cycle 1]: 4.836e-05, [2]
        [frontend_op_eliminate]: 1.824e-05
        [inline_after_opt_a]: 1.833e-05
[cconv]: 1.899e-05
[loop_unroll]: 0.00041622
[jit_opt_after_cconv]: 0.00014744, [1]
    [Cycle 1]: 0.00014124, [11]
        [c_1]: 2.419e-05
        [parameter_eliminate]: 2.71999e-06
        [updatestate_depend_eliminate]: 5.96e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [cse]: 1.965e-05
        [call_graph_tuple_transform]: 1.995e-05
        [tuple_list_get_item_eliminator]: 5.86e-06
        [none_parameter_eliminate]: 1.15001e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.84e-06
[remove_dup_value]: 1.08e-05
[partial_unused_args_eliminate]: 1.88002e-06
[environ_conv]: 5.00999e-06
[add_recomputation]: 3.621e-05
[cse_after_recomputation]: 2.281e-05, [1]
    [Cycle 1]: 1.769e-05, [1]
        [cse]: 1.195e-05
[auto_monad_reorder]: 2.229e-05
[get_jit_bprop_graph]: 1.67001e-06
[rewriter_after_jit_bprop_graph]: 4.48999e-06
[opt_after_jit_grad]: 0.00045778
[symbol_engine_optimizer]: 7.192e-05, [1]
    [Cycle 1]: 6.556e-05, [6]
        [build]: 2.86e-06
        [elim_shapecalc]: 8.15e-06
        [elim_not_effective]: 1.284e-05
        [opt_reshape]: 5.97999e-06
        [fold_const_symbol]: 8.45001e-06
        [renormalize]: 3.69997e-07
[validate]: 3.064e-05
Sums
    bootstrap                            :     0.000283s :  3.47%
    type_inference                       :     0.004318s : 52.87%
    event_method                         :     0.000020s :  0.25%
    auto_monad                           :     0.000055s :  0.67%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.39%
    rewriter_before_opt_a                :     0.000082s :  1.01%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000058s :  0.71%
    jit_opt_a.loop_unroll                :     0.000040s :  0.49%
    jit_opt_a.a_1                        :     0.000704s :  8.62%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.28%
    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        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  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 :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  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 :  0.31%
    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          :     0.000020s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000587s :  7.19%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.22%
    jit_opt_a.cse                        :     0.000035s :  0.43%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000031s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000474s :  5.80%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.22%
    cconv                                :     0.000019s :  0.23%
    loop_unroll                          :     0.000416s :  5.10%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.30%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.24%
    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 :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000011s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000036s :  0.44%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000022s :  0.27%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000458s :  5.60%
    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 :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.38%


Time group info:
------[substitution.]   0.000195    28
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000005s :      4: substitution.graph_param_transform
83.12% :     0.000162s :      4: substitution.inline
 1.25% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.67% :     0.000005s :      4: substitution.replace_old_param
 6.99% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004264     2
78.59% :     0.003351s :      1: type_inference.infer
21.41% :     0.000913s :      1: type_inference.specialize
------[replace.]   0.000079     8
54.53% :     0.000043s :      4: replace.inline
45.47% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000170     8
93.28% :     0.000159s :      4: match.inline
 6.72% :     0.000011s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   912
 1.27% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     13: predicate.addn_check_dump
 1.47% :     0.000002s :     13: predicate.addn_zero_filter
 1.78% :     0.000002s :     13: predicate.arithmetic_simplify
 1.39% :     0.000002s :     13: predicate.cast_eliminate
 0.53% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     13: predicate.compare_switch_simplify
 1.40% :     0.000002s :     13: predicate.depend_value_elim
 1.19% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.54% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     13: predicate.environ_get_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 0.86% :     0.000001s :      8: predicate.get_grad_eliminate
 0.29% :     0.000000s :      4: predicate.graph_param_transform
 5.16% :     0.000007s :     29: predicate.inline
 0.88% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      8: predicate.less_batch_normalization
 1.67% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     21: predicate.load_eliminater
 1.17% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     13: predicate.merge_addn
 1.18% :     0.000002s :     13: predicate.minmaximum_grad
 1.40% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.54% :     0.000003s :     21: predicate.partial_eliminate
 1.19% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.04% :     0.000003s :     25: predicate.replace_applicator
 0.61% :     0.000001s :      8: predicate.replace_old_param
 0.43% :     0.000001s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.16% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      8: predicate.special_op_eliminate
 1.02% :     0.000001s :      8: predicate.specialize_transform
 1.31% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     21: predicate.switch_defer_inline
 2.48% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.68% :     0.000010s :     67: predicate.switch_simplify
 1.37% :     0.000002s :     13: predicate.tile_eliminate
 1.28% :     0.000002s :     13: predicate.transpose_eliminate
 1.39% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000002s :     21: predicate.updatestate_pure_node_eliminater
 3.38% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     13: predicate.value_based_eliminate
 0.36% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000562    12
44.58% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.93% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.50% :     0.000267s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013322    72
 0.29% :     0.000039s :      1: add_recomputation
 0.44% :     0.000058s :      1: auto_monad
 0.19% :     0.000025s :      1: auto_monad_reorder
 2.21% :     0.000294s :      1: bootstrap
 0.16% :     0.000022s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.18% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.12% :     0.004945s :      1: jit_opt_a
 1.13% :     0.000150s :      1: jit_opt_after_cconv
 0.43% :     0.000057s :      1: jit_opt_b
 3.18% :     0.000423s :      1: loop_unroll
 3.61% :     0.000481s :      1: mutable_eliminate
 6.99% :     0.000932s :     26: opt.transform.jit_opt_a
 0.39% :     0.000053s :      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
 0.11% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.49% :     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% :     0.000035s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000013s :      1: remove_dup_value
 2.31% :     0.000308s :      1: renormalize.infer
 2.05% :     0.000273s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000034s :      1: rewriter_after_opt_a
 0.64% :     0.000085s :      1: rewriter_before_opt_a
 0.56% :     0.000074s :      1: symbol_engine_optimizer
32.49% :     0.004329s :      1: type_inference


TotalTime = 0.011899, [30]
[bootstrap]: 0.00042188
[type_inference]: 0.00446522
[event_method]: 2.017e-05
[auto_monad]: 5.985e-05
[graph_reusing]: 5.12e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.271e-05
[rewriter_before_opt_a]: 8.449e-05
[expand_dump_flag]: 2.94001e-06
[jit_opt_a]: 0.00486502, [2]
    [Cycle 1]: 0.00158292, [27]
        [switch_simplify]: 5.373e-05
        [loop_unroll]: 3.534e-05
        [a_1]: 0.00062699
        [with_stream_mark]: 1.488e-05
        [recompute_prepare]: 7.94002e-06
        [updatestate_depend_eliminate]: 4.04002e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 7.28e-06
        [updatestate_useless_node_eliminater]: 6.44999e-06
        [accelerated_algorithm]: 6.41998e-06
        [meta_shard_fg_expand]: 2.42001e-06
        [get_grad_eliminate_]: 6.21e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.359e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.22001e-06
        [replace_old_param]: 1.04e-05
        [inline_without_move]: 6.12001e-06
        [renormalize]: 0.00053892
        [add_forward_monad_depend]: 4.65001e-06
        [auto_monad_grad]: 2.11e-06
        [auto_monad_eliminator]: 1.203e-05
        [cse]: 2.357e-05
        [replace_applicator]: 1.205e-05
    [Cycle 2]: 0.00034174, [27]
        [switch_simplify]: 6.41e-06
        [loop_unroll]: 5.68002e-06
        [a_1]: 0.00011082
        [with_stream_mark]: 8.03001e-06
        [recompute_prepare]: 5.69e-06
        [updatestate_depend_eliminate]: 3.01001e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 5.48002e-06
        [accelerated_algorithm]: 5.68002e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.277e-05
        [j_node_and_user_rematch]: 8.33999e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 8.52e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 6.06e-06
        [cse]: 1.274e-05
        [replace_applicator]: 5.61e-06
[py_interpret_to_execute_after_opt_a]: 8.37e-06
[rewriter_after_opt_a]: 3.19e-05
[convert_after_rewriter]: 6.11e-06
[order_py_execute_after_rewriter]: 4.93001e-06
[mutable_eliminate]: 0.00045991
[jit_opt_b]: 5.16e-05, [1]
    [Cycle 1]: 4.515e-05, [2]
        [frontend_op_eliminate]: 1.718e-05
        [inline_after_opt_a]: 1.628e-05
[cconv]: 2.302e-05
[loop_unroll]: 0.00041568
[jit_opt_after_cconv]: 0.00014068, [1]
    [Cycle 1]: 0.00013464, [11]
        [c_1]: 2.387e-05
        [parameter_eliminate]: 2.34999e-06
        [updatestate_depend_eliminate]: 5.49e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [cse]: 1.884e-05
        [call_graph_tuple_transform]: 1.83e-05
        [tuple_list_get_item_eliminator]: 5.86998e-06
        [none_parameter_eliminate]: 1.06002e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.85002e-06
[remove_dup_value]: 1.275e-05
[partial_unused_args_eliminate]: 1.86e-06
[environ_conv]: 5.61003e-06
[add_recomputation]: 4.57e-05
[cse_after_recomputation]: 2.309e-05, [1]
    [Cycle 1]: 1.76e-05, [1]
        [cse]: 1.194e-05
[auto_monad_reorder]: 1.531e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.40999e-06
[opt_after_jit_grad]: 0.00043699
[symbol_engine_optimizer]: 7.198e-05, [1]
    [Cycle 1]: 6.605e-05, [6]
        [build]: 2.56e-06
        [elim_shapecalc]: 7.75998e-06
        [elim_not_effective]: 1.334e-05
        [opt_reshape]: 6.02999e-06
        [fold_const_symbol]: 9.12001e-06
        [renormalize]: 4.19997e-07
[validate]: 3.151e-05
Sums
    bootstrap                            :     0.000422s :  5.01%
    type_inference                       :     0.004465s : 53.00%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000060s :  0.71%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.39%
    rewriter_before_opt_a                :     0.000084s :  1.00%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000060s :  0.71%
    jit_opt_a.loop_unroll                :     0.000041s :  0.49%
    jit_opt_a.a_1                        :     0.000738s :  8.76%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.27%
    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        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000011s :  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 :  0.31%
    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          :     0.000019s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000539s :  6.40%
    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         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000460s :  5.46%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.19%
    cconv                                :     0.000023s :  0.27%
    loop_unroll                          :     0.000416s :  4.93%
    jit_opt_after_cconv.c_1              :     0.000024s :  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              :     0.000019s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  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  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000013s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000046s :  0.54%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000437s :  5.19%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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 :  0.00%
    validate                             :     0.000032s :  0.37%


Time group info:
------[substitution.]   0.000186    28
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000005s :      4: substitution.graph_param_transform
81.11% :     0.000151s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.45% :     0.000005s :      4: substitution.replace_old_param
 8.32% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004411     2
79.59% :     0.003511s :      1: type_inference.infer
20.41% :     0.000900s :      1: type_inference.specialize
------[replace.]   0.000080     8
54.02% :     0.000043s :      4: replace.inline
45.98% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     8
91.80% :     0.000148s :      4: match.inline
 8.20% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   912
 1.56% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.98% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     13: predicate.addn_check_dump
 1.27% :     0.000002s :     13: predicate.addn_zero_filter
 1.83% :     0.000002s :     13: predicate.arithmetic_simplify
 1.40% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.18% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.48% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     13: predicate.environ_get_eliminate
 1.23% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.97% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.52% :     0.000007s :     29: predicate.inline
 1.00% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      8: predicate.less_batch_normalization
 1.65% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.08% :     0.000003s :     21: predicate.load_eliminater
 1.33% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.19% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     13: predicate.merge_addn
 1.16% :     0.000002s :     13: predicate.minmaximum_grad
 1.33% :     0.000002s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.63% :     0.000003s :     21: predicate.partial_eliminate
 1.27% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.06% :     0.000003s :     25: predicate.replace_applicator
 0.60% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.21% :     0.000002s :     13: predicate.reshape_eliminate
 1.18% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      8: predicate.special_op_eliminate
 1.03% :     0.000001s :      8: predicate.specialize_transform
 1.33% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     21: predicate.switch_defer_inline
 2.21% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.51% :     0.000010s :     67: predicate.switch_simplify
 1.41% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.45% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000551    12
47.16% :     0.000260s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.88% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.96% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013484    72
 0.36% :     0.000049s :      1: add_recomputation
 0.47% :     0.000063s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 3.21% :     0.000433s :      1: bootstrap
 0.19% :     0.000026s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.17% :     0.000023s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
36.10% :     0.004868s :      1: jit_opt_a
 1.06% :     0.000144s :      1: jit_opt_after_cconv
 0.40% :     0.000054s :      1: jit_opt_b
 3.14% :     0.000423s :      1: loop_unroll
 3.47% :     0.000468s :      1: mutable_eliminate
 7.17% :     0.000967s :     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% :     0.000012s :      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% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.30% :     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
 0.26% :     0.000036s :      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
 2.16% :     0.000291s :      1: renormalize.infer
 1.79% :     0.000241s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000035s :      1: rewriter_after_opt_a
 0.65% :     0.000087s :      1: rewriter_before_opt_a
 0.55% :     0.000075s :      1: symbol_engine_optimizer
33.19% :     0.004476s :      1: type_inference


TotalTime = 0.0402987, [33]
[bootstrap]: 0.00026493
[type_inference]: 0.0213544
[event_method]: 1.944e-05
[auto_monad]: 0.00015361
[graph_reusing]: 6.69001e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 0.00012233
[rewriter_before_opt_a]: 0.00010002
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.00587532, [2]
    [Cycle 1]: 0.00218334, [27]
        [switch_simplify]: 5.941e-05
        [loop_unroll]: 3.726e-05
        [a_1]: 0.00068056
        [with_stream_mark]: 1.807e-05
        [recompute_prepare]: 1.081e-05
        [updatestate_depend_eliminate]: 6.46e-06
        [updatestate_assign_eliminate]: 5.07e-06
        [updatestate_loads_eliminate]: 4.83001e-06
        [parameter_eliminate]: 2.10002e-06
        [specialize_transform]: 9.43002e-06
        [updatestate_useless_node_eliminater]: 1.128e-05
        [accelerated_algorithm]: 8.74e-06
        [meta_shard_fg_expand]: 2.86e-06
        [get_grad_eliminate_]: 7.95998e-06
        [merge_forward]: 5.10001e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.945e-05
        [j_node_and_user_rematch]: 1.29e-05
        [meta_fg_expand]: 3.55e-06
        [replace_old_param]: 1.08e-05
        [inline_without_move]: 7.95e-06
        [renormalize]: 0.00100197
        [add_forward_monad_depend]: 5.51e-06
        [auto_monad_grad]: 1.87999e-06
        [auto_monad_eliminator]: 2.058e-05
        [cse]: 3.8e-05
        [replace_applicator]: 1.578e-05
    [Cycle 2]: 0.00045788, [27]
        [switch_simplify]: 9.49999e-06
        [loop_unroll]: 8.09002e-06
        [a_1]: 0.00016355
        [with_stream_mark]: 1.043e-05
        [recompute_prepare]: 8.26002e-06
        [updatestate_depend_eliminate]: 4.87e-06
        [updatestate_assign_eliminate]: 4.51002e-06
        [updatestate_loads_eliminate]: 4.27e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 8.10999e-06
        [updatestate_useless_node_eliminater]: 1.073e-05
        [accelerated_algorithm]: 7.83999e-06
        [meta_shard_fg_expand]: 1.66998e-06
        [get_grad_eliminate_]: 7.46999e-06
        [merge_forward]: 4.28999e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.574e-05
        [j_node_and_user_rematch]: 1.28e-05
        [meta_fg_expand]: 2.78003e-06
        [replace_old_param]: 9.29998e-06
        [inline_without_move]: 7.46999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 1.019e-05
        [cse]: 1.983e-05
        [replace_applicator]: 7.86001e-06
[py_interpret_to_execute_after_opt_a]: 1.085e-05
[rewriter_after_opt_a]: 0.00028647
[convert_after_rewriter]: 1.097e-05
[order_py_execute_after_rewriter]: 7.65e-06
[mutable_eliminate]: 0.00052681
[jit_opt_b]: 6.628e-05, [1]
    [Cycle 1]: 5.918e-05, [2]
        [frontend_op_eliminate]: 2.361e-05
        [inline_after_opt_a]: 2.419e-05
[cconv]: 2.659e-05
[loop_unroll]: 0.00044698
[jit_opt_after_cconv]: 0.00018935, [1]
    [Cycle 1]: 0.00018229, [11]
        [c_1]: 4.379e-05
        [parameter_eliminate]: 3.21001e-06
        [updatestate_depend_eliminate]: 8.1e-06
        [updatestate_assign_eliminate]: 4.38999e-06
        [updatestate_loads_eliminate]: 3.73001e-06
        [cse]: 2.622e-05
        [call_graph_tuple_transform]: 2.223e-05
        [tuple_list_get_item_eliminator]: 8.37e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 9.09e-06
[remove_dup_value]: 1.876e-05
[partial_unused_args_eliminate]: 2.49999e-06
[environ_conv]: 9.57001e-06
[add_recomputation]: 6.303e-05
[cse_after_recomputation]: 2.983e-05, [1]
    [Cycle 1]: 2.342e-05, [1]
        [cse]: 1.75e-05
[auto_monad_reorder]: 2.657e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 5.85002e-06
[opt_after_jit_grad]: 0.00051016
[symbol_engine_optimizer]: 0.0001044, [1]
    [Cycle 1]: 9.758e-05, [6]
        [build]: 1.043e-05
        [elim_shapecalc]: 1.292e-05
        [elim_not_effective]: 1.925e-05
        [opt_reshape]: 9.49999e-06
        [fold_const_symbol]: 1.41e-05
        [renormalize]: 4.50003e-07
[validate]: 5.535e-05
[backend_pass]: 1.03001e-06
[task_emit]: 0.00978207
[execute]: 6.33998e-06
Sums
    bootstrap                            :     0.000265s :  0.73%
    type_inference                       :     0.021354s : 58.59%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000154s :  0.42%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000122s :  0.34%
    rewriter_before_opt_a                :     0.000100s :  0.27%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000844s :  2.32%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  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 :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001002s :  2.75%
    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      :     0.000031s :  0.08%
    jit_opt_a.cse                        :     0.000058s :  0.16%
    jit_opt_a.replace_applicator         :     0.000024s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000286s :  0.79%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000527s :  1.45%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000027s :  0.07%
    loop_unroll                          :     0.000447s :  1.23%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.12%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000063s :  0.17%
    cse_after_recomputation.cse          :     0.000017s :  0.05%
    auto_monad_reorder                   :     0.000027s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000510s :  1.40%
    symbol_engine_optimizer.build        :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    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                             :     0.000055s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009782s : 26.84%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000255    47
 3.09% :     0.000008s :      2: substitution.depend_value_elim
 1.07% :     0.000003s :      4: substitution.elim_not_effective
 0.84% :     0.000002s :      4: substitution.fold_const_symbol
 2.66% :     0.000007s :      5: substitution.graph_param_transform
74.82% :     0.000191s :      4: substitution.inline
 1.67% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.92% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.45% :     0.000004s :      2: substitution.replace_old_param
 5.05% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.15% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.29% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021288     2
93.65% :     0.019937s :      1: type_inference.infer
 6.35% :     0.001351s :      1: type_inference.specialize
------[replace.]   0.000088     6
65.65% :     0.000058s :      4: replace.inline
34.35% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
94.12% :     0.000188s :      4: match.inline
 5.88% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1021
 1.27% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.56% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     15: predicate.addn_check_dump
 1.28% :     0.000002s :     15: predicate.addn_zero_filter
 2.03% :     0.000003s :     15: predicate.arithmetic_simplify
 1.63% :     0.000003s :     15: predicate.cast_eliminate
 0.49% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     15: predicate.compare_switch_simplify
 1.43% :     0.000002s :     15: predicate.depend_value_elim
 1.15% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.94% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.65% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     15: predicate.environ_get_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.04% :     0.000002s :     10: predicate.get_grad_eliminate
 0.26% :     0.000000s :      5: predicate.graph_param_transform
 5.25% :     0.000009s :     31: predicate.inline
 1.06% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :     10: predicate.less_batch_normalization
 1.57% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     22: predicate.load_eliminater
 1.38% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.45% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.57% :     0.000003s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.40% :     0.000004s :     22: predicate.partial_eliminate
 1.22% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.67% :     0.000003s :     15: predicate.reduce_eliminate
 1.57% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.71% :     0.000003s :     27: predicate.replace_applicator
 0.59% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.28% :     0.000002s :     15: predicate.reshape_eliminate
 1.21% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.08% :     0.000002s :     10: predicate.special_op_eliminate
 1.15% :     0.000002s :     10: predicate.specialize_transform
 1.52% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.21% :     0.000004s :     21: predicate.switch_defer_inline
 2.06% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.51% :     0.000011s :     68: predicate.switch_simplify
 1.20% :     0.000002s :     15: predicate.tile_eliminate
 1.22% :     0.000002s :     15: predicate.transpose_eliminate
 1.56% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.81% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.20% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000003s :     15: predicate.value_based_eliminate
 0.47% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000760    13
44.80% :     0.000340s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.13% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.06% :     0.000380s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042637    76
 0.15% :     0.000066s :      1: add_recomputation
 0.37% :     0.000159s :      1: auto_monad
 0.07% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000278s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.08% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.05% :     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% :     0.000009s :      1: graph_reusing
13.79% :     0.005878s :      1: jit_opt_a
 0.45% :     0.000192s :      1: jit_opt_after_cconv
 0.16% :     0.000069s :      1: jit_opt_b
 1.07% :     0.000456s :      1: loop_unroll
 1.26% :     0.000536s :      1: mutable_eliminate
 2.68% :     0.001141s :     26: opt.transform.jit_opt_a
 0.19% :     0.000080s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000041s :      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
 0.08% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.22% :     0.000520s :      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
 0.30% :     0.000127s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.41% :     0.000603s :      1: renormalize.infer
 0.92% :     0.000391s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.68% :     0.000291s :      1: rewriter_after_opt_a
 0.24% :     0.000104s :      1: rewriter_before_opt_a
 0.25% :     0.000107s :      1: symbol_engine_optimizer
22.97% :     0.009794s :      1: task_emit
50.12% :     0.021368s :      1: type_inference
 0.18% :     0.000077s :      1: validate


TotalTime = 0.0121503, [30]
[bootstrap]: 0.0002551
[type_inference]: 0.00450837
[event_method]: 2.051e-05
[auto_monad]: 6.368e-05
[graph_reusing]: 5.14e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 3.342e-05
[rewriter_before_opt_a]: 8.673e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00509118, [2]
    [Cycle 1]: 0.00163253, [27]
        [switch_simplify]: 5.596e-05
        [loop_unroll]: 3.513e-05
        [a_1]: 0.00060348
        [with_stream_mark]: 1.529e-05
        [recompute_prepare]: 8.35001e-06
        [updatestate_depend_eliminate]: 3.89002e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.73998e-06
        [updatestate_useless_node_eliminater]: 6.16e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.47001e-06
        [get_grad_eliminate_]: 6.02999e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.521e-05
        [j_node_and_user_rematch]: 9.36e-06
        [meta_fg_expand]: 2.61999e-06
        [replace_old_param]: 1.088e-05
        [inline_without_move]: 5.85002e-06
        [renormalize]: 0.00059765
        [add_forward_monad_depend]: 4.80001e-06
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.404e-05
        [cse]: 3.159e-05
        [replace_applicator]: 1.205e-05
    [Cycle 2]: 0.00035126, [27]
        [switch_simplify]: 6.54999e-06
        [loop_unroll]: 5.51e-06
        [a_1]: 0.00011251
        [with_stream_mark]: 8.72998e-06
        [recompute_prepare]: 5.96e-06
        [updatestate_depend_eliminate]: 3.04001e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.69999e-06
        [updatestate_useless_node_eliminater]: 5.52001e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 1.28002e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 2.57001e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.351e-05
        [j_node_and_user_rematch]: 8.25e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 9.38002e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 5.86e-06
        [cse]: 1.249e-05
        [replace_applicator]: 5.85002e-06
[py_interpret_to_execute_after_opt_a]: 9.12999e-06
[rewriter_after_opt_a]: 3.465e-05
[convert_after_rewriter]: 6.62002e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00051137
[jit_opt_b]: 5.423e-05, [1]
    [Cycle 1]: 4.771e-05, [2]
        [frontend_op_eliminate]: 1.887e-05
        [inline_after_opt_a]: 1.701e-05
[cconv]: 2.658e-05
[loop_unroll]: 0.00040706
[jit_opt_after_cconv]: 0.00014763, [1]
    [Cycle 1]: 0.00014147, [11]
        [c_1]: 2.428e-05
        [parameter_eliminate]: 2.65997e-06
        [updatestate_depend_eliminate]: 5.67999e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [cse]: 2.052e-05
        [call_graph_tuple_transform]: 1.96e-05
        [tuple_list_get_item_eliminator]: 6.29001e-06
        [none_parameter_eliminate]: 1.75001e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.93002e-06
[remove_dup_value]: 1.378e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 4.625e-05
[cse_after_recomputation]: 2.276e-05, [1]
    [Cycle 1]: 1.742e-05, [1]
        [cse]: 1.176e-05
[auto_monad_reorder]: 1.724e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 5.62999e-06
[opt_after_jit_grad]: 0.00050217
[symbol_engine_optimizer]: 7.583e-05, [1]
    [Cycle 1]: 6.991e-05, [6]
        [build]: 3.07002e-06
        [elim_shapecalc]: 8.26002e-06
        [elim_not_effective]: 1.355e-05
        [opt_reshape]: 6.36e-06
        [fold_const_symbol]: 1.001e-05
        [renormalize]: 4.69998e-07
[validate]: 3.46e-05
Sums
    bootstrap                            :     0.000255s :  3.00%
    type_inference                       :     0.004508s : 53.04%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000064s :  0.75%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.39%
    rewriter_before_opt_a                :     0.000087s :  1.02%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.74%
    jit_opt_a.loop_unroll                :     0.000041s :  0.48%
    jit_opt_a.a_1                        :     0.000716s :  8.42%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.17%
    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 :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000011s :  0.13%
    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 :     0.000029s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000020s :  0.24%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000598s :  7.03%
    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 :  0.23%
    jit_opt_a.cse                        :     0.000044s :  0.52%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    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                    :     0.000511s :  6.02%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000027s :  0.31%
    loop_unroll                          :     0.000407s :  4.79%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.29%
    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 :     0.000003s :  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 :  0.07%
    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 :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000046s :  0.54%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000017s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000502s :  5.91%
    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 :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000035s :  0.41%


Time group info:
------[substitution.]   0.000203    28
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000002s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      4: substitution.graph_param_transform
81.38% :     0.000165s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.42% :     0.000005s :      4: substitution.replace_old_param
 8.16% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004455     2
80.50% :     0.003586s :      1: type_inference.infer
19.50% :     0.000869s :      1: type_inference.specialize
------[replace.]   0.000079     8
55.52% :     0.000044s :      4: replace.inline
44.48% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000177     8
91.71% :     0.000162s :      4: match.inline
 8.29% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   912
 1.38% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     13: predicate.addn_check_dump
 1.28% :     0.000002s :     13: predicate.addn_zero_filter
 1.81% :     0.000002s :     13: predicate.arithmetic_simplify
 1.38% :     0.000002s :     13: predicate.cast_eliminate
 0.52% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     13: predicate.compare_switch_simplify
 1.18% :     0.000002s :     13: predicate.depend_value_elim
 1.21% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      4: predicate.elim_not_effective
 0.51% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     13: predicate.environ_get_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.50% :     0.000007s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.13% :     0.000002s :      8: predicate.less_batch_normalization
 1.66% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     21: predicate.load_eliminater
 1.27% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.13% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.87% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     13: predicate.merge_addn
 1.12% :     0.000001s :     13: predicate.minmaximum_grad
 1.65% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.58% :     0.000003s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     13: predicate.reduce_eliminate
 1.65% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.97% :     0.000003s :     25: predicate.replace_applicator
 0.70% :     0.000001s :      8: predicate.replace_old_param
 0.40% :     0.000001s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.25% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     13: predicate.same_eliminate
 0.70% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      8: predicate.special_op_eliminate
 0.95% :     0.000001s :      8: predicate.specialize_transform
 1.30% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     21: predicate.switch_defer_inline
 2.22% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.60% :     0.000010s :     67: predicate.switch_simplify
 1.15% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.44% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.32% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.95% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000577    12
49.56% :     0.000286s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.96% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.48% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013777    72
 0.36% :     0.000049s :      1: add_recomputation
 0.49% :     0.000067s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 1.94% :     0.000267s :      1: bootstrap
 0.21% :     0.000029s :      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
 0.17% :     0.000024s :      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
36.97% :     0.005094s :      1: jit_opt_a
 1.09% :     0.000150s :      1: jit_opt_after_cconv
 0.41% :     0.000057s :      1: jit_opt_b
 3.01% :     0.000415s :      1: loop_unroll
 3.77% :     0.000520s :      1: mutable_eliminate
 6.90% :     0.000951s :     26: opt.transform.jit_opt_a
 0.38% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000030s :      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% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.70% :     0.000509s :      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
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.31% :     0.000318s :      1: renormalize.infer
 1.98% :     0.000272s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000037s :      1: rewriter_after_opt_a
 0.65% :     0.000089s :      1: rewriter_before_opt_a
 0.57% :     0.000078s :      1: symbol_engine_optimizer
32.80% :     0.004519s :      1: type_inference


TotalTime = 0.0409795, [33]
[bootstrap]: 0.00030669
[type_inference]: 0.0217044
[event_method]: 1.942e-05
[auto_monad]: 0.00015714
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 0.00011597
[rewriter_before_opt_a]: 9.381e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.00596337, [2]
    [Cycle 1]: 0.00223972, [27]
        [switch_simplify]: 5.985e-05
        [loop_unroll]: 3.61e-05
        [a_1]: 0.00072046
        [with_stream_mark]: 2.05e-05
        [recompute_prepare]: 1.059e-05
        [updatestate_depend_eliminate]: 6.48e-06
        [updatestate_assign_eliminate]: 6.24999e-06
        [updatestate_loads_eliminate]: 4.75999e-06
        [parameter_eliminate]: 2.15002e-06
        [specialize_transform]: 9.04e-06
        [updatestate_useless_node_eliminater]: 1.113e-05
        [accelerated_algorithm]: 8.48999e-06
        [meta_shard_fg_expand]: 2.87002e-06
        [get_grad_eliminate_]: 8.03001e-06
        [merge_forward]: 5.22e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.971e-05
        [j_node_and_user_rematch]: 1.333e-05
        [meta_fg_expand]: 3.23998e-06
        [replace_old_param]: 1.099e-05
        [inline_without_move]: 7.95e-06
        [renormalize]: 0.00101518
        [add_forward_monad_depend]: 5.62001e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 2.081e-05
        [cse]: 3.601e-05
        [replace_applicator]: 1.682e-05
    [Cycle 2]: 0.00045874, [27]
        [switch_simplify]: 9.19e-06
        [loop_unroll]: 7.65998e-06
        [a_1]: 0.00016353
        [with_stream_mark]: 1.17e-05
        [recompute_prepare]: 7.79002e-06
        [updatestate_depend_eliminate]: 4.94003e-06
        [updatestate_assign_eliminate]: 4.73001e-06
        [updatestate_loads_eliminate]: 4.4e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 8.11002e-06
        [updatestate_useless_node_eliminater]: 1.082e-05
        [accelerated_algorithm]: 7.7e-06
        [meta_shard_fg_expand]: 1.76e-06
        [get_grad_eliminate_]: 7.29001e-06
        [merge_forward]: 4.17003e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.639e-05
        [j_node_and_user_rematch]: 1.252e-05
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 9.47001e-06
        [inline_without_move]: 7.48e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 8.59989e-07
        [auto_monad_eliminator]: 1.111e-05
        [cse]: 1.914e-05
        [replace_applicator]: 7.82e-06
[py_interpret_to_execute_after_opt_a]: 1.216e-05
[rewriter_after_opt_a]: 0.00029272
[convert_after_rewriter]: 1.077e-05
[order_py_execute_after_rewriter]: 7.32997e-06
[mutable_eliminate]: 0.000556
[jit_opt_b]: 6.715e-05, [1]
    [Cycle 1]: 6.013e-05, [2]
        [frontend_op_eliminate]: 2.414e-05
        [inline_after_opt_a]: 2.402e-05
[cconv]: 2.556e-05
[loop_unroll]: 0.00044355
[jit_opt_after_cconv]: 0.00026267, [1]
    [Cycle 1]: 0.00025609, [11]
        [c_1]: 0.00010994
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 8.29002e-06
        [updatestate_assign_eliminate]: 4.45999e-06
        [updatestate_loads_eliminate]: 3.76999e-06
        [cse]: 2.698e-05
        [call_graph_tuple_transform]: 2.333e-05
        [tuple_list_get_item_eliminator]: 8.50001e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 8.87e-06
[remove_dup_value]: 1.84e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 1.064e-05
[add_recomputation]: 6.201e-05
[cse_after_recomputation]: 3.013e-05, [1]
    [Cycle 1]: 2.421e-05, [1]
        [cse]: 1.808e-05
[auto_monad_reorder]: 2.681e-05
[get_jit_bprop_graph]: 1.97001e-06
[rewriter_after_jit_bprop_graph]: 5.03002e-06
[opt_after_jit_grad]: 0.00050599
[symbol_engine_optimizer]: 0.00013133, [1]
    [Cycle 1]: 0.00012499, [6]
        [build]: 3.82e-05
        [elim_shapecalc]: 1.333e-05
        [elim_not_effective]: 1.911e-05
        [opt_reshape]: 8.79e-06
        [fold_const_symbol]: 1.305e-05
        [renormalize]: 4.30009e-07
[validate]: 5.421e-05
[backend_pass]: 1.10001e-06
[task_emit]: 0.00985668
[execute]: 8.54e-06
Sums
    bootstrap                            :     0.000307s :  0.83%
    type_inference                       :     0.021704s : 58.53%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000157s :  0.42%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000116s :  0.31%
    rewriter_before_opt_a                :     0.000094s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.19%
    jit_opt_a.loop_unroll                :     0.000044s :  0.12%
    jit_opt_a.a_1                        :     0.000884s :  2.38%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001015s :  2.74%
    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      :     0.000032s :  0.09%
    jit_opt_a.cse                        :     0.000055s :  0.15%
    jit_opt_a.replace_applicator         :     0.000025s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000293s :  0.79%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000556s :  1.50%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000444s :  1.20%
    jit_opt_after_cconv.c_1              :     0.000110s :  0.30%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  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 :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000018s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000062s :  0.17%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000027s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000506s :  1.36%
    symbol_engine_optimizer.build        :     0.000038s :  0.10%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000054s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009857s : 26.58%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000271    47
 3.32% :     0.000009s :      2: substitution.depend_value_elim
 1.09% :     0.000003s :      4: substitution.elim_not_effective
 0.71% :     0.000002s :      4: substitution.fold_const_symbol
 2.49% :     0.000007s :      5: substitution.graph_param_transform
75.60% :     0.000205s :      4: substitution.inline
 1.71% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.66% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.29% :     0.000004s :      2: substitution.replace_old_param
 4.90% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.11% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.10% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021634     2
93.67% :     0.020264s :      1: type_inference.infer
 6.33% :     0.001370s :      1: type_inference.specialize
------[replace.]   0.000092     6
67.29% :     0.000062s :      4: replace.inline
32.71% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
94.37% :     0.000202s :      4: match.inline
 5.63% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1021
 1.30% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.25% :     0.000002s :     15: predicate.addn_zero_filter
 1.91% :     0.000003s :     15: predicate.arithmetic_simplify
 1.30% :     0.000002s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.30% :     0.000002s :     15: predicate.compare_switch_simplify
 1.36% :     0.000002s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.71% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.04% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 5.24% :     0.000009s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :     10: predicate.less_batch_normalization
 1.37% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.99% :     0.000003s :     22: predicate.load_eliminater
 1.40% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.39% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.21% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     15: predicate.merge_addn
 1.14% :     0.000002s :     15: predicate.minmaximum_grad
 1.44% :     0.000002s :      5: predicate.mutable_eliminate
 0.58% :     0.000001s :      5: predicate.opt_reshape
 2.32% :     0.000004s :     22: predicate.partial_eliminate
 1.23% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.73% :     0.000003s :     15: predicate.reduce_eliminate
 1.51% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.68% :     0.000003s :     27: predicate.replace_applicator
 0.70% :     0.000001s :     10: predicate.replace_old_param
 0.31% :     0.000001s :      5: predicate.reset_defer_inline
 1.20% :     0.000002s :     15: predicate.reshape_eliminate
 1.31% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.91% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     15: predicate.same_eliminate
 0.64% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.21% :     0.000002s :     10: predicate.special_op_eliminate
 1.11% :     0.000002s :     10: predicate.specialize_transform
 1.48% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.70% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.17% :     0.000004s :     21: predicate.switch_defer_inline
 1.91% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.00% :     0.000012s :     68: predicate.switch_simplify
 1.27% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.53% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 2.24% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000777    13
44.78% :     0.000348s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.67% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.55% :     0.000385s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043426    76
 0.15% :     0.000065s :      1: add_recomputation
 0.37% :     0.000162s :      1: auto_monad
 0.07% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.74% :     0.000321s :      1: bootstrap
 0.07% :     0.000028s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.05% :     0.000024s :      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% :     0.000008s :      1: graph_reusing
13.74% :     0.005967s :      1: jit_opt_a
 0.61% :     0.000266s :      1: jit_opt_after_cconv
 0.16% :     0.000070s :      1: jit_opt_b
 1.04% :     0.000452s :      1: loop_unroll
 1.30% :     0.000565s :      1: mutable_eliminate
 2.71% :     0.001178s :     26: opt.transform.jit_opt_a
 0.34% :     0.000147s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000041s :      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
 0.07% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000515s :      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
 0.28% :     0.000120s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.39% :     0.000602s :      1: renormalize.infer
 0.93% :     0.000405s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.68% :     0.000297s :      1: rewriter_after_opt_a
 0.23% :     0.000098s :      1: rewriter_before_opt_a
 0.31% :     0.000134s :      1: symbol_engine_optimizer
22.73% :     0.009869s :      1: task_emit
50.02% :     0.021720s :      1: type_inference
 0.18% :     0.000078s :      1: validate


TotalTime = 0.0140451, [30]
[bootstrap]: 0.00032818
[type_inference]: 0.0051232
[event_method]: 2.076e-05
[auto_monad]: 6.268e-05
[graph_reusing]: 5.62999e-06
[pre_auto_parallel]: 2.25002e-06
[py_interpret_to_execute]: 3.688e-05
[rewriter_before_opt_a]: 9.561e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00593434, [2]
    [Cycle 1]: 0.00207651, [27]
        [switch_simplify]: 6.513e-05
        [loop_unroll]: 3.548e-05
        [a_1]: 0.00074367
        [with_stream_mark]: 2.22e-05
        [recompute_prepare]: 8.92999e-06
        [updatestate_depend_eliminate]: 3.88999e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 3.24001e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 7.07997e-06
        [updatestate_useless_node_eliminater]: 5.94e-06
        [accelerated_algorithm]: 6.41e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 6.04001e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.438e-05
        [j_node_and_user_rematch]: 1.031e-05
        [meta_fg_expand]: 2.66999e-06
        [replace_old_param]: 1.13e-05
        [inline_without_move]: 6.16998e-06
        [renormalize]: 0.0008557
        [add_forward_monad_depend]: 6.03998e-06
        [auto_monad_grad]: 2.73e-06
        [auto_monad_eliminator]: 1.521e-05
        [cse]: 3.59e-05
        [replace_applicator]: 1.204e-05
    [Cycle 2]: 0.00036479, [27]
        [switch_simplify]: 6.58e-06
        [loop_unroll]: 5.87001e-06
        [a_1]: 0.00011453
        [with_stream_mark]: 9.29e-06
        [recompute_prepare]: 5.67001e-06
        [updatestate_depend_eliminate]: 3.29001e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.74999e-06
        [updatestate_useless_node_eliminater]: 5.64e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.59998e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.4e-05
        [j_node_and_user_rematch]: 9.23002e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 1.023e-05
        [inline_without_move]: 5.51e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 1.19e-06
        [auto_monad_eliminator]: 6.05002e-06
        [cse]: 1.37e-05
        [replace_applicator]: 5.83997e-06
[py_interpret_to_execute_after_opt_a]: 1.25e-05
[rewriter_after_opt_a]: 3.68e-05
[convert_after_rewriter]: 6.76e-06
[order_py_execute_after_rewriter]: 5.31002e-06
[mutable_eliminate]: 0.00068372
[jit_opt_b]: 6.18e-05, [1]
    [Cycle 1]: 5.416e-05, [2]
        [frontend_op_eliminate]: 2.032e-05
        [inline_after_opt_a]: 2.147e-05
[cconv]: 2.682e-05
[loop_unroll]: 0.00047243
[jit_opt_after_cconv]: 0.00015792, [1]
    [Cycle 1]: 0.00015131, [11]
        [c_1]: 2.608e-05
        [parameter_eliminate]: 3.34001e-06
        [updatestate_depend_eliminate]: 6.59999e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [cse]: 2.3e-05
        [call_graph_tuple_transform]: 2.064e-05
        [tuple_list_get_item_eliminator]: 6.11e-06
        [none_parameter_eliminate]: 1.79e-06
        [renormalize]: 1.20001e-06
        [switch_simplify]: 6.44999e-06
[remove_dup_value]: 1.497e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 5.79999e-06
[add_recomputation]: 7.185e-05
[cse_after_recomputation]: 2.437e-05, [1]
    [Cycle 1]: 1.888e-05, [1]
        [cse]: 1.289e-05
[auto_monad_reorder]: 1.798e-05
[get_jit_bprop_graph]: 2.20002e-06
[rewriter_after_jit_bprop_graph]: 6.63e-06
[opt_after_jit_grad]: 0.00053914
[symbol_engine_optimizer]: 7.859e-05, [1]
    [Cycle 1]: 7.197e-05, [6]
        [build]: 4.2e-06
        [elim_shapecalc]: 9.17001e-06
        [elim_not_effective]: 1.396e-05
        [opt_reshape]: 6.39999e-06
        [fold_const_symbol]: 9.76e-06
        [renormalize]: 6.39993e-07
[validate]: 4.167e-05
Sums
    bootstrap                            :     0.000328s :  3.29%
    type_inference                       :     0.005123s : 51.41%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000063s :  0.63%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.37%
    rewriter_before_opt_a                :     0.000096s :  0.96%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000072s :  0.72%
    jit_opt_a.loop_unroll                :     0.000041s :  0.41%
    jit_opt_a.a_1                        :     0.000858s :  8.61%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.32%
    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 :     0.000006s :  0.06%
    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 :     0.000012s :  0.12%
    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_        :     0.000012s :  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 :     0.000028s :  0.28%
    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        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000856s :  8.59%
    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 :  0.21%
    jit_opt_a.cse                        :     0.000050s :  0.50%
    jit_opt_a.replace_applicator         :     0.000018s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    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                    :     0.000684s :  6.86%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.22%
    cconv                                :     0.000027s :  0.27%
    loop_unroll                          :     0.000472s :  4.74%
    jit_opt_after_cconv.c_1              :     0.000026s :  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              :     0.000023s :  0.23%
    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                    :     0.000072s :  0.72%
    cse_after_recomputation.cse          :     0.000013s :  0.13%
    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                   :     0.000539s :  5.41%
    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 :  0.14%
    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                             :     0.000042s :  0.42%


Time group info:
------[substitution.]   0.000241    28
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000006s :      4: substitution.graph_param_transform
82.31% :     0.000198s :      4: substitution.inline
 1.66% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.82% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.65% :     0.000006s :      4: substitution.replace_old_param
 7.57% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005062     2
82.32% :     0.004167s :      1: type_inference.infer
17.68% :     0.000895s :      1: type_inference.specialize
------[replace.]   0.000162     8
32.26% :     0.000052s :      4: replace.inline
67.74% :     0.000109s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000211     8
92.37% :     0.000195s :      4: match.inline
 7.63% :     0.000016s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000144   912
 1.22% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.15% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.07% :     0.000002s :     13: predicate.addn_check_dump
 1.41% :     0.000002s :     13: predicate.addn_zero_filter
 2.18% :     0.000003s :     13: predicate.arithmetic_simplify
 1.77% :     0.000003s :     13: predicate.cast_eliminate
 0.55% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     13: predicate.compare_switch_simplify
 1.18% :     0.000002s :     13: predicate.depend_value_elim
 1.14% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      4: predicate.elim_not_effective
 0.47% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000001s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 5.41% :     0.000008s :     29: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.07% :     0.000002s :      8: predicate.less_batch_normalization
 1.63% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     21: predicate.load_eliminater
 1.57% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.98% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.84% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000002s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.37% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.27% :     0.000003s :     21: predicate.partial_eliminate
 1.23% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.86% :     0.000003s :     13: predicate.reduce_eliminate
 1.67% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.89% :     0.000003s :     25: predicate.replace_applicator
 0.69% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000001s :      4: predicate.reset_defer_inline
 1.18% :     0.000002s :     13: predicate.reshape_eliminate
 1.23% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     13: predicate.same_eliminate
 0.54% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.38% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.57% :     0.000004s :     21: predicate.switch_defer_inline
 2.15% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.77% :     0.000011s :     67: predicate.switch_simplify
 1.31% :     0.000002s :     13: predicate.tile_eliminate
 1.25% :     0.000002s :     13: predicate.transpose_eliminate
 1.49% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000648    12
47.86% :     0.000310s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.65% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.49% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016087    72
 0.47% :     0.000075s :      1: add_recomputation
 0.41% :     0.000066s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 2.13% :     0.000342s :      1: bootstrap
 0.18% :     0.000030s :      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% :     0.000024s :      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
36.91% :     0.005938s :      1: jit_opt_a
 1.00% :     0.000161s :      1: jit_opt_after_cconv
 0.40% :     0.000065s :      1: jit_opt_b
 2.99% :     0.000481s :      1: loop_unroll
 4.31% :     0.000693s :      1: mutable_eliminate
 6.88% :     0.001107s :     26: opt.transform.jit_opt_a
 0.35% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000035s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.40% :     0.000547s :      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% :     0.000040s :      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
 2.84% :     0.000457s :      1: renormalize.infer
 2.43% :     0.000390s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000040s :      1: rewriter_after_opt_a
 0.61% :     0.000099s :      1: rewriter_before_opt_a
 0.50% :     0.000081s :      1: symbol_engine_optimizer
31.92% :     0.005134s :      1: type_inference


TotalTime = 0.0120682, [30]
[bootstrap]: 0.00030837
[type_inference]: 0.00451431
[event_method]: 2.019e-05
[auto_monad]: 4.335e-05
[graph_reusing]: 3.81999e-06
[pre_auto_parallel]: 1.81e-06
[py_interpret_to_execute]: 3.401e-05
[rewriter_before_opt_a]: 7.945e-05
[expand_dump_flag]: 1.98002e-06
[jit_opt_a]: 0.00505983, [2]
    [Cycle 1]: 0.00173432, [27]
        [switch_simplify]: 5.545e-05
        [loop_unroll]: 3.639e-05
        [a_1]: 0.00059154
        [with_stream_mark]: 1.468e-05
        [recompute_prepare]: 8.33001e-06
        [updatestate_depend_eliminate]: 3.91001e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 7.05e-06
        [updatestate_useless_node_eliminater]: 6.02001e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 6.17001e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.279e-05
        [j_node_and_user_rematch]: 9.51998e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 9.52999e-06
        [inline_without_move]: 6.17999e-06
        [renormalize]: 0.00072702
        [add_forward_monad_depend]: 5.12e-06
        [auto_monad_grad]: 1.37e-06
        [auto_monad_eliminator]: 1.212e-05
        [cse]: 2.206e-05
        [replace_applicator]: 1.365e-05
    [Cycle 2]: 0.00035905, [27]
        [switch_simplify]: 7.08e-06
        [loop_unroll]: 6.28e-06
        [a_1]: 0.00011558
        [with_stream_mark]: 8.77999e-06
        [recompute_prepare]: 6.21998e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 5.61998e-06
        [accelerated_algorithm]: 5.97999e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 5.39998e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.331e-05
        [j_node_and_user_rematch]: 8.51002e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 9.32001e-06
        [inline_without_move]: 5.33002e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 6.06e-06
        [cse]: 1.307e-05
        [replace_applicator]: 6.14001e-06
[py_interpret_to_execute_after_opt_a]: 9.74e-06
[rewriter_after_opt_a]: 2.857e-05
[convert_after_rewriter]: 5.61e-06
[order_py_execute_after_rewriter]: 4.44998e-06
[mutable_eliminate]: 0.00050783
[jit_opt_b]: 5.767e-05, [1]
    [Cycle 1]: 4.783e-05, [2]
        [frontend_op_eliminate]: 1.887e-05
        [inline_after_opt_a]: 1.727e-05
[cconv]: 1.521e-05
[loop_unroll]: 0.00042217
[jit_opt_after_cconv]: 0.00014755, [1]
    [Cycle 1]: 0.0001414, [11]
        [c_1]: 2.634e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.84e-06
        [updatestate_assign_eliminate]: 2.81999e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 2.118e-05
        [call_graph_tuple_transform]: 1.769e-05
        [tuple_list_get_item_eliminator]: 6.17001e-06
        [none_parameter_eliminate]: 9.70002e-07
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.99e-06
[remove_dup_value]: 8.32998e-06
[partial_unused_args_eliminate]: 1.53002e-06
[environ_conv]: 4.46002e-06
[add_recomputation]: 3.076e-05
[cse_after_recomputation]: 2.383e-05, [1]
    [Cycle 1]: 1.864e-05, [1]
        [cse]: 1.284e-05
[auto_monad_reorder]: 1.211e-05
[get_jit_bprop_graph]: 1.44998e-06
[rewriter_after_jit_bprop_graph]: 4.51002e-06
[opt_after_jit_grad]: 0.00046126
[symbol_engine_optimizer]: 7.287e-05, [1]
    [Cycle 1]: 6.694e-05, [6]
        [build]: 2.49999e-06
        [elim_shapecalc]: 8.64e-06
        [elim_not_effective]: 1.332e-05
        [opt_reshape]: 6.26e-06
        [fold_const_symbol]: 8.75001e-06
        [renormalize]: 2.40019e-07
[validate]: 2.785e-05
Sums
    bootstrap                            :     0.000308s :  3.61%
    type_inference                       :     0.004514s : 52.80%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000043s :  0.51%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.40%
    rewriter_before_opt_a                :     0.000079s :  0.93%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.73%
    jit_opt_a.loop_unroll                :     0.000043s :  0.50%
    jit_opt_a.a_1                        :     0.000707s :  8.27%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.27%
    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 :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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_        :     0.000012s :  0.14%
    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 :     0.000026s :  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          :     0.000019s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000727s :  8.51%
    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      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000035s :  0.41%
    jit_opt_a.replace_applicator         :     0.000020s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000029s :  0.33%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000508s :  5.94%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000015s :  0.18%
    loop_unroll                          :     0.000422s :  4.94%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.31%
    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 :  0.03%
    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 :     0.000006s :  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  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000008s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000031s :  0.36%
    cse_after_recomputation.cse          :     0.000013s :  0.15%
    auto_monad_reorder                   :     0.000012s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000461s :  5.40%
    symbol_engine_optimizer.build        :     0.000002s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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 :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.33%


Time group info:
------[substitution.]   0.000177    28
 0.74% :     0.000001s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.17% :     0.000004s :      4: substitution.graph_param_transform
83.68% :     0.000148s :      4: substitution.inline
 1.63% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000004s :      4: substitution.replace_old_param
 6.70% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004465     2
81.06% :     0.003619s :      1: type_inference.infer
18.94% :     0.000846s :      1: type_inference.specialize
------[replace.]   0.000083     8
55.65% :     0.000046s :      4: replace.inline
44.35% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     8
93.60% :     0.000145s :      4: match.inline
 6.40% :     0.000010s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   912
 1.29% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.15% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     13: predicate.addn_check_dump
 1.20% :     0.000002s :     13: predicate.addn_zero_filter
 1.86% :     0.000003s :     13: predicate.arithmetic_simplify
 1.41% :     0.000002s :     13: predicate.cast_eliminate
 0.52% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.19% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.58% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     13: predicate.environ_get_eliminate
 1.18% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 1.07% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.26% :     0.000007s :     29: predicate.inline
 0.93% :     0.000001s :      8: predicate.inline_without_move
 0.45% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      8: predicate.less_batch_normalization
 1.57% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.11% :     0.000003s :     21: predicate.load_eliminater
 1.24% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.20% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     13: predicate.merge_addn
 1.11% :     0.000002s :     13: predicate.minmaximum_grad
 1.74% :     0.000002s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.51% :     0.000003s :     21: predicate.partial_eliminate
 1.37% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     13: predicate.reduce_eliminate
 1.81% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.07% :     0.000003s :     25: predicate.replace_applicator
 0.61% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000000s :      4: predicate.reset_defer_inline
 1.49% :     0.000002s :     13: predicate.reshape_eliminate
 1.18% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     13: predicate.same_eliminate
 0.70% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.37% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     21: predicate.switch_defer_inline
 2.21% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.54% :     0.000010s :     67: predicate.switch_simplify
 1.33% :     0.000002s :     13: predicate.tile_eliminate
 1.20% :     0.000002s :     13: predicate.transpose_eliminate
 1.41% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.86% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     13: predicate.value_based_eliminate
 0.43% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000574    12
47.51% :     0.000273s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.25% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.24% :     0.000271s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013826    72
 0.24% :     0.000034s :      1: add_recomputation
 0.34% :     0.000047s :      1: auto_monad
 0.11% :     0.000015s :      1: auto_monad_reorder
 2.34% :     0.000324s :      1: bootstrap
 0.13% :     0.000018s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.17% :     0.000024s :      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
36.62% :     0.005063s :      1: jit_opt_a
 1.09% :     0.000150s :      1: jit_opt_after_cconv
 0.44% :     0.000060s :      1: jit_opt_b
 3.11% :     0.000429s :      1: loop_unroll
 3.73% :     0.000515s :      1: mutable_eliminate
 6.82% :     0.000943s :     26: opt.transform.jit_opt_a
 0.38% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000030s :      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
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.39% :     0.000469s :      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% :     0.000037s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000011s :      1: remove_dup_value
 2.37% :     0.000328s :      1: renormalize.infer
 2.84% :     0.000393s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000031s :      1: rewriter_after_opt_a
 0.59% :     0.000082s :      1: rewriter_before_opt_a
 0.55% :     0.000075s :      1: symbol_engine_optimizer
32.73% :     0.004525s :      1: type_inference


TotalTime = 0.0438799, [33]
[bootstrap]: 0.00028359
[type_inference]: 0.0218843
[event_method]: 2.126e-05
[auto_monad]: 0.00014901
[graph_reusing]: 6.48e-06
[pre_auto_parallel]: 2.84001e-06
[py_interpret_to_execute]: 0.00012067
[rewriter_before_opt_a]: 9.811e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00619529, [2]
    [Cycle 1]: 0.00233591, [27]
        [switch_simplify]: 6.05e-05
        [loop_unroll]: 3.665e-05
        [a_1]: 0.0007099
        [with_stream_mark]: 2.162e-05
        [recompute_prepare]: 1.099e-05
        [updatestate_depend_eliminate]: 5.97001e-06
        [updatestate_assign_eliminate]: 5.60001e-06
        [updatestate_loads_eliminate]: 4.94e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 9.25999e-06
        [updatestate_useless_node_eliminater]: 1.154e-05
        [accelerated_algorithm]: 8.47998e-06
        [meta_shard_fg_expand]: 3.09001e-06
        [get_grad_eliminate_]: 8.08001e-06
        [merge_forward]: 4.89e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.938e-05
        [j_node_and_user_rematch]: 1.402e-05
        [meta_fg_expand]: 3.65998e-06
        [replace_old_param]: 1.204e-05
        [inline_without_move]: 8.27998e-06
        [renormalize]: 0.00111302
        [add_forward_monad_depend]: 5.45001e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 2.132e-05
        [cse]: 3.951e-05
        [replace_applicator]: 1.551e-05
    [Cycle 2]: 0.0004581, [27]
        [switch_simplify]: 9.17999e-06
        [loop_unroll]: 8.17e-06
        [a_1]: 0.00016377
        [with_stream_mark]: 1.067e-05
        [recompute_prepare]: 8.09002e-06
        [updatestate_depend_eliminate]: 4.89e-06
        [updatestate_assign_eliminate]: 4.48999e-06
        [updatestate_loads_eliminate]: 3.85998e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 8.27998e-06
        [updatestate_useless_node_eliminater]: 1.067e-05
        [accelerated_algorithm]: 8.13999e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 7.38e-06
        [merge_forward]: 4.65999e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.617e-05
        [j_node_and_user_rematch]: 1.231e-05
        [meta_fg_expand]: 2.59999e-06
        [replace_old_param]: 9.94001e-06
        [inline_without_move]: 7.57002e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 9.99001e-06
        [cse]: 1.955e-05
        [replace_applicator]: 7.88001e-06
[py_interpret_to_execute_after_opt_a]: 1.163e-05
[rewriter_after_opt_a]: 0.00031777
[convert_after_rewriter]: 1.279e-05
[order_py_execute_after_rewriter]: 6.83e-06
[mutable_eliminate]: 0.00055412
[jit_opt_b]: 6.54e-05, [1]
    [Cycle 1]: 5.856e-05, [2]
        [frontend_op_eliminate]: 2.304e-05
        [inline_after_opt_a]: 2.364e-05
[cconv]: 2.694e-05
[loop_unroll]: 0.00044326
[jit_opt_after_cconv]: 0.00018816, [1]
    [Cycle 1]: 0.0001815, [11]
        [c_1]: 4.456e-05
        [parameter_eliminate]: 2.94001e-06
        [updatestate_depend_eliminate]: 7.68001e-06
        [updatestate_assign_eliminate]: 4.12003e-06
        [updatestate_loads_eliminate]: 4.72998e-06
        [cse]: 2.629e-05
        [call_graph_tuple_transform]: 2.145e-05
        [tuple_list_get_item_eliminator]: 8.19998e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 8.70999e-06
[remove_dup_value]: 1.753e-05
[partial_unused_args_eliminate]: 2.01998e-06
[environ_conv]: 1.118e-05
[add_recomputation]: 6.477e-05
[cse_after_recomputation]: 3.062e-05, [1]
    [Cycle 1]: 2.468e-05, [1]
        [cse]: 1.88e-05
[auto_monad_reorder]: 2.751e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 5.05001e-06
[opt_after_jit_grad]: 0.00049164
[symbol_engine_optimizer]: 9.901e-05, [1]
    [Cycle 1]: 9.235e-05, [6]
        [build]: 1.162e-05
        [elim_shapecalc]: 1.162e-05
        [elim_not_effective]: 1.783e-05
        [opt_reshape]: 8.64e-06
        [fold_const_symbol]: 1.44e-05
        [renormalize]: 4.69998e-07
[validate]: 5.546e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.0124558
[execute]: 9.16998e-06
Sums
    bootstrap                            :     0.000284s :  0.71%
    type_inference                       :     0.021884s : 54.92%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000149s :  0.37%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000121s :  0.30%
    rewriter_before_opt_a                :     0.000098s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000070s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000874s :  2.19%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001113s :  2.79%
    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      :     0.000031s :  0.08%
    jit_opt_a.cse                        :     0.000059s :  0.15%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000318s :  0.80%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000554s :  1.39%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000027s :  0.07%
    loop_unroll                          :     0.000443s :  1.11%
    jit_opt_after_cconv.c_1              :     0.000045s :  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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.05%
    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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000018s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000065s :  0.16%
    cse_after_recomputation.cse          :     0.000019s :  0.05%
    auto_monad_reorder                   :     0.000028s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000492s :  1.23%
    symbol_engine_optimizer.build        :     0.000012s :  0.03%
    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 :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000055s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012456s : 31.26%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000277    47
 2.85% :     0.000008s :      2: substitution.depend_value_elim
 0.93% :     0.000003s :      4: substitution.elim_not_effective
 0.74% :     0.000002s :      4: substitution.fold_const_symbol
 2.45% :     0.000007s :      5: substitution.graph_param_transform
76.88% :     0.000213s :      4: substitution.inline
 1.63% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.47% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.61% :     0.000004s :      2: substitution.replace_old_param
 4.51% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 2.90% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.03% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021814     2
93.55% :     0.020407s :      1: type_inference.infer
 6.45% :     0.001407s :      1: type_inference.specialize
------[replace.]   0.000090     6
67.20% :     0.000061s :      4: replace.inline
32.80% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     6
94.87% :     0.000210s :      4: match.inline
 5.13% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000167  1021
 1.36% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.26% :     0.000002s :     15: predicate.addn_zero_filter
 1.88% :     0.000003s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.46% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     15: predicate.compare_switch_simplify
 1.37% :     0.000002s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.90% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.63% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     15: predicate.environ_get_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.06% :     0.000002s :     10: predicate.get_grad_eliminate
 0.23% :     0.000000s :      5: predicate.graph_param_transform
 5.52% :     0.000009s :     31: predicate.inline
 1.11% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.50% :     0.000003s :     10: predicate.less_batch_normalization
 1.54% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.11% :     0.000004s :     22: predicate.load_eliminater
 1.17% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.51% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.78% :     0.000003s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.39% :     0.000004s :     22: predicate.partial_eliminate
 1.18% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.66% :     0.000003s :     15: predicate.reduce_eliminate
 1.59% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.72% :     0.000003s :     27: predicate.replace_applicator
 0.71% :     0.000001s :     10: predicate.replace_old_param
 0.36% :     0.000001s :      5: predicate.reset_defer_inline
 1.20% :     0.000002s :     15: predicate.reshape_eliminate
 1.27% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.03% :     0.000002s :     10: predicate.special_op_eliminate
 1.11% :     0.000002s :     10: predicate.specialize_transform
 1.48% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.15% :     0.000004s :     21: predicate.switch_defer_inline
 1.94% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.63% :     0.000011s :     68: predicate.switch_simplify
 1.26% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.58% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.87% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.28% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     15: predicate.value_based_eliminate
 0.45% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000837    13
45.10% :     0.000378s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.31% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.58% :     0.000407s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046344    76
 0.15% :     0.000068s :      1: add_recomputation
 0.33% :     0.000154s :      1: auto_monad
 0.07% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.64% :     0.000298s :      1: bootstrap
 0.06% :     0.000030s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.06% :     0.000026s :      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
13.38% :     0.006199s :      1: jit_opt_a
 0.41% :     0.000191s :      1: jit_opt_after_cconv
 0.15% :     0.000068s :      1: jit_opt_b
 0.97% :     0.000451s :      1: loop_unroll
 1.21% :     0.000563s :      1: mutable_eliminate
 2.53% :     0.001172s :     26: opt.transform.jit_opt_a
 0.17% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000040s :      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% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.08% :     0.000500s :      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
 0.27% :     0.000125s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000020s :      1: remove_dup_value
 1.48% :     0.000687s :      1: renormalize.infer
 0.90% :     0.000417s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.70% :     0.000322s :      1: rewriter_after_opt_a
 0.22% :     0.000102s :      1: rewriter_before_opt_a
 0.22% :     0.000102s :      1: symbol_engine_optimizer
26.91% :     0.012472s :      1: task_emit
47.25% :     0.021900s :      1: type_inference
 0.17% :     0.000078s :      1: validate


TotalTime = 0.014123, [30]
[bootstrap]: 0.0003143
[type_inference]: 0.0053393
[event_method]: 2.054e-05
[auto_monad]: 6.728e-05
[graph_reusing]: 5.99999e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.993e-05
[rewriter_before_opt_a]: 9.039e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00581224, [2]
    [Cycle 1]: 0.00202139, [27]
        [switch_simplify]: 6.679e-05
        [loop_unroll]: 3.701e-05
        [a_1]: 0.00066138
        [with_stream_mark]: 2.095e-05
        [recompute_prepare]: 9.54e-06
        [updatestate_depend_eliminate]: 4.79998e-06
        [updatestate_assign_eliminate]: 3.49001e-06
        [updatestate_loads_eliminate]: 3.52002e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 7.4e-06
        [updatestate_useless_node_eliminater]: 6.16e-06
        [accelerated_algorithm]: 6.41e-06
        [meta_shard_fg_expand]: 2.49999e-06
        [get_grad_eliminate_]: 5.96e-06
        [merge_forward]: 3.78001e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.522e-05
        [j_node_and_user_rematch]: 1.084e-05
        [meta_fg_expand]: 2.88e-06
        [replace_old_param]: 1.131e-05
        [inline_without_move]: 6.34999e-06
        [renormalize]: 0.00086089
        [add_forward_monad_depend]: 7.88999e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 2.103e-05
        [cse]: 3.359e-05
        [replace_applicator]: 1.739e-05
    [Cycle 2]: 0.00038683, [27]
        [switch_simplify]: 8.05999e-06
        [loop_unroll]: 6.33e-06
        [a_1]: 0.00012164
        [with_stream_mark]: 1.36e-05
        [recompute_prepare]: 6.26998e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 2.63998e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 5.98002e-06
        [updatestate_useless_node_eliminater]: 6.20002e-06
        [accelerated_algorithm]: 6.28e-06
        [meta_shard_fg_expand]: 1.74998e-06
        [get_grad_eliminate_]: 5.72999e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.618e-05
        [j_node_and_user_rematch]: 9.89999e-06
        [meta_fg_expand]: 2.11998e-06
        [replace_old_param]: 9.93998e-06
        [inline_without_move]: 5.84e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.17999e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 6.68003e-06
        [cse]: 1.577e-05
        [replace_applicator]: 6.56999e-06
[py_interpret_to_execute_after_opt_a]: 1.322e-05
[rewriter_after_opt_a]: 3.998e-05
[convert_after_rewriter]: 7.64002e-06
[order_py_execute_after_rewriter]: 5.23002e-06
[mutable_eliminate]: 0.00069691
[jit_opt_b]: 6.116e-05, [1]
    [Cycle 1]: 5.28e-05, [2]
        [frontend_op_eliminate]: 2.138e-05
        [inline_after_opt_a]: 1.827e-05
[cconv]: 2.889e-05
[loop_unroll]: 0.00046302
[jit_opt_after_cconv]: 0.00017281, [1]
    [Cycle 1]: 0.00016492, [11]
        [c_1]: 2.67e-05
        [parameter_eliminate]: 4.2e-06
        [updatestate_depend_eliminate]: 9.07001e-06
        [updatestate_assign_eliminate]: 3.74002e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.893e-05
        [call_graph_tuple_transform]: 2.354e-05
        [tuple_list_get_item_eliminator]: 6.76999e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 6.38e-06
[remove_dup_value]: 1.554e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 6.56999e-06
[add_recomputation]: 5.732e-05
[cse_after_recomputation]: 2.68e-05, [1]
    [Cycle 1]: 2.041e-05, [1]
        [cse]: 1.436e-05
[auto_monad_reorder]: 1.969e-05
[get_jit_bprop_graph]: 2.29001e-06
[rewriter_after_jit_bprop_graph]: 6.49001e-06
[opt_after_jit_grad]: 0.00049075
[symbol_engine_optimizer]: 8.093e-05, [1]
    [Cycle 1]: 7.423e-05, [6]
        [build]: 3.98999e-06
        [elim_shapecalc]: 8.10999e-06
        [elim_not_effective]: 1.547e-05
        [opt_reshape]: 7.24001e-06
        [fold_const_symbol]: 1.024e-05
        [renormalize]: 6.39993e-07
[validate]: 4.474e-05
Sums
    bootstrap                            :     0.000314s :  3.11%
    type_inference                       :     0.005339s : 52.85%
    event_method                         :     0.000021s :  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              :     0.000040s :  0.40%
    rewriter_before_opt_a                :     0.000090s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000075s :  0.74%
    jit_opt_a.loop_unroll                :     0.000043s :  0.43%
    jit_opt_a.a_1                        :     0.000783s :  7.75%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000016s :  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 :     0.000006s :  0.06%
    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 :     0.000012s :  0.12%
    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_        :     0.000012s :  0.12%
    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 :     0.000031s :  0.31%
    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          :     0.000021s :  0.21%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000861s :  8.52%
    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      :     0.000028s :  0.27%
    jit_opt_a.cse                        :     0.000049s :  0.49%
    jit_opt_a.replace_applicator         :     0.000024s :  0.24%
    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                    :     0.000697s :  6.90%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.18%
    cconv                                :     0.000029s :  0.29%
    loop_unroll                          :     0.000463s :  4.58%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.26%
    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 :     0.000004s :  0.04%
    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 :     0.000024s :  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  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000007s :  0.07%
    add_recomputation                    :     0.000057s :  0.57%
    cse_after_recomputation.cse          :     0.000014s :  0.14%
    auto_monad_reorder                   :     0.000020s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000491s :  4.86%
    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  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.44%


Time group info:
------[substitution.]   0.000229    28
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 2.81% :     0.000006s :      4: substitution.graph_param_transform
80.96% :     0.000185s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.57% :     0.000006s :      4: substitution.replace_old_param
 8.15% :     0.000019s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005271     2
81.04% :     0.004272s :      1: type_inference.infer
18.96% :     0.000999s :      1: type_inference.specialize
------[replace.]   0.000090     8
56.78% :     0.000051s :      4: replace.inline
43.22% :     0.000039s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     8
91.73% :     0.000182s :      4: match.inline
 8.27% :     0.000016s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000150   912
 1.26% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.05% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.03% :     0.000002s :     13: predicate.addn_check_dump
 1.35% :     0.000002s :     13: predicate.addn_zero_filter
 1.95% :     0.000003s :     13: predicate.arithmetic_simplify
 1.24% :     0.000002s :     13: predicate.cast_eliminate
 0.49% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     13: predicate.compare_switch_simplify
 1.13% :     0.000002s :     13: predicate.depend_value_elim
 1.05% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      4: predicate.elim_not_effective
 0.53% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.07% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     13: predicate.environ_get_eliminate
 1.03% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      4: predicate.fold_const_symbol
 1.09% :     0.000002s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.35% :     0.000008s :     29: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.12% :     0.000002s :      8: predicate.less_batch_normalization
 1.79% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     21: predicate.load_eliminater
 1.24% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.12% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.77% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000002s :     13: predicate.merge_addn
 1.11% :     0.000002s :     13: predicate.minmaximum_grad
 1.62% :     0.000002s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.36% :     0.000004s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 2.02% :     0.000003s :     13: predicate.reduce_eliminate
 1.51% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.87% :     0.000003s :     25: predicate.replace_applicator
 0.59% :     0.000001s :      8: predicate.replace_old_param
 0.51% :     0.000001s :      4: predicate.reset_defer_inline
 1.19% :     0.000002s :     13: predicate.reshape_eliminate
 1.14% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 1.19% :     0.000002s :      4: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      8: predicate.special_op_eliminate
 0.95% :     0.000001s :      8: predicate.specialize_transform
 1.56% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.61% :     0.000004s :     21: predicate.switch_defer_inline
 2.10% :     0.000003s :     21: predicate.switch_layer_defer_inline
 8.16% :     0.000012s :     67: predicate.switch_simplify
 1.23% :     0.000002s :     13: predicate.tile_eliminate
 1.26% :     0.000002s :     13: predicate.transpose_eliminate
 1.37% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.47% :     0.000007s :     25: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.72% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000005s :     29: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000676    12
46.96% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.84% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.20% :     0.000326s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016092    72
 0.38% :     0.000060s :      1: add_recomputation
 0.44% :     0.000070s :      1: auto_monad
 0.14% :     0.000023s :      1: auto_monad_reorder
 2.04% :     0.000329s :      1: bootstrap
 0.20% :     0.000032s :      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% :     0.000024s :      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
36.14% :     0.005816s :      1: jit_opt_a
 1.09% :     0.000176s :      1: jit_opt_after_cconv
 0.40% :     0.000064s :      1: jit_opt_b
 2.93% :     0.000472s :      1: loop_unroll
 4.39% :     0.000707s :      1: mutable_eliminate
 6.50% :     0.001046s :     26: opt.transform.jit_opt_a
 0.37% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000033s :      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% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.11% :     0.000500s :      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% :     0.000043s :      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
 2.89% :     0.000465s :      1: renormalize.infer
 2.40% :     0.000386s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000043s :      1: rewriter_after_opt_a
 0.58% :     0.000093s :      1: rewriter_before_opt_a
 0.52% :     0.000084s :      1: symbol_engine_optimizer
33.25% :     0.005351s :      1: type_inference


TotalTime = 0.0135418, [30]
[bootstrap]: 0.00029511
[type_inference]: 0.00497436
[event_method]: 2.106e-05
[auto_monad]: 4.941e-05
[graph_reusing]: 4.38999e-06
[pre_auto_parallel]: 1.78002e-06
[py_interpret_to_execute]: 3.971e-05
[rewriter_before_opt_a]: 8.366e-05
[expand_dump_flag]: 2.63998e-06
[jit_opt_a]: 0.00564134, [2]
    [Cycle 1]: 0.00188868, [27]
        [switch_simplify]: 6.21e-05
        [loop_unroll]: 3.656e-05
        [a_1]: 0.00064612
        [with_stream_mark]: 1.625e-05
        [recompute_prepare]: 8.06001e-06
        [updatestate_depend_eliminate]: 4.08001e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 7.5e-06
        [updatestate_useless_node_eliminater]: 6.19001e-06
        [accelerated_algorithm]: 7.21001e-06
        [meta_shard_fg_expand]: 1.86998e-06
        [get_grad_eliminate_]: 6.14999e-06
        [merge_forward]: 2.84001e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.386e-05
        [j_node_and_user_rematch]: 1.052e-05
        [meta_fg_expand]: 2.37001e-06
        [replace_old_param]: 1.025e-05
        [inline_without_move]: 6.43e-06
        [renormalize]: 0.00077927
        [add_forward_monad_depend]: 6.34001e-06
        [auto_monad_grad]: 1.92001e-06
        [auto_monad_eliminator]: 1.336e-05
        [cse]: 2.905e-05
        [replace_applicator]: 1.512e-05
    [Cycle 2]: 0.00045411, [27]
        [switch_simplify]: 6.79001e-06
        [loop_unroll]: 6.23e-06
        [a_1]: 0.00012241
        [with_stream_mark]: 1.084e-05
        [recompute_prepare]: 6.68e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.12999e-06
        [specialize_transform]: 5.88002e-06
        [updatestate_useless_node_eliminater]: 5.97001e-06
        [accelerated_algorithm]: 7.11001e-06
        [meta_shard_fg_expand]: 1.65001e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.471e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 1.035e-05
        [inline_without_move]: 5.71e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.74e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 6.95998e-06
        [cse]: 1.597e-05
        [replace_applicator]: 5.81003e-06
[py_interpret_to_execute_after_opt_a]: 1.318e-05
[rewriter_after_opt_a]: 3.44e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 5.05999e-06
[mutable_eliminate]: 0.00059675
[jit_opt_b]: 6.736e-05, [1]
    [Cycle 1]: 5.992e-05, [2]
        [frontend_op_eliminate]: 2.012e-05
        [inline_after_opt_a]: 2.004e-05
[cconv]: 2.383e-05
[loop_unroll]: 0.0005225
[jit_opt_after_cconv]: 0.00017858, [1]
    [Cycle 1]: 0.00017052, [11]
        [c_1]: 2.859e-05
        [parameter_eliminate]: 4.16001e-06
        [updatestate_depend_eliminate]: 6.72002e-06
        [updatestate_assign_eliminate]: 2.99001e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [cse]: 3.328e-05
        [call_graph_tuple_transform]: 2.21e-05
        [tuple_list_get_item_eliminator]: 7.05e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 6.60002e-06
[remove_dup_value]: 1.118e-05
[partial_unused_args_eliminate]: 1.77001e-06
[environ_conv]: 5.20001e-06
[add_recomputation]: 4.566e-05
[cse_after_recomputation]: 2.731e-05, [1]
    [Cycle 1]: 2.131e-05, [1]
        [cse]: 1.538e-05
[auto_monad_reorder]: 1.575e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 5.57001e-06
[opt_after_jit_grad]: 0.0005653
[symbol_engine_optimizer]: 8.077e-05, [1]
    [Cycle 1]: 7.383e-05, [6]
        [build]: 3.7e-06
        [elim_shapecalc]: 9.17999e-06
        [elim_not_effective]: 1.427e-05
        [opt_reshape]: 7.31001e-06
        [fold_const_symbol]: 9.91e-06
        [renormalize]: 6.30011e-07
[validate]: 3.604e-05
Sums
    bootstrap                            :     0.000295s :  3.09%
    type_inference                       :     0.004974s : 52.10%
    event_method                         :     0.000021s :  0.22%
    auto_monad                           :     0.000049s :  0.52%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.42%
    rewriter_before_opt_a                :     0.000084s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.72%
    jit_opt_a.loop_unroll                :     0.000043s :  0.45%
    jit_opt_a.a_1                        :     0.000769s :  8.05%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.28%
    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 :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.12%
    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 :     0.000029s :  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          :     0.000021s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000779s :  8.16%
    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      :     0.000020s :  0.21%
    jit_opt_a.cse                        :     0.000045s :  0.47%
    jit_opt_a.replace_applicator         :     0.000021s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000034s :  0.36%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000597s :  6.25%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.21%
    cconv                                :     0.000024s :  0.25%
    loop_unroll                          :     0.000523s :  5.47%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.30%
    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              :     0.000033s :  0.35%
    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 :  0.07%
    remove_dup_value                     :     0.000011s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000046s :  0.48%
    cse_after_recomputation.cse          :     0.000015s :  0.16%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000565s :  5.92%
    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  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.38%


Time group info:
------[substitution.]   0.000218    28
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.39% :     0.000005s :      4: substitution.graph_param_transform
83.00% :     0.000181s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.39% :     0.000005s :      4: substitution.replace_old_param
 7.20% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004913     2
80.95% :     0.003977s :      1: type_inference.infer
19.05% :     0.000936s :      1: type_inference.specialize
------[replace.]   0.000092     8
56.41% :     0.000052s :      4: replace.inline
43.59% :     0.000040s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     8
92.86% :     0.000178s :      4: match.inline
 7.14% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000148   912
 1.22% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     13: predicate.addn_check_dump
 1.24% :     0.000002s :     13: predicate.addn_zero_filter
 1.98% :     0.000003s :     13: predicate.arithmetic_simplify
 1.57% :     0.000002s :     13: predicate.cast_eliminate
 0.50% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.05% :     0.000002s :     13: predicate.compare_switch_simplify
 1.16% :     0.000002s :     13: predicate.depend_value_elim
 1.10% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      4: predicate.elim_not_effective
 0.49% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.07% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.49% :     0.000002s :     13: predicate.environ_get_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 1.06% :     0.000002s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.11% :     0.000008s :     29: predicate.inline
 0.92% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      8: predicate.less_batch_normalization
 1.77% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.01% :     0.000003s :     21: predicate.load_eliminater
 1.93% :     0.000003s :      4: predicate.loop_unroll_after_grad
 4.27% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.85% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000002s :     13: predicate.merge_addn
 1.07% :     0.000002s :     13: predicate.minmaximum_grad
 1.76% :     0.000003s :      4: predicate.mutable_eliminate
 0.51% :     0.000001s :      4: predicate.opt_reshape
 2.34% :     0.000003s :     21: predicate.partial_eliminate
 1.33% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     13: predicate.reduce_eliminate
 1.64% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.86% :     0.000003s :     25: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000001s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.10% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     13: predicate.same_eliminate
 0.64% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      8: predicate.special_op_eliminate
 0.95% :     0.000001s :      8: predicate.specialize_transform
 1.43% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.38% :     0.000004s :     21: predicate.switch_defer_inline
 2.01% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.45% :     0.000011s :     67: predicate.switch_simplify
 1.30% :     0.000002s :     13: predicate.tile_eliminate
 1.14% :     0.000002s :     13: predicate.transpose_eliminate
 1.45% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.72% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000640    12
46.63% :     0.000298s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.93% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.44% :     0.000310s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015433    72
 0.32% :     0.000049s :      1: add_recomputation
 0.34% :     0.000053s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.02% :     0.000311s :      1: bootstrap
 0.17% :     0.000026s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000030s :      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
 0.04% :     0.000007s :      1: graph_reusing
36.58% :     0.005645s :      1: jit_opt_a
 1.18% :     0.000182s :      1: jit_opt_after_cconv
 0.46% :     0.000070s :      1: jit_opt_b
 3.44% :     0.000532s :      1: loop_unroll
 3.93% :     0.000606s :      1: mutable_eliminate
 6.62% :     0.001022s :     26: opt.transform.jit_opt_a
 0.39% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000033s :      4: opt.transform.jit_opt_b
 0.12% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.19% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.74% :     0.000577s :      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
 0.28% :     0.000043s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000014s :      1: remove_dup_value
 2.61% :     0.000402s :      1: renormalize.infer
 2.39% :     0.000369s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000037s :      1: rewriter_after_opt_a
 0.56% :     0.000086s :      1: rewriter_before_opt_a
 0.54% :     0.000084s :      1: symbol_engine_optimizer
32.32% :     0.004988s :      1: type_inference


TotalTime = 0.0148774, [30]
[bootstrap]: 0.00030721
[type_inference]: 0.005538
[event_method]: 2.269e-05
[auto_monad]: 6.614e-05
[graph_reusing]: 6.04999e-06
[pre_auto_parallel]: 2.41998e-06
[py_interpret_to_execute]: 4.591e-05
[rewriter_before_opt_a]: 8.946e-05
[expand_dump_flag]: 2.88998e-06
[jit_opt_a]: 0.00615726, [2]
    [Cycle 1]: 0.0020826, [27]
        [switch_simplify]: 6.481e-05
        [loop_unroll]: 3.654e-05
        [a_1]: 0.00069117
        [with_stream_mark]: 2.24e-05
        [recompute_prepare]: 8.67e-06
        [updatestate_depend_eliminate]: 4.69002e-06
        [updatestate_assign_eliminate]: 3.8e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 2.19999e-06
        [specialize_transform]: 7.23e-06
        [updatestate_useless_node_eliminater]: 6.17999e-06
        [accelerated_algorithm]: 6.47001e-06
        [meta_shard_fg_expand]: 2.54999e-06
        [get_grad_eliminate_]: 6.29001e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.435e-05
        [j_node_and_user_rematch]: 1.042e-05
        [meta_fg_expand]: 2.38998e-06
        [replace_old_param]: 1.214e-05
        [inline_without_move]: 6.28e-06
        [renormalize]: 0.00090802
        [add_forward_monad_depend]: 5.92001e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.47e-05
        [cse]: 3.284e-05
        [replace_applicator]: 1.545e-05
    [Cycle 2]: 0.00037555, [27]
        [switch_simplify]: 7.01001e-06
        [loop_unroll]: 6.14999e-06
        [a_1]: 0.00011908
        [with_stream_mark]: 1.122e-05
        [recompute_prepare]: 6.16e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 5.74e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 5.65001e-06
        [merge_forward]: 3.3e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.497e-05
        [j_node_and_user_rematch]: 9.23002e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 1.032e-05
        [inline_without_move]: 5.64e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.02998e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 7.06001e-06
        [cse]: 1.369e-05
        [replace_applicator]: 6.12001e-06
[py_interpret_to_execute_after_opt_a]: 1.66e-05
[rewriter_after_opt_a]: 3.712e-05
[convert_after_rewriter]: 7.36999e-06
[order_py_execute_after_rewriter]: 5.44998e-06
[mutable_eliminate]: 0.00085477
[jit_opt_b]: 5.825e-05, [1]
    [Cycle 1]: 5.076e-05, [2]
        [frontend_op_eliminate]: 1.948e-05
        [inline_after_opt_a]: 1.925e-05
[cconv]: 2.999e-05
[loop_unroll]: 0.00048225
[jit_opt_after_cconv]: 0.00017102, [1]
    [Cycle 1]: 0.00016432, [11]
        [c_1]: 2.936e-05
        [parameter_eliminate]: 3.18e-06
        [updatestate_depend_eliminate]: 7.51999e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [cse]: 2.804e-05
        [call_graph_tuple_transform]: 2.292e-05
        [tuple_list_get_item_eliminator]: 6.51999e-06
        [none_parameter_eliminate]: 1.82001e-06
        [renormalize]: 2.19996e-07
        [switch_simplify]: 6.28e-06
[remove_dup_value]: 1.581e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 6.46999e-06
[add_recomputation]: 5.405e-05
[cse_after_recomputation]: 2.574e-05, [1]
    [Cycle 1]: 1.969e-05, [1]
        [cse]: 1.355e-05
[auto_monad_reorder]: 1.879e-05
[get_jit_bprop_graph]: 2.15002e-06
[rewriter_after_jit_bprop_graph]: 8.61002e-06
[opt_after_jit_grad]: 0.00051691
[symbol_engine_optimizer]: 8.48e-05, [1]
    [Cycle 1]: 7.775e-05, [6]
        [build]: 4.26001e-06
        [elim_shapecalc]: 9.05001e-06
        [elim_not_effective]: 1.706e-05
        [opt_reshape]: 6.96999e-06
        [fold_const_symbol]: 1.047e-05
        [renormalize]: 9.60019e-07
[validate]: 5.184e-05
Sums
    bootstrap                            :     0.000307s :  2.91%
    type_inference                       :     0.005538s : 52.44%
    event_method                         :     0.000023s :  0.21%
    auto_monad                           :     0.000066s :  0.63%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.43%
    rewriter_before_opt_a                :     0.000089s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000072s :  0.68%
    jit_opt_a.loop_unroll                :     0.000043s :  0.40%
    jit_opt_a.a_1                        :     0.000810s :  7.67%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.32%
    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 :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.11%
    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_        :     0.000012s :  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 :  0.28%
    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          :     0.000022s :  0.21%
    jit_opt_a.inline_without_move        :     0.000012s :  0.11%
    jit_opt_a.renormalize                :     0.000908s :  8.60%
    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      :     0.000022s :  0.21%
    jit_opt_a.cse                        :     0.000047s :  0.44%
    jit_opt_a.replace_applicator         :     0.000022s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.16%
    rewriter_after_opt_a                 :     0.000037s :  0.35%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000855s :  8.09%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.18%
    cconv                                :     0.000030s :  0.28%
    loop_unroll                          :     0.000482s :  4.57%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.28%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    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 :     0.000007s :  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  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000054s :  0.51%
    cse_after_recomputation.cse          :     0.000014s :  0.13%
    auto_monad_reorder                   :     0.000019s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.08%
    opt_after_jit_grad                   :     0.000517s :  4.89%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    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 :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000052s :  0.49%


Time group info:
------[substitution.]   0.000262    28
 1.01% :     0.000003s :      2: substitution.elim_not_effective
 0.62% :     0.000002s :      2: substitution.fold_const_symbol
 2.40% :     0.000006s :      4: substitution.graph_param_transform
83.46% :     0.000219s :      4: substitution.inline
 1.51% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.49% :     0.000007s :      4: substitution.replace_old_param
 6.66% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005469     2
81.36% :     0.004449s :      1: type_inference.infer
18.64% :     0.001020s :      1: type_inference.specialize
------[replace.]   0.000095     8
57.44% :     0.000055s :      4: replace.inline
42.56% :     0.000041s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000232     8
93.29% :     0.000216s :      4: match.inline
 6.71% :     0.000016s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000146   912
 1.24% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.09% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.03% :     0.000002s :     13: predicate.addn_check_dump
 1.41% :     0.000002s :     13: predicate.addn_zero_filter
 1.90% :     0.000003s :     13: predicate.arithmetic_simplify
 1.44% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     13: predicate.compare_switch_simplify
 1.23% :     0.000002s :     13: predicate.depend_value_elim
 1.05% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      4: predicate.elim_not_effective
 0.62% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     13: predicate.environ_get_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 1.03% :     0.000002s :      8: predicate.get_grad_eliminate
 0.32% :     0.000000s :      4: predicate.graph_param_transform
 5.35% :     0.000008s :     29: predicate.inline
 0.91% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      8: predicate.less_batch_normalization
 1.72% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     21: predicate.load_eliminater
 1.74% :     0.000003s :      4: predicate.loop_unroll_after_grad
 4.21% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     13: predicate.merge_addn
 1.08% :     0.000002s :     13: predicate.minmaximum_grad
 1.74% :     0.000003s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.33% :     0.000003s :     21: predicate.partial_eliminate
 1.15% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.72% :     0.000003s :     13: predicate.reduce_eliminate
 1.74% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.95% :     0.000003s :     25: predicate.replace_applicator
 0.94% :     0.000001s :      8: predicate.replace_old_param
 0.38% :     0.000001s :      4: predicate.reset_defer_inline
 1.18% :     0.000002s :     13: predicate.reshape_eliminate
 1.13% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     13: predicate.same_eliminate
 0.58% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.05% :     0.000002s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.37% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.40% :     0.000004s :     21: predicate.switch_defer_inline
 2.11% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.31% :     0.000011s :     67: predicate.switch_simplify
 1.53% :     0.000002s :     13: predicate.tile_eliminate
 1.18% :     0.000002s :     13: predicate.transpose_eliminate
 1.46% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.91% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.72% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000776    12
50.09% :     0.000389s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.14% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.77% :     0.000340s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016914    72
 0.34% :     0.000057s :      1: add_recomputation
 0.41% :     0.000070s :      1: auto_monad
 0.13% :     0.000021s :      1: auto_monad_reorder
 1.89% :     0.000320s :      1: bootstrap
 0.19% :     0.000033s :      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
 0.16% :     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
36.42% :     0.006161s :      1: jit_opt_a
 1.03% :     0.000174s :      1: jit_opt_after_cconv
 0.36% :     0.000061s :      1: jit_opt_b
 2.90% :     0.000491s :      1: loop_unroll
 5.11% :     0.000865s :      1: mutable_eliminate
 6.30% :     0.001065s :     26: opt.transform.jit_opt_a
 0.36% :     0.000061s :      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
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000040s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000527s :      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% :     0.000049s :      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
 3.02% :     0.000511s :      1: renormalize.infer
 2.29% :     0.000388s :      1: renormalize.specialize
 0.06% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000040s :      1: rewriter_after_opt_a
 0.55% :     0.000093s :      1: rewriter_before_opt_a
 0.52% :     0.000088s :      1: symbol_engine_optimizer
32.83% :     0.005553s :      1: type_inference


TotalTime = 0.0453006, [33]
[bootstrap]: 0.0002873
[type_inference]: 0.0226198
[event_method]: 2.319e-05
[auto_monad]: 0.00016522
[graph_reusing]: 6.48e-06
[pre_auto_parallel]: 2.84999e-06
[py_interpret_to_execute]: 0.00013481
[rewriter_before_opt_a]: 9.981e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00652675, [2]
    [Cycle 1]: 0.00238723, [27]
        [switch_simplify]: 6.768e-05
        [loop_unroll]: 3.695e-05
        [a_1]: 0.00073005
        [with_stream_mark]: 1.968e-05
        [recompute_prepare]: 1.135e-05
        [updatestate_depend_eliminate]: 5.92999e-06
        [updatestate_assign_eliminate]: 5.71003e-06
        [updatestate_loads_eliminate]: 5.38002e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 9.23002e-06
        [updatestate_useless_node_eliminater]: 1.124e-05
        [accelerated_algorithm]: 8.74998e-06
        [meta_shard_fg_expand]: 3.08e-06
        [get_grad_eliminate_]: 8.18999e-06
        [merge_forward]: 5.28002e-06
        [cell_reuse_recompute_pass]: 1.86e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.004e-05
        [j_node_and_user_rematch]: 1.392e-05
        [meta_fg_expand]: 3.83999e-06
        [replace_old_param]: 1.191e-05
        [inline_without_move]: 8.02e-06
        [renormalize]: 0.00112642
        [add_forward_monad_depend]: 6.54001e-06
        [auto_monad_grad]: 2.47001e-06
        [auto_monad_eliminator]: 2.199e-05
        [cse]: 4.043e-05
        [replace_applicator]: 1.755e-05
    [Cycle 2]: 0.00050352, [27]
        [switch_simplify]: 9.01002e-06
        [loop_unroll]: 8.33999e-06
        [a_1]: 0.00016833
        [with_stream_mark]: 1.323e-05
        [recompute_prepare]: 8.23001e-06
        [updatestate_depend_eliminate]: 5.42999e-06
        [updatestate_assign_eliminate]: 4.63001e-06
        [updatestate_loads_eliminate]: 3.86999e-06
        [parameter_eliminate]: 1.59998e-06
        [specialize_transform]: 7.92003e-06
        [updatestate_useless_node_eliminater]: 1.046e-05
        [accelerated_algorithm]: 7.82e-06
        [meta_shard_fg_expand]: 2.00002e-06
        [get_grad_eliminate_]: 7.72998e-06
        [merge_forward]: 4.56002e-06
        [cell_reuse_recompute_pass]: 2.41e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.7e-05
        [j_node_and_user_rematch]: 1.252e-05
        [meta_fg_expand]: 2.84001e-06
        [replace_old_param]: 1.047e-05
        [inline_without_move]: 2.406e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.68002e-06
        [auto_monad_grad]: 1.22999e-06
        [auto_monad_eliminator]: 1.054e-05
        [cse]: 3.154e-05
        [replace_applicator]: 8.47e-06
[py_interpret_to_execute_after_opt_a]: 1.301e-05
[rewriter_after_opt_a]: 0.00029504
[convert_after_rewriter]: 1.153e-05
[order_py_execute_after_rewriter]: 7.97e-06
[mutable_eliminate]: 0.00067766
[jit_opt_b]: 6.944e-05, [1]
    [Cycle 1]: 6.239e-05, [2]
        [frontend_op_eliminate]: 2.505e-05
        [inline_after_opt_a]: 2.51e-05
[cconv]: 2.522e-05
[loop_unroll]: 0.00047131
[jit_opt_after_cconv]: 0.00020098, [1]
    [Cycle 1]: 0.00019386, [11]
        [c_1]: 4.596e-05
        [parameter_eliminate]: 3.12002e-06
        [updatestate_depend_eliminate]: 8.40999e-06
        [updatestate_assign_eliminate]: 4.53001e-06
        [updatestate_loads_eliminate]: 3.91999e-06
        [cse]: 3.265e-05
        [call_graph_tuple_transform]: 2.278e-05
        [tuple_list_get_item_eliminator]: 8.42e-06
        [none_parameter_eliminate]: 1.71002e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 9.69e-06
[remove_dup_value]: 2.212e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 1.301e-05
[add_recomputation]: 7.127e-05
[cse_after_recomputation]: 3.078e-05, [1]
    [Cycle 1]: 2.462e-05, [1]
        [cse]: 1.869e-05
[auto_monad_reorder]: 2.531e-05
[get_jit_bprop_graph]: 1.84e-06
[rewriter_after_jit_bprop_graph]: 5.51002e-06
[opt_after_jit_grad]: 0.00053528
[symbol_engine_optimizer]: 0.00010912, [1]
    [Cycle 1]: 0.00010235, [6]
        [build]: 1.463e-05
        [elim_shapecalc]: 1.223e-05
        [elim_not_effective]: 2.035e-05
        [opt_reshape]: 9.65002e-06
        [fold_const_symbol]: 1.477e-05
        [renormalize]: 8.29983e-07
[validate]: 6.347e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.0125525
[execute]: 5.88998e-06
Sums
    bootstrap                            :     0.000287s :  0.70%
    type_inference                       :     0.022620s : 55.15%
    event_method                         :     0.000023s :  0.06%
    auto_monad                           :     0.000165s :  0.40%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000135s :  0.33%
    rewriter_before_opt_a                :     0.000100s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000898s :  2.19%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.05%
    jit_opt_a.inline_without_move        :     0.000032s :  0.08%
    jit_opt_a.renormalize                :     0.001126s :  2.75%
    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      :     0.000033s :  0.08%
    jit_opt_a.cse                        :     0.000072s :  0.18%
    jit_opt_a.replace_applicator         :     0.000026s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000295s :  0.72%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000678s :  1.65%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000025s :  0.06%
    loop_unroll                          :     0.000471s :  1.15%
    jit_opt_after_cconv.c_1              :     0.000046s :  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 :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000022s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000071s :  0.17%
    cse_after_recomputation.cse          :     0.000019s :  0.05%
    auto_monad_reorder                   :     0.000025s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000535s :  1.30%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000063s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012552s : 30.60%
    execute                              :     0.000006s :  0.01%


Time group info:
------[substitution.]   0.000287    47
 3.30% :     0.000010s :      2: substitution.depend_value_elim
 1.13% :     0.000003s :      4: substitution.elim_not_effective
 0.84% :     0.000002s :      4: substitution.fold_const_symbol
 2.26% :     0.000006s :      5: substitution.graph_param_transform
76.02% :     0.000219s :      4: substitution.inline
 1.66% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.51% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.73% :     0.000005s :      2: substitution.replace_old_param
 4.33% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.14% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.08% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022543     2
92.76% :     0.020912s :      1: type_inference.infer
 7.24% :     0.001631s :      1: type_inference.specialize
------[replace.]   0.000097     6
67.54% :     0.000066s :      4: replace.inline
32.46% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     6
95.02% :     0.000216s :      4: match.inline
 4.98% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000192  1021
 1.16% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.43% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.04% :     0.000002s :     15: predicate.addn_check_dump
 1.23% :     0.000002s :     15: predicate.addn_zero_filter
 1.93% :     0.000004s :     15: predicate.arithmetic_simplify
 1.32% :     0.000003s :     15: predicate.cast_eliminate
 0.53% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     15: predicate.compare_switch_simplify
 1.26% :     0.000002s :     15: predicate.depend_value_elim
 1.11% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.08% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.07% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.90% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      5: predicate.elim_not_effective
 0.57% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.12% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000001s :      5: predicate.graph_param_transform
 4.85% :     0.000009s :     31: predicate.inline
 9.17% :     0.000018s :     10: predicate.inline_without_move
 0.38% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.33% :     0.000003s :     10: predicate.less_batch_normalization
 1.37% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.73% :     0.000003s :     22: predicate.load_eliminater
 1.35% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.19% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.66% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000002s :     15: predicate.merge_addn
 1.01% :     0.000002s :     15: predicate.minmaximum_grad
 1.41% :     0.000003s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.18% :     0.000004s :     22: predicate.partial_eliminate
 1.12% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.66% :     0.000003s :     15: predicate.reduce_eliminate
 1.30% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.60% :     0.000003s :     27: predicate.replace_applicator
 0.74% :     0.000001s :     10: predicate.replace_old_param
 0.31% :     0.000001s :      5: predicate.reset_defer_inline
 1.15% :     0.000002s :     15: predicate.reshape_eliminate
 1.19% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.42% :     0.000003s :     15: predicate.same_eliminate
 0.53% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.95% :     0.000002s :     10: predicate.special_op_eliminate
 0.92% :     0.000002s :     10: predicate.specialize_transform
 1.27% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.94% :     0.000004s :     21: predicate.switch_defer_inline
 1.80% :     0.000003s :     21: predicate.switch_layer_defer_inline
 5.95% :     0.000011s :     68: predicate.switch_simplify
 1.16% :     0.000002s :     15: predicate.tile_eliminate
 1.20% :     0.000002s :     15: predicate.transpose_eliminate
 1.37% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.26% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.62% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000812    13
43.02% :     0.000349s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.20% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.78% :     0.000421s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047844    76
 0.16% :     0.000075s :      1: add_recomputation
 0.36% :     0.000171s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.63% :     0.000300s :      1: bootstrap
 0.06% :     0.000028s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.06% :     0.000028s :      1: event_method
 0.02% :     0.000010s :      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
13.65% :     0.006531s :      1: jit_opt_a
 0.43% :     0.000204s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 1.00% :     0.000480s :      1: loop_unroll
 1.43% :     0.000686s :      1: mutable_eliminate
 2.56% :     0.001224s :     26: opt.transform.jit_opt_a
 0.17% :     0.000083s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.14% :     0.000544s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000139s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000025s :      1: remove_dup_value
 1.40% :     0.000671s :      1: renormalize.infer
 0.93% :     0.000445s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000300s :      1: rewriter_after_opt_a
 0.22% :     0.000103s :      1: rewriter_before_opt_a
 0.23% :     0.000112s :      1: symbol_engine_optimizer
26.26% :     0.012565s :      1: task_emit
47.31% :     0.022637s :      1: type_inference
 0.19% :     0.000089s :      1: validate


TotalTime = 0.0123674, [30]
[bootstrap]: 0.00024365
[type_inference]: 0.00430508
[event_method]: 2.076e-05
[auto_monad]: 5.607e-05
[graph_reusing]: 5.22e-06
[pre_auto_parallel]: 2.63003e-06
[py_interpret_to_execute]: 3.906e-05
[rewriter_before_opt_a]: 8.636e-05
[expand_dump_flag]: 2.86e-06
[jit_opt_a]: 0.00552134, [2]
    [Cycle 1]: 0.00184315, [27]
        [switch_simplify]: 6.26e-05
        [loop_unroll]: 3.62e-05
        [a_1]: 0.00062671
        [with_stream_mark]: 1.838e-05
        [recompute_prepare]: 7.78999e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 3.48e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 7.36999e-06
        [updatestate_useless_node_eliminater]: 6.28e-06
        [accelerated_algorithm]: 6.41e-06
        [meta_shard_fg_expand]: 2.46998e-06
        [get_grad_eliminate_]: 5.93002e-06
        [merge_forward]: 4.41002e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.505e-05
        [j_node_and_user_rematch]: 1.048e-05
        [meta_fg_expand]: 3.09001e-06
        [replace_old_param]: 1.108e-05
        [inline_without_move]: 5.94e-06
        [renormalize]: 0.00076728
        [add_forward_monad_depend]: 5.34e-06
        [auto_monad_grad]: 2.59001e-06
        [auto_monad_eliminator]: 1.599e-05
        [cse]: 2.717e-05
        [replace_applicator]: 1.411e-05
    [Cycle 2]: 0.00036202, [27]
        [switch_simplify]: 7.39002e-06
        [loop_unroll]: 5.82001e-06
        [a_1]: 0.00011388
        [with_stream_mark]: 1.055e-05
        [recompute_prepare]: 5.64e-06
        [updatestate_depend_eliminate]: 3.23998e-06
        [updatestate_assign_eliminate]: 2.65997e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.09003e-06
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 5.82001e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.88002e-06
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 8.62998e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 9.79999e-06
        [inline_without_move]: 5.64e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 6.93998e-06
        [cse]: 1.221e-05
        [replace_applicator]: 6.08998e-06
[py_interpret_to_execute_after_opt_a]: 8.69998e-06
[rewriter_after_opt_a]: 3.32e-05
[convert_after_rewriter]: 8.89998e-06
[order_py_execute_after_rewriter]: 3.92998e-06
[mutable_eliminate]: 0.00053947
[jit_opt_b]: 5.297e-05, [1]
    [Cycle 1]: 4.591e-05, [2]
        [frontend_op_eliminate]: 1.775e-05
        [inline_after_opt_a]: 1.704e-05
[cconv]: 2.936e-05
[loop_unroll]: 0.00041666
[jit_opt_after_cconv]: 0.0001435, [1]
    [Cycle 1]: 0.00013744, [11]
        [c_1]: 2.504e-05
        [parameter_eliminate]: 1.96e-06
        [updatestate_depend_eliminate]: 4.80999e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 1.907e-05
        [call_graph_tuple_transform]: 1.791e-05
        [tuple_list_get_item_eliminator]: 6.34001e-06
        [none_parameter_eliminate]: 7.09988e-07
        [renormalize]: 3.29979e-07
        [switch_simplify]: 6.19001e-06
[remove_dup_value]: 9.32001e-06
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.32999e-06
[add_recomputation]: 4.395e-05
[cse_after_recomputation]: 2.309e-05, [1]
    [Cycle 1]: 1.756e-05, [1]
        [cse]: 1.176e-05
[auto_monad_reorder]: 1.152e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 4.87e-06
[opt_after_jit_grad]: 0.00048669
[symbol_engine_optimizer]: 7.248e-05, [1]
    [Cycle 1]: 6.697e-05, [6]
        [build]: 2.91e-06
        [elim_shapecalc]: 8.43999e-06
        [elim_not_effective]: 1.295e-05
        [opt_reshape]: 6.24001e-06
        [fold_const_symbol]: 8.97e-06
        [renormalize]: 3.60014e-07
[validate]: 3.553e-05
Sums
    bootstrap                            :     0.000244s :  2.87%
    type_inference                       :     0.004305s : 50.63%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000056s :  0.66%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.46%
    rewriter_before_opt_a                :     0.000086s :  1.02%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000070s :  0.82%
    jit_opt_a.loop_unroll                :     0.000042s :  0.49%
    jit_opt_a.a_1                        :     0.000741s :  8.71%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.09%
    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       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000011s :  0.13%
    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 :     0.000029s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.06%
    jit_opt_a.replace_old_param          :     0.000021s :  0.25%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000767s :  9.02%
    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      :     0.000023s :  0.27%
    jit_opt_a.cse                        :     0.000039s :  0.46%
    jit_opt_a.replace_applicator         :     0.000020s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000033s :  0.39%
    convert_after_rewriter               :     0.000009s :  0.10%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000539s :  6.34%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000029s :  0.35%
    loop_unroll                          :     0.000417s :  4.90%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.29%
    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 :     0.000003s :  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 :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000009s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.52%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000012s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000487s :  5.72%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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 :  0.00%
    validate                             :     0.000036s :  0.42%


Time group info:
------[substitution.]   0.000215    28
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.40% :     0.000001s :      2: substitution.fold_const_symbol
 1.81% :     0.000004s :      4: substitution.graph_param_transform
83.52% :     0.000180s :      4: substitution.inline
 1.72% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.28% :     0.000005s :      4: substitution.replace_old_param
 7.57% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004249     2
80.13% :     0.003405s :      1: type_inference.infer
19.87% :     0.000844s :      1: type_inference.specialize
------[replace.]   0.000082     8
53.78% :     0.000044s :      4: replace.inline
46.22% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     8
92.52% :     0.000177s :      4: match.inline
 7.48% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   912
 1.20% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.25% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     13: predicate.addn_check_dump
 1.14% :     0.000002s :     13: predicate.addn_zero_filter
 2.17% :     0.000003s :     13: predicate.arithmetic_simplify
 1.37% :     0.000002s :     13: predicate.cast_eliminate
 0.41% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     13: predicate.compare_switch_simplify
 1.23% :     0.000002s :     13: predicate.depend_value_elim
 1.11% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      4: predicate.elim_not_effective
 0.51% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.08% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     13: predicate.environ_get_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.94% :     0.000001s :      8: predicate.get_grad_eliminate
 0.30% :     0.000000s :      4: predicate.graph_param_transform
 5.47% :     0.000008s :     29: predicate.inline
 0.94% :     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
 1.59% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     21: predicate.load_eliminater
 1.30% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.33% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.36% :     0.000002s :     13: predicate.merge_addn
 1.13% :     0.000002s :     13: predicate.minmaximum_grad
 1.33% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.45% :     0.000003s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.75% :     0.000001s :      8: predicate.replace_old_param
 0.31% :     0.000000s :      4: predicate.reset_defer_inline
 1.26% :     0.000002s :     13: predicate.reshape_eliminate
 1.23% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.57% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.11% :     0.000002s :      8: predicate.special_op_eliminate
 1.01% :     0.000001s :      8: predicate.specialize_transform
 1.38% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.73% :     0.000004s :     21: predicate.switch_defer_inline
 2.18% :     0.000003s :     21: predicate.switch_layer_defer_inline
 8.19% :     0.000011s :     67: predicate.switch_simplify
 1.29% :     0.000002s :     13: predicate.tile_eliminate
 1.13% :     0.000002s :     13: predicate.transpose_eliminate
 1.42% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.19% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.80% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     13: predicate.value_based_eliminate
 0.34% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000582    12
46.12% :     0.000268s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.18% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.69% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014195    72
 0.33% :     0.000047s :      1: add_recomputation
 0.42% :     0.000059s :      1: auto_monad
 0.10% :     0.000014s :      1: auto_monad_reorder
 1.80% :     0.000255s :      1: bootstrap
 0.23% :     0.000032s :      1: cconv
 0.08% :     0.000011s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.17% :     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
38.92% :     0.005524s :      1: jit_opt_a
 1.03% :     0.000146s :      1: jit_opt_after_cconv
 0.39% :     0.000055s :      1: jit_opt_b
 2.98% :     0.000423s :      1: loop_unroll
 3.85% :     0.000547s :      1: mutable_eliminate
 6.96% :     0.000988s :     26: opt.transform.jit_opt_a
 0.37% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000029s :      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% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.48% :     0.000494s :      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% :     0.000042s :      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
 2.98% :     0.000422s :      1: renormalize.infer
 2.37% :     0.000337s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000036s :      1: rewriter_after_opt_a
 0.63% :     0.000089s :      1: rewriter_before_opt_a
 0.53% :     0.000075s :      1: symbol_engine_optimizer
30.40% :     0.004316s :      1: type_inference


TotalTime = 0.0122892, [30]
[bootstrap]: 0.00025932
[type_inference]: 0.00457194
[event_method]: 1.979e-05
[auto_monad]: 5.935e-05
[graph_reusing]: 5.46998e-06
[pre_auto_parallel]: 2.38998e-06
[py_interpret_to_execute]: 3.592e-05
[rewriter_before_opt_a]: 8.394e-05
[expand_dump_flag]: 2.79999e-06
[jit_opt_a]: 0.00520533, [2]
    [Cycle 1]: 0.0017601, [27]
        [switch_simplify]: 5.397e-05
        [loop_unroll]: 3.607e-05
        [a_1]: 0.00062855
        [with_stream_mark]: 1.598e-05
        [recompute_prepare]: 8.30999e-06
        [updatestate_depend_eliminate]: 4.11001e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 6.98e-06
        [updatestate_useless_node_eliminater]: 5.96e-06
        [accelerated_algorithm]: 6.31e-06
        [meta_shard_fg_expand]: 2.16998e-06
        [get_grad_eliminate_]: 6.00002e-06
        [merge_forward]: 3.55e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.451e-05
        [j_node_and_user_rematch]: 9.87001e-06
        [meta_fg_expand]: 2.84001e-06
        [replace_old_param]: 1.078e-05
        [inline_without_move]: 6.16e-06
        [renormalize]: 0.00071483
        [add_forward_monad_depend]: 4.47e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.338e-05
        [cse]: 2.149e-05
        [replace_applicator]: 1.405e-05
    [Cycle 2]: 0.00039634, [27]
        [switch_simplify]: 6.79001e-06
        [loop_unroll]: 5.97001e-06
        [a_1]: 0.0001448
        [with_stream_mark]: 9.82999e-06
        [recompute_prepare]: 6.35002e-06
        [updatestate_depend_eliminate]: 3.27002e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.25002e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 5.76003e-06
        [updatestate_useless_node_eliminater]: 5.80002e-06
        [accelerated_algorithm]: 6.42001e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 5.38002e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.366e-05
        [j_node_and_user_rematch]: 9.09998e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 9.29998e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.82999e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 7.01001e-06
        [cse]: 1.473e-05
        [replace_applicator]: 5.86e-06
[py_interpret_to_execute_after_opt_a]: 1.142e-05
[rewriter_after_opt_a]: 3.142e-05
[convert_after_rewriter]: 6.94999e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00048926
[jit_opt_b]: 5.335e-05, [1]
    [Cycle 1]: 4.76e-05, [2]
        [frontend_op_eliminate]: 1.82e-05
        [inline_after_opt_a]: 1.813e-05
[cconv]: 2.345e-05
[loop_unroll]: 0.00041127
[jit_opt_after_cconv]: 0.00015954, [1]
    [Cycle 1]: 0.00015344, [11]
        [c_1]: 2.527e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 6.26998e-06
        [updatestate_assign_eliminate]: 2.83998e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [cse]: 2.224e-05
        [call_graph_tuple_transform]: 2.089e-05
        [tuple_list_get_item_eliminator]: 6.71e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 9.50007e-07
        [switch_simplify]: 6.20002e-06
[remove_dup_value]: 1.377e-05
[partial_unused_args_eliminate]: 2.91999e-06
[environ_conv]: 6.16e-06
[add_recomputation]: 4.55e-05
[cse_after_recomputation]: 2.485e-05, [1]
    [Cycle 1]: 1.912e-05, [1]
        [cse]: 1.309e-05
[auto_monad_reorder]: 1.8e-05
[get_jit_bprop_graph]: 1.92999e-06
[rewriter_after_jit_bprop_graph]: 5.01002e-06
[opt_after_jit_grad]: 0.00046801
[symbol_engine_optimizer]: 7.507e-05, [1]
    [Cycle 1]: 6.925e-05, [6]
        [build]: 2.94001e-06
        [elim_shapecalc]: 8.60001e-06
        [elim_not_effective]: 1.377e-05
        [opt_reshape]: 6.34001e-06
        [fold_const_symbol]: 9.15999e-06
        [renormalize]: 4.40021e-07
[validate]: 3.713e-05
Sums
    bootstrap                            :     0.000259s :  2.98%
    type_inference                       :     0.004572s : 52.61%
    event_method                         :     0.000020s :  0.23%
    auto_monad                           :     0.000059s :  0.68%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.41%
    rewriter_before_opt_a                :     0.000084s :  0.97%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.70%
    jit_opt_a.loop_unroll                :     0.000042s :  0.48%
    jit_opt_a.a_1                        :     0.000773s :  8.90%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.30%
    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 :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    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  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.06%
    jit_opt_a.replace_old_param          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000715s :  8.23%
    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      :     0.000020s :  0.23%
    jit_opt_a.cse                        :     0.000036s :  0.42%
    jit_opt_a.replace_applicator         :     0.000020s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000031s :  0.36%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000489s :  5.63%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.21%
    cconv                                :     0.000023s :  0.27%
    loop_unroll                          :     0.000411s :  4.73%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.29%
    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              :     0.000022s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  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      :     0.000001s :  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                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000045s :  0.52%
    cse_after_recomputation.cse          :     0.000013s :  0.15%
    auto_monad_reorder                   :     0.000018s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000468s :  5.39%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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  :     0.000000s :  0.01%
    validate                             :     0.000037s :  0.43%


Time group info:
------[substitution.]   0.000219    28
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000006s :      4: substitution.graph_param_transform
83.13% :     0.000182s :      4: substitution.inline
 1.37% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.97% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000005s :      4: substitution.replace_old_param
 7.31% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004521     2
79.98% :     0.003615s :      1: type_inference.infer
20.02% :     0.000905s :      1: type_inference.specialize
------[replace.]   0.000085     8
57.28% :     0.000048s :      4: replace.inline
42.72% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000194     8
92.77% :     0.000180s :      4: match.inline
 7.23% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000165   912
 1.19% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.88% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     13: predicate.addn_check_dump
 1.05% :     0.000002s :     13: predicate.addn_zero_filter
 1.56% :     0.000003s :     13: predicate.arithmetic_simplify
 1.05% :     0.000002s :     13: predicate.cast_eliminate
 0.40% :     0.000001s :      4: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     13: predicate.compare_switch_simplify
 1.05% :     0.000002s :     13: predicate.depend_value_elim
 0.93% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.02% :     0.000002s :     13: predicate.dict_get_item_eliminator
 0.96% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      4: predicate.elim_not_effective
 0.41% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 0.99% :     0.000002s :     13: predicate.environ_add_const_eliminate
 0.98% :     0.000002s :     13: predicate.environ_get_add_eliminate
 0.96% :     0.000002s :     13: predicate.environ_get_depend_swap
18.71% :     0.000031s :     13: predicate.environ_get_eliminate
 0.96% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      4: predicate.fold_const_symbol
 0.83% :     0.000001s :      8: predicate.get_grad_eliminate
 0.22% :     0.000000s :      4: predicate.graph_param_transform
 4.31% :     0.000007s :     29: predicate.inline
 0.81% :     0.000001s :      8: predicate.inline_without_move
 0.36% :     0.000001s :      8: predicate.j_node_and_user_rematch
 0.87% :     0.000001s :      8: predicate.less_batch_normalization
 1.35% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.64% :     0.000003s :     21: predicate.load_eliminater
 1.29% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.51% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.51% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000002s :     13: predicate.merge_addn
 0.89% :     0.000001s :     13: predicate.minmaximum_grad
 1.45% :     0.000002s :      4: predicate.mutable_eliminate
 0.36% :     0.000001s :      4: predicate.opt_reshape
 2.11% :     0.000003s :     21: predicate.partial_eliminate
 0.98% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.31% :     0.000002s :     13: predicate.reduce_eliminate
 1.38% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.64% :     0.000003s :     25: predicate.replace_applicator
 0.60% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000001s :      4: predicate.reset_defer_inline
 1.05% :     0.000002s :     13: predicate.reshape_eliminate
 0.97% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.53% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.04% :     0.000002s :     13: predicate.same_eliminate
 0.49% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      8: predicate.special_op_eliminate
 0.79% :     0.000001s :      8: predicate.specialize_transform
 1.11% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 0.95% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.05% :     0.000003s :     21: predicate.switch_defer_inline
 1.78% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.25% :     0.000010s :     67: predicate.switch_simplify
 0.96% :     0.000002s :     13: predicate.tile_eliminate
 1.00% :     0.000002s :     13: predicate.transpose_eliminate
 1.35% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.19% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.26% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.55% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.49% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.39% :     0.000002s :     13: predicate.value_based_eliminate
 0.33% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.50% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000589    12
45.49% :     0.000268s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.12% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.39% :     0.000285s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014091    72
 0.34% :     0.000049s :      1: add_recomputation
 0.44% :     0.000063s :      1: auto_monad
 0.15% :     0.000021s :      1: auto_monad_reorder
 1.91% :     0.000268s :      1: bootstrap
 0.19% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000027s :      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% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
36.96% :     0.005208s :      1: jit_opt_a
 1.15% :     0.000162s :      1: jit_opt_after_cconv
 0.40% :     0.000056s :      1: jit_opt_b
 2.97% :     0.000418s :      1: loop_unroll
 3.52% :     0.000496s :      1: mutable_eliminate
 7.18% :     0.001012s :     26: opt.transform.jit_opt_a
 0.39% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000030s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.37% :     0.000475s :      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% :     0.000039s :      1: py_interpret_to_execute
 0.10% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.75% :     0.000388s :      1: renormalize.infer
 2.27% :     0.000319s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000035s :      1: rewriter_after_opt_a
 0.62% :     0.000087s :      1: rewriter_before_opt_a
 0.55% :     0.000078s :      1: symbol_engine_optimizer
32.51% :     0.004581s :      1: type_inference


TotalTime = 0.0469468, [33]
[bootstrap]: 0.00035265
[type_inference]: 0.024891
[event_method]: 2.173e-05
[auto_monad]: 0.00015875
[graph_reusing]: 6.98e-06
[pre_auto_parallel]: 3.48e-06
[py_interpret_to_execute]: 0.00014356
[rewriter_before_opt_a]: 0.00010146
[expand_dump_flag]: 3.47002e-06
[jit_opt_a]: 0.00669512, [2]
    [Cycle 1]: 0.0024407, [27]
        [switch_simplify]: 6.639e-05
        [loop_unroll]: 3.617e-05
        [a_1]: 0.00073491
        [with_stream_mark]: 2.106e-05
        [recompute_prepare]: 1.048e-05
        [updatestate_depend_eliminate]: 6.71e-06
        [updatestate_assign_eliminate]: 6.17999e-06
        [updatestate_loads_eliminate]: 4.48999e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 9.31e-06
        [updatestate_useless_node_eliminater]: 1.137e-05
        [accelerated_algorithm]: 9.39998e-06
        [meta_shard_fg_expand]: 2.79999e-06
        [get_grad_eliminate_]: 8.56002e-06
        [merge_forward]: 5.20999e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.057e-05
        [j_node_and_user_rematch]: 1.364e-05
        [meta_fg_expand]: 3.83001e-06
        [replace_old_param]: 1.236e-05
        [inline_without_move]: 8.17998e-06
        [renormalize]: 0.00117805
        [add_forward_monad_depend]: 5.76e-06
        [auto_monad_grad]: 2.74001e-06
        [auto_monad_eliminator]: 2.042e-05
        [cse]: 4.085e-05
        [replace_applicator]: 1.586e-05
    [Cycle 2]: 0.00046643, [27]
        [switch_simplify]: 8.70001e-06
        [loop_unroll]: 7.93999e-06
        [a_1]: 0.0001672
        [with_stream_mark]: 1.145e-05
        [recompute_prepare]: 8.43001e-06
        [updatestate_depend_eliminate]: 5.10999e-06
        [updatestate_assign_eliminate]: 5.49e-06
        [updatestate_loads_eliminate]: 3.76001e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 7.97998e-06
        [updatestate_useless_node_eliminater]: 1.057e-05
        [accelerated_algorithm]: 7.85e-06
        [meta_shard_fg_expand]: 1.97999e-06
        [get_grad_eliminate_]: 7.50998e-06
        [merge_forward]: 4.28001e-06
        [cell_reuse_recompute_pass]: 2.01998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.615e-05
        [j_node_and_user_rematch]: 1.301e-05
        [meta_fg_expand]: 2.76999e-06
        [replace_old_param]: 1.04e-05
        [inline_without_move]: 7.97e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 9.96e-06
        [cse]: 2.028e-05
        [replace_applicator]: 7.95e-06
[py_interpret_to_execute_after_opt_a]: 1.306e-05
[rewriter_after_opt_a]: 0.00028457
[convert_after_rewriter]: 1.201e-05
[order_py_execute_after_rewriter]: 7.75998e-06
[mutable_eliminate]: 0.00072045
[jit_opt_b]: 6.91e-05, [1]
    [Cycle 1]: 6.139e-05, [2]
        [frontend_op_eliminate]: 2.502e-05
        [inline_after_opt_a]: 2.425e-05
[cconv]: 4.706e-05
[loop_unroll]: 0.00049248
[jit_opt_after_cconv]: 0.00020238, [1]
    [Cycle 1]: 0.00019501, [11]
        [c_1]: 4.62e-05
        [parameter_eliminate]: 2.67001e-06
        [updatestate_depend_eliminate]: 8.57998e-06
        [updatestate_assign_eliminate]: 4.53001e-06
        [updatestate_loads_eliminate]: 3.88001e-06
        [cse]: 3.09e-05
        [call_graph_tuple_transform]: 2.38e-05
        [tuple_list_get_item_eliminator]: 8.32e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 9.06002e-06
[remove_dup_value]: 1.864e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 1.202e-05
[add_recomputation]: 6.612e-05
[cse_after_recomputation]: 3.191e-05, [1]
    [Cycle 1]: 2.48e-05, [1]
        [cse]: 1.849e-05
[auto_monad_reorder]: 2.522e-05
[get_jit_bprop_graph]: 1.83997e-06
[rewriter_after_jit_bprop_graph]: 5.54e-06
[opt_after_jit_grad]: 0.0005108
[symbol_engine_optimizer]: 0.00010579, [1]
    [Cycle 1]: 9.92e-05, [6]
        [build]: 1.259e-05
        [elim_shapecalc]: 1.212e-05
        [elim_not_effective]: 2.077e-05
        [opt_reshape]: 9.24998e-06
        [fold_const_symbol]: 1.491e-05
        [renormalize]: 3.19997e-07
[validate]: 6.972e-05
[backend_pass]: 1.14e-06
[task_emit]: 0.0115728
[execute]: 9.36e-06
Sums
    bootstrap                            :     0.000353s :  0.83%
    type_inference                       :     0.024891s : 58.64%
    event_method                         :     0.000022s :  0.05%
    auto_monad                           :     0.000159s :  0.37%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000144s :  0.34%
    rewriter_before_opt_a                :     0.000101s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.18%
    jit_opt_a.loop_unroll                :     0.000044s :  0.10%
    jit_opt_a.a_1                        :     0.000902s :  2.13%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    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 :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001178s :  2.78%
    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      :     0.000030s :  0.07%
    jit_opt_a.cse                        :     0.000061s :  0.14%
    jit_opt_a.replace_applicator         :     0.000024s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000285s :  0.67%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000720s :  1.70%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000047s :  0.11%
    loop_unroll                          :     0.000492s :  1.16%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000031s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000066s :  0.16%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000025s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000511s :  1.20%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.05%
    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                             :     0.000070s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011573s : 27.27%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000302    47
 3.10% :     0.000009s :      2: substitution.depend_value_elim
 0.97% :     0.000003s :      4: substitution.elim_not_effective
 0.82% :     0.000002s :      4: substitution.fold_const_symbol
 2.25% :     0.000007s :      5: substitution.graph_param_transform
76.36% :     0.000231s :      4: substitution.inline
 1.76% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.46% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.50% :     0.000005s :      2: substitution.replace_old_param
 4.84% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.11% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.82% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024820     2
94.01% :     0.023333s :      1: type_inference.infer
 5.99% :     0.001487s :      1: type_inference.specialize
------[replace.]   0.000094     6
65.52% :     0.000061s :      4: replace.inline
34.48% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000241     6
94.46% :     0.000228s :      4: match.inline
 5.54% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000176  1021
 1.32% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.24% :     0.000002s :     15: predicate.addn_check_dump
 1.42% :     0.000002s :     15: predicate.addn_zero_filter
 1.99% :     0.000003s :     15: predicate.arithmetic_simplify
 1.46% :     0.000003s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     15: predicate.compare_switch_simplify
 1.46% :     0.000003s :     15: predicate.depend_value_elim
 1.12% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      5: predicate.elim_not_effective
 0.65% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.53% :     0.000003s :     15: predicate.environ_get_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      5: predicate.fold_const_symbol
 1.13% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 4.93% :     0.000009s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :     10: predicate.less_batch_normalization
 1.46% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     22: predicate.load_eliminater
 1.31% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.30% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.07% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.15% :     0.000002s :     15: predicate.minmaximum_grad
 1.86% :     0.000003s :      5: predicate.mutable_eliminate
 0.53% :     0.000001s :      5: predicate.opt_reshape
 2.34% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     15: predicate.reduce_eliminate
 1.46% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.71% :     0.000003s :     27: predicate.replace_applicator
 0.67% :     0.000001s :     10: predicate.replace_old_param
 0.35% :     0.000001s :      5: predicate.reset_defer_inline
 1.48% :     0.000003s :     15: predicate.reshape_eliminate
 1.16% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.49% :     0.000003s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.13% :     0.000002s :     10: predicate.special_op_eliminate
 1.05% :     0.000002s :     10: predicate.specialize_transform
 1.52% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.39% :     0.000004s :     21: predicate.switch_defer_inline
 1.88% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.95% :     0.000012s :     68: predicate.switch_simplify
 1.30% :     0.000002s :     15: predicate.tile_eliminate
 1.18% :     0.000002s :     15: predicate.transpose_eliminate
 1.52% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.24% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     15: predicate.value_based_eliminate
 0.42% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000888    13
44.65% :     0.000397s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.59% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.76% :     0.000433s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049527    76
 0.14% :     0.000069s :      1: add_recomputation
 0.43% :     0.000212s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.75% :     0.000371s :      1: bootstrap
 0.10% :     0.000052s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000026s :      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% :     0.000010s :      1: graph_reusing
13.53% :     0.006699s :      1: jit_opt_a
 0.41% :     0.000205s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 1.01% :     0.000501s :      1: loop_unroll
 1.48% :     0.000731s :      1: mutable_eliminate
 2.44% :     0.001209s :     26: opt.transform.jit_opt_a
 0.17% :     0.000084s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.05% :     0.000519s :      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
 0.30% :     0.000148s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.42% :     0.000704s :      1: renormalize.infer
 0.94% :     0.000465s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.59% :     0.000290s :      1: rewriter_after_opt_a
 0.21% :     0.000105s :      1: rewriter_before_opt_a
 0.22% :     0.000109s :      1: symbol_engine_optimizer
23.41% :     0.011594s :      1: task_emit
50.28% :     0.024905s :      1: type_inference
 0.19% :     0.000096s :      1: validate


TotalTime = 0.0122051, [30]
[bootstrap]: 0.00026357
[type_inference]: 0.00459426
[event_method]: 2.155e-05
[auto_monad]: 6.584e-05
[graph_reusing]: 5.32999e-06
[pre_auto_parallel]: 2.56998e-06
[py_interpret_to_execute]: 3.67e-05
[rewriter_before_opt_a]: 0.00010123
[expand_dump_flag]: 2.97002e-06
[jit_opt_a]: 0.00511505, [2]
    [Cycle 1]: 0.00168518, [27]
        [switch_simplify]: 5.7e-05
        [loop_unroll]: 3.727e-05
        [a_1]: 0.00061899
        [with_stream_mark]: 1.323e-05
        [recompute_prepare]: 7.98999e-06
        [updatestate_depend_eliminate]: 3.31001e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 6.94999e-06
        [updatestate_useless_node_eliminater]: 6.17999e-06
        [accelerated_algorithm]: 6.56e-06
        [meta_shard_fg_expand]: 1.79e-06
        [get_grad_eliminate_]: 8e-06
        [merge_forward]: 3.26001e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.284e-05
        [j_node_and_user_rematch]: 9.92999e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 1.106e-05
        [inline_without_move]: 6.13998e-06
        [renormalize]: 0.00057328
        [add_forward_monad_depend]: 4.53999e-06
        [auto_monad_grad]: 1.73002e-06
        [auto_monad_eliminator]: 1.112e-05
        [cse]: 2.12e-05
        [replace_applicator]: 1.294e-05
    [Cycle 2]: 0.0003588, [27]
        [switch_simplify]: 6.94999e-06
        [loop_unroll]: 6.14001e-06
        [a_1]: 0.000115
        [with_stream_mark]: 8.55001e-06
        [recompute_prepare]: 5.74999e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.38998e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 5.99999e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 1.23002e-06
        [get_grad_eliminate_]: 5.35001e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.303e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 9.47001e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 6.33998e-06
        [cse]: 1.263e-05
        [replace_applicator]: 6.17001e-06
[py_interpret_to_execute_after_opt_a]: 9.21002e-06
[rewriter_after_opt_a]: 3.009e-05
[convert_after_rewriter]: 7.01999e-06
[order_py_execute_after_rewriter]: 4.49998e-06
[mutable_eliminate]: 0.00048015
[jit_opt_b]: 5.779e-05, [1]
    [Cycle 1]: 5.106e-05, [2]
        [frontend_op_eliminate]: 1.867e-05
        [inline_after_opt_a]: 2.102e-05
[cconv]: 1.584e-05
[loop_unroll]: 0.00042874
[jit_opt_after_cconv]: 0.00015207, [1]
    [Cycle 1]: 0.00014533, [11]
        [c_1]: 2.634e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.89999e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [cse]: 2.046e-05
        [call_graph_tuple_transform]: 1.903e-05
        [tuple_list_get_item_eliminator]: 6.79999e-06
        [none_parameter_eliminate]: 1.05001e-06
        [renormalize]: 2.9002e-07
        [switch_simplify]: 6.44001e-06
[remove_dup_value]: 8.69e-06
[partial_unused_args_eliminate]: 1.70001e-06
[environ_conv]: 4.32e-06
[add_recomputation]: 3.259e-05
[cse_after_recomputation]: 2.312e-05, [1]
    [Cycle 1]: 1.794e-05, [1]
        [cse]: 1.212e-05
[auto_monad_reorder]: 1.256e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 4.74998e-06
[opt_after_jit_grad]: 0.00046036
[symbol_engine_optimizer]: 7.4e-05, [1]
    [Cycle 1]: 6.803e-05, [6]
        [build]: 2.57001e-06
        [elim_shapecalc]: 8.17e-06
        [elim_not_effective]: 1.323e-05
        [opt_reshape]: 6.07999e-06
        [fold_const_symbol]: 9.14e-06
        [renormalize]: 4.10015e-07
[validate]: 2.999e-05
Sums
    bootstrap                            :     0.000264s :  3.10%
    type_inference                       :     0.004594s : 54.04%
    event_method                         :     0.000022s :  0.25%
    auto_monad                           :     0.000066s :  0.77%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.43%
    rewriter_before_opt_a                :     0.000101s :  1.19%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000064s :  0.75%
    jit_opt_a.loop_unroll                :     0.000043s :  0.51%
    jit_opt_a.a_1                        :     0.000734s :  8.63%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000014s :  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 :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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_        :     0.000013s :  0.16%
    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 :  0.30%
    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          :     0.000021s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000573s :  6.74%
    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 :  0.21%
    jit_opt_a.cse                        :     0.000034s :  0.40%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000030s :  0.35%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000480s :  5.65%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.25%
    cconv                                :     0.000016s :  0.19%
    loop_unroll                          :     0.000429s :  5.04%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.31%
    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              :     0.000020s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.08%
    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 :  0.08%
    remove_dup_value                     :     0.000009s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000033s :  0.38%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    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                   :     0.000460s :  5.41%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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 :  0.00%
    validate                             :     0.000030s :  0.35%


Time group info:
------[substitution.]   0.000204    28
 0.63% :     0.000001s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.03% :     0.000004s :      4: substitution.graph_param_transform
85.42% :     0.000174s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.39% :     0.000005s :      4: substitution.replace_old_param
 5.77% :     0.000012s :      4: 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.]   0.000083     8
55.20% :     0.000046s :      4: replace.inline
44.80% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     8
94.60% :     0.000171s :      4: match.inline
 5.40% :     0.000010s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.19% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.08% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     13: predicate.addn_check_dump
 1.25% :     0.000002s :     13: predicate.addn_zero_filter
 1.92% :     0.000003s :     13: predicate.arithmetic_simplify
 1.30% :     0.000002s :     13: predicate.cast_eliminate
 0.48% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     13: predicate.compare_switch_simplify
 1.21% :     0.000002s :     13: predicate.depend_value_elim
 1.14% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.17% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      4: predicate.elim_not_effective
 0.46% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     13: predicate.environ_get_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      4: predicate.fold_const_symbol
 0.93% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.54% :     0.000008s :     29: predicate.inline
 0.99% :     0.000001s :      8: predicate.inline_without_move
 0.47% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      8: predicate.less_batch_normalization
 1.95% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.02% :     0.000003s :     21: predicate.load_eliminater
 1.26% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.25% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.51% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.52% :     0.000003s :     21: predicate.partial_eliminate
 1.15% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     13: predicate.reduce_eliminate
 1.56% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.93% :     0.000003s :     25: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.39% :     0.000001s :      4: predicate.reset_defer_inline
 1.17% :     0.000002s :     13: predicate.reshape_eliminate
 1.18% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     13: predicate.same_eliminate
 0.56% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      8: predicate.special_op_eliminate
 1.01% :     0.000001s :      8: predicate.specialize_transform
 1.59% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     21: predicate.switch_defer_inline
 2.23% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.88% :     0.000011s :     67: predicate.switch_simplify
 1.17% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.53% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.85% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000570    12
47.13% :     0.000268s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.08% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.79% :     0.000267s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013843    72
 0.26% :     0.000035s :      1: add_recomputation
 0.50% :     0.000069s :      1: auto_monad
 0.11% :     0.000015s :      1: auto_monad_reorder
 1.99% :     0.000275s :      1: bootstrap
 0.14% :     0.000019s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.18% :     0.000025s :      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
36.98% :     0.005119s :      1: jit_opt_a
 1.12% :     0.000155s :      1: jit_opt_after_cconv
 0.44% :     0.000061s :      1: jit_opt_b
 3.15% :     0.000436s :      1: loop_unroll
 3.53% :     0.000488s :      1: mutable_eliminate
 7.04% :     0.000975s :     26: opt.transform.jit_opt_a
 0.39% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.24% :     0.000033s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.38% :     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
 0.29% :     0.000040s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000011s :      1: remove_dup_value
 2.15% :     0.000298s :      1: renormalize.infer
 1.94% :     0.000269s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000033s :      1: rewriter_after_opt_a
 0.75% :     0.000104s :      1: rewriter_before_opt_a
 0.55% :     0.000077s :      1: symbol_engine_optimizer
33.26% :     0.004605s :      1: type_inference


TotalTime = 0.0121695, [30]
[bootstrap]: 0.00026135
[type_inference]: 0.00452501
[event_method]: 2.205e-05
[auto_monad]: 6.327e-05
[graph_reusing]: 5.30001e-06
[pre_auto_parallel]: 2.22001e-06
[py_interpret_to_execute]: 3.648e-05
[rewriter_before_opt_a]: 8.619e-05
[expand_dump_flag]: 2.66e-06
[jit_opt_a]: 0.00508227, [2]
    [Cycle 1]: 0.00168126, [27]
        [switch_simplify]: 5.706e-05
        [loop_unroll]: 3.756e-05
        [a_1]: 0.00063232
        [with_stream_mark]: 1.621e-05
        [recompute_prepare]: 8.50999e-06
        [updatestate_depend_eliminate]: 3.78999e-06
        [updatestate_assign_eliminate]: 3.15002e-06
        [updatestate_loads_eliminate]: 2.88998e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 6.90002e-06
        [updatestate_useless_node_eliminater]: 6.12999e-06
        [accelerated_algorithm]: 6.16998e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 3.06001e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.395e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 1.121e-05
        [inline_without_move]: 5.95002e-06
        [renormalize]: 0.00062298
        [add_forward_monad_depend]: 5.15999e-06
        [auto_monad_grad]: 2.35002e-06
        [auto_monad_eliminator]: 1.37e-05
        [cse]: 2.252e-05
        [replace_applicator]: 1.263e-05
    [Cycle 2]: 0.00035931, [27]
        [switch_simplify]: 6.90002e-06
        [loop_unroll]: 5.91e-06
        [a_1]: 0.00011445
        [with_stream_mark]: 8.74e-06
        [recompute_prepare]: 5.78997e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.31998e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 5.76e-06
        [accelerated_algorithm]: 6.23e-06
        [meta_shard_fg_expand]: 1.34998e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.243e-05
        [j_node_and_user_rematch]: 8.65001e-06
        [meta_fg_expand]: 1.94999e-06
        [replace_old_param]: 9.61e-06
        [inline_without_move]: 5.40001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.93002e-06
        [cse]: 1.331e-05
        [replace_applicator]: 5.92999e-06
[py_interpret_to_execute_after_opt_a]: 1.047e-05
[rewriter_after_opt_a]: 3.365e-05
[convert_after_rewriter]: 6.80002e-06
[order_py_execute_after_rewriter]: 5.44e-06
[mutable_eliminate]: 0.00047241
[jit_opt_b]: 5.338e-05, [1]
    [Cycle 1]: 4.637e-05, [2]
        [frontend_op_eliminate]: 1.807e-05
        [inline_after_opt_a]: 1.708e-05
[cconv]: 2.173e-05
[loop_unroll]: 0.00041278
[jit_opt_after_cconv]: 0.0001498, [1]
    [Cycle 1]: 0.00014375, [11]
        [c_1]: 2.455e-05
        [parameter_eliminate]: 2.26998e-06
        [updatestate_depend_eliminate]: 5.27999e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [cse]: 2.049e-05
        [call_graph_tuple_transform]: 2.065e-05
        [tuple_list_get_item_eliminator]: 6.33998e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 6.24001e-06
[remove_dup_value]: 1.456e-05
[partial_unused_args_eliminate]: 2.35002e-06
[environ_conv]: 5.65001e-06
[add_recomputation]: 4.086e-05
[cse_after_recomputation]: 2.354e-05, [1]
    [Cycle 1]: 1.81e-05, [1]
        [cse]: 1.258e-05
[auto_monad_reorder]: 1.669e-05
[get_jit_bprop_graph]: 1.57999e-06
[rewriter_after_jit_bprop_graph]: 4.82e-06
[opt_after_jit_grad]: 0.00052697
[symbol_engine_optimizer]: 8.253e-05, [1]
    [Cycle 1]: 7.63e-05, [6]
        [build]: 3.63e-06
        [elim_shapecalc]: 9.84999e-06
        [elim_not_effective]: 1.516e-05
        [opt_reshape]: 7.23e-06
        [fold_const_symbol]: 1.019e-05
        [renormalize]: 4.00003e-07
[validate]: 3.786e-05
Sums
    bootstrap                            :     0.000261s :  3.05%
    type_inference                       :     0.004525s : 52.81%
    event_method                         :     0.000022s :  0.26%
    auto_monad                           :     0.000063s :  0.74%
    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                :     0.000086s :  1.01%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000064s :  0.75%
    jit_opt_a.loop_unroll                :     0.000043s :  0.51%
    jit_opt_a.a_1                        :     0.000747s :  8.71%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.17%
    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 :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000011s :  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 :  0.31%
    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          :     0.000021s :  0.24%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000623s :  7.27%
    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                        :     0.000036s :  0.42%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000034s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000472s :  5.51%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000022s :  0.25%
    loop_unroll                          :     0.000413s :  4.82%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.29%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  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                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000041s :  0.48%
    cse_after_recomputation.cse          :     0.000013s :  0.15%
    auto_monad_reorder                   :     0.000017s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000527s :  6.15%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.44%


Time group info:
------[substitution.]   0.000210    28
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000006s :      4: substitution.graph_param_transform
81.99% :     0.000172s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.46% :     0.000005s :      4: substitution.replace_old_param
 7.68% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004470     2
79.45% :     0.003551s :      1: type_inference.infer
20.55% :     0.000919s :      1: type_inference.specialize
------[replace.]   0.000086     8
56.31% :     0.000048s :      4: replace.inline
43.69% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     8
92.36% :     0.000169s :      4: match.inline
 7.64% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   912
 1.33% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     13: predicate.addn_check_dump
 1.18% :     0.000002s :     13: predicate.addn_zero_filter
 1.89% :     0.000003s :     13: predicate.arithmetic_simplify
 1.41% :     0.000002s :     13: predicate.cast_eliminate
 0.45% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.10% :     0.000002s :     13: predicate.compare_switch_simplify
 1.15% :     0.000002s :     13: predicate.depend_value_elim
 1.15% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      4: predicate.elim_not_effective
 0.57% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     13: predicate.environ_get_eliminate
 1.07% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.30% :     0.000000s :      4: predicate.fold_const_symbol
 0.89% :     0.000001s :      8: predicate.get_grad_eliminate
 0.39% :     0.000001s :      4: predicate.graph_param_transform
 5.49% :     0.000008s :     29: predicate.inline
 0.89% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :      8: predicate.less_batch_normalization
 1.59% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.91% :     0.000003s :     21: predicate.load_eliminater
 1.39% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.40% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     13: predicate.merge_addn
 1.08% :     0.000002s :     13: predicate.minmaximum_grad
 1.76% :     0.000002s :      4: predicate.mutable_eliminate
 0.50% :     0.000001s :      4: predicate.opt_reshape
 2.56% :     0.000004s :     21: predicate.partial_eliminate
 1.19% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     13: predicate.reduce_eliminate
 1.61% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.02% :     0.000003s :     25: predicate.replace_applicator
 0.64% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.23% :     0.000002s :     13: predicate.reshape_eliminate
 1.12% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     13: predicate.same_eliminate
 0.57% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.15% :     0.000002s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.40% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     21: predicate.switch_defer_inline
 2.17% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.85% :     0.000011s :     67: predicate.switch_simplify
 1.20% :     0.000002s :     13: predicate.tile_eliminate
 1.20% :     0.000002s :     13: predicate.transpose_eliminate
 1.46% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000578    12
45.73% :     0.000264s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.82% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.44% :     0.000280s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013860    72
 0.32% :     0.000044s :      1: add_recomputation
 0.48% :     0.000067s :      1: auto_monad
 0.14% :     0.000019s :      1: auto_monad_reorder
 1.96% :     0.000272s :      1: bootstrap
 0.18% :     0.000024s :      1: cconv
 0.09% :     0.000013s :      1: convert_after_rewriter
 0.19% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.18% :     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
36.69% :     0.005085s :      1: jit_opt_a
 1.10% :     0.000153s :      1: jit_opt_after_cconv
 0.40% :     0.000056s :      1: jit_opt_b
 3.03% :     0.000421s :      1: loop_unroll
 3.46% :     0.000480s :      1: mutable_eliminate
 7.12% :     0.000986s :     26: opt.transform.jit_opt_a
 0.39% :     0.000055s :      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
 0.18% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.28% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.86% :     0.000535s :      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
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.36% :     0.000328s :      1: renormalize.infer
 2.08% :     0.000288s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000037s :      1: rewriter_after_opt_a
 0.64% :     0.000089s :      1: rewriter_before_opt_a
 0.62% :     0.000085s :      1: symbol_engine_optimizer
32.72% :     0.004536s :      1: type_inference


TotalTime = 0.013028, [30]
[bootstrap]: 0.0002762
[type_inference]: 0.00466937
[event_method]: 2.087e-05
[auto_monad]: 6.555e-05
[graph_reusing]: 5.49e-06
[pre_auto_parallel]: 2.34001e-06
[py_interpret_to_execute]: 3.732e-05
[rewriter_before_opt_a]: 8.897e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.0055967, [2]
    [Cycle 1]: 0.00197481, [27]
        [switch_simplify]: 6.101e-05
        [loop_unroll]: 3.596e-05
        [a_1]: 0.00064899
        [with_stream_mark]: 2.191e-05
        [recompute_prepare]: 9.12001e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 3.07002e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 7.06001e-06
        [updatestate_useless_node_eliminater]: 6.16e-06
        [accelerated_algorithm]: 6.39001e-06
        [meta_shard_fg_expand]: 2.25002e-06
        [get_grad_eliminate_]: 6.04001e-06
        [merge_forward]: 4.04997e-06
        [cell_reuse_recompute_pass]: 1.26002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.533e-05
        [j_node_and_user_rematch]: 1.014e-05
        [meta_fg_expand]: 2.44999e-06
        [replace_old_param]: 7.661e-05
        [inline_without_move]: 9.24998e-06
        [renormalize]: 0.0007984
        [add_forward_monad_depend]: 5.71998e-06
        [auto_monad_grad]: 2.68998e-06
        [auto_monad_eliminator]: 1.49e-05
        [cse]: 2.44e-05
        [replace_applicator]: 1.281e-05
    [Cycle 2]: 0.0003649, [27]
        [switch_simplify]: 6.73e-06
        [loop_unroll]: 5.95002e-06
        [a_1]: 0.00011605
        [with_stream_mark]: 1.114e-05
        [recompute_prepare]: 5.87001e-06
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 2.68003e-06
        [updatestate_loads_eliminate]: 2.25002e-06
        [parameter_eliminate]: 1.36002e-06
        [specialize_transform]: 5.70001e-06
        [updatestate_useless_node_eliminater]: 6.11998e-06
        [accelerated_algorithm]: 6.06998e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 5.49e-06
        [merge_forward]: 3.14999e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.386e-05
        [j_node_and_user_rematch]: 9.20001e-06
        [meta_fg_expand]: 2.28002e-06
        [replace_old_param]: 8.94e-06
        [inline_without_move]: 5.74999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 6.02001e-06
        [cse]: 1.366e-05
        [replace_applicator]: 6.08998e-06
[py_interpret_to_execute_after_opt_a]: 1.37e-05
[rewriter_after_opt_a]: 3.887e-05
[convert_after_rewriter]: 7.33e-06
[order_py_execute_after_rewriter]: 5.14998e-06
[mutable_eliminate]: 0.00061412
[jit_opt_b]: 5.571e-05, [1]
    [Cycle 1]: 4.873e-05, [2]
        [frontend_op_eliminate]: 1.84e-05
        [inline_after_opt_a]: 1.803e-05
[cconv]: 2.622e-05
[loop_unroll]: 0.00045095
[jit_opt_after_cconv]: 0.00015232, [1]
    [Cycle 1]: 0.00014571, [11]
        [c_1]: 2.529e-05
        [parameter_eliminate]: 2.87002e-06
        [updatestate_depend_eliminate]: 6.00002e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 2.232e-05
        [call_graph_tuple_transform]: 1.979e-05
        [tuple_list_get_item_eliminator]: 6.31e-06
        [none_parameter_eliminate]: 1.15999e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 6.10002e-06
[remove_dup_value]: 8.35001e-06
[partial_unused_args_eliminate]: 1.62999e-06
[environ_conv]: 4.55999e-06
[add_recomputation]: 4.334e-05
[cse_after_recomputation]: 2.417e-05, [1]
    [Cycle 1]: 1.859e-05, [1]
        [cse]: 1.279e-05
[auto_monad_reorder]: 1.259e-05
[get_jit_bprop_graph]: 1.71e-06
[rewriter_after_jit_bprop_graph]: 5.91e-06
[opt_after_jit_grad]: 0.00050578
[symbol_engine_optimizer]: 7.84e-05, [1]
    [Cycle 1]: 7.107e-05, [6]
        [build]: 4.30999e-06
        [elim_shapecalc]: 8.82999e-06
        [elim_not_effective]: 1.414e-05
        [opt_reshape]: 6.26e-06
        [fold_const_symbol]: 8.84e-06
        [renormalize]: 1.20001e-06
[validate]: 4.22e-05
Sums
    bootstrap                            :     0.000276s :  3.01%
    type_inference                       :     0.004669s : 50.82%
    event_method                         :     0.000021s :  0.23%
    auto_monad                           :     0.000066s :  0.71%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.41%
    rewriter_before_opt_a                :     0.000089s :  0.97%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000068s :  0.74%
    jit_opt_a.loop_unroll                :     0.000042s :  0.46%
    jit_opt_a.a_1                        :     0.000765s :  8.33%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000015s :  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 :     0.000005s :  0.06%
    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 :     0.000012s :  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_        :     0.000012s :  0.13%
    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 :     0.000029s :  0.32%
    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          :     0.000086s :  0.93%
    jit_opt_a.inline_without_move        :     0.000015s :  0.16%
    jit_opt_a.renormalize                :     0.000798s :  8.69%
    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      :     0.000021s :  0.23%
    jit_opt_a.cse                        :     0.000038s :  0.41%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.15%
    rewriter_after_opt_a                 :     0.000039s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000614s :  6.68%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000026s :  0.29%
    loop_unroll                          :     0.000451s :  4.91%
    jit_opt_after_cconv.c_1              :     0.000025s :  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 :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  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 :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000008s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000043s :  0.47%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000013s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000506s :  5.51%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    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 :  0.07%
    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.]   0.000231    28
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 1.82% :     0.000004s :      4: substitution.graph_param_transform
83.27% :     0.000192s :      4: substitution.inline
 1.41% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.86% :     0.000007s :      4: substitution.replace_old_param
 7.40% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004606     2
80.27% :     0.003697s :      1: type_inference.infer
19.73% :     0.000909s :      1: type_inference.specialize
------[replace.]   0.000089     8
57.45% :     0.000051s :      4: replace.inline
42.55% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000204     8
92.68% :     0.000189s :      4: match.inline
 7.32% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   912
 1.28% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.10% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     13: predicate.addn_check_dump
 1.32% :     0.000002s :     13: predicate.addn_zero_filter
 1.68% :     0.000002s :     13: predicate.arithmetic_simplify
 1.41% :     0.000002s :     13: predicate.cast_eliminate
 0.45% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     13: predicate.compare_switch_simplify
 1.17% :     0.000002s :     13: predicate.depend_value_elim
 1.13% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      4: predicate.elim_not_effective
 0.57% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.33% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     13: predicate.environ_get_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 0.98% :     0.000001s :      8: predicate.get_grad_eliminate
 0.32% :     0.000000s :      4: predicate.graph_param_transform
 5.59% :     0.000008s :     29: predicate.inline
 1.35% :     0.000002s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.15% :     0.000002s :      8: predicate.less_batch_normalization
 1.78% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     21: predicate.load_eliminater
 1.23% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.18% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 1.52% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.55% :     0.000004s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     13: predicate.reduce_eliminate
 1.88% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.03% :     0.000003s :     25: predicate.replace_applicator
 0.70% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000000s :      4: predicate.reset_defer_inline
 1.18% :     0.000002s :     13: predicate.reshape_eliminate
 1.18% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     13: predicate.same_eliminate
 0.57% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :      8: predicate.special_op_eliminate
 0.95% :     0.000001s :      8: predicate.specialize_transform
 1.26% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.59% :     0.000004s :     21: predicate.switch_defer_inline
 2.20% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.52% :     0.000010s :     67: predicate.switch_simplify
 1.19% :     0.000002s :     13: predicate.tile_eliminate
 1.13% :     0.000002s :     13: predicate.transpose_eliminate
 1.45% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     13: predicate.value_based_eliminate
 0.35% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000605    12
45.46% :     0.000275s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.78% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.76% :     0.000289s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014977    72
 0.31% :     0.000046s :      1: add_recomputation
 0.46% :     0.000069s :      1: auto_monad
 0.10% :     0.000016s :      1: auto_monad_reorder
 1.93% :     0.000290s :      1: bootstrap
 0.19% :     0.000029s :      1: cconv
 0.09% :     0.000013s :      1: convert_after_rewriter
 0.18% :     0.000026s :      1: cse_after_recomputation
 0.04% :     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
 0.05% :     0.000008s :      1: graph_reusing
37.39% :     0.005600s :      1: jit_opt_a
 1.04% :     0.000155s :      1: jit_opt_after_cconv
 0.39% :     0.000058s :      1: jit_opt_b
 3.07% :     0.000459s :      1: loop_unroll
 4.16% :     0.000623s :      1: mutable_eliminate
 7.20% :     0.001078s :     26: opt.transform.jit_opt_a
 0.36% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000030s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.44% :     0.000515s :      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% :     0.000040s :      1: py_interpret_to_execute
 0.11% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000011s :      1: remove_dup_value
 2.99% :     0.000447s :      1: renormalize.infer
 2.29% :     0.000343s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000042s :      1: rewriter_after_opt_a
 0.61% :     0.000092s :      1: rewriter_before_opt_a
 0.54% :     0.000081s :      1: symbol_engine_optimizer
31.25% :     0.004680s :      1: type_inference


TotalTime = 0.0460101, [33]
[bootstrap]: 0.0003181
[type_inference]: 0.0250306
[event_method]: 2.069e-05
[auto_monad]: 0.00015855
[graph_reusing]: 6.74001e-06
[pre_auto_parallel]: 2.99001e-06
[py_interpret_to_execute]: 0.00012732
[rewriter_before_opt_a]: 9.929e-05
[expand_dump_flag]: 3.85e-06
[jit_opt_a]: 0.00640666, [2]
    [Cycle 1]: 0.00241109, [27]
        [switch_simplify]: 6.498e-05
        [loop_unroll]: 3.658e-05
        [a_1]: 0.0007218
        [with_stream_mark]: 1.989e-05
        [recompute_prepare]: 1.074e-05
        [updatestate_depend_eliminate]: 6.01e-06
        [updatestate_assign_eliminate]: 6.20002e-06
        [updatestate_loads_eliminate]: 4.67e-06
        [parameter_eliminate]: 2.24999e-06
        [specialize_transform]: 9.75002e-06
        [updatestate_useless_node_eliminater]: 1.18e-05
        [accelerated_algorithm]: 8.69e-06
        [meta_shard_fg_expand]: 2.93e-06
        [get_grad_eliminate_]: 9.10999e-06
        [merge_forward]: 5.93002e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.039e-05
        [j_node_and_user_rematch]: 1.366e-05
        [meta_fg_expand]: 3.73001e-06
        [replace_old_param]: 1.201e-05
        [inline_without_move]: 8.32e-06
        [renormalize]: 0.00115679
        [add_forward_monad_depend]: 6.53998e-06
        [auto_monad_grad]: 1.90001e-06
        [auto_monad_eliminator]: 2.281e-05
        [cse]: 4.32e-05
        [replace_applicator]: 1.766e-05
    [Cycle 2]: 0.00047266, [27]
        [switch_simplify]: 9.14e-06
        [loop_unroll]: 7.96001e-06
        [a_1]: 0.00017095
        [with_stream_mark]: 1.278e-05
        [recompute_prepare]: 8.2e-06
        [updatestate_depend_eliminate]: 5.14998e-06
        [updatestate_assign_eliminate]: 4.70999e-06
        [updatestate_loads_eliminate]: 3.98001e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 8.05999e-06
        [updatestate_useless_node_eliminater]: 1.074e-05
        [accelerated_algorithm]: 7.97998e-06
        [meta_shard_fg_expand]: 1.84998e-06
        [get_grad_eliminate_]: 7.64002e-06
        [merge_forward]: 4.44998e-06
        [cell_reuse_recompute_pass]: 2.08998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.667e-05
        [j_node_and_user_rematch]: 1.255e-05
        [meta_fg_expand]: 3.03e-06
        [replace_old_param]: 1.118e-05
        [inline_without_move]: 7.45e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 1.021e-05
        [cse]: 2.118e-05
        [replace_applicator]: 7.72002e-06
[py_interpret_to_execute_after_opt_a]: 1.278e-05
[rewriter_after_opt_a]: 0.00028693
[convert_after_rewriter]: 1.144e-05
[order_py_execute_after_rewriter]: 7.02002e-06
[mutable_eliminate]: 0.00059109
[jit_opt_b]: 7.002e-05, [1]
    [Cycle 1]: 6.293e-05, [2]
        [frontend_op_eliminate]: 2.494e-05
        [inline_after_opt_a]: 2.518e-05
[cconv]: 2.582e-05
[loop_unroll]: 0.00048697
[jit_opt_after_cconv]: 0.00020945, [1]
    [Cycle 1]: 0.00020133, [11]
        [c_1]: 4.934e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 8.70001e-06
        [updatestate_assign_eliminate]: 4.61002e-06
        [updatestate_loads_eliminate]: 4.20999e-06
        [cse]: 3.102e-05
        [call_graph_tuple_transform]: 2.359e-05
        [tuple_list_get_item_eliminator]: 9.29e-06
        [none_parameter_eliminate]: 1.87001e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 9.69e-06
[remove_dup_value]: 2.113e-05
[partial_unused_args_eliminate]: 2.94999e-06
[environ_conv]: 1.218e-05
[add_recomputation]: 7.593e-05
[cse_after_recomputation]: 3.363e-05, [1]
    [Cycle 1]: 2.712e-05, [1]
        [cse]: 2.073e-05
[auto_monad_reorder]: 2.751e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 6.61999e-06
[opt_after_jit_grad]: 0.00059419
[symbol_engine_optimizer]: 0.0001147, [1]
    [Cycle 1]: 0.00010729, [6]
        [build]: 1.314e-05
        [elim_shapecalc]: 1.366e-05
        [elim_not_effective]: 2.246e-05
        [opt_reshape]: 9.77999e-06
        [fold_const_symbol]: 1.509e-05
        [renormalize]: 6.39993e-07
[validate]: 6.159e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.0109469
[execute]: 7.32002e-06
Sums
    bootstrap                            :     0.000318s :  0.76%
    type_inference                       :     0.025031s : 59.84%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000159s :  0.38%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000127s :  0.30%
    rewriter_before_opt_a                :     0.000099s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.18%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000893s :  2.13%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001157s :  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      :     0.000033s :  0.08%
    jit_opt_a.cse                        :     0.000064s :  0.15%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000287s :  0.69%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000591s :  1.41%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000026s :  0.06%
    loop_unroll                          :     0.000487s :  1.16%
    jit_opt_after_cconv.c_1              :     0.000049s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000031s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000076s :  0.18%
    cse_after_recomputation.cse          :     0.000021s :  0.05%
    auto_monad_reorder                   :     0.000028s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000594s :  1.42%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000062s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010947s : 26.17%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000285    47
 3.28% :     0.000009s :      2: substitution.depend_value_elim
 1.18% :     0.000003s :      4: substitution.elim_not_effective
 0.89% :     0.000003s :      4: substitution.fold_const_symbol
 2.45% :     0.000007s :      5: substitution.graph_param_transform
74.78% :     0.000213s :      4: substitution.inline
 1.68% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.68% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.88% :     0.000005s :      2: substitution.replace_old_param
 4.47% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.21% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.49% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024956     2
93.96% :     0.023449s :      1: type_inference.infer
 6.04% :     0.001507s :      1: type_inference.specialize
------[replace.]   0.000096     6
66.07% :     0.000064s :      4: replace.inline
33.93% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
94.81% :     0.000210s :      4: match.inline
 5.19% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000176  1021
 1.21% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.48% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.33% :     0.000002s :     15: predicate.addn_zero_filter
 2.02% :     0.000004s :     15: predicate.arithmetic_simplify
 1.44% :     0.000003s :     15: predicate.cast_eliminate
 0.46% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.24% :     0.000002s :     15: predicate.compare_switch_simplify
 1.50% :     0.000003s :     15: predicate.depend_value_elim
 1.29% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.27% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      5: predicate.elim_not_effective
 0.66% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.07% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.60% :     0.000003s :     15: predicate.environ_get_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.04% :     0.000002s :     10: predicate.get_grad_eliminate
 0.29% :     0.000001s :      5: predicate.graph_param_transform
 5.01% :     0.000009s :     31: predicate.inline
 1.04% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :     10: predicate.less_batch_normalization
 1.80% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.88% :     0.000003s :     22: predicate.load_eliminater
 1.58% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.36% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.15% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 1.46% :     0.000003s :      5: predicate.mutable_eliminate
 0.59% :     0.000001s :      5: predicate.opt_reshape
 2.35% :     0.000004s :     22: predicate.partial_eliminate
 1.26% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.83% :     0.000003s :     15: predicate.reduce_eliminate
 1.47% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.69% :     0.000003s :     27: predicate.replace_applicator
 0.64% :     0.000001s :     10: predicate.replace_old_param
 0.40% :     0.000001s :      5: predicate.reset_defer_inline
 1.33% :     0.000002s :     15: predicate.reshape_eliminate
 1.22% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     15: predicate.same_eliminate
 0.54% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :     10: predicate.special_op_eliminate
 1.01% :     0.000002s :     10: predicate.specialize_transform
 1.41% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.17% :     0.000004s :     21: predicate.switch_defer_inline
 1.80% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.41% :     0.000011s :     68: predicate.switch_simplify
 1.25% :     0.000002s :     15: predicate.tile_eliminate
 1.20% :     0.000002s :     15: predicate.transpose_eliminate
 1.52% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.86% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.15% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000003s :     15: predicate.value_based_eliminate
 0.53% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000868    13
47.60% :     0.000413s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.65% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.75% :     0.000414s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048574    76
 0.16% :     0.000079s :      1: add_recomputation
 0.34% :     0.000164s :      1: auto_monad
 0.06% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000336s :      1: bootstrap
 0.06% :     0.000028s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000036s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.05% :     0.000025s :      1: event_method
 0.02% :     0.000012s :      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
13.20% :     0.006410s :      1: jit_opt_a
 0.44% :     0.000212s :      1: jit_opt_after_cconv
 0.15% :     0.000073s :      1: jit_opt_b
 1.02% :     0.000495s :      1: loop_unroll
 1.23% :     0.000600s :      1: mutable_eliminate
 2.47% :     0.001201s :     26: opt.transform.jit_opt_a
 0.18% :     0.000087s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000057s :      4: opt.transform.symbol_engine_opt
 1.24% :     0.000605s :      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
 0.27% :     0.000131s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000024s :      1: remove_dup_value
 1.33% :     0.000646s :      1: renormalize.infer
 1.03% :     0.000502s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.60% :     0.000292s :      1: rewriter_after_opt_a
 0.21% :     0.000103s :      1: rewriter_before_opt_a
 0.24% :     0.000118s :      1: symbol_engine_optimizer
22.57% :     0.010963s :      1: task_emit
51.56% :     0.025043s :      1: type_inference
 0.18% :     0.000086s :      1: validate


TotalTime = 0.0126562, [30]
[bootstrap]: 0.00028325
[type_inference]: 0.004741
[event_method]: 2.059e-05
[auto_monad]: 6.256e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 3.582e-05
[rewriter_before_opt_a]: 8.855e-05
[expand_dump_flag]: 3.43e-06
[jit_opt_a]: 0.00530586, [2]
    [Cycle 1]: 0.00183543, [27]
        [switch_simplify]: 5.793e-05
        [loop_unroll]: 3.56e-05
        [a_1]: 0.00063062
        [with_stream_mark]: 1.92e-05
        [recompute_prepare]: 8.48001e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 3.29001e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 6.84999e-06
        [updatestate_useless_node_eliminater]: 6.31998e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 6.34001e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.483e-05
        [j_node_and_user_rematch]: 9.67001e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 1.085e-05
        [inline_without_move]: 6.03002e-06
        [renormalize]: 0.00076002
        [add_forward_monad_depend]: 5.67999e-06
        [auto_monad_grad]: 2.24999e-06
        [auto_monad_eliminator]: 1.467e-05
        [cse]: 2.94e-05
        [replace_applicator]: 1.435e-05
    [Cycle 2]: 0.00036422, [27]
        [switch_simplify]: 7.01999e-06
        [loop_unroll]: 6.09999e-06
        [a_1]: 0.00011467
        [with_stream_mark]: 1.151e-05
        [recompute_prepare]: 5.89e-06
        [updatestate_depend_eliminate]: 3.21999e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.61e-06
        [updatestate_useless_node_eliminater]: 5.49e-06
        [accelerated_algorithm]: 5.93002e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 5.49e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 8.74003e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 9.20999e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 1.30001e-06
        [auto_monad_eliminator]: 6.31e-06
        [cse]: 1.353e-05
        [replace_applicator]: 5.97001e-06
[py_interpret_to_execute_after_opt_a]: 1.136e-05
[rewriter_after_opt_a]: 3.813e-05
[convert_after_rewriter]: 7.53e-06
[order_py_execute_after_rewriter]: 5.19998e-06
[mutable_eliminate]: 0.00055071
[jit_opt_b]: 5.524e-05, [1]
    [Cycle 1]: 4.763e-05, [2]
        [frontend_op_eliminate]: 1.835e-05
        [inline_after_opt_a]: 1.767e-05
[cconv]: 2.658e-05
[loop_unroll]: 0.00042097
[jit_opt_after_cconv]: 0.00015627, [1]
    [Cycle 1]: 0.00015, [11]
        [c_1]: 2.505e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 6.04001e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [cse]: 2.638e-05
        [call_graph_tuple_transform]: 2.036e-05
        [tuple_list_get_item_eliminator]: 6.41998e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 6.34999e-06
[remove_dup_value]: 1.488e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 6.45002e-06
[add_recomputation]: 4.932e-05
[cse_after_recomputation]: 2.439e-05, [1]
    [Cycle 1]: 1.885e-05, [1]
        [cse]: 1.32e-05
[auto_monad_reorder]: 1.733e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 5.02999e-06
[opt_after_jit_grad]: 0.00044113
[symbol_engine_optimizer]: 7.66e-05, [1]
    [Cycle 1]: 7.037e-05, [6]
        [build]: 2.70002e-06
        [elim_shapecalc]: 8.72e-06
        [elim_not_effective]: 1.422e-05
        [opt_reshape]: 6.38998e-06
        [fold_const_symbol]: 9.95002e-06
        [renormalize]: 6.50005e-07
[validate]: 3.904e-05
Sums
    bootstrap                            :     0.000283s :  3.15%
    type_inference                       :     0.004741s : 52.73%
    event_method                         :     0.000021s :  0.23%
    auto_monad                           :     0.000063s :  0.70%
    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                :     0.000089s :  0.98%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000065s :  0.72%
    jit_opt_a.loop_unroll                :     0.000042s :  0.46%
    jit_opt_a.a_1                        :     0.000745s :  8.29%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.34%
    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 :     0.000005s :  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 :     0.000012s :  0.13%
    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_        :     0.000012s :  0.13%
    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 :     0.000029s :  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          :     0.000020s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000760s :  8.45%
    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      :     0.000021s :  0.23%
    jit_opt_a.cse                        :     0.000043s :  0.48%
    jit_opt_a.replace_applicator         :     0.000020s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000038s :  0.42%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000551s :  6.13%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000027s :  0.30%
    loop_unroll                          :     0.000421s :  4.68%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.28%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000026s :  0.29%
    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      :     0.000000s :  0.00%
    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 :  0.07%
    add_recomputation                    :     0.000049s :  0.55%
    cse_after_recomputation.cse          :     0.000013s :  0.15%
    auto_monad_reorder                   :     0.000017s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000441s :  4.91%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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 :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.43%


Time group info:
------[substitution.]   0.000219    28
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.46% :     0.000005s :      4: substitution.graph_param_transform
82.39% :     0.000180s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.31% :     0.000005s :      4: substitution.replace_old_param
 7.57% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004686     2
80.98% :     0.003795s :      1: type_inference.infer
19.02% :     0.000891s :      1: type_inference.specialize
------[replace.]   0.000083     8
55.49% :     0.000046s :      4: replace.inline
44.51% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     8
92.49% :     0.000177s :      4: match.inline
 7.51% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.31% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.11% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     13: predicate.addn_check_dump
 1.16% :     0.000002s :     13: predicate.addn_zero_filter
 1.76% :     0.000002s :     13: predicate.arithmetic_simplify
 1.32% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     13: predicate.compare_switch_simplify
 1.20% :     0.000002s :     13: predicate.depend_value_elim
 1.11% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      4: predicate.elim_not_effective
 0.47% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     13: predicate.environ_get_eliminate
 1.21% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.98% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.56% :     0.000008s :     29: predicate.inline
 0.88% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.12% :     0.000002s :      8: predicate.less_batch_normalization
 1.58% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     21: predicate.load_eliminater
 1.27% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.29% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.89% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     13: predicate.merge_addn
 1.17% :     0.000002s :     13: predicate.minmaximum_grad
 1.58% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.70% :     0.000004s :     21: predicate.partial_eliminate
 1.19% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     13: predicate.reduce_eliminate
 1.67% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.77% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.38% :     0.000002s :     13: predicate.reshape_eliminate
 1.16% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     13: predicate.same_eliminate
 0.66% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      8: predicate.special_op_eliminate
 1.03% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.68% :     0.000011s :     67: predicate.switch_simplify
 1.24% :     0.000002s :     13: predicate.tile_eliminate
 1.19% :     0.000002s :     13: predicate.transpose_eliminate
 1.52% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.39% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000594    12
47.26% :     0.000281s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.96% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.78% :     0.000278s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014478    72
 0.36% :     0.000052s :      1: add_recomputation
 0.46% :     0.000066s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 2.03% :     0.000294s :      1: bootstrap
 0.20% :     0.000029s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000027s :      1: cse_after_recomputation
 0.06% :     0.000009s :      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
36.67% :     0.005309s :      1: jit_opt_a
 1.10% :     0.000159s :      1: jit_opt_after_cconv
 0.40% :     0.000058s :      1: jit_opt_b
 2.96% :     0.000429s :      1: loop_unroll
 3.86% :     0.000560s :      1: mutable_eliminate
 6.81% :     0.000986s :     26: opt.transform.jit_opt_a
 0.38% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000030s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.10% :     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
 0.27% :     0.000039s :      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
 3.03% :     0.000438s :      1: renormalize.infer
 2.17% :     0.000314s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000041s :      1: rewriter_after_opt_a
 0.63% :     0.000091s :      1: rewriter_before_opt_a
 0.55% :     0.000079s :      1: symbol_engine_optimizer
32.82% :     0.004751s :      1: type_inference


TotalTime = 0.0152349, [30]
[bootstrap]: 0.00031859
[type_inference]: 0.00583951
[event_method]: 2.269e-05
[auto_monad]: 7.221e-05
[graph_reusing]: 1.794e-05
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 4.748e-05
[rewriter_before_opt_a]: 9.388e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.00624958, [2]
    [Cycle 1]: 0.00200204, [27]
        [switch_simplify]: 6.623e-05
        [loop_unroll]: 3.682e-05
        [a_1]: 0.00069964
        [with_stream_mark]: 2.119e-05
        [recompute_prepare]: 1.126e-05
        [updatestate_depend_eliminate]: 4.66002e-06
        [updatestate_assign_eliminate]: 3.68e-06
        [updatestate_loads_eliminate]: 2.96001e-06
        [parameter_eliminate]: 2.39001e-06
        [specialize_transform]: 7.45e-06
        [updatestate_useless_node_eliminater]: 6.49001e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 2.45002e-06
        [get_grad_eliminate_]: 5.94999e-06
        [merge_forward]: 4.62e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.642e-05
        [j_node_and_user_rematch]: 9.87999e-06
        [meta_fg_expand]: 2.44999e-06
        [replace_old_param]: 1.274e-05
        [inline_without_move]: 6.28e-06
        [renormalize]: 0.00077971
        [add_forward_monad_depend]: 8.25999e-06
        [auto_monad_grad]: 3.03e-06
        [auto_monad_eliminator]: 1.931e-05
        [cse]: 3.646e-05
        [replace_applicator]: 1.898e-05
    [Cycle 2]: 0.00038979, [27]
        [switch_simplify]: 7.01001e-06
        [loop_unroll]: 5.68997e-06
        [a_1]: 0.00011342
        [with_stream_mark]: 1.251e-05
        [recompute_prepare]: 6.83998e-06
        [updatestate_depend_eliminate]: 4.12e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 5.77001e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 5.10001e-06
        [merge_forward]: 3.42002e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.467e-05
        [j_node_and_user_rematch]: 9.04e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 1.095e-05
        [inline_without_move]: 5.82999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.51998e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 8.35001e-06
        [cse]: 1.771e-05
        [replace_applicator]: 7.97e-06
[py_interpret_to_execute_after_opt_a]: 1.578e-05
[rewriter_after_opt_a]: 3.971e-05
[convert_after_rewriter]: 7.45e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00067805
[jit_opt_b]: 5.984e-05, [1]
    [Cycle 1]: 5.135e-05, [2]
        [frontend_op_eliminate]: 1.952e-05
        [inline_after_opt_a]: 1.861e-05
[cconv]: 2.689e-05
[loop_unroll]: 0.00055698
[jit_opt_after_cconv]: 0.00017673, [1]
    [Cycle 1]: 0.00016911, [11]
        [c_1]: 2.655e-05
        [parameter_eliminate]: 4.17e-06
        [updatestate_depend_eliminate]: 7.48e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.75997e-06
        [cse]: 3.231e-05
        [call_graph_tuple_transform]: 2.408e-05
        [tuple_list_get_item_eliminator]: 6.36e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 6.72002e-06
[remove_dup_value]: 1.47e-05
[partial_unused_args_eliminate]: 2.68e-06
[environ_conv]: 6.57002e-06
[add_recomputation]: 5.246e-05
[cse_after_recomputation]: 2.809e-05, [1]
    [Cycle 1]: 2.083e-05, [1]
        [cse]: 1.351e-05
[auto_monad_reorder]: 1.822e-05
[get_jit_bprop_graph]: 1.51002e-06
[rewriter_after_jit_bprop_graph]: 7.08e-06
[opt_after_jit_grad]: 0.00055353
[symbol_engine_optimizer]: 8.075e-05, [1]
    [Cycle 1]: 7.359e-05, [6]
        [build]: 3.86999e-06
        [elim_shapecalc]: 8.88002e-06
        [elim_not_effective]: 1.599e-05
        [opt_reshape]: 6.28002e-06
        [fold_const_symbol]: 9.69e-06
        [renormalize]: 3.7998e-07
[validate]: 4.404e-05
Sums
    bootstrap                            :     0.000319s :  2.97%
    type_inference                       :     0.005840s : 54.46%
    event_method                         :     0.000023s :  0.21%
    auto_monad                           :     0.000072s :  0.67%
    graph_reusing                        :     0.000018s :  0.17%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.44%
    rewriter_before_opt_a                :     0.000094s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000073s :  0.68%
    jit_opt_a.loop_unroll                :     0.000043s :  0.40%
    jit_opt_a.a_1                        :     0.000813s :  7.58%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.17%
    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       :     0.000013s :  0.12%
    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_        :     0.000011s :  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 :     0.000031s :  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          :     0.000024s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.11%
    jit_opt_a.renormalize                :     0.000780s :  7.27%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.26%
    jit_opt_a.cse                        :     0.000054s :  0.51%
    jit_opt_a.replace_applicator         :     0.000027s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.15%
    rewriter_after_opt_a                 :     0.000040s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000678s :  6.32%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.17%
    cconv                                :     0.000027s :  0.25%
    loop_unroll                          :     0.000557s :  5.19%
    jit_opt_after_cconv.c_1              :     0.000027s :  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 :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000032s :  0.30%
    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  :     0.000007s :  0.06%
    remove_dup_value                     :     0.000015s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000052s :  0.49%
    cse_after_recomputation.cse          :     0.000014s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000554s :  5.16%
    symbol_engine_optimizer.build        :     0.000004s :  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 :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.41%


Time group info:
------[substitution.]   0.000268    28
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000002s :      2: substitution.fold_const_symbol
 2.08% :     0.000006s :      4: substitution.graph_param_transform
84.26% :     0.000226s :      4: substitution.inline
 1.25% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.70% :     0.000007s :      4: substitution.replace_old_param
 6.46% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005755     2
80.90% :     0.004656s :      1: type_inference.infer
19.10% :     0.001099s :      1: type_inference.specialize
------[replace.]   0.000096     8
58.74% :     0.000056s :      4: replace.inline
41.26% :     0.000040s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000238     8
93.63% :     0.000223s :      4: match.inline
 6.37% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   912
 1.20% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     13: predicate.addn_check_dump
 1.28% :     0.000002s :     13: predicate.addn_zero_filter
 2.01% :     0.000003s :     13: predicate.arithmetic_simplify
 1.23% :     0.000002s :     13: predicate.cast_eliminate
 0.47% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.04% :     0.000001s :     13: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.17% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      4: predicate.elim_not_effective
 0.45% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.07% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     13: predicate.environ_get_eliminate
 1.07% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 0.83% :     0.000001s :      8: predicate.get_grad_eliminate
 0.41% :     0.000001s :      4: predicate.graph_param_transform
 5.41% :     0.000008s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.39% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      8: predicate.less_batch_normalization
 1.59% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.96% :     0.000003s :     21: predicate.load_eliminater
 1.68% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.12% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     13: predicate.merge_addn
 1.08% :     0.000002s :     13: predicate.minmaximum_grad
 1.88% :     0.000003s :      4: predicate.mutable_eliminate
 0.46% :     0.000001s :      4: predicate.opt_reshape
 2.38% :     0.000003s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     13: predicate.reduce_eliminate
 1.63% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.93% :     0.000003s :     25: predicate.replace_applicator
 0.75% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.39% :     0.000002s :     13: predicate.reshape_eliminate
 1.13% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     13: predicate.same_eliminate
 0.75% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      8: predicate.special_op_eliminate
 0.94% :     0.000001s :      8: predicate.specialize_transform
 1.21% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.49% :     0.000004s :     21: predicate.switch_defer_inline
 2.15% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.94% :     0.000011s :     67: predicate.switch_simplify
 1.21% :     0.000002s :     13: predicate.tile_eliminate
 1.60% :     0.000002s :     13: predicate.transpose_eliminate
 1.47% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.41% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     13: predicate.value_based_eliminate
 0.35% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000712    12
49.85% :     0.000355s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.97% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.17% :     0.000307s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017174    72
 0.33% :     0.000056s :      1: add_recomputation
 0.45% :     0.000077s :      1: auto_monad
 0.12% :     0.000021s :      1: auto_monad_reorder
 1.94% :     0.000333s :      1: bootstrap
 0.17% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000031s :      1: cse_after_recomputation
 0.05% :     0.000009s :      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
 0.12% :     0.000021s :      1: graph_reusing
36.41% :     0.006253s :      1: jit_opt_a
 1.04% :     0.000179s :      1: jit_opt_after_cconv
 0.36% :     0.000062s :      1: jit_opt_b
 3.30% :     0.000566s :      1: loop_unroll
 4.03% :     0.000692s :      1: mutable_eliminate
 6.26% :     0.001076s :     26: opt.transform.jit_opt_a
 0.34% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000032s :      4: opt.transform.jit_opt_b
 0.11% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.28% :     0.000563s :      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% :     0.000051s :      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
 2.49% :     0.000427s :      1: renormalize.infer
 2.00% :     0.000343s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000043s :      1: rewriter_after_opt_a
 0.57% :     0.000097s :      1: rewriter_before_opt_a
 0.49% :     0.000083s :      1: symbol_engine_optimizer
34.11% :     0.005858s :      1: type_inference


TotalTime = 0.0132727, [30]
[bootstrap]: 0.00028707
[type_inference]: 0.00495111
[event_method]: 2.068e-05
[auto_monad]: 6.474e-05
[graph_reusing]: 5.72001e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 3.935e-05
[rewriter_before_opt_a]: 8.389e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.00558388, [2]
    [Cycle 1]: 0.00186665, [27]
        [switch_simplify]: 5.364e-05
        [loop_unroll]: 3.573e-05
        [a_1]: 0.0006424
        [with_stream_mark]: 1.882e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 3.56999e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 6.98e-06
        [updatestate_useless_node_eliminater]: 6.14999e-06
        [accelerated_algorithm]: 6.19999e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 6.01998e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.483e-05
        [j_node_and_user_rematch]: 9.37999e-06
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 1.141e-05
        [inline_without_move]: 5.94999e-06
        [renormalize]: 0.00079085
        [add_forward_monad_depend]: 5.42001e-06
        [auto_monad_grad]: 2.88e-06
        [auto_monad_eliminator]: 1.368e-05
        [cse]: 3.031e-05
        [replace_applicator]: 1.405e-05
    [Cycle 2]: 0.00036163, [27]
        [switch_simplify]: 6.43e-06
        [loop_unroll]: 5.71e-06
        [a_1]: 0.00011383
        [with_stream_mark]: 1.096e-05
        [recompute_prepare]: 5.65001e-06
        [updatestate_depend_eliminate]: 3.25998e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.92001e-06
        [updatestate_useless_node_eliminater]: 5.72999e-06
        [accelerated_algorithm]: 6.04001e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 5.24998e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.415e-05
        [j_node_and_user_rematch]: 8.65999e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 9.87999e-06
        [inline_without_move]: 5.35001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 1.25999e-06
        [auto_monad_eliminator]: 5.86e-06
        [cse]: 1.338e-05
        [replace_applicator]: 5.79e-06
[py_interpret_to_execute_after_opt_a]: 2.064e-05
[rewriter_after_opt_a]: 4.488e-05
[convert_after_rewriter]: 7.53e-06
[order_py_execute_after_rewriter]: 5.06002e-06
[mutable_eliminate]: 0.00058411
[jit_opt_b]: 5.241e-05, [1]
    [Cycle 1]: 4.591e-05, [2]
        [frontend_op_eliminate]: 1.76e-05
        [inline_after_opt_a]: 1.698e-05
[cconv]: 2.587e-05
[loop_unroll]: 0.00043471
[jit_opt_after_cconv]: 0.00015487, [1]
    [Cycle 1]: 0.00014821, [11]
        [c_1]: 2.448e-05
        [parameter_eliminate]: 3.03998e-06
        [updatestate_depend_eliminate]: 5.00001e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [cse]: 2.17e-05
        [call_graph_tuple_transform]: 2.012e-05
        [tuple_list_get_item_eliminator]: 6.23e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 6.31e-06
[remove_dup_value]: 1.528e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 5.99999e-06
[add_recomputation]: 4.485e-05
[cse_after_recomputation]: 2.344e-05, [1]
    [Cycle 1]: 1.8e-05, [1]
        [cse]: 1.182e-05
[auto_monad_reorder]: 3.093e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 5.80002e-06
[opt_after_jit_grad]: 0.00049476
[symbol_engine_optimizer]: 7.64e-05, [1]
    [Cycle 1]: 7.034e-05, [6]
        [build]: 3.21999e-06
        [elim_shapecalc]: 8.37e-06
        [elim_not_effective]: 1.394e-05
        [opt_reshape]: 6.12999e-06
        [fold_const_symbol]: 9.83002e-06
        [renormalize]: 4.19997e-07
[validate]: 3.841e-05
Sums
    bootstrap                            :     0.000287s :  3.07%
    type_inference                       :     0.004951s : 52.93%
    event_method                         :     0.000021s :  0.22%
    auto_monad                           :     0.000065s :  0.69%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.42%
    rewriter_before_opt_a                :     0.000084s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000060s :  0.64%
    jit_opt_a.loop_unroll                :     0.000041s :  0.44%
    jit_opt_a.a_1                        :     0.000756s :  8.08%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000014s :  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        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  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          :     0.000021s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000791s :  8.45%
    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      :     0.000020s :  0.21%
    jit_opt_a.cse                        :     0.000044s :  0.47%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.22%
    rewriter_after_opt_a                 :     0.000045s :  0.48%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000584s :  6.24%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000026s :  0.28%
    loop_unroll                          :     0.000435s :  4.65%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.26%
    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 :     0.000003s :  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 :  0.22%
    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 :  0.00%
    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                    :     0.000045s :  0.48%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000031s :  0.33%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000495s :  5.29%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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 :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.41%


Time group info:
------[substitution.]   0.000234    28
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000002s :      2: substitution.fold_const_symbol
 2.40% :     0.000006s :      4: substitution.graph_param_transform
79.94% :     0.000187s :      4: substitution.inline
 1.16% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.54% :     0.000006s :      4: substitution.replace_old_param
10.36% :     0.000024s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004891     2
80.89% :     0.003956s :      1: type_inference.infer
19.11% :     0.000935s :      1: type_inference.specialize
------[replace.]   0.000086     8
56.76% :     0.000049s :      4: replace.inline
43.24% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000207     8
89.40% :     0.000185s :      4: match.inline
10.60% :     0.000022s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   912
 1.32% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     13: predicate.addn_check_dump
 1.15% :     0.000002s :     13: predicate.addn_zero_filter
 1.82% :     0.000002s :     13: predicate.arithmetic_simplify
 1.38% :     0.000002s :     13: predicate.cast_eliminate
 0.53% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     13: predicate.compare_switch_simplify
 1.24% :     0.000002s :     13: predicate.depend_value_elim
 1.18% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      4: predicate.elim_not_effective
 0.53% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     13: predicate.environ_get_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 0.95% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.49% :     0.000007s :     29: predicate.inline
 0.91% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      8: predicate.less_batch_normalization
 1.67% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.99% :     0.000003s :     21: predicate.load_eliminater
 1.32% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.14% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 1.41% :     0.000002s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.49% :     0.000003s :     21: predicate.partial_eliminate
 1.16% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     13: predicate.reduce_eliminate
 1.63% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.96% :     0.000003s :     25: predicate.replace_applicator
 1.00% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.15% :     0.000002s :     13: predicate.reshape_eliminate
 1.21% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     13: predicate.same_eliminate
 0.64% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     21: predicate.switch_defer_inline
 2.21% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.19% :     0.000010s :     67: predicate.switch_simplify
 1.24% :     0.000002s :     13: predicate.tile_eliminate
 1.45% :     0.000002s :     13: predicate.transpose_eliminate
 1.35% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     13: predicate.value_based_eliminate
 0.42% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000630    12
48.08% :     0.000303s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.31% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.61% :     0.000287s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015127    72
 0.31% :     0.000048s :      1: add_recomputation
 0.45% :     0.000068s :      1: auto_monad
 0.22% :     0.000034s :      1: auto_monad_reorder
 2.00% :     0.000303s :      1: bootstrap
 0.19% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000026s :      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
 0.05% :     0.000008s :      1: graph_reusing
36.94% :     0.005587s :      1: jit_opt_a
 1.04% :     0.000158s :      1: jit_opt_after_cconv
 0.36% :     0.000055s :      1: jit_opt_b
 2.92% :     0.000442s :      1: loop_unroll
 3.91% :     0.000592s :      1: mutable_eliminate
 6.56% :     0.000993s :     26: opt.transform.jit_opt_a
 0.36% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000029s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      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
 0.23% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.31% :     0.000501s :      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% :     0.000042s :      1: py_interpret_to_execute
 0.16% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.95% :     0.000446s :      1: renormalize.infer
 2.23% :     0.000337s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.32% :     0.000048s :      1: rewriter_after_opt_a
 0.57% :     0.000087s :      1: rewriter_before_opt_a
 0.52% :     0.000079s :      1: symbol_engine_optimizer
32.79% :     0.004961s :      1: type_inference


TotalTime = 0.0141512, [30]
[bootstrap]: 0.00032382
[type_inference]: 0.0054866
[event_method]: 2.301e-05
[auto_monad]: 6.934e-05
[graph_reusing]: 6.31998e-06
[pre_auto_parallel]: 2.22001e-06
[py_interpret_to_execute]: 4.228e-05
[rewriter_before_opt_a]: 8.979e-05
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.00575688, [2]
    [Cycle 1]: 0.00191164, [27]
        [switch_simplify]: 5.643e-05
        [loop_unroll]: 3.503e-05
        [a_1]: 0.00072305
        [with_stream_mark]: 1.911e-05
        [recompute_prepare]: 1.092e-05
        [updatestate_depend_eliminate]: 3.54002e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 7.25e-06
        [updatestate_useless_node_eliminater]: 5.94999e-06
        [accelerated_algorithm]: 6.44999e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 3.84002e-06
        [cell_reuse_recompute_pass]: 8.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.425e-05
        [j_node_and_user_rematch]: 9.71e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 1.263e-05
        [inline_without_move]: 6.01998e-06
        [renormalize]: 0.00073091
        [add_forward_monad_depend]: 7.16999e-06
        [auto_monad_grad]: 2.81999e-06
        [auto_monad_eliminator]: 1.554e-05
        [cse]: 2.301e-05
        [replace_applicator]: 1.706e-05
    [Cycle 2]: 0.00037942, [27]
        [switch_simplify]: 7.15e-06
        [loop_unroll]: 5.99e-06
        [a_1]: 0.00011504
        [with_stream_mark]: 1.021e-05
        [recompute_prepare]: 6.31e-06
        [updatestate_depend_eliminate]: 2.93998e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 6.56e-06
        [updatestate_useless_node_eliminater]: 5.57999e-06
        [accelerated_algorithm]: 6.26e-06
        [meta_shard_fg_expand]: 1.57999e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.23e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.425e-05
        [j_node_and_user_rematch]: 9.97999e-06
        [meta_fg_expand]: 2.56998e-06
        [replace_old_param]: 1.022e-05
        [inline_without_move]: 5.50001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.60001e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 8.13001e-06
        [cse]: 1.596e-05
        [replace_applicator]: 6.99001e-06
[py_interpret_to_execute_after_opt_a]: 1.229e-05
[rewriter_after_opt_a]: 3.694e-05
[convert_after_rewriter]: 7.88001e-06
[order_py_execute_after_rewriter]: 5.30999e-06
[mutable_eliminate]: 0.00060811
[jit_opt_b]: 5.507e-05, [1]
    [Cycle 1]: 4.837e-05, [2]
        [frontend_op_eliminate]: 1.799e-05
        [inline_after_opt_a]: 1.778e-05
[cconv]: 2.327e-05
[loop_unroll]: 0.00048342
[jit_opt_after_cconv]: 0.0001589, [1]
    [Cycle 1]: 0.00015213, [11]
        [c_1]: 2.537e-05
        [parameter_eliminate]: 2.96001e-06
        [updatestate_depend_eliminate]: 6.86999e-06
        [updatestate_assign_eliminate]: 2.65997e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [cse]: 2.39e-05
        [call_graph_tuple_transform]: 2.022e-05
        [tuple_list_get_item_eliminator]: 6.71e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 6.83e-06
[remove_dup_value]: 1.285e-05
[partial_unused_args_eliminate]: 2.31998e-06
[environ_conv]: 5.54e-06
[add_recomputation]: 4.77e-05
[cse_after_recomputation]: 2.509e-05, [1]
    [Cycle 1]: 1.935e-05, [1]
        [cse]: 1.247e-05
[auto_monad_reorder]: 1.752e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 5.29998e-06
[opt_after_jit_grad]: 0.00055076
[symbol_engine_optimizer]: 7.855e-05, [1]
    [Cycle 1]: 7.169e-05, [6]
        [build]: 3.29001e-06
        [elim_shapecalc]: 8.57e-06
        [elim_not_effective]: 1.478e-05
        [opt_reshape]: 6.17999e-06
        [fold_const_symbol]: 9.14003e-06
        [renormalize]: 3.59985e-07
[validate]: 3.767e-05
Sums
    bootstrap                            :     0.000324s :  3.21%
    type_inference                       :     0.005487s : 54.41%
    event_method                         :     0.000023s :  0.23%
    auto_monad                           :     0.000069s :  0.69%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.42%
    rewriter_before_opt_a                :     0.000090s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000064s :  0.63%
    jit_opt_a.loop_unroll                :     0.000041s :  0.41%
    jit_opt_a.a_1                        :     0.000838s :  8.31%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.17%
    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        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000014s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.28%
    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          :     0.000023s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.11%
    jit_opt_a.renormalize                :     0.000731s :  7.25%
    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      :     0.000024s :  0.23%
    jit_opt_a.cse                        :     0.000039s :  0.39%
    jit_opt_a.replace_applicator         :     0.000024s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.12%
    rewriter_after_opt_a                 :     0.000037s :  0.37%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000608s :  6.03%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.18%
    cconv                                :     0.000023s :  0.23%
    loop_unroll                          :     0.000483s :  4.79%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.25%
    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 :  0.02%
    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 :     0.000007s :  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  :     0.000007s :  0.07%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000048s :  0.47%
    cse_after_recomputation.cse          :     0.000012s :  0.12%
    auto_monad_reorder                   :     0.000018s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000551s :  5.46%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.37%


Time group info:
------[substitution.]   0.000282    28
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.92% :     0.000005s :      4: substitution.graph_param_transform
85.97% :     0.000242s :      4: substitution.inline
 0.89% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.61% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000006s :      4: substitution.replace_old_param
 6.13% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005400     2
81.28% :     0.004389s :      1: type_inference.infer
18.72% :     0.001011s :      1: type_inference.specialize
------[replace.]   0.000095     8
57.92% :     0.000055s :      4: replace.inline
42.08% :     0.000040s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000254     8
93.97% :     0.000239s :      4: match.inline
 6.03% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   912
 1.29% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.19% :     0.000002s :     13: predicate.addn_check_dump
 1.32% :     0.000002s :     13: predicate.addn_zero_filter
 2.58% :     0.000004s :     13: predicate.arithmetic_simplify
 1.63% :     0.000002s :     13: predicate.cast_eliminate
 0.42% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     13: predicate.compare_switch_simplify
 1.16% :     0.000002s :     13: predicate.depend_value_elim
 1.19% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      4: predicate.elim_not_effective
 0.45% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.39% :     0.000002s :     13: predicate.environ_get_eliminate
 1.27% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.88% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.22% :     0.000007s :     29: predicate.inline
 0.84% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      8: predicate.less_batch_normalization
 1.58% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.01% :     0.000003s :     21: predicate.load_eliminater
 1.17% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.98% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000002s :     13: predicate.merge_addn
 1.08% :     0.000002s :     13: predicate.minmaximum_grad
 1.31% :     0.000002s :      4: predicate.mutable_eliminate
 0.41% :     0.000001s :      4: predicate.opt_reshape
 2.35% :     0.000003s :     21: predicate.partial_eliminate
 1.14% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.79% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.93% :     0.000003s :     25: predicate.replace_applicator
 0.58% :     0.000001s :      8: predicate.replace_old_param
 0.39% :     0.000001s :      4: predicate.reset_defer_inline
 1.17% :     0.000002s :     13: predicate.reshape_eliminate
 1.22% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.46% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.89% :     0.000004s :     21: predicate.switch_defer_inline
 2.14% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.78% :     0.000011s :     67: predicate.switch_simplify
 1.19% :     0.000002s :     13: predicate.tile_eliminate
 1.75% :     0.000002s :     13: predicate.transpose_eliminate
 1.40% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.47% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000674    12
49.72% :     0.000335s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.22% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.06% :     0.000304s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016046    72
 0.32% :     0.000051s :      1: add_recomputation
 0.46% :     0.000074s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 2.12% :     0.000340s :      1: bootstrap
 0.16% :     0.000026s :      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
 0.17% :     0.000027s :      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
35.90% :     0.005760s :      1: jit_opt_a
 1.01% :     0.000162s :      1: jit_opt_after_cconv
 0.36% :     0.000058s :      1: jit_opt_b
 3.06% :     0.000492s :      1: loop_unroll
 3.84% :     0.000617s :      1: mutable_eliminate
 6.77% :     0.001086s :     26: opt.transform.jit_opt_a
 0.35% :     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
 0.22% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.48% :     0.000559s :      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% :     0.000046s :      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
 2.50% :     0.000401s :      1: renormalize.infer
 1.99% :     0.000320s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000040s :      1: rewriter_after_opt_a
 0.58% :     0.000094s :      1: rewriter_before_opt_a
 0.51% :     0.000081s :      1: symbol_engine_optimizer
34.26% :     0.005497s :      1: type_inference


TotalTime = 0.0468292, [33]
[bootstrap]: 0.00031737
[type_inference]: 0.0244302
[event_method]: 2.451e-05
[auto_monad]: 0.00017137
[graph_reusing]: 7.41001e-06
[pre_auto_parallel]: 2.89999e-06
[py_interpret_to_execute]: 0.00015222
[rewriter_before_opt_a]: 0.00011004
[expand_dump_flag]: 4.12e-06
[jit_opt_a]: 0.00699543, [2]
    [Cycle 1]: 0.0027665, [27]
        [switch_simplify]: 6.068e-05
        [loop_unroll]: 3.613e-05
        [a_1]: 0.00071879
        [with_stream_mark]: 2.071e-05
        [recompute_prepare]: 1.165e-05
        [updatestate_depend_eliminate]: 6.48e-06
        [updatestate_assign_eliminate]: 5.00999e-06
        [updatestate_loads_eliminate]: 5.07e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 9.15999e-06
        [updatestate_useless_node_eliminater]: 1.184e-05
        [accelerated_algorithm]: 9.32001e-06
        [meta_shard_fg_expand]: 2.71999e-06
        [get_grad_eliminate_]: 8.23999e-06
        [merge_forward]: 4.65001e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.681e-05
        [j_node_and_user_rematch]: 1.431e-05
        [meta_fg_expand]: 3.31999e-06
        [replace_old_param]: 1.269e-05
        [inline_without_move]: 8.42e-06
        [renormalize]: 0.00151942
        [add_forward_monad_depend]: 6.41e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 2.245e-05
        [cse]: 3.518e-05
        [replace_applicator]: 1.832e-05
    [Cycle 2]: 0.0004923, [27]
        [switch_simplify]: 9.37001e-06
        [loop_unroll]: 8.37e-06
        [a_1]: 0.00017677
        [with_stream_mark]: 1.55e-05
        [recompute_prepare]: 8.25999e-06
        [updatestate_depend_eliminate]: 5.23002e-06
        [updatestate_assign_eliminate]: 5.56e-06
        [updatestate_loads_eliminate]: 3.83001e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 8.35001e-06
        [updatestate_useless_node_eliminater]: 1.189e-05
        [accelerated_algorithm]: 8.55001e-06
        [meta_shard_fg_expand]: 2.64001e-06
        [get_grad_eliminate_]: 7.93001e-06
        [merge_forward]: 4.93001e-06
        [cell_reuse_recompute_pass]: 2.68998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.765e-05
        [j_node_and_user_rematch]: 1.297e-05
        [meta_fg_expand]: 2.89001e-06
        [replace_old_param]: 1.059e-05
        [inline_without_move]: 7.69002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49998e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 1.082e-05
        [cse]: 2.101e-05
        [replace_applicator]: 7.91001e-06
[py_interpret_to_execute_after_opt_a]: 1.431e-05
[rewriter_after_opt_a]: 0.00030785
[convert_after_rewriter]: 1.166e-05
[order_py_execute_after_rewriter]: 8.05e-06
[mutable_eliminate]: 0.00077826
[jit_opt_b]: 7.491e-05, [1]
    [Cycle 1]: 6.664e-05, [2]
        [frontend_op_eliminate]: 2.545e-05
        [inline_after_opt_a]: 2.856e-05
[cconv]: 3.351e-05
[loop_unroll]: 0.00052678
[jit_opt_after_cconv]: 0.00021863, [1]
    [Cycle 1]: 0.00021098, [11]
        [c_1]: 5.079e-05
        [parameter_eliminate]: 4.47e-06
        [updatestate_depend_eliminate]: 8.01001e-06
        [updatestate_assign_eliminate]: 4.91002e-06
        [updatestate_loads_eliminate]: 4.63001e-06
        [cse]: 4.124e-05
        [call_graph_tuple_transform]: 2.395e-05
        [tuple_list_get_item_eliminator]: 8.55999e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 1.04998e-06
        [switch_simplify]: 9.31e-06
[remove_dup_value]: 2.048e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 1.18e-05
[add_recomputation]: 7.385e-05
[cse_after_recomputation]: 6.636e-05, [1]
    [Cycle 1]: 5.886e-05, [1]
        [cse]: 1.991e-05
[auto_monad_reorder]: 2.687e-05
[get_jit_bprop_graph]: 2.08002e-06
[rewriter_after_jit_bprop_graph]: 8.08999e-06
[opt_after_jit_grad]: 0.00057273
[symbol_engine_optimizer]: 0.00011199, [1]
    [Cycle 1]: 0.00010415, [6]
        [build]: 1.675e-05
        [elim_shapecalc]: 1.204e-05
        [elim_not_effective]: 2.062e-05
        [opt_reshape]: 9.35001e-06
        [fold_const_symbol]: 1.493e-05
        [renormalize]: 6.39993e-07
[validate]: 6.326e-05
[backend_pass]: 1.31002e-06
[task_emit]: 0.0114186
[execute]: 9.64e-06
Sums
    bootstrap                            :     0.000317s :  0.75%
    type_inference                       :     0.024430s : 57.66%
    event_method                         :     0.000025s :  0.06%
    auto_monad                           :     0.000171s :  0.40%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000152s :  0.36%
    rewriter_before_opt_a                :     0.000110s :  0.26%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000070s :  0.17%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000896s :  2.11%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  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 :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    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        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001519s :  3.59%
    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      :     0.000033s :  0.08%
    jit_opt_a.cse                        :     0.000056s :  0.13%
    jit_opt_a.replace_applicator         :     0.000026s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000308s :  0.73%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000778s :  1.84%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000029s :  0.07%
    cconv                                :     0.000034s :  0.08%
    loop_unroll                          :     0.000527s :  1.24%
    jit_opt_after_cconv.c_1              :     0.000051s :  0.12%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000041s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000074s :  0.17%
    cse_after_recomputation.cse          :     0.000020s :  0.05%
    auto_monad_reorder                   :     0.000027s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000573s :  1.35%
    symbol_engine_optimizer.build        :     0.000017s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000063s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011419s : 26.95%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000279    47
 3.56% :     0.000010s :      2: substitution.depend_value_elim
 1.01% :     0.000003s :      4: substitution.elim_not_effective
 0.78% :     0.000002s :      4: substitution.fold_const_symbol
 2.53% :     0.000007s :      5: substitution.graph_param_transform
74.44% :     0.000208s :      4: substitution.inline
 1.82% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.48% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.65% :     0.000005s :      2: substitution.replace_old_param
 4.24% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.86% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.63% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024344     2
93.26% :     0.022704s :      1: type_inference.infer
 6.74% :     0.001640s :      1: type_inference.specialize
------[replace.]   0.000094     6
65.31% :     0.000061s :      4: replace.inline
34.69% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000216     6
95.04% :     0.000205s :      4: match.inline
 4.96% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1021
 1.32% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.50% :     0.000003s :     15: predicate.addn_zero_filter
 2.11% :     0.000004s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.46% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     15: predicate.compare_switch_simplify
 1.28% :     0.000002s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      5: predicate.elim_not_effective
 0.52% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.14% :     0.000002s :     10: predicate.get_grad_eliminate
 0.33% :     0.000001s :      5: predicate.graph_param_transform
 4.77% :     0.000008s :     31: predicate.inline
 1.27% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.46% :     0.000003s :     10: predicate.less_batch_normalization
 1.37% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.03% :     0.000004s :     22: predicate.load_eliminater
 1.34% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.32% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.16% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     15: predicate.merge_addn
 1.22% :     0.000002s :     15: predicate.minmaximum_grad
 1.98% :     0.000004s :      5: predicate.mutable_eliminate
 0.55% :     0.000001s :      5: predicate.opt_reshape
 2.24% :     0.000004s :     22: predicate.partial_eliminate
 1.65% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.71% :     0.000003s :     15: predicate.reduce_eliminate
 1.41% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.59% :     0.000003s :     27: predicate.replace_applicator
 0.68% :     0.000001s :     10: predicate.replace_old_param
 0.64% :     0.000001s :      5: predicate.reset_defer_inline
 1.21% :     0.000002s :     15: predicate.reshape_eliminate
 1.35% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.09% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.02% :     0.000002s :     10: predicate.special_op_eliminate
 1.10% :     0.000002s :     10: predicate.specialize_transform
 1.42% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.62% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.17% :     0.000004s :     21: predicate.switch_defer_inline
 1.87% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.33% :     0.000011s :     68: predicate.switch_simplify
 1.26% :     0.000002s :     15: predicate.tile_eliminate
 1.42% :     0.000003s :     15: predicate.transpose_eliminate
 1.54% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 2.11% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.75% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.34% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000003s :     15: predicate.value_based_eliminate
 0.45% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000903    13
43.07% :     0.000389s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.81% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.11% :     0.000470s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049755    76
 0.16% :     0.000078s :      1: add_recomputation
 0.36% :     0.000177s :      1: auto_monad
 0.06% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000331s :      1: bootstrap
 0.07% :     0.000036s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.14% :     0.000069s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.06% :     0.000029s :      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% :     0.000010s :      1: graph_reusing
14.07% :     0.006999s :      1: jit_opt_a
 0.45% :     0.000222s :      1: jit_opt_after_cconv
 0.16% :     0.000078s :      1: jit_opt_b
 1.08% :     0.000538s :      1: loop_unroll
 1.59% :     0.000792s :      1: mutable_eliminate
 2.42% :     0.001204s :     26: opt.transform.jit_opt_a
 0.18% :     0.000089s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000046s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.17% :     0.000583s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000157s :      1: py_interpret_to_execute
 0.03% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000024s :      1: remove_dup_value
 1.96% :     0.000973s :      1: renormalize.infer
 1.08% :     0.000536s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000313s :      1: rewriter_after_opt_a
 0.23% :     0.000115s :      1: rewriter_before_opt_a
 0.23% :     0.000115s :      1: symbol_engine_optimizer
22.99% :     0.011439s :      1: task_emit
49.14% :     0.024451s :      1: type_inference
 0.18% :     0.000089s :      1: validate


TotalTime = 0.0125927, [30]
[bootstrap]: 0.00026319
[type_inference]: 0.00451683
[event_method]: 2.075e-05
[auto_monad]: 5.076e-05
[graph_reusing]: 4.80999e-06
[pre_auto_parallel]: 2.14e-06
[py_interpret_to_execute]: 3.748e-05
[rewriter_before_opt_a]: 8.243e-05
[expand_dump_flag]: 2.39001e-06
[jit_opt_a]: 0.00539934, [2]
    [Cycle 1]: 0.00178576, [27]
        [switch_simplify]: 5.4e-05
        [loop_unroll]: 3.467e-05
        [a_1]: 0.00062868
        [with_stream_mark]: 1.488e-05
        [recompute_prepare]: 8.12e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.44e-06
        [specialize_transform]: 7.03e-06
        [updatestate_useless_node_eliminater]: 6.82002e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.84e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.233e-05
        [j_node_and_user_rematch]: 9.94999e-06
        [meta_fg_expand]: 1.91003e-06
        [replace_old_param]: 1.031e-05
        [inline_without_move]: 6.04001e-06
        [renormalize]: 0.00072828
        [add_forward_monad_depend]: 4.18999e-06
        [auto_monad_grad]: 2.04e-06
        [auto_monad_eliminator]: 1.325e-05
        [cse]: 2.657e-05
        [replace_applicator]: 1.419e-05
    [Cycle 2]: 0.00035856, [27]
        [switch_simplify]: 6.64001e-06
        [loop_unroll]: 5.87001e-06
        [a_1]: 0.00011229
        [with_stream_mark]: 1.054e-05
        [recompute_prepare]: 5.79999e-06
        [updatestate_depend_eliminate]: 2.93998e-06
        [updatestate_assign_eliminate]: 2.43002e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.62999e-06
        [updatestate_useless_node_eliminater]: 5.61e-06
        [accelerated_algorithm]: 5.87999e-06
        [meta_shard_fg_expand]: 1.75001e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 3.06999e-06
        [cell_reuse_recompute_pass]: 1.60001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.501e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 9.91e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 1.60013e-07
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 8.59989e-07
        [auto_monad_eliminator]: 6.60997e-06
        [cse]: 1.22e-05
        [replace_applicator]: 5.87999e-06
[py_interpret_to_execute_after_opt_a]: 1.095e-05
[rewriter_after_opt_a]: 3.239e-05
[convert_after_rewriter]: 6.71e-06
[order_py_execute_after_rewriter]: 7.46001e-06
[mutable_eliminate]: 0.00056563
[jit_opt_b]: 6.982e-05, [1]
    [Cycle 1]: 6.279e-05, [2]
        [frontend_op_eliminate]: 1.999e-05
        [inline_after_opt_a]: 1.673e-05
[cconv]: 1.787e-05
[loop_unroll]: 0.00042579
[jit_opt_after_cconv]: 0.00015469, [1]
    [Cycle 1]: 0.00014814, [11]
        [c_1]: 2.526e-05
        [parameter_eliminate]: 2.74999e-06
        [updatestate_depend_eliminate]: 5.46e-06
        [updatestate_assign_eliminate]: 5.54e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [cse]: 2.392e-05
        [call_graph_tuple_transform]: 1.956e-05
        [tuple_list_get_item_eliminator]: 6.19001e-06
        [none_parameter_eliminate]: 1.04e-06
        [renormalize]: 2.40019e-07
        [switch_simplify]: 6.21e-06
[remove_dup_value]: 9.32999e-06
[partial_unused_args_eliminate]: 1.76e-06
[environ_conv]: 4.60999e-06
[add_recomputation]: 3.415e-05
[cse_after_recomputation]: 2.243e-05, [1]
    [Cycle 1]: 1.723e-05, [1]
        [cse]: 1.173e-05
[auto_monad_reorder]: 1.322e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 6.21e-06
[opt_after_jit_grad]: 0.00049637
[symbol_engine_optimizer]: 0.00013192, [1]
    [Cycle 1]: 0.00012572, [6]
        [build]: 3.71999e-06
        [elim_shapecalc]: 8.95001e-06
        [elim_not_effective]: 1.702e-05
        [opt_reshape]: 6.70998e-06
        [fold_const_symbol]: 9.46e-06
        [renormalize]: 6.10016e-07
[validate]: 3.76e-05
Sums
    bootstrap                            :     0.000263s :  3.02%
    type_inference                       :     0.004517s : 51.90%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000051s :  0.58%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.43%
    rewriter_before_opt_a                :     0.000082s :  0.95%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.70%
    jit_opt_a.loop_unroll                :     0.000041s :  0.47%
    jit_opt_a.a_1                        :     0.000741s :  8.51%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.29%
    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       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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_        :     0.000011s :  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 :     0.000027s :  0.31%
    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          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000728s :  8.37%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.23%
    jit_opt_a.cse                        :     0.000039s :  0.45%
    jit_opt_a.replace_applicator         :     0.000020s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000032s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000007s :  0.09%
    mutable_eliminate                    :     0.000566s :  6.50%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000018s :  0.21%
    loop_unroll                          :     0.000426s :  4.89%
    jit_opt_after_cconv.c_1              :     0.000025s :  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 :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000024s :  0.27%
    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 :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000009s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000034s :  0.39%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000013s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000496s :  5.70%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.20%
    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                             :     0.000038s :  0.43%


Time group info:
------[substitution.]   0.000199    28
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.23% :     0.000004s :      4: substitution.graph_param_transform
82.86% :     0.000165s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.08% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.65% :     0.000005s :      4: substitution.replace_old_param
 6.92% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004458     2
79.85% :     0.003560s :      1: type_inference.infer
20.15% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000083     8
56.46% :     0.000047s :      4: replace.inline
43.54% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     8
93.16% :     0.000162s :      4: match.inline
 6.84% :     0.000012s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   912
 1.27% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     13: predicate.addn_check_dump
 1.25% :     0.000002s :     13: predicate.addn_zero_filter
 1.91% :     0.000003s :     13: predicate.arithmetic_simplify
 1.39% :     0.000002s :     13: predicate.cast_eliminate
 0.48% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     13: predicate.compare_switch_simplify
 1.10% :     0.000002s :     13: predicate.depend_value_elim
 1.10% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.17% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      4: predicate.elim_not_effective
 0.47% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000002s :     13: predicate.environ_add_const_eliminate
 6.33% :     0.000009s :     13: predicate.environ_get_add_eliminate
 1.06% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     13: predicate.environ_get_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000001s :      8: predicate.get_grad_eliminate
 0.23% :     0.000000s :      4: predicate.graph_param_transform
 5.12% :     0.000007s :     29: predicate.inline
 0.91% :     0.000001s :      8: predicate.inline_without_move
 0.37% :     0.000001s :      8: predicate.j_node_and_user_rematch
 0.98% :     0.000001s :      8: predicate.less_batch_normalization
 1.58% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.87% :     0.000003s :     21: predicate.load_eliminater
 1.24% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.77% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.85% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     13: predicate.merge_addn
 1.10% :     0.000002s :     13: predicate.minmaximum_grad
 1.37% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.27% :     0.000003s :     21: predicate.partial_eliminate
 1.17% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     13: predicate.reduce_eliminate
 1.49% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.85% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.91% :     0.000003s :     25: predicate.replace_applicator
 0.75% :     0.000001s :      8: predicate.replace_old_param
 0.29% :     0.000000s :      4: predicate.reset_defer_inline
 1.08% :     0.000002s :     13: predicate.reshape_eliminate
 1.12% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.12% :     0.000002s :     13: predicate.same_eliminate
 0.55% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      8: predicate.special_op_eliminate
 0.89% :     0.000001s :      8: predicate.specialize_transform
 1.35% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     21: predicate.switch_defer_inline
 2.24% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.34% :     0.000011s :     67: predicate.switch_simplify
 1.10% :     0.000002s :     13: predicate.tile_eliminate
 1.54% :     0.000002s :     13: predicate.transpose_eliminate
 1.31% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     21: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.36% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000570    12
45.72% :     0.000261s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.10% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.18% :     0.000280s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014381    72
 0.26% :     0.000037s :      1: add_recomputation
 0.38% :     0.000054s :      1: auto_monad
 0.11% :     0.000016s :      1: auto_monad_reorder
 1.92% :     0.000277s :      1: bootstrap
 0.14% :     0.000020s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.17% :     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
37.57% :     0.005403s :      1: jit_opt_a
 1.10% :     0.000158s :      1: jit_opt_after_cconv
 0.51% :     0.000073s :      1: jit_opt_b
 3.01% :     0.000434s :      1: loop_unroll
 3.99% :     0.000574s :      1: mutable_eliminate
 6.80% :     0.000978s :     26: opt.transform.jit_opt_a
 0.38% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000031s :      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
 0.18% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.51% :     0.000504s :      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
 0.28% :     0.000041s :      1: py_interpret_to_execute
 0.10% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000012s :      1: remove_dup_value
 2.80% :     0.000403s :      1: renormalize.infer
 2.21% :     0.000317s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000035s :      1: rewriter_after_opt_a
 0.59% :     0.000085s :      1: rewriter_before_opt_a
 0.94% :     0.000135s :      1: symbol_engine_optimizer
31.49% :     0.004529s :      1: type_inference


TotalTime = 0.0143092, [30]
[bootstrap]: 0.00031115
[type_inference]: 0.00562417
[event_method]: 2.336e-05
[auto_monad]: 6.814e-05
[graph_reusing]: 5.83997e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 4.129e-05
[rewriter_before_opt_a]: 9.182e-05
[expand_dump_flag]: 3.21999e-06
[jit_opt_a]: 0.00587659, [2]
    [Cycle 1]: 0.00192403, [27]
        [switch_simplify]: 6.312e-05
        [loop_unroll]: 3.522e-05
        [a_1]: 0.00065356
        [with_stream_mark]: 2.018e-05
        [recompute_prepare]: 8.69e-06
        [updatestate_depend_eliminate]: 4.66997e-06
        [updatestate_assign_eliminate]: 3.53e-06
        [updatestate_loads_eliminate]: 2.91999e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 7.02002e-06
        [updatestate_useless_node_eliminater]: 6.09999e-06
        [accelerated_algorithm]: 6.24999e-06
        [meta_shard_fg_expand]: 2.38002e-06
        [get_grad_eliminate_]: 5.77001e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.525e-05
        [j_node_and_user_rematch]: 1.001e-05
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 1.106e-05
        [inline_without_move]: 6.08998e-06
        [renormalize]: 0.00078645
        [add_forward_monad_depend]: 6.53e-06
        [auto_monad_grad]: 2.17001e-06
        [auto_monad_eliminator]: 1.526e-05
        [cse]: 3.632e-05
        [replace_applicator]: 1.513e-05
    [Cycle 2]: 0.00036821, [27]
        [switch_simplify]: 6.79999e-06
        [loop_unroll]: 6.49001e-06
        [a_1]: 0.00011329
        [with_stream_mark]: 1.219e-05
        [recompute_prepare]: 6.10002e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.57999e-06
        [updatestate_useless_node_eliminater]: 5.53002e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.84998e-06
        [get_grad_eliminate_]: 5.25999e-06
        [merge_forward]: 3.88999e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.493e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 2.45002e-06
        [replace_old_param]: 9.44e-06
        [inline_without_move]: 5.48002e-06
        [renormalize]: 1.19995e-07
        [add_forward_monad_depend]: 1.67001e-06
        [auto_monad_grad]: 1.34003e-06
        [auto_monad_eliminator]: 6.27001e-06
        [cse]: 1.529e-05
        [replace_applicator]: 5.54e-06
[py_interpret_to_execute_after_opt_a]: 1.143e-05
[rewriter_after_opt_a]: 3.741e-05
[convert_after_rewriter]: 7.08998e-06
[order_py_execute_after_rewriter]: 5.36998e-06
[mutable_eliminate]: 0.00065405
[jit_opt_b]: 5.656e-05, [1]
    [Cycle 1]: 4.897e-05, [2]
        [frontend_op_eliminate]: 1.908e-05
        [inline_after_opt_a]: 1.83e-05
[cconv]: 2.608e-05
[loop_unroll]: 0.00042625
[jit_opt_after_cconv]: 0.00015911, [1]
    [Cycle 1]: 0.00015264, [11]
        [c_1]: 2.464e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 5.61998e-06
        [updatestate_assign_eliminate]: 2.65002e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [cse]: 2.485e-05
        [call_graph_tuple_transform]: 2.135e-05
        [tuple_list_get_item_eliminator]: 9.54e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 6.42001e-06
[remove_dup_value]: 1.474e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 6.12001e-06
[add_recomputation]: 5.167e-05
[cse_after_recomputation]: 2.454e-05, [1]
    [Cycle 1]: 1.875e-05, [1]
        [cse]: 1.304e-05
[auto_monad_reorder]: 1.837e-05
[get_jit_bprop_graph]: 1.95001e-06
[rewriter_after_jit_bprop_graph]: 5.49e-06
[opt_after_jit_grad]: 0.00046338
[symbol_engine_optimizer]: 7.441e-05, [1]
    [Cycle 1]: 6.845e-05, [6]
        [build]: 3.90998e-06
        [elim_shapecalc]: 8.06001e-06
        [elim_not_effective]: 1.319e-05
        [opt_reshape]: 6.26998e-06
        [fold_const_symbol]: 9.49999e-06
        [renormalize]: 3.9002e-07
[validate]: 4.078e-05
Sums
    bootstrap                            :     0.000311s :  3.07%
    type_inference                       :     0.005624s : 55.53%
    event_method                         :     0.000023s :  0.23%
    auto_monad                           :     0.000068s :  0.67%
    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                :     0.000092s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000070s :  0.69%
    jit_opt_a.loop_unroll                :     0.000042s :  0.41%
    jit_opt_a.a_1                        :     0.000767s :  7.57%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.32%
    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 :     0.000007s :  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       :     0.000013s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.30%
    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          :     0.000020s :  0.20%
    jit_opt_a.inline_without_move        :     0.000012s :  0.11%
    jit_opt_a.renormalize                :     0.000787s :  7.77%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.21%
    jit_opt_a.cse                        :     0.000052s :  0.51%
    jit_opt_a.replace_applicator         :     0.000021s :  0.20%
    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                    :     0.000654s :  6.46%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.18%
    cconv                                :     0.000026s :  0.26%
    loop_unroll                          :     0.000426s :  4.21%
    jit_opt_after_cconv.c_1              :     0.000025s :  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 :     0.000003s :  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 :     0.000021s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000010s :  0.09%
    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 :  0.51%
    cse_after_recomputation.cse          :     0.000013s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000463s :  4.58%
    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 :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.40%


Time group info:
------[substitution.]   0.000240    28
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000006s :      4: substitution.graph_param_transform
83.28% :     0.000200s :      4: substitution.inline
 1.29% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000005s :      4: substitution.replace_old_param
 7.15% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005547     2
81.25% :     0.004508s :      1: type_inference.infer
18.75% :     0.001040s :      1: type_inference.specialize
------[replace.]   0.000087     8
58.30% :     0.000051s :      4: replace.inline
41.70% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     8
92.81% :     0.000197s :      4: match.inline
 7.19% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   912
 1.27% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     13: predicate.addn_check_dump
 1.25% :     0.000002s :     13: predicate.addn_zero_filter
 2.18% :     0.000003s :     13: predicate.arithmetic_simplify
 1.39% :     0.000002s :     13: predicate.cast_eliminate
 0.54% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.10% :     0.000002s :     13: predicate.compare_switch_simplify
 1.13% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      4: predicate.elim_not_effective
 0.51% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.92% :     0.000001s :      8: predicate.get_grad_eliminate
 0.39% :     0.000001s :      4: predicate.graph_param_transform
 5.24% :     0.000007s :     29: predicate.inline
 0.93% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :      8: predicate.less_batch_normalization
 1.62% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.08% :     0.000003s :     21: predicate.load_eliminater
 1.25% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.23% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     13: predicate.merge_addn
 1.06% :     0.000001s :     13: predicate.minmaximum_grad
 1.62% :     0.000002s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.46% :     0.000003s :     21: predicate.partial_eliminate
 1.17% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     13: predicate.reduce_eliminate
 1.76% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.00% :     0.000003s :     25: predicate.replace_applicator
 0.73% :     0.000001s :      8: predicate.replace_old_param
 0.49% :     0.000001s :      4: predicate.reset_defer_inline
 1.28% :     0.000002s :     13: predicate.reshape_eliminate
 1.16% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.38% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     21: predicate.switch_defer_inline
 2.16% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.55% :     0.000010s :     67: predicate.switch_simplify
 1.14% :     0.000002s :     13: predicate.tile_eliminate
 1.13% :     0.000002s :     13: predicate.transpose_eliminate
 1.49% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000715    12
51.29% :     0.000366s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.93% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.78% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016180    72
 0.34% :     0.000054s :      1: add_recomputation
 0.44% :     0.000072s :      1: auto_monad
 0.13% :     0.000021s :      1: auto_monad_reorder
 2.04% :     0.000330s :      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
 0.17% :     0.000027s :      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
36.34% :     0.005880s :      1: jit_opt_a
 1.00% :     0.000162s :      1: jit_opt_after_cconv
 0.37% :     0.000060s :      1: jit_opt_b
 2.68% :     0.000433s :      1: loop_unroll
 4.11% :     0.000664s :      1: mutable_eliminate
 6.26% :     0.001013s :     26: opt.transform.jit_opt_a
 0.36% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000031s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      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
 2.91% :     0.000471s :      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% :     0.000044s :      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
 2.79% :     0.000451s :      1: renormalize.infer
 2.02% :     0.000326s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000040s :      1: rewriter_after_opt_a
 0.59% :     0.000095s :      1: rewriter_before_opt_a
 0.48% :     0.000077s :      1: symbol_engine_optimizer
34.84% :     0.005636s :      1: type_inference


TotalTime = 0.0439251, [33]
[bootstrap]: 0.00028797
[type_inference]: 0.0229612
[event_method]: 2.449e-05
[auto_monad]: 0.00016814
[graph_reusing]: 7.06001e-06
[pre_auto_parallel]: 3.73999e-06
[py_interpret_to_execute]: 0.0001302
[rewriter_before_opt_a]: 0.00010097
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00677292, [2]
    [Cycle 1]: 0.00255234, [27]
        [switch_simplify]: 6.992e-05
        [loop_unroll]: 3.638e-05
        [a_1]: 0.00073801
        [with_stream_mark]: 2.331e-05
        [recompute_prepare]: 1.227e-05
        [updatestate_depend_eliminate]: 6.16e-06
        [updatestate_assign_eliminate]: 5.59e-06
        [updatestate_loads_eliminate]: 5.14e-06
        [parameter_eliminate]: 2.00002e-06
        [specialize_transform]: 9.32999e-06
        [updatestate_useless_node_eliminater]: 1.188e-05
        [accelerated_algorithm]: 9.12001e-06
        [meta_shard_fg_expand]: 2.93998e-06
        [get_grad_eliminate_]: 8.64998e-06
        [merge_forward]: 5.59998e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.938e-05
        [j_node_and_user_rematch]: 1.373e-05
        [meta_fg_expand]: 3.98001e-06
        [replace_old_param]: 1.207e-05
        [inline_without_move]: 8.55999e-06
        [renormalize]: 0.00125631
        [add_forward_monad_depend]: 6.25002e-06
        [auto_monad_grad]: 3.12002e-06
        [auto_monad_eliminator]: 2.116e-05
        [cse]: 4.443e-05
        [replace_applicator]: 1.901e-05
    [Cycle 2]: 0.00047872, [27]
        [switch_simplify]: 9.05999e-06
        [loop_unroll]: 7.93001e-06
        [a_1]: 0.00016863
        [with_stream_mark]: 1.259e-05
        [recompute_prepare]: 8.57e-06
        [updatestate_depend_eliminate]: 5.25001e-06
        [updatestate_assign_eliminate]: 4.53001e-06
        [updatestate_loads_eliminate]: 3.93001e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 8.15999e-06
        [updatestate_useless_node_eliminater]: 1.118e-05
        [accelerated_algorithm]: 8.43001e-06
        [meta_shard_fg_expand]: 2.41e-06
        [get_grad_eliminate_]: 7.60998e-06
        [merge_forward]: 5.17999e-06
        [cell_reuse_recompute_pass]: 2.37001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.762e-05
        [j_node_and_user_rematch]: 1.271e-05
        [meta_fg_expand]: 3.48e-06
        [replace_old_param]: 1.052e-05
        [inline_without_move]: 7.89002e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 1.55999e-06
        [auto_monad_eliminator]: 1.049e-05
        [cse]: 2.199e-05
        [replace_applicator]: 8.27e-06
[py_interpret_to_execute_after_opt_a]: 1.351e-05
[rewriter_after_opt_a]: 0.00028827
[convert_after_rewriter]: 1.098e-05
[order_py_execute_after_rewriter]: 7.61999e-06
[mutable_eliminate]: 0.00074224
[jit_opt_b]: 7.221e-05, [1]
    [Cycle 1]: 6.404e-05, [2]
        [frontend_op_eliminate]: 2.552e-05
        [inline_after_opt_a]: 2.531e-05
[cconv]: 2.807e-05
[loop_unroll]: 0.00048483
[jit_opt_after_cconv]: 0.00020621, [1]
    [Cycle 1]: 0.00019934, [11]
        [c_1]: 4.679e-05
        [parameter_eliminate]: 3.18e-06
        [updatestate_depend_eliminate]: 8.45999e-06
        [updatestate_assign_eliminate]: 4.43999e-06
        [updatestate_loads_eliminate]: 4.13001e-06
        [cse]: 3.503e-05
        [call_graph_tuple_transform]: 2.262e-05
        [tuple_list_get_item_eliminator]: 8.54e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 1.067e-05
[remove_dup_value]: 1.977e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 1.065e-05
[add_recomputation]: 6.657e-05
[cse_after_recomputation]: 3.047e-05, [1]
    [Cycle 1]: 2.415e-05, [1]
        [cse]: 1.785e-05
[auto_monad_reorder]: 2.479e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 6.54001e-06
[opt_after_jit_grad]: 0.00054617
[symbol_engine_optimizer]: 0.00010527, [1]
    [Cycle 1]: 9.808e-05, [6]
        [build]: 1.319e-05
        [elim_shapecalc]: 1.152e-05
        [elim_not_effective]: 2.044e-05
        [opt_reshape]: 8.75999e-06
        [fold_const_symbol]: 1.377e-05
        [renormalize]: 5.8001e-07
[validate]: 5.729e-05
[backend_pass]: 1.04003e-06
[task_emit]: 0.010498
[execute]: 9.05001e-06
Sums
    bootstrap                            :     0.000288s :  0.73%
    type_inference                       :     0.022961s : 58.12%
    event_method                         :     0.000024s :  0.06%
    auto_monad                           :     0.000168s :  0.43%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000130s :  0.33%
    rewriter_before_opt_a                :     0.000101s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000079s :  0.20%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000907s :  2.29%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    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  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001256s :  3.18%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000032s :  0.08%
    jit_opt_a.cse                        :     0.000066s :  0.17%
    jit_opt_a.replace_applicator         :     0.000027s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000288s :  0.73%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000742s :  1.88%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000028s :  0.07%
    loop_unroll                          :     0.000485s :  1.23%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.12%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000011s :  0.03%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000067s :  0.17%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000025s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000546s :  1.38%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000057s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010498s : 26.57%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000295    47
 3.51% :     0.000010s :      2: substitution.depend_value_elim
 1.09% :     0.000003s :      4: substitution.elim_not_effective
 0.77% :     0.000002s :      4: substitution.fold_const_symbol
 2.18% :     0.000006s :      5: substitution.graph_param_transform
75.82% :     0.000224s :      4: substitution.inline
 1.77% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.35% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.62% :     0.000005s :      2: substitution.replace_old_param
 4.47% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.13% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.29% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022884     2
92.74% :     0.021222s :      1: type_inference.infer
 7.26% :     0.001662s :      1: type_inference.specialize
------[replace.]   0.000098     6
65.16% :     0.000064s :      4: replace.inline
34.84% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000233     6
94.82% :     0.000221s :      4: match.inline
 5.18% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000176  1021
 1.17% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.20% :     0.000002s :     15: predicate.addn_zero_filter
 2.35% :     0.000004s :     15: predicate.arithmetic_simplify
 1.26% :     0.000002s :     15: predicate.cast_eliminate
 0.80% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.19% :     0.000002s :     15: predicate.compare_switch_simplify
 1.37% :     0.000002s :     15: predicate.depend_value_elim
 1.19% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.96% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      5: predicate.elim_not_effective
 0.55% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.07% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.36% :     0.000001s :      5: predicate.graph_param_transform
 5.10% :     0.000009s :     31: predicate.inline
 1.04% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.47% :     0.000003s :     10: predicate.less_batch_normalization
 1.33% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.82% :     0.000003s :     22: predicate.load_eliminater
 1.30% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.38% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.13% :     0.000002s :     15: predicate.minmaximum_grad
 1.67% :     0.000003s :      5: predicate.mutable_eliminate
 0.47% :     0.000001s :      5: predicate.opt_reshape
 2.25% :     0.000004s :     22: predicate.partial_eliminate
 1.13% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.68% :     0.000003s :     15: predicate.reduce_eliminate
 1.70% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.67% :     0.000003s :     27: predicate.replace_applicator
 0.64% :     0.000001s :     10: predicate.replace_old_param
 0.36% :     0.000001s :      5: predicate.reset_defer_inline
 1.24% :     0.000002s :     15: predicate.reshape_eliminate
 1.17% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :     10: predicate.special_op_eliminate
 1.05% :     0.000002s :     10: predicate.specialize_transform
 1.51% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.07% :     0.000004s :     21: predicate.switch_defer_inline
 1.88% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.91% :     0.000012s :     68: predicate.switch_simplify
 1.25% :     0.000002s :     15: predicate.tile_eliminate
 1.24% :     0.000002s :     15: predicate.transpose_eliminate
 1.49% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 2.08% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.70% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.51% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000886    13
42.69% :     0.000378s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.08% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.23% :     0.000463s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046594    76
 0.15% :     0.000070s :      1: add_recomputation
 0.37% :     0.000174s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000305s :      1: bootstrap
 0.07% :     0.000031s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.06% :     0.000029s :      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% :     0.000010s :      1: graph_reusing
14.54% :     0.006776s :      1: jit_opt_a
 0.45% :     0.000209s :      1: jit_opt_after_cconv
 0.16% :     0.000075s :      1: jit_opt_b
 1.06% :     0.000494s :      1: loop_unroll
 1.61% :     0.000752s :      1: mutable_eliminate
 2.62% :     0.001222s :     26: opt.transform.jit_opt_a
 0.18% :     0.000084s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000556s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.29% :     0.000134s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.68% :     0.000781s :      1: renormalize.infer
 1.00% :     0.000465s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000293s :      1: rewriter_after_opt_a
 0.23% :     0.000105s :      1: rewriter_before_opt_a
 0.23% :     0.000108s :      1: symbol_engine_optimizer
22.56% :     0.010514s :      1: task_emit
49.32% :     0.022980s :      1: type_inference
 0.18% :     0.000082s :      1: validate


TotalTime = 0.012558, [30]
[bootstrap]: 0.00027807
[type_inference]: 0.0048371
[event_method]: 2.193e-05
[auto_monad]: 8.17e-05
[graph_reusing]: 6.68e-06
[pre_auto_parallel]: 2.60002e-06
[py_interpret_to_execute]: 4.297e-05
[rewriter_before_opt_a]: 8.975e-05
[expand_dump_flag]: 3.66001e-06
[jit_opt_a]: 0.00512561, [2]
    [Cycle 1]: 0.00164671, [27]
        [switch_simplify]: 5.589e-05
        [loop_unroll]: 3.486e-05
        [a_1]: 0.00061939
        [with_stream_mark]: 1.731e-05
        [recompute_prepare]: 8.22998e-06
        [updatestate_depend_eliminate]: 4.03001e-06
        [updatestate_assign_eliminate]: 3.66001e-06
        [updatestate_loads_eliminate]: 3.21999e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 7.27002e-06
        [updatestate_useless_node_eliminater]: 5.83997e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 2.10002e-06
        [get_grad_eliminate_]: 6.12001e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.506e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 1.154e-05
        [inline_without_move]: 5.96e-06
        [renormalize]: 0.00059293
        [add_forward_monad_depend]: 5.51e-06
        [auto_monad_grad]: 1.74998e-06
        [auto_monad_eliminator]: 1.396e-05
        [cse]: 3.206e-05
        [replace_applicator]: 1.264e-05
    [Cycle 2]: 0.00035461, [27]
        [switch_simplify]: 6.71e-06
        [loop_unroll]: 5.72001e-06
        [a_1]: 0.00011171
        [with_stream_mark]: 9.71e-06
        [recompute_prepare]: 5.52999e-06
        [updatestate_depend_eliminate]: 3.21999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.318e-05
        [j_node_and_user_rematch]: 8.38999e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 9.54e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 7.08e-06
        [cse]: 1.327e-05
        [replace_applicator]: 5.88002e-06
[py_interpret_to_execute_after_opt_a]: 9.22001e-06
[rewriter_after_opt_a]: 3.442e-05
[convert_after_rewriter]: 6.94001e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00050108
[jit_opt_b]: 5.397e-05, [1]
    [Cycle 1]: 4.746e-05, [2]
        [frontend_op_eliminate]: 1.831e-05
        [inline_after_opt_a]: 1.775e-05
[cconv]: 2.348e-05
[loop_unroll]: 0.00042937
[jit_opt_after_cconv]: 0.00014965, [1]
    [Cycle 1]: 0.00014343, [11]
        [c_1]: 2.465e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.94999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [cse]: 1.981e-05
        [call_graph_tuple_transform]: 2.003e-05
        [tuple_list_get_item_eliminator]: 5.84e-06
        [none_parameter_eliminate]: 1.97999e-06
        [renormalize]: 3.09985e-07
        [switch_simplify]: 6.26e-06
[remove_dup_value]: 1.486e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 5.98998e-06
[add_recomputation]: 4.774e-05
[cse_after_recomputation]: 2.311e-05, [1]
    [Cycle 1]: 1.802e-05, [1]
        [cse]: 1.235e-05
[auto_monad_reorder]: 1.789e-05
[get_jit_bprop_graph]: 1.64998e-06
[rewriter_after_jit_bprop_graph]: 4.48999e-06
[opt_after_jit_grad]: 0.0004581
[symbol_engine_optimizer]: 7.333e-05, [1]
    [Cycle 1]: 6.744e-05, [6]
        [build]: 2.93e-06
        [elim_shapecalc]: 8.26002e-06
        [elim_not_effective]: 1.309e-05
        [opt_reshape]: 6.08002e-06
        [fold_const_symbol]: 9.15001e-06
        [renormalize]: 4.19997e-07
[validate]: 3.493e-05
Sums
    bootstrap                            :     0.000278s :  3.14%
    type_inference                       :     0.004837s : 54.54%
    event_method                         :     0.000022s :  0.25%
    auto_monad                           :     0.000082s :  0.92%
    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                :     0.000090s :  1.01%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.71%
    jit_opt_a.loop_unroll                :     0.000041s :  0.46%
    jit_opt_a.a_1                        :     0.000731s :  8.24%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.30%
    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 :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.13%
    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 :  0.13%
    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 :  0.32%
    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          :     0.000021s :  0.24%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000593s :  6.69%
    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      :     0.000021s :  0.24%
    jit_opt_a.cse                        :     0.000045s :  0.51%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000501s :  5.65%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000023s :  0.26%
    loop_unroll                          :     0.000429s :  4.84%
    jit_opt_after_cconv.c_1              :     0.000025s :  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              :     0.000020s :  0.22%
    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      :     0.000000s :  0.00%
    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 :  0.07%
    add_recomputation                    :     0.000048s :  0.54%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000458s :  5.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.39%


Time group info:
------[substitution.]   0.000216    28
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000006s :      4: substitution.graph_param_transform
82.74% :     0.000179s :      4: substitution.inline
 1.38% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.38% :     0.000005s :      4: substitution.replace_old_param
 7.42% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004769     2
79.97% :     0.003814s :      1: type_inference.infer
20.03% :     0.000955s :      1: type_inference.specialize
------[replace.]   0.000080     8
56.87% :     0.000046s :      4: replace.inline
43.13% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     8
92.56% :     0.000176s :      4: match.inline
 7.44% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   912
 1.30% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     13: predicate.addn_check_dump
 1.38% :     0.000002s :     13: predicate.addn_zero_filter
 1.92% :     0.000003s :     13: predicate.arithmetic_simplify
 1.31% :     0.000002s :     13: predicate.cast_eliminate
 0.42% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     13: predicate.compare_switch_simplify
 1.18% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.49% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.93% :     0.000001s :      8: predicate.get_grad_eliminate
 0.29% :     0.000000s :      4: predicate.graph_param_transform
 6.05% :     0.000008s :     29: predicate.inline
 0.98% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      8: predicate.less_batch_normalization
 1.67% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     21: predicate.load_eliminater
 1.27% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.07% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     13: predicate.merge_addn
 1.10% :     0.000002s :     13: predicate.minmaximum_grad
 1.45% :     0.000002s :      4: predicate.mutable_eliminate
 0.49% :     0.000001s :      4: predicate.opt_reshape
 2.55% :     0.000003s :     21: predicate.partial_eliminate
 1.17% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     13: predicate.reduce_eliminate
 1.60% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.79% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.90% :     0.000003s :     25: predicate.replace_applicator
 0.88% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.14% :     0.000002s :     13: predicate.reshape_eliminate
 1.43% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      8: predicate.special_op_eliminate
 1.19% :     0.000002s :      8: predicate.specialize_transform
 1.24% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     21: predicate.switch_defer_inline
 2.41% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.59% :     0.000010s :     67: predicate.switch_simplify
 1.13% :     0.000002s :     13: predicate.tile_eliminate
 1.15% :     0.000002s :     13: predicate.transpose_eliminate
 1.35% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.87% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000606    12
50.38% :     0.000305s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.64% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.98% :     0.000273s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014186    72
 0.36% :     0.000050s :      1: add_recomputation
 0.61% :     0.000086s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 2.09% :     0.000296s :      1: bootstrap
 0.19% :     0.000026s :      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
 0.18% :     0.000026s :      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
36.15% :     0.005128s :      1: jit_opt_a
 1.07% :     0.000152s :      1: jit_opt_after_cconv
 0.40% :     0.000057s :      1: jit_opt_b
 3.08% :     0.000437s :      1: loop_unroll
 3.58% :     0.000509s :      1: mutable_eliminate
 6.81% :     0.000965s :     26: opt.transform.jit_opt_a
 0.38% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000030s :      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
 0.16% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.28% :     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
 0.32% :     0.000046s :      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
 2.20% :     0.000312s :      1: renormalize.infer
 1.93% :     0.000274s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000037s :      1: rewriter_after_opt_a
 0.65% :     0.000092s :      1: rewriter_before_opt_a
 0.54% :     0.000076s :      1: symbol_engine_optimizer
34.20% :     0.004851s :      1: type_inference


TotalTime = 0.0117919, [30]
[bootstrap]: 0.00025419
[type_inference]: 0.00434
[event_method]: 2.053e-05
[auto_monad]: 4.315e-05
[graph_reusing]: 3.58e-06
[pre_auto_parallel]: 1.77999e-06
[py_interpret_to_execute]: 3.37e-05
[rewriter_before_opt_a]: 7.966e-05
[expand_dump_flag]: 1.96e-06
[jit_opt_a]: 0.00500784, [2]
    [Cycle 1]: 0.00169009, [27]
        [switch_simplify]: 5.206e-05
        [loop_unroll]: 3.631e-05
        [a_1]: 0.00059931
        [with_stream_mark]: 1.418e-05
        [recompute_prepare]: 7.95e-06
        [updatestate_depend_eliminate]: 3.41999e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 6.46e-06
        [updatestate_useless_node_eliminater]: 6.07999e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 6.05002e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 8.59989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.216e-05
        [j_node_and_user_rematch]: 9.37001e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 9.59e-06
        [inline_without_move]: 5.92999e-06
        [renormalize]: 0.00064699
        [add_forward_monad_depend]: 4.67e-06
        [auto_monad_grad]: 1.68002e-06
        [auto_monad_eliminator]: 1.116e-05
        [cse]: 2.346e-05
        [replace_applicator]: 1.263e-05
    [Cycle 2]: 0.00035426, [27]
        [switch_simplify]: 6.64999e-06
        [loop_unroll]: 5.98998e-06
        [a_1]: 0.00011267
        [with_stream_mark]: 9.12001e-06
        [recompute_prepare]: 5.63002e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.52999e-06
        [updatestate_useless_node_eliminater]: 5.91998e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 5.21002e-06
        [merge_forward]: 2.67001e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.286e-05
        [j_node_and_user_rematch]: 8.50999e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 9.10999e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 6.02001e-06
        [cse]: 1.324e-05
        [replace_applicator]: 5.72001e-06
[py_interpret_to_execute_after_opt_a]: 1.031e-05
[rewriter_after_opt_a]: 3.134e-05
[convert_after_rewriter]: 5.96e-06
[order_py_execute_after_rewriter]: 4.15e-06
[mutable_eliminate]: 0.0005131
[jit_opt_b]: 5.379e-05, [1]
    [Cycle 1]: 4.763e-05, [2]
        [frontend_op_eliminate]: 1.824e-05
        [inline_after_opt_a]: 1.693e-05
[cconv]: 1.746e-05
[loop_unroll]: 0.0004096
[jit_opt_after_cconv]: 0.0001497, [1]
    [Cycle 1]: 0.00014353, [11]
        [c_1]: 2.494e-05
        [parameter_eliminate]: 2.87002e-06
        [updatestate_depend_eliminate]: 6.08998e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 2.102e-05
        [call_graph_tuple_transform]: 1.816e-05
        [tuple_list_get_item_eliminator]: 6.17999e-06
        [none_parameter_eliminate]: 1.34003e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 6.29999e-06
[remove_dup_value]: 9.07999e-06
[partial_unused_args_eliminate]: 1.99e-06
[environ_conv]: 4.37e-06
[add_recomputation]: 3.537e-05
[cse_after_recomputation]: 2.366e-05, [1]
    [Cycle 1]: 1.838e-05, [1]
        [cse]: 1.249e-05
[auto_monad_reorder]: 1.341e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 4.52003e-06
[opt_after_jit_grad]: 0.00045741
[symbol_engine_optimizer]: 7.3e-05, [1]
    [Cycle 1]: 6.712e-05, [6]
        [build]: 2.73e-06
        [elim_shapecalc]: 8.60001e-06
        [elim_not_effective]: 1.31e-05
        [opt_reshape]: 6.04001e-06
        [fold_const_symbol]: 8.90001e-06
        [renormalize]: 6.10016e-07
[validate]: 3.265e-05
Sums
    bootstrap                            :     0.000254s :  3.08%
    type_inference                       :     0.004340s : 52.67%
    event_method                         :     0.000021s :  0.25%
    auto_monad                           :     0.000043s :  0.52%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.41%
    rewriter_before_opt_a                :     0.000080s :  0.97%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.71%
    jit_opt_a.loop_unroll                :     0.000042s :  0.51%
    jit_opt_a.a_1                        :     0.000712s :  8.64%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000014s :  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        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.15%
    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_        :     0.000011s :  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 :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000647s :  7.85%
    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      :     0.000017s :  0.21%
    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 :  0.13%
    rewriter_after_opt_a                 :     0.000031s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000513s :  6.23%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000017s :  0.21%
    loop_unroll                          :     0.000410s :  4.97%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.30%
    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 :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000009s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000035s :  0.43%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000013s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000457s :  5.55%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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                             :     0.000033s :  0.40%


Time group info:
------[substitution.]   0.000184    28
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.10% :     0.000004s :      4: substitution.graph_param_transform
83.29% :     0.000153s :      4: substitution.inline
 1.34% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000004s :      4: substitution.replace_old_param
 7.24% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004292     2
80.10% :     0.003438s :      1: type_inference.infer
19.90% :     0.000854s :      1: type_inference.specialize
------[replace.]   0.000087     8
59.93% :     0.000052s :      4: replace.inline
40.07% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     8
92.98% :     0.000151s :      4: match.inline
 7.02% :     0.000011s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   912
 1.21% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     13: predicate.addn_check_dump
 1.34% :     0.000002s :     13: predicate.addn_zero_filter
 2.08% :     0.000003s :     13: predicate.arithmetic_simplify
 1.28% :     0.000002s :     13: predicate.cast_eliminate
 0.53% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.13% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      4: predicate.elim_not_effective
 0.53% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     13: predicate.environ_get_eliminate
 1.12% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.91% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.13% :     0.000007s :     29: predicate.inline
 0.92% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      8: predicate.less_batch_normalization
 1.66% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.12% :     0.000003s :     21: predicate.load_eliminater
 1.21% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.39% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 1.59% :     0.000002s :      4: predicate.mutable_eliminate
 0.43% :     0.000001s :      4: predicate.opt_reshape
 2.52% :     0.000003s :     21: predicate.partial_eliminate
 1.28% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     13: predicate.reduce_eliminate
 1.72% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.00% :     0.000003s :     25: predicate.replace_applicator
 0.65% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.27% :     0.000002s :     13: predicate.reshape_eliminate
 1.24% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.35% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     21: predicate.switch_defer_inline
 2.26% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.39% :     0.000010s :     67: predicate.switch_simplify
 1.21% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.36% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     13: predicate.value_based_eliminate
 0.35% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000571    12
46.80% :     0.000267s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.07% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.13% :     0.000275s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013460    72
 0.29% :     0.000039s :      1: add_recomputation
 0.35% :     0.000047s :      1: auto_monad
 0.12% :     0.000016s :      1: auto_monad_reorder
 1.98% :     0.000267s :      1: bootstrap
 0.15% :     0.000020s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.18% :     0.000024s :      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
37.23% :     0.005011s :      1: jit_opt_a
 1.13% :     0.000152s :      1: jit_opt_after_cconv
 0.42% :     0.000056s :      1: jit_opt_b
 3.10% :     0.000417s :      1: loop_unroll
 3.87% :     0.000521s :      1: mutable_eliminate
 7.00% :     0.000942s :     26: opt.transform.jit_opt_a
 0.38% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000029s :      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
 0.18% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.46% :     0.000465s :      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
 0.27% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000012s :      1: remove_dup_value
 2.62% :     0.000353s :      1: renormalize.infer
 2.13% :     0.000287s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000034s :      1: rewriter_after_opt_a
 0.61% :     0.000082s :      1: rewriter_before_opt_a
 0.56% :     0.000076s :      1: symbol_engine_optimizer
32.32% :     0.004351s :      1: type_inference


TotalTime = 0.0116476, [30]
[bootstrap]: 0.00025034
[type_inference]: 0.00428492
[event_method]: 1.981e-05
[auto_monad]: 7.426e-05
[graph_reusing]: 5.12e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 3.57e-05
[rewriter_before_opt_a]: 8.774e-05
[expand_dump_flag]: 2.51998e-06
[jit_opt_a]: 0.00490327, [2]
    [Cycle 1]: 0.00159393, [27]
        [switch_simplify]: 5.151e-05
        [loop_unroll]: 3.491e-05
        [a_1]: 0.00058368
        [with_stream_mark]: 1.319e-05
        [recompute_prepare]: 8.15e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 2.88998e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 1.21002e-06
        [specialize_transform]: 6.94999e-06
        [updatestate_useless_node_eliminater]: 5.97999e-06
        [accelerated_algorithm]: 6.18002e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 5.89999e-06
        [merge_forward]: 3.14001e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.25e-05
        [j_node_and_user_rematch]: 1.913e-05
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 1.103e-05
        [inline_without_move]: 6.07001e-06
        [renormalize]: 0.00058643
        [add_forward_monad_depend]: 5.09998e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.257e-05
        [cse]: 2.464e-05
        [replace_applicator]: 1.356e-05
    [Cycle 2]: 0.00037212, [27]
        [switch_simplify]: 7.43999e-06
        [loop_unroll]: 6.42001e-06
        [a_1]: 0.00012426
        [with_stream_mark]: 9.32999e-06
        [recompute_prepare]: 6.16998e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.66998e-06
        [updatestate_useless_node_eliminater]: 5.73002e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 5.78002e-06
        [merge_forward]: 2.69999e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.406e-05
        [j_node_and_user_rematch]: 9.25999e-06
        [meta_fg_expand]: 1.71998e-06
        [replace_old_param]: 9.65002e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 6.44001e-06
        [cse]: 1.251e-05
        [replace_applicator]: 6.03002e-06
[py_interpret_to_execute_after_opt_a]: 1.034e-05
[rewriter_after_opt_a]: 3.227e-05
[convert_after_rewriter]: 6.69999e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00048683
[jit_opt_b]: 5.377e-05, [1]
    [Cycle 1]: 4.625e-05, [2]
        [frontend_op_eliminate]: 1.801e-05
        [inline_after_opt_a]: 1.707e-05
[cconv]: 2.163e-05
[loop_unroll]: 0.00040772
[jit_opt_after_cconv]: 0.00015141, [1]
    [Cycle 1]: 0.00014472, [11]
        [c_1]: 2.474e-05
        [parameter_eliminate]: 2.86999e-06
        [updatestate_depend_eliminate]: 8.13999e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [cse]: 2.047e-05
        [call_graph_tuple_transform]: 1.971e-05
        [tuple_list_get_item_eliminator]: 6.13998e-06
        [none_parameter_eliminate]: 1.56002e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 6.06998e-06
[remove_dup_value]: 1.394e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 5.16002e-06
[add_recomputation]: 4.235e-05
[cse_after_recomputation]: 2.235e-05, [1]
    [Cycle 1]: 1.703e-05, [1]
        [cse]: 1.127e-05
[auto_monad_reorder]: 1.419e-05
[get_jit_bprop_graph]: 1.60001e-06
[rewriter_after_jit_bprop_graph]: 4.08001e-06
[opt_after_jit_grad]: 0.00044333
[symbol_engine_optimizer]: 7.284e-05, [1]
    [Cycle 1]: 6.689e-05, [6]
        [build]: 2.76e-06
        [elim_shapecalc]: 8.35999e-06
        [elim_not_effective]: 1.333e-05
        [opt_reshape]: 6.17999e-06
        [fold_const_symbol]: 8.82e-06
        [renormalize]: 4.50003e-07
[validate]: 2.982e-05
Sums
    bootstrap                            :     0.000250s :  3.07%
    type_inference                       :     0.004285s : 52.53%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000074s :  0.91%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.44%
    rewriter_before_opt_a                :     0.000088s :  1.08%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000059s :  0.72%
    jit_opt_a.loop_unroll                :     0.000041s :  0.51%
    jit_opt_a.a_1                        :     0.000708s :  8.68%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.18%
    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        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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_        :     0.000012s :  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 :     0.000027s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.35%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000021s :  0.25%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000587s :  7.19%
    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      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000037s :  0.46%
    jit_opt_a.replace_applicator         :     0.000020s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    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                    :     0.000487s :  5.97%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000022s :  0.27%
    loop_unroll                          :     0.000408s :  5.00%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.30%
    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 :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  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 :  0.08%
    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 :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000042s :  0.52%
    cse_after_recomputation.cse          :     0.000011s :  0.14%
    auto_monad_reorder                   :     0.000014s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000443s :  5.43%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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                             :     0.000030s :  0.37%


Time group info:
------[substitution.]   0.000185    28
 1.21% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 3.06% :     0.000006s :      4: substitution.graph_param_transform
80.22% :     0.000148s :      4: substitution.inline
 2.03% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.26% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.71% :     0.000005s :      4: substitution.replace_old_param
 7.97% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004236     2
80.29% :     0.003401s :      1: type_inference.infer
19.71% :     0.000835s :      1: type_inference.specialize
------[replace.]   0.000079     8
56.29% :     0.000045s :      4: replace.inline
43.71% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     8
91.82% :     0.000146s :      4: match.inline
 8.18% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   912
 1.27% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.12% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.17% :     0.000002s :     13: predicate.addn_check_dump
 1.22% :     0.000002s :     13: predicate.addn_zero_filter
 1.81% :     0.000002s :     13: predicate.arithmetic_simplify
 1.46% :     0.000002s :     13: predicate.cast_eliminate
 0.40% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.13% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.59% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.21% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 1.03% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.56% :     0.000007s :     29: predicate.inline
 0.98% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      8: predicate.less_batch_normalization
 1.67% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.03% :     0.000003s :     21: predicate.load_eliminater
 1.30% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.27% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     13: predicate.merge_addn
 1.14% :     0.000002s :     13: predicate.minmaximum_grad
 1.33% :     0.000002s :      4: predicate.mutable_eliminate
 0.41% :     0.000001s :      4: predicate.opt_reshape
 2.49% :     0.000003s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     13: predicate.reduce_eliminate
 1.59% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.03% :     0.000003s :     25: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.41% :     0.000001s :      4: predicate.reset_defer_inline
 1.19% :     0.000002s :     13: predicate.reshape_eliminate
 1.46% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     13: predicate.same_eliminate
 0.57% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     21: predicate.switch_defer_inline
 2.27% :     0.000003s :     21: predicate.switch_layer_defer_inline
 8.08% :     0.000011s :     67: predicate.switch_simplify
 1.24% :     0.000002s :     13: predicate.tile_eliminate
 1.39% :     0.000002s :     13: predicate.transpose_eliminate
 1.48% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     13: predicate.value_based_eliminate
 0.35% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
47.12% :     0.000258s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.12% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.76% :     0.000262s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013270    72
 0.34% :     0.000045s :      1: add_recomputation
 0.59% :     0.000078s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 1.98% :     0.000262s :      1: bootstrap
 0.18% :     0.000024s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.17% :     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
36.97% :     0.004906s :      1: jit_opt_a
 1.16% :     0.000154s :      1: jit_opt_after_cconv
 0.43% :     0.000056s :      1: jit_opt_b
 3.13% :     0.000416s :      1: loop_unroll
 3.73% :     0.000495s :      1: mutable_eliminate
 7.18% :     0.000953s :     26: opt.transform.jit_opt_a
 0.40% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.40% :     0.000451s :      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% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000039s :      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
 2.29% :     0.000304s :      1: renormalize.infer
 2.08% :     0.000276s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000035s :      1: rewriter_after_opt_a
 0.69% :     0.000091s :      1: rewriter_before_opt_a
 0.57% :     0.000075s :      1: symbol_engine_optimizer
32.37% :     0.004295s :      1: type_inference


TotalTime = 0.0137287, [30]
[bootstrap]: 0.0002881
[type_inference]: 0.00502049
[event_method]: 2.108e-05
[auto_monad]: 6.751e-05
[graph_reusing]: 6.07999e-06
[pre_auto_parallel]: 2.69001e-06
[py_interpret_to_execute]: 4.218e-05
[rewriter_before_opt_a]: 9.055e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00586462, [2]
    [Cycle 1]: 0.0020006, [27]
        [switch_simplify]: 6.689e-05
        [loop_unroll]: 3.554e-05
        [a_1]: 0.00065556
        [with_stream_mark]: 2.203e-05
        [recompute_prepare]: 1.045e-05
        [updatestate_depend_eliminate]: 4.13001e-06
        [updatestate_assign_eliminate]: 3.78999e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 8.07e-06
        [updatestate_useless_node_eliminater]: 6.06998e-06
        [accelerated_algorithm]: 6.93e-06
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 5.82999e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.649e-05
        [j_node_and_user_rematch]: 1.001e-05
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 1.171e-05
        [inline_without_move]: 6.33e-06
        [renormalize]: 0.00084415
        [add_forward_monad_depend]: 6.10002e-06
        [auto_monad_grad]: 3.04999e-06
        [auto_monad_eliminator]: 1.589e-05
        [cse]: 3.566e-05
        [replace_applicator]: 1.697e-05
    [Cycle 2]: 0.00037366, [27]
        [switch_simplify]: 7.15e-06
        [loop_unroll]: 5.91e-06
        [a_1]: 0.00011866
        [with_stream_mark]: 1.086e-05
        [recompute_prepare]: 5.91e-06
        [updatestate_depend_eliminate]: 3.02002e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [parameter_eliminate]: 1.99999e-06
        [specialize_transform]: 6.40002e-06
        [updatestate_useless_node_eliminater]: 6.14001e-06
        [accelerated_algorithm]: 6.29001e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 5.50001e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.48e-05
        [j_node_and_user_rematch]: 8.70001e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 9.96e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.00999e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 6.07999e-06
        [cse]: 1.395e-05
        [replace_applicator]: 5.54998e-06
[py_interpret_to_execute_after_opt_a]: 1.386e-05
[rewriter_after_opt_a]: 3.84e-05
[convert_after_rewriter]: 7.73001e-06
[order_py_execute_after_rewriter]: 5.09e-06
[mutable_eliminate]: 0.00066139
[jit_opt_b]: 5.595e-05, [1]
    [Cycle 1]: 4.733e-05, [2]
        [frontend_op_eliminate]: 1.868e-05
        [inline_after_opt_a]: 1.743e-05
[cconv]: 2.597e-05
[loop_unroll]: 0.00044492
[jit_opt_after_cconv]: 0.00015727, [1]
    [Cycle 1]: 0.00015084, [11]
        [c_1]: 2.546e-05
        [parameter_eliminate]: 2.53003e-06
        [updatestate_depend_eliminate]: 6.16e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.54e-05
        [call_graph_tuple_transform]: 2.162e-05
        [tuple_list_get_item_eliminator]: 6.22001e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 6.16998e-06
[remove_dup_value]: 1.67e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 5.74e-06
[add_recomputation]: 5.11e-05
[cse_after_recomputation]: 2.393e-05, [1]
    [Cycle 1]: 1.837e-05, [1]
        [cse]: 1.237e-05
[auto_monad_reorder]: 1.764e-05
[get_jit_bprop_graph]: 1.93997e-06
[rewriter_after_jit_bprop_graph]: 5.99999e-06
[opt_after_jit_grad]: 0.00049453
[symbol_engine_optimizer]: 7.892e-05, [1]
    [Cycle 1]: 7.259e-05, [6]
        [build]: 4.57e-06
        [elim_shapecalc]: 8.79e-06
        [elim_not_effective]: 1.411e-05
        [opt_reshape]: 6.39001e-06
        [fold_const_symbol]: 9.59999e-06
        [renormalize]: 6.09987e-07
[validate]: 4.04e-05
Sums
    bootstrap                            :     0.000288s :  2.99%
    type_inference                       :     0.005020s : 52.12%
    event_method                         :     0.000021s :  0.22%
    auto_monad                           :     0.000068s :  0.70%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000042s :  0.44%
    rewriter_before_opt_a                :     0.000091s :  0.94%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000074s :  0.77%
    jit_opt_a.loop_unroll                :     0.000041s :  0.43%
    jit_opt_a.a_1                        :     0.000774s :  8.04%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.34%
    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 :     0.000006s :  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       :     0.000014s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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_        :     0.000011s :  0.12%
    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 :  0.32%
    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          :     0.000022s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000844s :  8.76%
    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 :  0.23%
    jit_opt_a.cse                        :     0.000050s :  0.51%
    jit_opt_a.replace_applicator         :     0.000023s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.14%
    rewriter_after_opt_a                 :     0.000038s :  0.40%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000661s :  6.87%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000026s :  0.27%
    loop_unroll                          :     0.000445s :  4.62%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.26%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000025s :  0.26%
    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  :     0.000006s :  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                    :     0.000051s :  0.53%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000495s :  5.13%
    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 :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.42%


Time group info:
------[substitution.]   0.000231    28
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000006s :      4: substitution.graph_param_transform
82.02% :     0.000190s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.64% :     0.000006s :      4: substitution.replace_old_param
 7.58% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004950     2
81.12% :     0.004015s :      1: type_inference.infer
18.88% :     0.000935s :      1: type_inference.specialize
------[replace.]   0.000091     8
55.96% :     0.000051s :      4: replace.inline
44.04% :     0.000040s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000203     8
92.23% :     0.000187s :      4: match.inline
 7.77% :     0.000016s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   912
 1.29% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.03% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     13: predicate.addn_check_dump
 1.31% :     0.000002s :     13: predicate.addn_zero_filter
 2.00% :     0.000003s :     13: predicate.arithmetic_simplify
 1.33% :     0.000002s :     13: predicate.cast_eliminate
 0.46% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.06% :     0.000001s :     13: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.12% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      4: predicate.elim_not_effective
 0.52% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.08% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.13% :     0.000002s :     13: predicate.environ_get_eliminate
 1.07% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 1.09% :     0.000002s :      8: predicate.get_grad_eliminate
 0.24% :     0.000000s :      4: predicate.graph_param_transform
 5.43% :     0.000008s :     29: predicate.inline
 0.93% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      8: predicate.less_batch_normalization
 1.57% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.96% :     0.000003s :     21: predicate.load_eliminater
 1.28% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     42: predicate.loop_unroll_before_grad
 1.85% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     13: predicate.merge_addn
 1.09% :     0.000002s :     13: predicate.minmaximum_grad
 1.55% :     0.000002s :      4: predicate.mutable_eliminate
 0.56% :     0.000001s :      4: predicate.opt_reshape
 2.35% :     0.000003s :     21: predicate.partial_eliminate
 1.14% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     13: predicate.reduce_eliminate
 1.63% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.06% :     0.000003s :     25: predicate.replace_applicator
 0.89% :     0.000001s :      8: predicate.replace_old_param
 0.31% :     0.000000s :      4: predicate.reset_defer_inline
 1.24% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.61% :     0.000002s :     13: predicate.same_eliminate
 0.62% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      8: predicate.special_op_eliminate
 1.07% :     0.000001s :      8: predicate.specialize_transform
 1.49% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.90% :     0.000004s :     21: predicate.switch_defer_inline
 2.11% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.49% :     0.000010s :     67: predicate.switch_simplify
 1.29% :     0.000002s :     13: predicate.tile_eliminate
 1.19% :     0.000002s :     13: predicate.transpose_eliminate
 1.32% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.54% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000642    12
50.17% :     0.000322s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.05% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.78% :     0.000288s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015677    72
 0.34% :     0.000054s :      1: add_recomputation
 0.45% :     0.000071s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 1.93% :     0.000302s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000026s :      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
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.43% :     0.005869s :      1: jit_opt_a
 1.02% :     0.000160s :      1: jit_opt_after_cconv
 0.38% :     0.000059s :      1: jit_opt_b
 2.89% :     0.000453s :      1: loop_unroll
 4.28% :     0.000671s :      1: mutable_eliminate
 6.58% :     0.001031s :     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
 0.09% :     0.000013s :      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
 0.23% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.21% :     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% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000045s :      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
 3.07% :     0.000482s :      1: renormalize.infer
 2.25% :     0.000353s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000042s :      1: rewriter_after_opt_a
 0.60% :     0.000093s :      1: rewriter_before_opt_a
 0.52% :     0.000081s :      1: symbol_engine_optimizer
32.13% :     0.005036s :      1: type_inference


TotalTime = 0.058443, [33]
[bootstrap]: 0.00037962
[type_inference]: 0.035931
[event_method]: 2.311e-05
[auto_monad]: 0.0001708
[graph_reusing]: 6.88e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 0.00012001
[rewriter_before_opt_a]: 9.935e-05
[expand_dump_flag]: 3.2e-06
[jit_opt_a]: 0.00663134, [2]
    [Cycle 1]: 0.00238071, [27]
        [switch_simplify]: 6.273e-05
        [loop_unroll]: 3.614e-05
        [a_1]: 0.00072519
        [with_stream_mark]: 2.455e-05
        [recompute_prepare]: 1.122e-05
        [updatestate_depend_eliminate]: 6.49001e-06
        [updatestate_assign_eliminate]: 7.36999e-06
        [updatestate_loads_eliminate]: 4.68001e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 9.44998e-06
        [updatestate_useless_node_eliminater]: 1.221e-05
        [accelerated_algorithm]: 1.03e-05
        [meta_shard_fg_expand]: 3.3e-06
        [get_grad_eliminate_]: 8.40001e-06
        [merge_forward]: 5.07e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.047e-05
        [j_node_and_user_rematch]: 1.344e-05
        [meta_fg_expand]: 4.13001e-06
        [replace_old_param]: 1.208e-05
        [inline_without_move]: 8.28999e-06
        [renormalize]: 0.00112544
        [add_forward_monad_depend]: 5.81e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 2.174e-05
        [cse]: 3.222e-05
        [replace_applicator]: 1.683e-05
    [Cycle 2]: 0.00051661, [27]
        [switch_simplify]: 8.91997e-06
        [loop_unroll]: 7.97998e-06
        [a_1]: 0.00016806
        [with_stream_mark]: 1.387e-05
        [recompute_prepare]: 8.2e-06
        [updatestate_depend_eliminate]: 5.25999e-06
        [updatestate_assign_eliminate]: 4.85001e-06
        [updatestate_loads_eliminate]: 4.31002e-06
        [parameter_eliminate]: 1.12999e-06
        [specialize_transform]: 8.3e-06
        [updatestate_useless_node_eliminater]: 1.061e-05
        [accelerated_algorithm]: 8.03001e-06
        [meta_shard_fg_expand]: 2.36998e-06
        [get_grad_eliminate_]: 7.38999e-06
        [merge_forward]: 4.54002e-06
        [cell_reuse_recompute_pass]: 1.84998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.734e-05
        [j_node_and_user_rematch]: 1.272e-05
        [meta_fg_expand]: 2.77002e-06
        [replace_old_param]: 1.038e-05
        [inline_without_move]: 7.50998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.34998e-06
        [auto_monad_grad]: 1.06002e-06
        [auto_monad_eliminator]: 4.938e-05
        [cse]: 2.356e-05
        [replace_applicator]: 8.43999e-06
[py_interpret_to_execute_after_opt_a]: 1.537e-05
[rewriter_after_opt_a]: 0.000326
[convert_after_rewriter]: 1.256e-05
[order_py_execute_after_rewriter]: 7.45e-06
[mutable_eliminate]: 0.00069699
[jit_opt_b]: 6.881e-05, [1]
    [Cycle 1]: 6.118e-05, [2]
        [frontend_op_eliminate]: 2.459e-05
        [inline_after_opt_a]: 2.457e-05
[cconv]: 2.902e-05
[loop_unroll]: 0.00047941
[jit_opt_after_cconv]: 0.00020367, [1]
    [Cycle 1]: 0.00019622, [11]
        [c_1]: 4.723e-05
        [parameter_eliminate]: 2.76999e-06
        [updatestate_depend_eliminate]: 8.85001e-06
        [updatestate_assign_eliminate]: 4.58999e-06
        [updatestate_loads_eliminate]: 4.17e-06
        [cse]: 3.154e-05
        [call_graph_tuple_transform]: 2.377e-05
        [tuple_list_get_item_eliminator]: 8.44002e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 9.32001e-06
[remove_dup_value]: 2.182e-05
[partial_unused_args_eliminate]: 2.29999e-06
[environ_conv]: 1.132e-05
[add_recomputation]: 7.243e-05
[cse_after_recomputation]: 3.19e-05, [1]
    [Cycle 1]: 2.481e-05, [1]
        [cse]: 1.852e-05
[auto_monad_reorder]: 2.782e-05
[get_jit_bprop_graph]: 2.27001e-06
[rewriter_after_jit_bprop_graph]: 6.69999e-06
[opt_after_jit_grad]: 0.00053836
[symbol_engine_optimizer]: 0.00011057, [1]
    [Cycle 1]: 0.0001033, [6]
        [build]: 1.521e-05
        [elim_shapecalc]: 1.245e-05
        [elim_not_effective]: 2.077e-05
        [opt_reshape]: 9.34998e-06
        [fold_const_symbol]: 1.408e-05
        [renormalize]: 4.99975e-07
[validate]: 6.199e-05
[backend_pass]: 1.22e-06
[task_emit]: 0.0121031
[execute]: 8.27998e-06
Sums
    bootstrap                            :     0.000380s :  0.70%
    type_inference                       :     0.035931s : 66.49%
    event_method                         :     0.000023s :  0.04%
    auto_monad                           :     0.000171s :  0.32%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000120s :  0.22%
    rewriter_before_opt_a                :     0.000099s :  0.18%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.13%
    jit_opt_a.loop_unroll                :     0.000044s :  0.08%
    jit_opt_a.a_1                        :     0.000893s :  1.65%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.03%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000022s :  0.04%
    jit_opt_a.inline_without_move        :     0.000016s :  0.03%
    jit_opt_a.renormalize                :     0.001126s :  2.08%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000071s :  0.13%
    jit_opt_a.cse                        :     0.000056s :  0.10%
    jit_opt_a.replace_applicator         :     0.000025s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.03%
    rewriter_after_opt_a                 :     0.000326s :  0.60%
    convert_after_rewriter               :     0.000013s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000697s :  1.29%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.05%
    cconv                                :     0.000029s :  0.05%
    loop_unroll                          :     0.000479s :  0.89%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000032s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.04%
    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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000022s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000011s :  0.02%
    add_recomputation                    :     0.000072s :  0.13%
    cse_after_recomputation.cse          :     0.000019s :  0.03%
    auto_monad_reorder                   :     0.000028s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000538s :  1.00%
    symbol_engine_optimizer.build        :     0.000015s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000062s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012103s : 22.40%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000285    47
 3.57% :     0.000010s :      2: substitution.depend_value_elim
 0.97% :     0.000003s :      4: substitution.elim_not_effective
 0.73% :     0.000002s :      4: substitution.fold_const_symbol
 2.38% :     0.000007s :      5: substitution.graph_param_transform
75.62% :     0.000215s :      4: substitution.inline
 1.60% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.75% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.50% :     0.000004s :      2: substitution.replace_old_param
 4.49% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.05% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.34% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.035837     2
95.44% :     0.034203s :      1: type_inference.infer
 4.56% :     0.001634s :      1: type_inference.specialize
------[replace.]   0.000095     6
67.53% :     0.000064s :      4: replace.inline
32.47% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000224     6
94.78% :     0.000212s :      4: match.inline
 5.22% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1021
 1.27% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.42% :     0.000002s :     15: predicate.addn_zero_filter
 1.74% :     0.000003s :     15: predicate.arithmetic_simplify
 1.40% :     0.000002s :     15: predicate.cast_eliminate
 0.54% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     15: predicate.compare_switch_simplify
 1.43% :     0.000002s :     15: predicate.depend_value_elim
 1.11% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.05% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      5: predicate.elim_not_effective
 0.63% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     15: predicate.environ_get_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.35% :     0.000001s :      5: predicate.graph_param_transform
 4.81% :     0.000008s :     31: predicate.inline
 1.12% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.70% :     0.000003s :     10: predicate.less_batch_normalization
 1.41% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     22: predicate.load_eliminater
 1.53% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.33% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.20% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     15: predicate.merge_addn
 1.17% :     0.000002s :     15: predicate.minmaximum_grad
 1.50% :     0.000003s :      5: predicate.mutable_eliminate
 0.55% :     0.000001s :      5: predicate.opt_reshape
 2.31% :     0.000004s :     22: predicate.partial_eliminate
 1.31% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.68% :     0.000003s :     15: predicate.reduce_eliminate
 1.52% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.74% :     0.000003s :     27: predicate.replace_applicator
 0.89% :     0.000002s :     10: predicate.replace_old_param
 0.37% :     0.000001s :      5: predicate.reset_defer_inline
 1.20% :     0.000002s :     15: predicate.reshape_eliminate
 1.48% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.94% :     0.000002s :     10: predicate.special_op_eliminate
 1.04% :     0.000002s :     10: predicate.specialize_transform
 1.47% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.14% :     0.000004s :     21: predicate.switch_defer_inline
 2.15% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.60% :     0.000011s :     68: predicate.switch_simplify
 1.33% :     0.000002s :     15: predicate.tile_eliminate
 1.22% :     0.000002s :     15: predicate.transpose_eliminate
 1.59% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.91% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.23% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     15: predicate.value_based_eliminate
 0.48% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000970    13
49.86% :     0.000484s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.44% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.70% :     0.000443s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.060962    76
 0.12% :     0.000076s :      1: add_recomputation
 0.29% :     0.000177s :      1: auto_monad
 0.05% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000406s :      1: bootstrap
 0.05% :     0.000032s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.06% :     0.000034s :      1: cse_after_recomputation
 0.02% :     0.000014s :      1: environ_conv
 0.04% :     0.000027s :      1: event_method
 0.02% :     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
10.88% :     0.006634s :      1: jit_opt_a
 0.34% :     0.000207s :      1: jit_opt_after_cconv
 0.12% :     0.000072s :      1: jit_opt_b
 0.80% :     0.000487s :      1: loop_unroll
 1.16% :     0.000707s :      1: mutable_eliminate
 1.97% :     0.001200s :     26: opt.transform.jit_opt_a
 0.14% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000042s :      4: opt.transform.jit_opt_b
 0.03% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000052s :      4: opt.transform.symbol_engine_opt
 0.90% :     0.000548s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000124s :      1: py_interpret_to_execute
 0.03% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000025s :      1: remove_dup_value
 1.06% :     0.000647s :      1: renormalize.infer
 0.77% :     0.000470s :      1: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.54% :     0.000331s :      1: rewriter_after_opt_a
 0.17% :     0.000104s :      1: rewriter_before_opt_a
 0.19% :     0.000114s :      1: symbol_engine_optimizer
19.88% :     0.012117s :      1: task_emit
58.96% :     0.035945s :      1: type_inference
 0.15% :     0.000088s :      1: validate


TotalTime = 0.0150535, [30]
[bootstrap]: 0.00031389
[type_inference]: 0.0057564
[event_method]: 2.437e-05
[auto_monad]: 7.76e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.93998e-06
[py_interpret_to_execute]: 4.396e-05
[rewriter_before_opt_a]: 8.993e-05
[expand_dump_flag]: 3.6e-06
[jit_opt_a]: 0.00612066, [2]
    [Cycle 1]: 0.00198174, [27]
        [switch_simplify]: 6.382e-05
        [loop_unroll]: 3.534e-05
        [a_1]: 0.00068957
        [with_stream_mark]: 2.286e-05
        [recompute_prepare]: 1.259e-05
        [updatestate_depend_eliminate]: 4.13999e-06
        [updatestate_assign_eliminate]: 3.75e-06
        [updatestate_loads_eliminate]: 2.97002e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 8.15e-06
        [updatestate_useless_node_eliminater]: 5.94999e-06
        [accelerated_algorithm]: 6.32001e-06
        [meta_shard_fg_expand]: 2.64999e-06
        [get_grad_eliminate_]: 5.71e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.789e-05
        [j_node_and_user_rematch]: 1.091e-05
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 1.182e-05
        [inline_without_move]: 6.11e-06
        [renormalize]: 0.00078816
        [add_forward_monad_depend]: 6.81001e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.752e-05
        [cse]: 3.429e-05
        [replace_applicator]: 1.648e-05
    [Cycle 2]: 0.0003879, [27]
        [switch_simplify]: 7.13998e-06
        [loop_unroll]: 6.04001e-06
        [a_1]: 0.0001139
        [with_stream_mark]: 1.266e-05
        [recompute_prepare]: 6.96001e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.79999e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 5.72001e-06
        [updatestate_useless_node_eliminater]: 5.66003e-06
        [accelerated_algorithm]: 6.01998e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.29998e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 1.98002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.71e-05
        [j_node_and_user_rematch]: 9.80002e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 9.56e-06
        [inline_without_move]: 5.64e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 1.49998e-06
        [auto_monad_eliminator]: 7.82e-06
        [cse]: 1.657e-05
        [replace_applicator]: 7.55e-06
[py_interpret_to_execute_after_opt_a]: 1.336e-05
[rewriter_after_opt_a]: 4.014e-05
[convert_after_rewriter]: 7.94002e-06
[order_py_execute_after_rewriter]: 6.01e-06
[mutable_eliminate]: 0.00069361
[jit_opt_b]: 5.837e-05, [1]
    [Cycle 1]: 5.021e-05, [2]
        [frontend_op_eliminate]: 1.926e-05
        [inline_after_opt_a]: 1.865e-05
[cconv]: 2.692e-05
[loop_unroll]: 0.00052943
[jit_opt_after_cconv]: 0.00017465, [1]
    [Cycle 1]: 0.00016639, [11]
        [c_1]: 2.619e-05
        [parameter_eliminate]: 3.16999e-06
        [updatestate_depend_eliminate]: 9.22999e-06
        [updatestate_assign_eliminate]: 3.45e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [cse]: 3.189e-05
        [call_graph_tuple_transform]: 2.294e-05
        [tuple_list_get_item_eliminator]: 6.05002e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.89e-06
[remove_dup_value]: 1.536e-05
[partial_unused_args_eliminate]: 2.60002e-06
[environ_conv]: 5.72999e-06
[add_recomputation]: 5.066e-05
[cse_after_recomputation]: 2.909e-05, [1]
    [Cycle 1]: 2.216e-05, [1]
        [cse]: 1.372e-05
[auto_monad_reorder]: 2.005e-05
[get_jit_bprop_graph]: 2.21998e-06
[rewriter_after_jit_bprop_graph]: 6.74001e-06
[opt_after_jit_grad]: 0.00059677
[symbol_engine_optimizer]: 8.214e-05, [1]
    [Cycle 1]: 7.501e-05, [6]
        [build]: 4.86997e-06
        [elim_shapecalc]: 9.67001e-06
        [elim_not_effective]: 1.329e-05
        [opt_reshape]: 6.80002e-06
        [fold_const_symbol]: 9.42001e-06
        [renormalize]: 6.50005e-07
[validate]: 4.407e-05
Sums
    bootstrap                            :     0.000314s :  2.95%
    type_inference                       :     0.005756s : 54.09%
    event_method                         :     0.000024s :  0.23%
    auto_monad                           :     0.000078s :  0.73%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.41%
    rewriter_before_opt_a                :     0.000090s :  0.85%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000071s :  0.67%
    jit_opt_a.loop_unroll                :     0.000041s :  0.39%
    jit_opt_a.a_1                        :     0.000803s :  7.55%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.18%
    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 :     0.000006s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000014s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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 :  0.10%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.33%
    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          :     0.000021s :  0.20%
    jit_opt_a.inline_without_move        :     0.000012s :  0.11%
    jit_opt_a.renormalize                :     0.000788s :  7.41%
    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      :     0.000025s :  0.24%
    jit_opt_a.cse                        :     0.000051s :  0.48%
    jit_opt_a.replace_applicator         :     0.000024s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.13%
    rewriter_after_opt_a                 :     0.000040s :  0.38%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000694s :  6.52%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.18%
    cconv                                :     0.000027s :  0.25%
    loop_unroll                          :     0.000529s :  4.97%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.25%
    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 :     0.000003s :  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 :  0.06%
    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                     :     0.000015s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000051s :  0.48%
    cse_after_recomputation.cse          :     0.000014s :  0.13%
    auto_monad_reorder                   :     0.000020s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000597s :  5.61%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.09%
    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 :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.41%


Time group info:
------[substitution.]   0.000266    28
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.38% :     0.000006s :      4: substitution.graph_param_transform
84.59% :     0.000225s :      4: substitution.inline
 1.52% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.78% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000006s :      4: substitution.replace_old_param
 6.36% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005672     2
81.54% :     0.004624s :      1: type_inference.infer
18.46% :     0.001047s :      1: type_inference.specialize
------[replace.]   0.000094     8
56.08% :     0.000053s :      4: replace.inline
43.92% :     0.000041s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000237     8
93.61% :     0.000222s :      4: match.inline
 6.39% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   912
 1.21% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     13: predicate.addn_check_dump
 1.32% :     0.000002s :     13: predicate.addn_zero_filter
 1.84% :     0.000003s :     13: predicate.arithmetic_simplify
 1.19% :     0.000002s :     13: predicate.cast_eliminate
 0.42% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     13: predicate.compare_switch_simplify
 1.15% :     0.000002s :     13: predicate.depend_value_elim
 1.09% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.11% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.44% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     13: predicate.environ_get_eliminate
 1.07% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.89% :     0.000001s :      8: predicate.get_grad_eliminate
 0.37% :     0.000001s :      4: predicate.graph_param_transform
 5.73% :     0.000008s :     29: predicate.inline
 0.91% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :      8: predicate.less_batch_normalization
 1.74% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     21: predicate.load_eliminater
 1.22% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.11% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     13: predicate.merge_addn
 1.14% :     0.000002s :     13: predicate.minmaximum_grad
 1.53% :     0.000002s :      4: predicate.mutable_eliminate
 0.53% :     0.000001s :      4: predicate.opt_reshape
 2.35% :     0.000003s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     13: predicate.reduce_eliminate
 1.58% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.04% :     0.000003s :     25: predicate.replace_applicator
 0.96% :     0.000001s :      8: predicate.replace_old_param
 0.45% :     0.000001s :      4: predicate.reset_defer_inline
 1.41% :     0.000002s :     13: predicate.reshape_eliminate
 1.21% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 1.24% :     0.000002s :      4: predicate.row_tensor_eliminate
 1.17% :     0.000002s :     13: predicate.same_eliminate
 0.79% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.39% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.65% :     0.000004s :     21: predicate.switch_defer_inline
 2.05% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.43% :     0.000010s :     67: predicate.switch_simplify
 1.16% :     0.000002s :     13: predicate.tile_eliminate
 1.19% :     0.000002s :     13: predicate.transpose_eliminate
 1.46% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.60% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 2.07% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     13: predicate.value_based_eliminate
 0.36% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000729    12
51.25% :     0.000373s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.52% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.23% :     0.000330s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016969    72
 0.32% :     0.000055s :      1: add_recomputation
 0.48% :     0.000081s :      1: auto_monad
 0.14% :     0.000023s :      1: auto_monad_reorder
 1.95% :     0.000331s :      1: bootstrap
 0.17% :     0.000030s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.19% :     0.000031s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.17% :     0.000028s :      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
36.09% :     0.006124s :      1: jit_opt_a
 1.05% :     0.000178s :      1: jit_opt_after_cconv
 0.37% :     0.000062s :      1: jit_opt_b
 3.18% :     0.000539s :      1: loop_unroll
 4.15% :     0.000704s :      1: mutable_eliminate
 6.28% :     0.001066s :     26: opt.transform.jit_opt_a
 0.34% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000031s :      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% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.57% :     0.000606s :      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
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.66% :     0.000451s :      1: renormalize.infer
 1.93% :     0.000328s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000044s :      1: rewriter_after_opt_a
 0.55% :     0.000093s :      1: rewriter_before_opt_a
 0.50% :     0.000085s :      1: symbol_engine_optimizer
34.03% :     0.005774s :      1: type_inference


TotalTime = 0.0115977, [30]
[bootstrap]: 0.00024791
[type_inference]: 0.00436328
[event_method]: 1.993e-05
[auto_monad]: 4.353e-05
[graph_reusing]: 3.58e-06
[pre_auto_parallel]: 1.71998e-06
[py_interpret_to_execute]: 3.446e-05
[rewriter_before_opt_a]: 7.939e-05
[expand_dump_flag]: 1.74998e-06
[jit_opt_a]: 0.00486891, [2]
    [Cycle 1]: 0.00157182, [27]
        [switch_simplify]: 5.138e-05
        [loop_unroll]: 3.524e-05
        [a_1]: 0.00057005
        [with_stream_mark]: 1.205e-05
        [recompute_prepare]: 8.13999e-06
        [updatestate_depend_eliminate]: 3.31001e-06
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 6.76e-06
        [updatestate_useless_node_eliminater]: 6.07999e-06
        [accelerated_algorithm]: 6.24999e-06
        [meta_shard_fg_expand]: 1.57999e-06
        [get_grad_eliminate_]: 5.74999e-06
        [merge_forward]: 2.79001e-06
        [cell_reuse_recompute_pass]: 8.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.213e-05
        [j_node_and_user_rematch]: 9.17999e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 9.32999e-06
        [inline_without_move]: 5.88998e-06
        [renormalize]: 0.00060379
        [add_forward_monad_depend]: 3.86999e-06
        [auto_monad_grad]: 1.29998e-06
        [auto_monad_eliminator]: 1.024e-05
        [cse]: 2.219e-05
        [replace_applicator]: 1.307e-05
    [Cycle 2]: 0.00035122, [27]
        [switch_simplify]: 6.79001e-06
        [loop_unroll]: 5.66998e-06
        [a_1]: 0.00011376
        [with_stream_mark]: 8.92e-06
        [recompute_prepare]: 5.69999e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.65001e-06
        [updatestate_useless_node_eliminater]: 5.53002e-06
        [accelerated_algorithm]: 5.87001e-06
        [meta_shard_fg_expand]: 1.19998e-06
        [get_grad_eliminate_]: 5.08002e-06
        [merge_forward]: 2.75002e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.323e-05
        [j_node_and_user_rematch]: 8.37998e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 8.75001e-06
        [inline_without_move]: 5.40001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 5.86e-06
        [cse]: 1.299e-05
        [replace_applicator]: 5.89e-06
[py_interpret_to_execute_after_opt_a]: 9.12001e-06
[rewriter_after_opt_a]: 2.762e-05
[convert_after_rewriter]: 5.44998e-06
[order_py_execute_after_rewriter]: 4.18001e-06
[mutable_eliminate]: 0.00046875
[jit_opt_b]: 5.747e-05, [1]
    [Cycle 1]: 4.801e-05, [2]
        [frontend_op_eliminate]: 1.897e-05
        [inline_after_opt_a]: 1.715e-05
[cconv]: 1.534e-05
[loop_unroll]: 0.00040825
[jit_opt_after_cconv]: 0.00014431, [1]
    [Cycle 1]: 0.00013794, [11]
        [c_1]: 2.505e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 4.95999e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 1.969e-05
        [call_graph_tuple_transform]: 1.785e-05
        [tuple_list_get_item_eliminator]: 6.04999e-06
        [none_parameter_eliminate]: 9.09989e-07
        [renormalize]: 6.50005e-07
        [switch_simplify]: 6.04001e-06
[remove_dup_value]: 7.9e-06
[partial_unused_args_eliminate]: 4.63999e-06
[environ_conv]: 4.52e-06
[add_recomputation]: 3.232e-05
[cse_after_recomputation]: 2.441e-05, [1]
    [Cycle 1]: 1.888e-05, [1]
        [cse]: 1.297e-05
[auto_monad_reorder]: 1.246e-05
[get_jit_bprop_graph]: 1.44998e-06
[rewriter_after_jit_bprop_graph]: 3.9e-06
[opt_after_jit_grad]: 0.00044829
[symbol_engine_optimizer]: 7.201e-05, [1]
    [Cycle 1]: 6.593e-05, [6]
        [build]: 2.17999e-06
        [elim_shapecalc]: 8.52998e-06
        [elim_not_effective]: 1.24e-05
        [opt_reshape]: 6.19999e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 6.19999e-07
[validate]: 2.861e-05
Sums
    bootstrap                            :     0.000248s :  3.06%
    type_inference                       :     0.004363s : 53.84%
    event_method                         :     0.000020s :  0.25%
    auto_monad                           :     0.000044s :  0.54%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.43%
    rewriter_before_opt_a                :     0.000079s :  0.98%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.72%
    jit_opt_a.loop_unroll                :     0.000041s :  0.50%
    jit_opt_a.a_1                        :     0.000684s :  8.44%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000014s :  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        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  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_        :     0.000011s :  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 :     0.000025s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000604s :  7.45%
    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      :     0.000016s :  0.20%
    jit_opt_a.cse                        :     0.000035s :  0.43%
    jit_opt_a.replace_applicator         :     0.000019s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000028s :  0.34%
    convert_after_rewriter               :     0.000005s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000469s :  5.78%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000015s :  0.19%
    loop_unroll                          :     0.000408s :  5.04%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.31%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  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      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000008s :  0.10%
    partial_unused_args_eliminate        :     0.000005s :  0.06%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000032s :  0.40%
    cse_after_recomputation.cse          :     0.000013s :  0.16%
    auto_monad_reorder                   :     0.000012s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000448s :  5.53%
    symbol_engine_optimizer.build        :     0.000002s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    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 :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000029s :  0.35%


Time group info:
------[substitution.]   0.000167    28
 0.79% :     0.000001s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000004s :      4: substitution.graph_param_transform
82.98% :     0.000139s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.55% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.28% :     0.000004s :      4: substitution.replace_old_param
 6.66% :     0.000011s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004311     2
80.43% :     0.003467s :      1: type_inference.infer
19.57% :     0.000844s :      1: type_inference.specialize
------[replace.]   0.000078     8
55.34% :     0.000043s :      4: replace.inline
44.66% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000146     8
93.63% :     0.000136s :      4: match.inline
 6.37% :     0.000009s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   912
 1.17% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     13: predicate.addn_check_dump
 1.29% :     0.000002s :     13: predicate.addn_zero_filter
 1.77% :     0.000002s :     13: predicate.arithmetic_simplify
 1.62% :     0.000002s :     13: predicate.cast_eliminate
 0.38% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     13: predicate.compare_switch_simplify
 1.16% :     0.000002s :     13: predicate.depend_value_elim
 1.21% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.49% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.28% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     13: predicate.environ_get_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.94% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.37% :     0.000007s :     29: predicate.inline
 0.90% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.14% :     0.000002s :      8: predicate.less_batch_normalization
 1.71% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.10% :     0.000003s :     21: predicate.load_eliminater
 1.18% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.18% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     13: predicate.merge_addn
 1.25% :     0.000002s :     13: predicate.minmaximum_grad
 1.53% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.41% :     0.000003s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     13: predicate.reduce_eliminate
 1.63% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.99% :     0.000003s :     25: predicate.replace_applicator
 0.61% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.26% :     0.000002s :     13: predicate.reshape_eliminate
 1.35% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.31% :     0.000002s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.30% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     21: predicate.switch_defer_inline
 2.14% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.75% :     0.000010s :     67: predicate.switch_simplify
 1.12% :     0.000001s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.53% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     13: predicate.value_based_eliminate
 0.46% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000557    12
46.50% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.69% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.81% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013199    72
 0.27% :     0.000035s :      1: add_recomputation
 0.36% :     0.000047s :      1: auto_monad
 0.12% :     0.000015s :      1: auto_monad_reorder
 1.96% :     0.000259s :      1: bootstrap
 0.14% :     0.000018s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.20% :     0.000027s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.18% :     0.000023s :      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
36.91% :     0.004872s :      1: jit_opt_a
 1.11% :     0.000147s :      1: jit_opt_after_cconv
 0.46% :     0.000060s :      1: jit_opt_b
 3.15% :     0.000416s :      1: loop_unroll
 3.62% :     0.000477s :      1: mutable_eliminate
 6.91% :     0.000912s :     26: opt.transform.jit_opt_a
 0.39% :     0.000052s :      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% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.45% :     0.000456s :      1: opt_after_jit_grad
 0.05% :     0.000006s :      1: order_py_execute_after_rewriter
 0.05% :     0.000007s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000037s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000010s :      1: remove_dup_value
 2.27% :     0.000300s :      1: renormalize.infer
 2.25% :     0.000297s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000030s :      1: rewriter_after_opt_a
 0.62% :     0.000082s :      1: rewriter_before_opt_a
 0.57% :     0.000075s :      1: symbol_engine_optimizer
33.15% :     0.004375s :      1: type_inference


TotalTime = 0.0119738, [30]
[bootstrap]: 0.00025541
[type_inference]: 0.00460135
[event_method]: 2.124e-05
[auto_monad]: 5.076e-05
[graph_reusing]: 4.72998e-06
[pre_auto_parallel]: 1.96e-06
[py_interpret_to_execute]: 3.49e-05
[rewriter_before_opt_a]: 8.556e-05
[expand_dump_flag]: 2.69999e-06
[jit_opt_a]: 0.00495555, [2]
    [Cycle 1]: 0.00157509, [27]
        [switch_simplify]: 5.087e-05
        [loop_unroll]: 3.437e-05
        [a_1]: 0.00058651
        [with_stream_mark]: 1.277e-05
        [recompute_prepare]: 7.79002e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 6.94001e-06
        [updatestate_useless_node_eliminater]: 5.86998e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.47999e-06
        [get_grad_eliminate_]: 5.68997e-06
        [merge_forward]: 2.62001e-06
        [cell_reuse_recompute_pass]: 9.79984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.18e-05
        [j_node_and_user_rematch]: 9.12999e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 1.032e-05
        [inline_without_move]: 6.19001e-06
        [renormalize]: 0.00059548
        [add_forward_monad_depend]: 4.45e-06
        [auto_monad_grad]: 1.66998e-06
        [auto_monad_eliminator]: 1.127e-05
        [cse]: 2.169e-05
        [replace_applicator]: 1.174e-05
    [Cycle 2]: 0.00034976, [27]
        [switch_simplify]: 6.41998e-06
        [loop_unroll]: 5.61e-06
        [a_1]: 0.00011153
        [with_stream_mark]: 8.45999e-06
        [recompute_prepare]: 5.77999e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.62001e-06
        [updatestate_useless_node_eliminater]: 5.51e-06
        [accelerated_algorithm]: 5.99999e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.17999e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.384e-05
        [j_node_and_user_rematch]: 8.75999e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 9.66998e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 6.10002e-06
        [cse]: 1.219e-05
        [replace_applicator]: 5.73002e-06
[py_interpret_to_execute_after_opt_a]: 9.24e-06
[rewriter_after_opt_a]: 2.858e-05
[convert_after_rewriter]: 5.77999e-06
[order_py_execute_after_rewriter]: 4.48999e-06
[mutable_eliminate]: 0.00047768
[jit_opt_b]: 5.399e-05, [1]
    [Cycle 1]: 4.774e-05, [2]
        [frontend_op_eliminate]: 1.768e-05
        [inline_after_opt_a]: 1.732e-05
[cconv]: 1.744e-05
[loop_unroll]: 0.0004245
[jit_opt_after_cconv]: 0.00014986, [1]
    [Cycle 1]: 0.00014322, [11]
        [c_1]: 2.489e-05
        [parameter_eliminate]: 2.89001e-06
        [updatestate_depend_eliminate]: 5.51e-06
        [updatestate_assign_eliminate]: 2.53003e-06
        [updatestate_loads_eliminate]: 2.96999e-06
        [cse]: 1.991e-05
        [call_graph_tuple_transform]: 1.851e-05
        [tuple_list_get_item_eliminator]: 8.87e-06
        [none_parameter_eliminate]: 1.12e-06
        [renormalize]: 3.20026e-07
        [switch_simplify]: 6.14999e-06
[remove_dup_value]: 1.016e-05
[partial_unused_args_eliminate]: 1.55001e-06
[environ_conv]: 4.78001e-06
[add_recomputation]: 3.366e-05
[cse_after_recomputation]: 2.319e-05, [1]
    [Cycle 1]: 1.81e-05, [1]
        [cse]: 1.232e-05
[auto_monad_reorder]: 1.277e-05
[get_jit_bprop_graph]: 1.47999e-06
[rewriter_after_jit_bprop_graph]: 4.43999e-06
[opt_after_jit_grad]: 0.00044442
[symbol_engine_optimizer]: 7.356e-05, [1]
    [Cycle 1]: 6.756e-05, [6]
        [build]: 2.51998e-06
        [elim_shapecalc]: 7.77002e-06
        [elim_not_effective]: 1.318e-05
        [opt_reshape]: 6.20002e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 8.09989e-07
[validate]: 2.984e-05
Sums
    bootstrap                            :     0.000255s :  3.04%
    type_inference                       :     0.004601s : 54.74%
    event_method                         :     0.000021s :  0.25%
    auto_monad                           :     0.000051s :  0.60%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.42%
    rewriter_before_opt_a                :     0.000086s :  1.02%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000057s :  0.68%
    jit_opt_a.loop_unroll                :     0.000040s :  0.48%
    jit_opt_a.a_1                        :     0.000698s :  8.30%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.25%
    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 :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  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 :  0.03%
    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 :     0.000026s :  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          :     0.000020s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000596s :  7.08%
    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      :     0.000017s :  0.21%
    jit_opt_a.cse                        :     0.000034s :  0.40%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000029s :  0.34%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000478s :  5.68%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000017s :  0.21%
    loop_unroll                          :     0.000425s :  5.05%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.30%
    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 :     0.000003s :  0.04%
    jit_opt_after_cconv.cse              :     0.000020s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.11%
    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 :  0.07%
    remove_dup_value                     :     0.000010s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000034s :  0.40%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000013s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000444s :  5.29%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    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                             :     0.000030s :  0.35%


Time group info:
------[substitution.]   0.000186    28
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.23% :     0.000004s :      4: substitution.graph_param_transform
83.30% :     0.000155s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000004s :      4: substitution.replace_old_param
 6.96% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004544     2
80.59% :     0.003663s :      1: type_inference.infer
19.41% :     0.000882s :      1: type_inference.specialize
------[replace.]   0.000078     8
56.24% :     0.000044s :      4: replace.inline
43.76% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     8
93.30% :     0.000152s :      4: match.inline
 6.70% :     0.000011s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   912
 1.25% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.20% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     13: predicate.addn_check_dump
 1.28% :     0.000002s :     13: predicate.addn_zero_filter
 1.75% :     0.000002s :     13: predicate.arithmetic_simplify
 1.42% :     0.000002s :     13: predicate.cast_eliminate
 0.46% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     13: predicate.compare_switch_simplify
 1.23% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000001s :     13: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      4: predicate.elim_not_effective
 0.54% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.46% :     0.000002s :     13: predicate.environ_get_eliminate
 1.13% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.97% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.11% :     0.000007s :     29: predicate.inline
 0.96% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      8: predicate.less_batch_normalization
 1.72% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.02% :     0.000003s :     21: predicate.load_eliminater
 1.28% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.22% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.40% :     0.000002s :      4: predicate.mutable_eliminate
 0.51% :     0.000001s :      4: predicate.opt_reshape
 2.58% :     0.000003s :     21: predicate.partial_eliminate
 1.21% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     13: predicate.reduce_eliminate
 1.86% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.93% :     0.000003s :     25: predicate.replace_applicator
 0.64% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000000s :      4: predicate.reset_defer_inline
 1.26% :     0.000002s :     13: predicate.reshape_eliminate
 1.21% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.30% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     21: predicate.switch_defer_inline
 2.21% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.48% :     0.000010s :     67: predicate.switch_simplify
 1.24% :     0.000002s :     13: predicate.tile_eliminate
 1.28% :     0.000002s :     13: predicate.transpose_eliminate
 1.45% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.98% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000569    12
45.64% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.27% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.09% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013578    72
 0.27% :     0.000036s :      1: add_recomputation
 0.40% :     0.000054s :      1: auto_monad
 0.11% :     0.000015s :      1: auto_monad_reorder
 1.97% :     0.000267s :      1: bootstrap
 0.15% :     0.000020s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.19% :     0.000025s :      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
36.52% :     0.004958s :      1: jit_opt_a
 1.12% :     0.000153s :      1: jit_opt_after_cconv
 0.42% :     0.000057s :      1: jit_opt_b
 3.18% :     0.000432s :      1: loop_unroll
 3.57% :     0.000485s :      1: mutable_eliminate
 6.80% :     0.000923s :     26: opt.transform.jit_opt_a
 0.41% :     0.000055s :      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% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.33% :     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
 0.28% :     0.000038s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000013s :      1: remove_dup_value
 2.32% :     0.000315s :      1: renormalize.infer
 2.02% :     0.000274s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000031s :      1: rewriter_after_opt_a
 0.65% :     0.000089s :      1: rewriter_before_opt_a
 0.56% :     0.000076s :      1: symbol_engine_optimizer
33.96% :     0.004612s :      1: type_inference


TotalTime = 0.0119534, [30]
[bootstrap]: 0.00024862
[type_inference]: 0.00430316
[event_method]: 2.038e-05
[auto_monad]: 5.071e-05
[graph_reusing]: 4.05998e-06
[pre_auto_parallel]: 1.89999e-06
[py_interpret_to_execute]: 3.531e-05
[rewriter_before_opt_a]: 8.159e-05
[expand_dump_flag]: 2.67001e-06
[jit_opt_a]: 0.00518164, [2]
    [Cycle 1]: 0.00176471, [27]
        [switch_simplify]: 5.201e-05
        [loop_unroll]: 3.486e-05
        [a_1]: 0.00059788
        [with_stream_mark]: 1.598e-05
        [recompute_prepare]: 8.38001e-06
        [updatestate_depend_eliminate]: 3.41999e-06
        [updatestate_assign_eliminate]: 2.81999e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 6.88e-06
        [updatestate_useless_node_eliminater]: 5.94999e-06
        [accelerated_algorithm]: 6.33e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 6.04001e-06
        [merge_forward]: 3.36001e-06
        [cell_reuse_recompute_pass]: 8.39995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 9.74999e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 1.002e-05
        [inline_without_move]: 6.09001e-06
        [renormalize]: 0.00063567
        [add_forward_monad_depend]: 5.37001e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.265e-05
        [cse]: 0.00013549
        [replace_applicator]: 1.422e-05
    [Cycle 2]: 0.00035152, [27]
        [switch_simplify]: 6.58998e-06
        [loop_unroll]: 5.84e-06
        [a_1]: 0.0001124
        [with_stream_mark]: 1.018e-05
        [recompute_prepare]: 5.58002e-06
        [updatestate_depend_eliminate]: 3.06999e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.50001e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 5.76998e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 2.89001e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.354e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 9.00001e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.55001e-06
        [auto_monad_grad]: 1.54e-06
        [auto_monad_eliminator]: 6.69001e-06
        [cse]: 1.416e-05
        [replace_applicator]: 5.68002e-06
[py_interpret_to_execute_after_opt_a]: 1.048e-05
[rewriter_after_opt_a]: 3.096e-05
[convert_after_rewriter]: 6.26998e-06
[order_py_execute_after_rewriter]: 4.85999e-06
[mutable_eliminate]: 0.00051534
[jit_opt_b]: 5.218e-05, [1]
    [Cycle 1]: 4.61e-05, [2]
        [frontend_op_eliminate]: 1.79e-05
        [inline_after_opt_a]: 1.71e-05
[cconv]: 2.057e-05
[loop_unroll]: 0.00041878
[jit_opt_after_cconv]: 0.00014426, [1]
    [Cycle 1]: 0.00013875, [11]
        [c_1]: 2.429e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 5.74999e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 2.015e-05
        [call_graph_tuple_transform]: 1.927e-05
        [tuple_list_get_item_eliminator]: 5.83002e-06
        [none_parameter_eliminate]: 1.09e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 6.36e-06
[remove_dup_value]: 1.071e-05
[partial_unused_args_eliminate]: 1.64e-06
[environ_conv]: 4.62e-06
[add_recomputation]: 3.775e-05
[cse_after_recomputation]: 2.328e-05, [1]
    [Cycle 1]: 1.784e-05, [1]
        [cse]: 1.227e-05
[auto_monad_reorder]: 1.528e-05
[get_jit_bprop_graph]: 1.91998e-06
[rewriter_after_jit_bprop_graph]: 4.3e-06
[opt_after_jit_grad]: 0.00046155
[symbol_engine_optimizer]: 7.427e-05, [1]
    [Cycle 1]: 6.814e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 8.57e-06
        [elim_not_effective]: 1.379e-05
        [opt_reshape]: 6.28e-06
        [fold_const_symbol]: 9.22999e-06
        [renormalize]: 5.8001e-07
[validate]: 3.481e-05
Sums
    bootstrap                            :     0.000249s :  2.98%
    type_inference                       :     0.004303s : 51.54%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000051s :  0.61%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.42%
    rewriter_before_opt_a                :     0.000082s :  0.98%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000059s :  0.70%
    jit_opt_a.loop_unroll                :     0.000041s :  0.49%
    jit_opt_a.a_1                        :     0.000710s :  8.51%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000014s :  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 :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    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_        :     0.000011s :  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 :     0.000027s :  0.32%
    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          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000636s :  7.61%
    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      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000150s :  1.79%
    jit_opt_a.replace_applicator         :     0.000020s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000031s :  0.37%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000515s :  6.17%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000021s :  0.25%
    loop_unroll                          :     0.000419s :  5.02%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.29%
    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              :     0.000020s :  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 :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000011s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000038s :  0.45%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    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                   :     0.000462s :  5.53%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    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 :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.42%


Time group info:
------[substitution.]   0.000196    28
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.41% :     0.000005s :      4: substitution.graph_param_transform
82.62% :     0.000162s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000004s :      4: substitution.replace_old_param
 7.65% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004251     2
79.84% :     0.003395s :      1: type_inference.infer
20.16% :     0.000857s :      1: type_inference.specialize
------[replace.]   0.000079     8
57.43% :     0.000045s :      4: replace.inline
42.57% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     8
92.50% :     0.000159s :      4: match.inline
 7.50% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   912
 1.16% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     13: predicate.addn_check_dump
 1.33% :     0.000002s :     13: predicate.addn_zero_filter
 1.85% :     0.000002s :     13: predicate.arithmetic_simplify
 1.30% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.13% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      4: predicate.elim_not_effective
 0.44% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     13: predicate.environ_get_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      4: predicate.fold_const_symbol
 1.02% :     0.000001s :      8: predicate.get_grad_eliminate
 0.30% :     0.000000s :      4: predicate.graph_param_transform
 5.12% :     0.000007s :     29: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      8: predicate.less_batch_normalization
 1.79% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.15% :     0.000003s :     21: predicate.load_eliminater
 1.18% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.26% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 1.36% :     0.000002s :      4: predicate.mutable_eliminate
 0.51% :     0.000001s :      4: predicate.opt_reshape
 2.49% :     0.000003s :     21: predicate.partial_eliminate
 1.27% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     13: predicate.reduce_eliminate
 1.70% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.11% :     0.000003s :     25: predicate.replace_applicator
 0.71% :     0.000001s :      8: predicate.replace_old_param
 0.36% :     0.000000s :      4: predicate.reset_defer_inline
 1.32% :     0.000002s :     13: predicate.reshape_eliminate
 1.22% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      8: predicate.special_op_eliminate
 0.94% :     0.000001s :      8: predicate.specialize_transform
 1.28% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.69% :     0.000010s :     67: predicate.switch_simplify
 1.22% :     0.000002s :     13: predicate.tile_eliminate
 1.43% :     0.000002s :     13: predicate.transpose_eliminate
 1.48% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.96% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     13: 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.]   0.000560    12
46.81% :     0.000262s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.06% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.13% :     0.000269s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013607    72
 0.30% :     0.000040s :      1: add_recomputation
 0.40% :     0.000054s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 1.90% :     0.000258s :      1: bootstrap
 0.17% :     0.000023s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.17% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
38.10% :     0.005185s :      1: jit_opt_a
 1.08% :     0.000147s :      1: jit_opt_after_cconv
 0.40% :     0.000055s :      1: jit_opt_b
 3.13% :     0.000426s :      1: loop_unroll
 3.84% :     0.000523s :      1: mutable_eliminate
 6.92% :     0.000942s :     26: opt.transform.jit_opt_a
 0.39% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000029s :      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
 0.18% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.45% :     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% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000038s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000013s :      1: remove_dup_value
 2.52% :     0.000343s :      1: renormalize.infer
 2.10% :     0.000286s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000034s :      1: rewriter_after_opt_a
 0.62% :     0.000084s :      1: rewriter_before_opt_a
 0.57% :     0.000077s :      1: symbol_engine_optimizer
31.70% :     0.004313s :      1: type_inference


TotalTime = 0.0500438, [33]
[bootstrap]: 0.00035764
[type_inference]: 0.0251384
[event_method]: 2.545e-05
[auto_monad]: 0.0001761
[graph_reusing]: 6.61999e-06
[pre_auto_parallel]: 3.15998e-06
[py_interpret_to_execute]: 0.000141
[rewriter_before_opt_a]: 0.0001056
[expand_dump_flag]: 3.90998e-06
[jit_opt_a]: 0.00769348, [2]
    [Cycle 1]: 0.00284592, [27]
        [switch_simplify]: 7.826e-05
        [loop_unroll]: 3.569e-05
        [a_1]: 0.00083095
        [with_stream_mark]: 3e-05
        [recompute_prepare]: 1.319e-05
        [updatestate_depend_eliminate]: 6.75998e-06
        [updatestate_assign_eliminate]: 5.59e-06
        [updatestate_loads_eliminate]: 5.35001e-06
        [parameter_eliminate]: 2.13998e-06
        [specialize_transform]: 9.19e-06
        [updatestate_useless_node_eliminater]: 1.49e-05
        [accelerated_algorithm]: 8.87999e-06
        [meta_shard_fg_expand]: 3.38e-06
        [get_grad_eliminate_]: 8.84998e-06
        [merge_forward]: 6.13998e-06
        [cell_reuse_recompute_pass]: 1.86003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.171e-05
        [j_node_and_user_rematch]: 1.606e-05
        [meta_fg_expand]: 4.22e-06
        [replace_old_param]: 6.082e-05
        [inline_without_move]: 1.023e-05
        [renormalize]: 0.0013453
        [add_forward_monad_depend]: 8.25e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 2.624e-05
        [cse]: 4.503e-05
        [replace_applicator]: 2.27e-05
    [Cycle 2]: 0.00053868, [27]
        [switch_simplify]: 9.82999e-06
        [loop_unroll]: 8.45001e-06
        [a_1]: 0.00017496
        [with_stream_mark]: 1.861e-05
        [recompute_prepare]: 9.46998e-06
        [updatestate_depend_eliminate]: 7.30998e-06
        [updatestate_assign_eliminate]: 4.79e-06
        [updatestate_loads_eliminate]: 4.15999e-06
        [parameter_eliminate]: 1.62999e-06
        [specialize_transform]: 8.92999e-06
        [updatestate_useless_node_eliminater]: 1.211e-05
        [accelerated_algorithm]: 7.7e-06
        [meta_shard_fg_expand]: 2.93e-06
        [get_grad_eliminate_]: 7.98001e-06
        [merge_forward]: 5.79e-06
        [cell_reuse_recompute_pass]: 3.39001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.934e-05
        [j_node_and_user_rematch]: 1.261e-05
        [meta_fg_expand]: 3.04999e-06
        [replace_old_param]: 1.214e-05
        [inline_without_move]: 8.10999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 3.85998e-06
        [auto_monad_grad]: 1.89e-06
        [auto_monad_eliminator]: 1.694e-05
        [cse]: 2.806e-05
        [replace_applicator]: 1.161e-05
[py_interpret_to_execute_after_opt_a]: 1.923e-05
[rewriter_after_opt_a]: 0.00032173
[convert_after_rewriter]: 1.327e-05
[order_py_execute_after_rewriter]: 7.71001e-06
[mutable_eliminate]: 0.00100832
[jit_opt_b]: 7.615e-05, [1]
    [Cycle 1]: 6.651e-05, [2]
        [frontend_op_eliminate]: 2.652e-05
        [inline_after_opt_a]: 2.641e-05
[cconv]: 3.515e-05
[loop_unroll]: 0.00067253
[jit_opt_after_cconv]: 0.00023901, [1]
    [Cycle 1]: 0.00023062, [11]
        [c_1]: 5.385e-05
        [parameter_eliminate]: 5.60001e-06
        [updatestate_depend_eliminate]: 1.318e-05
        [updatestate_assign_eliminate]: 5.14998e-06
        [updatestate_loads_eliminate]: 4.50999e-06
        [cse]: 4.595e-05
        [call_graph_tuple_transform]: 2.542e-05
        [tuple_list_get_item_eliminator]: 8.62e-06
        [none_parameter_eliminate]: 1.95001e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 1.015e-05
[remove_dup_value]: 2.014e-05
[partial_unused_args_eliminate]: 2.79999e-06
[environ_conv]: 1.44e-05
[add_recomputation]: 7.587e-05
[cse_after_recomputation]: 4.006e-05, [1]
    [Cycle 1]: 3.304e-05, [1]
        [cse]: 2.388e-05
[auto_monad_reorder]: 3.155e-05
[get_jit_bprop_graph]: 2.41e-06
[rewriter_after_jit_bprop_graph]: 7.24001e-06
[opt_after_jit_grad]: 0.00068377
[symbol_engine_optimizer]: 0.00011557, [1]
    [Cycle 1]: 0.00010726, [6]
        [build]: 1.592e-05
        [elim_shapecalc]: 1.309e-05
        [elim_not_effective]: 2.043e-05
        [opt_reshape]: 8.72998e-06
        [fold_const_symbol]: 1.522e-05
        [renormalize]: 7.7e-07
[validate]: 7.058e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.0126665
[execute]: 7.83999e-06
Sums
    bootstrap                            :     0.000358s :  0.80%
    type_inference                       :     0.025138s : 55.89%
    event_method                         :     0.000025s :  0.06%
    auto_monad                           :     0.000176s :  0.39%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000141s :  0.31%
    rewriter_before_opt_a                :     0.000106s :  0.23%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000088s :  0.20%
    jit_opt_a.loop_unroll                :     0.000044s :  0.10%
    jit_opt_a.a_1                        :     0.001006s :  2.24%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  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 :     0.000027s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000012s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000073s :  0.16%
    jit_opt_a.inline_without_move        :     0.000018s :  0.04%
    jit_opt_a.renormalize                :     0.001345s :  2.99%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000043s :  0.10%
    jit_opt_a.cse                        :     0.000073s :  0.16%
    jit_opt_a.replace_applicator         :     0.000034s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.04%
    rewriter_after_opt_a                 :     0.000322s :  0.72%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.001008s :  2.24%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000035s :  0.08%
    loop_unroll                          :     0.000673s :  1.50%
    jit_opt_after_cconv.c_1              :     0.000054s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000046s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000020s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000014s :  0.03%
    add_recomputation                    :     0.000076s :  0.17%
    cse_after_recomputation.cse          :     0.000024s :  0.05%
    auto_monad_reorder                   :     0.000032s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000684s :  1.52%
    symbol_engine_optimizer.build        :     0.000016s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000071s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012666s : 28.16%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000414    47
 2.77% :     0.000011s :      2: substitution.depend_value_elim
 0.69% :     0.000003s :      4: substitution.elim_not_effective
 0.71% :     0.000003s :      4: substitution.fold_const_symbol
 1.88% :     0.000008s :      5: substitution.graph_param_transform
68.12% :     0.000282s :      4: substitution.inline
 1.33% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.05% :     0.000008s :      8: substitution.remove_not_recompute_node
12.85% :     0.000053s :      2: substitution.replace_old_param
 3.34% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.18% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.09% :     0.000013s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.025046     2
93.12% :     0.023323s :      1: type_inference.infer
 6.88% :     0.001723s :      1: type_inference.specialize
------[replace.]   0.000115     6
65.83% :     0.000076s :      4: replace.inline
34.17% :     0.000039s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000291     6
95.61% :     0.000278s :      4: match.inline
 4.39% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000189  1021
 1.81% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.24% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     15: predicate.addn_check_dump
 1.45% :     0.000003s :     15: predicate.addn_zero_filter
 2.06% :     0.000004s :     15: predicate.arithmetic_simplify
 1.35% :     0.000003s :     15: predicate.cast_eliminate
 0.43% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.31% :     0.000002s :     15: predicate.compare_switch_simplify
 1.49% :     0.000003s :     15: predicate.depend_value_elim
 1.01% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.29% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.61% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.04% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     15: predicate.environ_get_eliminate
 1.05% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.06% :     0.000002s :     10: predicate.get_grad_eliminate
 0.40% :     0.000001s :      5: predicate.graph_param_transform
 5.69% :     0.000011s :     31: predicate.inline
 1.38% :     0.000003s :     10: predicate.inline_without_move
 0.38% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.09% :     0.000002s :     10: predicate.less_batch_normalization
 1.26% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.23% :     0.000004s :     22: predicate.load_eliminater
 2.00% :     0.000004s :      5: predicate.loop_unroll_after_grad
 3.16% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.96% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 2.15% :     0.000004s :      5: predicate.mutable_eliminate
 0.44% :     0.000001s :      5: predicate.opt_reshape
 2.41% :     0.000005s :     22: predicate.partial_eliminate
 1.19% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.83% :     0.000003s :     15: predicate.reduce_eliminate
 1.28% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.73% :     0.000003s :     27: predicate.replace_applicator
 0.63% :     0.000001s :     10: predicate.replace_old_param
 0.42% :     0.000001s :      5: predicate.reset_defer_inline
 1.08% :     0.000002s :     15: predicate.reshape_eliminate
 1.10% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.86% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     15: predicate.same_eliminate
 0.87% :     0.000002s :     10: predicate.set_cell_output_no_recompute
 1.04% :     0.000002s :     10: predicate.special_op_eliminate
 1.02% :     0.000002s :     10: predicate.specialize_transform
 1.62% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.60% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.89% :     0.000004s :     21: predicate.switch_defer_inline
 1.66% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.84% :     0.000013s :     68: predicate.switch_simplify
 1.15% :     0.000002s :     15: predicate.tile_eliminate
 1.08% :     0.000002s :     15: predicate.transpose_eliminate
 1.31% :     0.000002s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.61% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.75% :     0.000007s :     32: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000003s :     15: predicate.value_based_eliminate
 0.37% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000949    13
42.33% :     0.000402s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.10% :     0.000067s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.57% :     0.000480s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.053007    76
 0.15% :     0.000081s :      1: add_recomputation
 0.34% :     0.000182s :      1: auto_monad
 0.07% :     0.000036s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.71% :     0.000374s :      1: bootstrap
 0.07% :     0.000038s :      1: cconv
 0.03% :     0.000017s :      1: convert_after_rewriter
 0.08% :     0.000043s :      1: cse_after_recomputation
 0.03% :     0.000017s :      1: environ_conv
 0.06% :     0.000030s :      1: event_method
 0.02% :     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
14.52% :     0.007697s :      1: jit_opt_a
 0.46% :     0.000243s :      1: jit_opt_after_cconv
 0.15% :     0.000079s :      1: jit_opt_b
 1.29% :     0.000684s :      1: loop_unroll
 1.93% :     0.001024s :      1: mutable_eliminate
 2.64% :     0.001398s :     26: opt.transform.jit_opt_a
 0.18% :     0.000094s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000044s :      4: opt.transform.jit_opt_b
 0.05% :     0.000024s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000038s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.31% :     0.000693s :      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
 0.28% :     0.000146s :      1: py_interpret_to_execute
 0.04% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000023s :      1: remove_dup_value
 1.56% :     0.000827s :      1: renormalize.infer
 0.96% :     0.000507s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.62% :     0.000329s :      1: rewriter_after_opt_a
 0.21% :     0.000111s :      1: rewriter_before_opt_a
 0.22% :     0.000118s :      1: symbol_engine_optimizer
23.92% :     0.012678s :      1: task_emit
47.46% :     0.025159s :      1: type_inference
 0.19% :     0.000101s :      1: validate


TotalTime = 0.0131558, [30]
[bootstrap]: 0.00028866
[type_inference]: 0.00489791
[event_method]: 2.109e-05
[auto_monad]: 6.426e-05
[graph_reusing]: 5.77001e-06
[pre_auto_parallel]: 1.99e-06
[py_interpret_to_execute]: 3.669e-05
[rewriter_before_opt_a]: 8.755e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.00563895, [2]
    [Cycle 1]: 0.00183059, [27]
        [switch_simplify]: 6.039e-05
        [loop_unroll]: 3.589e-05
        [a_1]: 0.00062133
        [with_stream_mark]: 2.012e-05
        [recompute_prepare]: 8.35001e-06
        [updatestate_depend_eliminate]: 4.11001e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 6.83e-06
        [updatestate_useless_node_eliminater]: 6.07999e-06
        [accelerated_algorithm]: 6.48e-06
        [meta_shard_fg_expand]: 2.62001e-06
        [get_grad_eliminate_]: 6.24001e-06
        [merge_forward]: 4.23001e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.555e-05
        [j_node_and_user_rematch]: 9.51998e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 1.097e-05
        [inline_without_move]: 6.02999e-06
        [renormalize]: 0.00076119
        [add_forward_monad_depend]: 5.51998e-06
        [auto_monad_grad]: 2.98998e-06
        [auto_monad_eliminator]: 1.393e-05
        [cse]: 2.801e-05
        [replace_applicator]: 1.365e-05
    [Cycle 2]: 0.00035649, [27]
        [switch_simplify]: 7.08e-06
        [loop_unroll]: 5.67999e-06
        [a_1]: 0.00011102
        [with_stream_mark]: 1.048e-05
        [recompute_prepare]: 5.84e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 5.71998e-06
        [updatestate_useless_node_eliminater]: 5.49e-06
        [accelerated_algorithm]: 5.71998e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 5.28002e-06
        [merge_forward]: 2.84999e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.426e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 9.15001e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 6.28e-06
        [cse]: 1.277e-05
        [replace_applicator]: 5.69999e-06
[py_interpret_to_execute_after_opt_a]: 1.244e-05
[rewriter_after_opt_a]: 3.655e-05
[convert_after_rewriter]: 7.05998e-06
[order_py_execute_after_rewriter]: 5.15001e-06
[mutable_eliminate]: 0.00055096
[jit_opt_b]: 5.293e-05, [1]
    [Cycle 1]: 4.654e-05, [2]
        [frontend_op_eliminate]: 1.747e-05
        [inline_after_opt_a]: 1.721e-05
[cconv]: 2.409e-05
[loop_unroll]: 0.00042002
[jit_opt_after_cconv]: 0.00015443, [1]
    [Cycle 1]: 0.00014806, [11]
        [c_1]: 2.447e-05
        [parameter_eliminate]: 2.88e-06
        [updatestate_depend_eliminate]: 5.72001e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 5.41998e-06
        [cse]: 2.238e-05
        [call_graph_tuple_transform]: 2.096e-05
        [tuple_list_get_item_eliminator]: 6.07001e-06
        [none_parameter_eliminate]: 1.96998e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 6.07001e-06
[remove_dup_value]: 1.469e-05
[partial_unused_args_eliminate]: 2.30002e-06
[environ_conv]: 5.91e-06
[add_recomputation]: 4.75e-05
[cse_after_recomputation]: 2.313e-05, [1]
    [Cycle 1]: 1.753e-05, [1]
        [cse]: 1.195e-05
[auto_monad_reorder]: 1.853e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 5.64998e-06
[opt_after_jit_grad]: 0.00046219
[symbol_engine_optimizer]: 7.443e-05, [1]
    [Cycle 1]: 6.876e-05, [6]
        [build]: 3.2e-06
        [elim_shapecalc]: 8.91997e-06
        [elim_not_effective]: 1.347e-05
        [opt_reshape]: 6.11998e-06
        [fold_const_symbol]: 9.34998e-06
        [renormalize]: 3.60014e-07
[validate]: 3.663e-05
Sums
    bootstrap                            :     0.000289s :  3.15%
    type_inference                       :     0.004898s : 53.52%
    event_method                         :     0.000021s :  0.23%
    auto_monad                           :     0.000064s :  0.70%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.40%
    rewriter_before_opt_a                :     0.000088s :  0.96%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000067s :  0.74%
    jit_opt_a.loop_unroll                :     0.000042s :  0.45%
    jit_opt_a.a_1                        :     0.000732s :  8.00%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.09%
    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       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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_        :     0.000012s :  0.13%
    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 :     0.000030s :  0.33%
    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          :     0.000020s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000761s :  8.32%
    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      :     0.000020s :  0.22%
    jit_opt_a.cse                        :     0.000041s :  0.45%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.14%
    rewriter_after_opt_a                 :     0.000037s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000551s :  6.02%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000024s :  0.26%
    loop_unroll                          :     0.000420s :  4.59%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.27%
    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 :     0.000005s :  0.06%
    jit_opt_after_cconv.cse              :     0.000022s :  0.24%
    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 :  0.07%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.52%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000019s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000462s :  5.05%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.40%


Time group info:
------[substitution.]   0.000209    28
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000006s :      4: substitution.graph_param_transform
81.38% :     0.000170s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.49% :     0.000005s :      4: substitution.replace_old_param
 7.98% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004833     2
80.64% :     0.003897s :      1: type_inference.infer
19.36% :     0.000935s :      1: type_inference.specialize
------[replace.]   0.000080     8
55.14% :     0.000044s :      4: replace.inline
44.86% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     8
91.92% :     0.000167s :      4: match.inline
 8.08% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   912
 1.31% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.94% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     13: predicate.addn_check_dump
 1.46% :     0.000002s :     13: predicate.addn_zero_filter
 1.65% :     0.000002s :     13: predicate.arithmetic_simplify
 1.26% :     0.000002s :     13: predicate.cast_eliminate
 0.42% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     13: predicate.compare_switch_simplify
 1.22% :     0.000002s :     13: predicate.depend_value_elim
 1.19% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      4: predicate.elim_not_effective
 0.54% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     13: predicate.environ_get_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 1.23% :     0.000002s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 5.50% :     0.000007s :     29: predicate.inline
 0.91% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.14% :     0.000002s :      8: predicate.less_batch_normalization
 1.70% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.01% :     0.000003s :     21: predicate.load_eliminater
 1.38% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.20% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     13: predicate.merge_addn
 1.10% :     0.000001s :     13: predicate.minmaximum_grad
 1.38% :     0.000002s :      4: predicate.mutable_eliminate
 0.42% :     0.000001s :      4: predicate.opt_reshape
 2.57% :     0.000003s :     21: predicate.partial_eliminate
 1.20% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     13: predicate.reduce_eliminate
 1.63% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.83% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.04% :     0.000003s :     25: predicate.replace_applicator
 0.75% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000000s :      4: predicate.reset_defer_inline
 1.22% :     0.000002s :     13: predicate.reshape_eliminate
 1.17% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     13: predicate.same_eliminate
 0.69% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.36% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     21: predicate.switch_defer_inline
 2.21% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.69% :     0.000010s :     67: predicate.switch_simplify
 1.16% :     0.000002s :     13: predicate.tile_eliminate
 1.27% :     0.000002s :     13: predicate.transpose_eliminate
 1.45% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000603    12
48.48% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.55% :     0.000281s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014971    72
 0.34% :     0.000050s :      1: add_recomputation
 0.45% :     0.000068s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 2.01% :     0.000300s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.17% :     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
37.69% :     0.005642s :      1: jit_opt_a
 1.05% :     0.000157s :      1: jit_opt_after_cconv
 0.37% :     0.000056s :      1: jit_opt_b
 2.85% :     0.000427s :      1: loop_unroll
 3.73% :     0.000558s :      1: mutable_eliminate
 6.52% :     0.000976s :     26: opt.transform.jit_opt_a
 0.36% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000029s :      4: opt.transform.jit_opt_b
 0.09% :     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
 0.23% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.14% :     0.000470s :      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% :     0.000039s :      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
 2.72% :     0.000407s :      1: renormalize.infer
 2.32% :     0.000347s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000039s :      1: rewriter_after_opt_a
 0.60% :     0.000090s :      1: rewriter_before_opt_a
 0.52% :     0.000077s :      1: symbol_engine_optimizer
32.80% :     0.004910s :      1: type_inference


TotalTime = 0.0132828, [30]
[bootstrap]: 0.00033424
[type_inference]: 0.00503939
[event_method]: 2.284e-05
[auto_monad]: 6.548e-05
[graph_reusing]: 6.06998e-06
[pre_auto_parallel]: 2.64001e-06
[py_interpret_to_execute]: 3.965e-05
[rewriter_before_opt_a]: 9.594e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.00558554, [2]
    [Cycle 1]: 0.00172901, [27]
        [switch_simplify]: 5.955e-05
        [loop_unroll]: 3.448e-05
        [a_1]: 0.00061281
        [with_stream_mark]: 1.601e-05
        [recompute_prepare]: 8.25e-06
        [updatestate_depend_eliminate]: 3.66001e-06
        [updatestate_assign_eliminate]: 3.81001e-06
        [updatestate_loads_eliminate]: 2.76999e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 6.82002e-06
        [updatestate_useless_node_eliminater]: 6.21e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.72001e-06
        [merge_forward]: 3.80998e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.509e-05
        [j_node_and_user_rematch]: 9.61998e-06
        [meta_fg_expand]: 2.61999e-06
        [replace_old_param]: 1.114e-05
        [inline_without_move]: 5.79e-06
        [renormalize]: 0.00067452
        [add_forward_monad_depend]: 4.67e-06
        [auto_monad_grad]: 2.67001e-06
        [auto_monad_eliminator]: 1.351e-05
        [cse]: 3.33e-05
        [replace_applicator]: 1.341e-05
    [Cycle 2]: 0.0003542, [27]
        [switch_simplify]: 6.49001e-06
        [loop_unroll]: 6.19999e-06
        [a_1]: 0.00011231
        [with_stream_mark]: 9.59999e-06
        [recompute_prepare]: 5.67999e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 5.71e-06
        [accelerated_algorithm]: 6.04999e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.375e-05
        [j_node_and_user_rematch]: 8.08999e-06
        [meta_fg_expand]: 1.94999e-06
        [replace_old_param]: 8.87999e-06
        [inline_without_move]: 5.37999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 5.95002e-06
        [cse]: 1.335e-05
        [replace_applicator]: 5.59998e-06
[py_interpret_to_execute_after_opt_a]: 1.142e-05
[rewriter_after_opt_a]: 3.598e-05
[convert_after_rewriter]: 6.74999e-06
[order_py_execute_after_rewriter]: 7.65998e-06
[mutable_eliminate]: 0.00050691
[jit_opt_b]: 5.346e-05, [1]
    [Cycle 1]: 4.62e-05, [2]
        [frontend_op_eliminate]: 1.814e-05
        [inline_after_opt_a]: 1.672e-05
[cconv]: 2.224e-05
[loop_unroll]: 0.00046107
[jit_opt_after_cconv]: 0.00014889, [1]
    [Cycle 1]: 0.00014248, [11]
        [c_1]: 2.451e-05
        [parameter_eliminate]: 2.09e-06
        [updatestate_depend_eliminate]: 5.39e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [cse]: 2.178e-05
        [call_graph_tuple_transform]: 1.951e-05
        [tuple_list_get_item_eliminator]: 6.24001e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 5.89e-06
[remove_dup_value]: 1.408e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 5.58002e-06
[add_recomputation]: 4.976e-05
[cse_after_recomputation]: 2.271e-05, [1]
    [Cycle 1]: 1.729e-05, [1]
        [cse]: 1.167e-05
[auto_monad_reorder]: 1.756e-05
[get_jit_bprop_graph]: 1.74e-06
[rewriter_after_jit_bprop_graph]: 4.57e-06
[opt_after_jit_grad]: 0.00044802
[symbol_engine_optimizer]: 7.202e-05, [1]
    [Cycle 1]: 6.578e-05, [6]
        [build]: 2.91e-06
        [elim_shapecalc]: 7.88999e-06
        [elim_not_effective]: 1.247e-05
        [opt_reshape]: 6.07999e-06
        [fold_const_symbol]: 9.04e-06
        [renormalize]: 3.80009e-07
[validate]: 3.525e-05
Sums
    bootstrap                            :     0.000334s :  3.62%
    type_inference                       :     0.005039s : 54.64%
    event_method                         :     0.000023s :  0.25%
    auto_monad                           :     0.000065s :  0.71%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.43%
    rewriter_before_opt_a                :     0.000096s :  1.04%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000066s :  0.72%
    jit_opt_a.loop_unroll                :     0.000041s :  0.44%
    jit_opt_a.a_1                        :     0.000725s :  7.86%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.28%
    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 :     0.000006s :  0.07%
    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 :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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_        :     0.000011s :  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 :     0.000029s :  0.31%
    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          :     0.000020s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000675s :  7.31%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.21%
    jit_opt_a.cse                        :     0.000047s :  0.51%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000036s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000008s :  0.08%
    mutable_eliminate                    :     0.000507s :  5.50%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000022s :  0.24%
    loop_unroll                          :     0.000461s :  5.00%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.27%
    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 :     0.000003s :  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 :  0.21%
    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      :     0.000001s :  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                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000050s :  0.54%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000448s :  4.86%
    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  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.38%


Time group info:
------[substitution.]   0.000210    28
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000006s :      4: substitution.graph_param_transform
81.94% :     0.000172s :      4: substitution.inline
 1.38% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.43% :     0.000005s :      4: substitution.replace_old_param
 7.98% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004966     2
81.51% :     0.004047s :      1: type_inference.infer
18.49% :     0.000918s :      1: type_inference.specialize
------[replace.]   0.000080     8
56.87% :     0.000045s :      4: replace.inline
43.13% :     0.000034s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     8
91.87% :     0.000170s :      4: match.inline
 8.13% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   912
 1.19% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.93% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     13: predicate.addn_check_dump
 1.29% :     0.000002s :     13: predicate.addn_zero_filter
 1.92% :     0.000003s :     13: predicate.arithmetic_simplify
 1.37% :     0.000002s :     13: predicate.cast_eliminate
 0.48% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.23% :     0.000002s :     13: predicate.compare_switch_simplify
 1.18% :     0.000002s :     13: predicate.depend_value_elim
 1.19% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      4: predicate.elim_not_effective
 0.44% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     13: predicate.environ_get_eliminate
 1.13% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.95% :     0.000001s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 5.38% :     0.000007s :     29: predicate.inline
 0.92% :     0.000001s :      8: predicate.inline_without_move
 0.49% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      8: predicate.less_batch_normalization
 1.66% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.03% :     0.000003s :     21: predicate.load_eliminater
 1.46% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.26% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     13: predicate.merge_addn
 1.14% :     0.000002s :     13: predicate.minmaximum_grad
 1.50% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.68% :     0.000004s :     21: predicate.partial_eliminate
 1.26% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     13: predicate.reduce_eliminate
 1.64% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.00% :     0.000003s :     25: predicate.replace_applicator
 0.75% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.32% :     0.000002s :     13: predicate.reshape_eliminate
 1.27% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.74% :     0.000010s :     67: predicate.switch_simplify
 1.42% :     0.000002s :     13: predicate.tile_eliminate
 1.23% :     0.000002s :     13: predicate.transpose_eliminate
 1.38% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.23% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000002s :     21: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000601    12
50.37% :     0.000303s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.61% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.03% :     0.000265s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014993    72
 0.35% :     0.000052s :      1: add_recomputation
 0.46% :     0.000069s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 2.34% :     0.000350s :      1: bootstrap
 0.16% :     0.000025s :      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
 0.18% :     0.000027s :      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
37.27% :     0.005589s :      1: jit_opt_a
 1.01% :     0.000152s :      1: jit_opt_after_cconv
 0.37% :     0.000056s :      1: jit_opt_b
 3.13% :     0.000469s :      1: loop_unroll
 3.43% :     0.000514s :      1: mutable_eliminate
 6.42% :     0.000963s :     26: opt.transform.jit_opt_a
 0.35% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000029s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      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
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.04% :     0.000455s :      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
 0.28% :     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
 2.39% :     0.000359s :      1: renormalize.infer
 2.06% :     0.000308s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000039s :      1: rewriter_after_opt_a
 0.66% :     0.000099s :      1: rewriter_before_opt_a
 0.50% :     0.000075s :      1: symbol_engine_optimizer
33.70% :     0.005052s :      1: type_inference


TotalTime = 0.0451672, [33]
[bootstrap]: 0.00030676
[type_inference]: 0.0234048
[event_method]: 2.048e-05
[auto_monad]: 0.00018787
[graph_reusing]: 5.95002e-06
[pre_auto_parallel]: 3.13e-06
[py_interpret_to_execute]: 0.00013532
[rewriter_before_opt_a]: 9.84e-05
[expand_dump_flag]: 3.65e-06
[jit_opt_a]: 0.00662405, [2]
    [Cycle 1]: 0.002523, [27]
        [switch_simplify]: 6.407e-05
        [loop_unroll]: 3.543e-05
        [a_1]: 0.00072606
        [with_stream_mark]: 2.221e-05
        [recompute_prepare]: 1.1e-05
        [updatestate_depend_eliminate]: 6.46999e-06
        [updatestate_assign_eliminate]: 5.46e-06
        [updatestate_loads_eliminate]: 4.94e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 9.31e-06
        [updatestate_useless_node_eliminater]: 1.186e-05
        [accelerated_algorithm]: 8.46002e-06
        [meta_shard_fg_expand]: 2.89001e-06
        [get_grad_eliminate_]: 8.07e-06
        [merge_forward]: 5.89e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 5.425e-05
        [j_node_and_user_rematch]: 1.461e-05
        [meta_fg_expand]: 4.16001e-06
        [replace_old_param]: 1.214e-05
        [inline_without_move]: 7.88001e-06
        [renormalize]: 0.00123451
        [add_forward_monad_depend]: 6.23e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 2.217e-05
        [cse]: 3.83e-05
        [replace_applicator]: 1.822e-05
    [Cycle 2]: 0.00058211, [27]
        [switch_simplify]: 8.90001e-06
        [loop_unroll]: 7.58001e-06
        [a_1]: 0.00016776
        [with_stream_mark]: 1.252e-05
        [recompute_prepare]: 8.06001e-06
        [updatestate_depend_eliminate]: 5.44e-06
        [updatestate_assign_eliminate]: 4.3e-06
        [updatestate_loads_eliminate]: 3.83001e-06
        [parameter_eliminate]: 1.31002e-06
        [specialize_transform]: 8.16002e-06
        [updatestate_useless_node_eliminater]: 1.077e-05
        [accelerated_algorithm]: 8.05e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 7.58999e-06
        [merge_forward]: 4.70001e-06
        [cell_reuse_recompute_pass]: 2.02001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.71e-05
        [j_node_and_user_rematch]: 1.251e-05
        [meta_fg_expand]: 8.39e-05
        [replace_old_param]: 1.471e-05
        [inline_without_move]: 8.65001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 2.04999e-06
        [auto_monad_grad]: 1.47999e-06
        [auto_monad_eliminator]: 1.221e-05
        [cse]: 2.347e-05
        [replace_applicator]: 8.69e-06
[py_interpret_to_execute_after_opt_a]: 1.286e-05
[rewriter_after_opt_a]: 0.00029474
[convert_after_rewriter]: 1.157e-05
[order_py_execute_after_rewriter]: 7.30998e-06
[mutable_eliminate]: 0.00067305
[jit_opt_b]: 6.833e-05, [1]
    [Cycle 1]: 6.049e-05, [2]
        [frontend_op_eliminate]: 2.445e-05
        [inline_after_opt_a]: 2.403e-05
[cconv]: 2.758e-05
[loop_unroll]: 0.00046269
[jit_opt_after_cconv]: 0.00019254, [1]
    [Cycle 1]: 0.00018622, [11]
        [c_1]: 4.471e-05
        [parameter_eliminate]: 2.88e-06
        [updatestate_depend_eliminate]: 9.37999e-06
        [updatestate_assign_eliminate]: 4.62998e-06
        [updatestate_loads_eliminate]: 4.02e-06
        [cse]: 2.839e-05
        [call_graph_tuple_transform]: 2.307e-05
        [tuple_list_get_item_eliminator]: 8.32e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 8.30999e-06
[remove_dup_value]: 1.941e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 1.01e-05
[add_recomputation]: 6.711e-05
[cse_after_recomputation]: 3.229e-05, [1]
    [Cycle 1]: 2.614e-05, [1]
        [cse]: 1.952e-05
[auto_monad_reorder]: 2.588e-05
[get_jit_bprop_graph]: 1.93997e-06
[rewriter_after_jit_bprop_graph]: 6.63e-06
[opt_after_jit_grad]: 0.00055882
[symbol_engine_optimizer]: 0.00010459, [1]
    [Cycle 1]: 9.779e-05, [6]
        [build]: 1.297e-05
        [elim_shapecalc]: 1.214e-05
        [elim_not_effective]: 1.931e-05
        [opt_reshape]: 9.14998e-06
        [fold_const_symbol]: 1.431e-05
        [renormalize]: 4.19997e-07
[validate]: 5.819e-05
[backend_pass]: 1.13001e-06
[task_emit]: 0.0115022
[execute]: 9.00001e-06
Sums
    bootstrap                            :     0.000307s :  0.75%
    type_inference                       :     0.023405s : 57.11%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000188s :  0.46%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000135s :  0.33%
    rewriter_before_opt_a                :     0.000098s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.18%
    jit_opt_a.loop_unroll                :     0.000043s :  0.10%
    jit_opt_a.a_1                        :     0.000894s :  2.18%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    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 :     0.000071s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000088s :  0.21%
    jit_opt_a.replace_old_param          :     0.000027s :  0.07%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001235s :  3.01%
    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      :     0.000034s :  0.08%
    jit_opt_a.cse                        :     0.000062s :  0.15%
    jit_opt_a.replace_applicator         :     0.000027s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000295s :  0.72%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000673s :  1.64%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000028s :  0.07%
    loop_unroll                          :     0.000463s :  1.13%
    jit_opt_after_cconv.c_1              :     0.000045s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000028s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.02%
    add_recomputation                    :     0.000067s :  0.16%
    cse_after_recomputation.cse          :     0.000020s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000559s :  1.36%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000058s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011502s : 28.07%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000290    47
 3.32% :     0.000010s :      2: substitution.depend_value_elim
 0.90% :     0.000003s :      4: substitution.elim_not_effective
 0.76% :     0.000002s :      4: substitution.fold_const_symbol
 2.24% :     0.000006s :      5: substitution.graph_param_transform
75.63% :     0.000219s :      4: substitution.inline
 1.72% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.78% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.89% :     0.000005s :      2: substitution.replace_old_param
 4.56% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.14% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.06% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023332     2
93.45% :     0.021804s :      1: type_inference.infer
 6.55% :     0.001528s :      1: type_inference.specialize
------[replace.]   0.000094     6
64.42% :     0.000060s :      4: replace.inline
35.58% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     6
94.73% :     0.000216s :      4: match.inline
 5.27% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000172  1021
 1.50% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     15: predicate.addn_check_dump
 1.30% :     0.000002s :     15: predicate.addn_zero_filter
 2.31% :     0.000004s :     15: predicate.arithmetic_simplify
 1.50% :     0.000003s :     15: predicate.cast_eliminate
 0.46% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     15: predicate.compare_switch_simplify
 1.48% :     0.000003s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.92% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.57% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     15: predicate.environ_get_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.38% :     0.000001s :      5: predicate.graph_param_transform
 5.17% :     0.000009s :     31: predicate.inline
 1.35% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :     10: predicate.less_batch_normalization
 1.51% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.94% :     0.000003s :     22: predicate.load_eliminater
 1.13% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.31% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     15: predicate.merge_addn
 1.13% :     0.000002s :     15: predicate.minmaximum_grad
 1.88% :     0.000003s :      5: predicate.mutable_eliminate
 0.51% :     0.000001s :      5: predicate.opt_reshape
 2.49% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.57% :     0.000003s :     15: predicate.reduce_eliminate
 1.39% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.86% :     0.000003s :     27: predicate.replace_applicator
 0.73% :     0.000001s :     10: predicate.replace_old_param
 0.35% :     0.000001s :      5: predicate.reset_defer_inline
 1.28% :     0.000002s :     15: predicate.reshape_eliminate
 1.23% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.02% :     0.000002s :     10: predicate.special_op_eliminate
 1.07% :     0.000002s :     10: predicate.specialize_transform
 1.41% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.11% :     0.000004s :     21: predicate.switch_defer_inline
 1.85% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.64% :     0.000011s :     68: predicate.switch_simplify
 1.34% :     0.000002s :     15: predicate.tile_eliminate
 1.41% :     0.000002s :     15: predicate.transpose_eliminate
 1.52% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.79% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.25% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000909    13
44.03% :     0.000400s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.60% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.37% :     0.000458s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047824    76
 0.15% :     0.000070s :      1: add_recomputation
 0.41% :     0.000194s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000321s :      1: bootstrap
 0.06% :     0.000030s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000035s :      1: cse_after_recomputation
 0.03% :     0.000012s :      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% :     0.000009s :      1: graph_reusing
13.86% :     0.006628s :      1: jit_opt_a
 0.41% :     0.000195s :      1: jit_opt_after_cconv
 0.15% :     0.000071s :      1: jit_opt_b
 0.98% :     0.000470s :      1: loop_unroll
 1.43% :     0.000682s :      1: mutable_eliminate
 2.59% :     0.001237s :     26: opt.transform.jit_opt_a
 0.17% :     0.000081s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000568s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000140s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.56% :     0.000744s :      1: renormalize.infer
 1.01% :     0.000482s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000300s :      1: rewriter_after_opt_a
 0.22% :     0.000103s :      1: rewriter_before_opt_a
 0.22% :     0.000107s :      1: symbol_engine_optimizer
24.09% :     0.011519s :      1: task_emit
48.97% :     0.023419s :      1: type_inference
 0.17% :     0.000083s :      1: validate


TotalTime = 0.0131542, [30]
[bootstrap]: 0.00027156
[type_inference]: 0.0048913
[event_method]: 2.13e-05
[auto_monad]: 6.529e-05
[graph_reusing]: 6.23e-06
[pre_auto_parallel]: 2.50002e-06
[py_interpret_to_execute]: 3.928e-05
[rewriter_before_opt_a]: 9.451e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00551609, [2]
    [Cycle 1]: 0.0018314, [27]
        [switch_simplify]: 6.302e-05
        [loop_unroll]: 3.546e-05
        [a_1]: 0.00065741
        [with_stream_mark]: 1.858e-05
        [recompute_prepare]: 8e-06
        [updatestate_depend_eliminate]: 3.92998e-06
        [updatestate_assign_eliminate]: 3.51999e-06
        [updatestate_loads_eliminate]: 3.20998e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 6.98998e-06
        [updatestate_useless_node_eliminater]: 6.04001e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.47001e-06
        [get_grad_eliminate_]: 5.92999e-06
        [merge_forward]: 3.86001e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.483e-05
        [j_node_and_user_rematch]: 9.99001e-06
        [meta_fg_expand]: 2.36998e-06
        [replace_old_param]: 1.083e-05
        [inline_without_move]: 5.77999e-06
        [renormalize]: 0.00071141
        [add_forward_monad_depend]: 5.10999e-06
        [auto_monad_grad]: 2.05002e-06
        [auto_monad_eliminator]: 1.563e-05
        [cse]: 3.382e-05
        [replace_applicator]: 1.403e-05
    [Cycle 2]: 0.00036299, [27]
        [switch_simplify]: 6.76e-06
        [loop_unroll]: 5.61998e-06
        [a_1]: 0.00011422
        [with_stream_mark]: 1.122e-05
        [recompute_prepare]: 5.86e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.23998e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.51998e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 1.92001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.452e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 1.02e-05
        [inline_without_move]: 5.33002e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 6.66999e-06
        [cse]: 1.24e-05
        [replace_applicator]: 6.17999e-06
[py_interpret_to_execute_after_opt_a]: 9.88998e-06
[rewriter_after_opt_a]: 3.739e-05
[convert_after_rewriter]: 6.76e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00055797
[jit_opt_b]: 5.655e-05, [1]
    [Cycle 1]: 4.919e-05, [2]
        [frontend_op_eliminate]: 1.858e-05
        [inline_after_opt_a]: 1.821e-05
[cconv]: 2.564e-05
[loop_unroll]: 0.00044559
[jit_opt_after_cconv]: 0.00015857, [1]
    [Cycle 1]: 0.00015183, [11]
        [c_1]: 2.462e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 6.48e-06
        [updatestate_assign_eliminate]: 2.81999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.416e-05
        [call_graph_tuple_transform]: 2.483e-05
        [tuple_list_get_item_eliminator]: 6.16e-06
        [none_parameter_eliminate]: 1.86003e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 6.56e-06
[remove_dup_value]: 1.511e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 2.868e-05
[add_recomputation]: 5.171e-05
[cse_after_recomputation]: 2.477e-05, [1]
    [Cycle 1]: 1.895e-05, [1]
        [cse]: 1.319e-05
[auto_monad_reorder]: 1.695e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 5.89999e-06
[opt_after_jit_grad]: 0.00049381
[symbol_engine_optimizer]: 7.543e-05, [1]
    [Cycle 1]: 6.899e-05, [6]
        [build]: 3.68999e-06
        [elim_shapecalc]: 8.53001e-06
        [elim_not_effective]: 1.324e-05
        [opt_reshape]: 6.16e-06
        [fold_const_symbol]: 9.46e-06
        [renormalize]: 3.89991e-07
[validate]: 5.865e-05
Sums
    bootstrap                            :     0.000272s :  2.93%
    type_inference                       :     0.004891s : 52.84%
    event_method                         :     0.000021s :  0.23%
    auto_monad                           :     0.000065s :  0.71%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.42%
    rewriter_before_opt_a                :     0.000095s :  1.02%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000070s :  0.75%
    jit_opt_a.loop_unroll                :     0.000041s :  0.44%
    jit_opt_a.a_1                        :     0.000772s :  8.34%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  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       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.12%
    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              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  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          :     0.000021s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000711s :  7.69%
    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      :     0.000022s :  0.24%
    jit_opt_a.cse                        :     0.000046s :  0.50%
    jit_opt_a.replace_applicator         :     0.000020s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000037s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000558s :  6.03%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000026s :  0.28%
    loop_unroll                          :     0.000446s :  4.81%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.27%
    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              :     0.000024s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.27%
    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  :     0.000007s :  0.07%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000029s :  0.31%
    add_recomputation                    :     0.000052s :  0.56%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000017s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000494s :  5.33%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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                             :     0.000059s :  0.63%


Time group info:
------[substitution.]   0.000214    28
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.89% :     0.000006s :      4: substitution.graph_param_transform
81.90% :     0.000175s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.62% :     0.000006s :      4: substitution.replace_old_param
 7.58% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004828     2
81.00% :     0.003910s :      1: type_inference.infer
19.00% :     0.000917s :      1: type_inference.specialize
------[replace.]   0.000086     8
55.66% :     0.000048s :      4: replace.inline
44.34% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000186     8
92.35% :     0.000172s :      4: match.inline
 7.65% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   912
 1.20% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.12% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     13: predicate.addn_check_dump
 1.26% :     0.000002s :     13: predicate.addn_zero_filter
 1.89% :     0.000003s :     13: predicate.arithmetic_simplify
 1.30% :     0.000002s :     13: predicate.cast_eliminate
 0.38% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.10% :     0.000002s :     13: predicate.compare_switch_simplify
 1.12% :     0.000002s :     13: predicate.depend_value_elim
 1.19% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.45% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     13: predicate.environ_get_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.95% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.25% :     0.000007s :     29: predicate.inline
 0.91% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      8: predicate.less_batch_normalization
 1.65% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.96% :     0.000003s :     21: predicate.load_eliminater
 1.38% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.04% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 2.08% :     0.000003s :      4: predicate.mutable_eliminate
 0.42% :     0.000001s :      4: predicate.opt_reshape
 2.57% :     0.000004s :     21: predicate.partial_eliminate
 1.17% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     13: predicate.reduce_eliminate
 1.63% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.99% :     0.000003s :     25: predicate.replace_applicator
 0.98% :     0.000001s :      8: predicate.replace_old_param
 0.54% :     0.000001s :      4: predicate.reset_defer_inline
 1.15% :     0.000002s :     13: predicate.reshape_eliminate
 1.35% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     13: predicate.same_eliminate
 0.56% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      8: predicate.special_op_eliminate
 1.00% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.56% :     0.000004s :     21: predicate.switch_defer_inline
 2.21% :     0.000003s :     21: predicate.switch_layer_defer_inline
 8.33% :     0.000012s :     67: predicate.switch_simplify
 1.20% :     0.000002s :     13: predicate.tile_eliminate
 1.18% :     0.000002s :     13: predicate.transpose_eliminate
 1.35% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.86% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     13: predicate.value_based_eliminate
 0.34% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000621    12
50.94% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.76% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.29% :     0.000275s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014947    72
 0.37% :     0.000055s :      1: add_recomputation
 0.46% :     0.000069s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 1.90% :     0.000284s :      1: bootstrap
 0.19% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000027s :      1: cse_after_recomputation
 0.21% :     0.000032s :      1: environ_conv
 0.17% :     0.000025s :      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
36.92% :     0.005519s :      1: jit_opt_a
 1.08% :     0.000162s :      1: jit_opt_after_cconv
 0.40% :     0.000059s :      1: jit_opt_b
 3.04% :     0.000454s :      1: loop_unroll
 3.81% :     0.000569s :      1: mutable_eliminate
 6.81% :     0.001017s :     26: opt.transform.jit_opt_a
 0.39% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000030s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.36% :     0.000502s :      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% :     0.000042s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.75% :     0.000410s :      1: renormalize.infer
 1.96% :     0.000293s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000040s :      1: rewriter_after_opt_a
 0.65% :     0.000098s :      1: rewriter_before_opt_a
 0.52% :     0.000078s :      1: symbol_engine_optimizer
32.81% :     0.004905s :      1: type_inference


TotalTime = 0.0120916, [30]
[bootstrap]: 0.00027147
[type_inference]: 0.00454485
[event_method]: 2.041e-05
[auto_monad]: 6.114e-05
[graph_reusing]: 5.51e-06
[pre_auto_parallel]: 2.21998e-06
[py_interpret_to_execute]: 3.729e-05
[rewriter_before_opt_a]: 8.738e-05
[expand_dump_flag]: 2.77002e-06
[jit_opt_a]: 0.00505016, [2]
    [Cycle 1]: 0.00166929, [27]
        [switch_simplify]: 5.73e-05
        [loop_unroll]: 3.569e-05
        [a_1]: 0.00061318
        [with_stream_mark]: 1.668e-05
        [recompute_prepare]: 8.28001e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 3.08998e-06
        [updatestate_loads_eliminate]: 2.75997e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 6.81999e-06
        [updatestate_useless_node_eliminater]: 6.07999e-06
        [accelerated_algorithm]: 6.49999e-06
        [meta_shard_fg_expand]: 2.44999e-06
        [get_grad_eliminate_]: 5.87999e-06
        [merge_forward]: 3.68999e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.507e-05
        [j_node_and_user_rematch]: 9.57001e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 1.071e-05
        [inline_without_move]: 5.93002e-06
        [renormalize]: 0.00062513
        [add_forward_monad_depend]: 5.41002e-06
        [auto_monad_grad]: 2.18998e-06
        [auto_monad_eliminator]: 1.318e-05
        [cse]: 2.716e-05
        [replace_applicator]: 1.207e-05
    [Cycle 2]: 0.00034605, [27]
        [switch_simplify]: 6.58e-06
        [loop_unroll]: 6.01998e-06
        [a_1]: 0.00011285
        [with_stream_mark]: 8.17e-06
        [recompute_prepare]: 5.76e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 5.78997e-06
        [updatestate_useless_node_eliminater]: 5.62999e-06
        [accelerated_algorithm]: 5.77999e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.28e-05
        [j_node_and_user_rematch]: 8.40001e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 8.57e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 9.80013e-07
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.55001e-06
        [cse]: 1.27e-05
        [replace_applicator]: 5.59e-06
[py_interpret_to_execute_after_opt_a]: 2.236e-05
[rewriter_after_opt_a]: 3.533e-05
[convert_after_rewriter]: 6.61e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00047875
[jit_opt_b]: 5.216e-05, [1]
    [Cycle 1]: 4.577e-05, [2]
        [frontend_op_eliminate]: 1.778e-05
        [inline_after_opt_a]: 1.704e-05
[cconv]: 2.238e-05
[loop_unroll]: 0.00041027
[jit_opt_after_cconv]: 0.00014625, [1]
    [Cycle 1]: 0.00013974, [11]
        [c_1]: 2.453e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 5.57999e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 1.996e-05
        [call_graph_tuple_transform]: 1.926e-05
        [tuple_list_get_item_eliminator]: 6.11998e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 6.33e-06
[remove_dup_value]: 1.376e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.38002e-06
[add_recomputation]: 4.758e-05
[cse_after_recomputation]: 2.24e-05, [1]
    [Cycle 1]: 1.73e-05, [1]
        [cse]: 1.195e-05
[auto_monad_reorder]: 1.586e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 4.74002e-06
[opt_after_jit_grad]: 0.00045237
[symbol_engine_optimizer]: 7.555e-05, [1]
    [Cycle 1]: 6.93e-05, [6]
        [build]: 3.17002e-06
        [elim_shapecalc]: 8.47998e-06
        [elim_not_effective]: 1.41e-05
        [opt_reshape]: 6.22001e-06
        [fold_const_symbol]: 9.21998e-06
        [renormalize]: 3.59985e-07
[validate]: 3.721e-05
Sums
    bootstrap                            :     0.000271s :  3.19%
    type_inference                       :     0.004545s : 53.37%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000061s :  0.72%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.44%
    rewriter_before_opt_a                :     0.000087s :  1.03%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000064s :  0.75%
    jit_opt_a.loop_unroll                :     0.000042s :  0.49%
    jit_opt_a.a_1                        :     0.000726s :  8.53%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.29%
    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 :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000011s :  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 :  0.33%
    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          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000625s :  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 :  0.22%
    jit_opt_a.cse                        :     0.000040s :  0.47%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000022s :  0.26%
    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                    :     0.000479s :  5.62%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000022s :  0.26%
    loop_unroll                          :     0.000410s :  4.82%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.29%
    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              :     0.000020s :  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      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    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 :  0.56%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    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                   :     0.000452s :  5.31%
    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 :  0.00%
    validate                             :     0.000037s :  0.44%


Time group info:
------[substitution.]   0.000204    28
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.46% :     0.000005s :      4: substitution.graph_param_transform
82.10% :     0.000168s :      4: substitution.inline
 1.38% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000004s :      4: substitution.replace_old_param
 7.96% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004491     2
79.84% :     0.003585s :      1: type_inference.infer
20.16% :     0.000905s :      1: type_inference.specialize
------[replace.]   0.000080     8
55.77% :     0.000045s :      4: replace.inline
44.23% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     8
92.00% :     0.000165s :      4: match.inline
 8.00% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   912
 1.30% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     13: predicate.addn_check_dump
 1.27% :     0.000002s :     13: predicate.addn_zero_filter
 1.82% :     0.000002s :     13: predicate.arithmetic_simplify
 1.26% :     0.000002s :     13: predicate.cast_eliminate
 0.45% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.10% :     0.000002s :     13: predicate.compare_switch_simplify
 1.21% :     0.000002s :     13: predicate.depend_value_elim
 1.15% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      4: predicate.elim_not_effective
 0.50% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.42% :     0.000002s :     13: predicate.environ_get_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000001s :      8: predicate.get_grad_eliminate
 0.34% :     0.000000s :      4: predicate.graph_param_transform
 5.83% :     0.000008s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.46% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      8: predicate.less_batch_normalization
 1.70% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.91% :     0.000003s :     21: predicate.load_eliminater
 1.21% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.28% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     13: predicate.merge_addn
 1.21% :     0.000002s :     13: predicate.minmaximum_grad
 1.39% :     0.000002s :      4: predicate.mutable_eliminate
 0.48% :     0.000001s :      4: predicate.opt_reshape
 2.62% :     0.000004s :     21: predicate.partial_eliminate
 1.14% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.00% :     0.000003s :     25: predicate.replace_applicator
 0.65% :     0.000001s :      8: predicate.replace_old_param
 0.36% :     0.000000s :      4: predicate.reset_defer_inline
 1.16% :     0.000002s :     13: predicate.reshape_eliminate
 1.20% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     13: predicate.same_eliminate
 0.69% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.59% :     0.000004s :     21: predicate.switch_defer_inline
 2.15% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.75% :     0.000011s :     67: predicate.switch_simplify
 1.39% :     0.000002s :     13: predicate.tile_eliminate
 1.19% :     0.000002s :     13: predicate.transpose_eliminate
 1.41% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.89% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000002s :     13: predicate.value_based_eliminate
 0.36% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000564    12
46.02% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.81% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.17% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013754    72
 0.37% :     0.000050s :      1: add_recomputation
 0.47% :     0.000065s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.06% :     0.000283s :      1: bootstrap
 0.18% :     0.000025s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.17% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
36.74% :     0.005053s :      1: jit_opt_a
 1.09% :     0.000149s :      1: jit_opt_after_cconv
 0.40% :     0.000055s :      1: jit_opt_b
 3.04% :     0.000418s :      1: loop_unroll
 3.54% :     0.000486s :      1: mutable_eliminate
 6.99% :     0.000961s :     26: opt.transform.jit_opt_a
 0.39% :     0.000053s :      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
 0.18% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.34% :     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
 0.29% :     0.000040s :      1: py_interpret_to_execute
 0.18% :     0.000025s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.48% :     0.000341s :      1: renormalize.infer
 2.01% :     0.000276s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000038s :      1: rewriter_after_opt_a
 0.66% :     0.000090s :      1: rewriter_before_opt_a
 0.57% :     0.000078s :      1: symbol_engine_optimizer
33.12% :     0.004555s :      1: type_inference


TotalTime = 0.0123152, [30]
[bootstrap]: 0.000266
[type_inference]: 0.00455676
[event_method]: 2.073e-05
[auto_monad]: 6.146e-05
[graph_reusing]: 5.64998e-06
[pre_auto_parallel]: 2.16998e-06
[py_interpret_to_execute]: 3.561e-05
[rewriter_before_opt_a]: 8.582e-05
[expand_dump_flag]: 2.83998e-06
[jit_opt_a]: 0.00515994, [2]
    [Cycle 1]: 0.0017314, [27]
        [switch_simplify]: 5.833e-05
        [loop_unroll]: 3.623e-05
        [a_1]: 0.0006269
        [with_stream_mark]: 1.937e-05
        [recompute_prepare]: 8.12998e-06
        [updatestate_depend_eliminate]: 4.25e-06
        [updatestate_assign_eliminate]: 3.63999e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 2.41e-06
        [specialize_transform]: 7.48e-06
        [updatestate_useless_node_eliminater]: 6.96999e-06
        [accelerated_algorithm]: 7.03e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 5.94999e-06
        [merge_forward]: 3.83999e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.453e-05
        [j_node_and_user_rematch]: 1.009e-05
        [meta_fg_expand]: 2.66999e-06
        [replace_old_param]: 1.096e-05
        [inline_without_move]: 5.82999e-06
        [renormalize]: 0.00065889
        [add_forward_monad_depend]: 6.23998e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.434e-05
        [cse]: 2.606e-05
        [replace_applicator]: 1.315e-05
    [Cycle 2]: 0.00036861, [27]
        [switch_simplify]: 6.91999e-06
        [loop_unroll]: 5.99e-06
        [a_1]: 0.000113
        [with_stream_mark]: 8.89e-06
        [recompute_prepare]: 5.87001e-06
        [updatestate_depend_eliminate]: 3.17002e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 5.81e-06
        [accelerated_algorithm]: 5.94999e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 1.211e-05
        [merge_forward]: 2.90998e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05
        [j_node_and_user_rematch]: 8.77e-06
        [meta_fg_expand]: 1.94999e-06
        [replace_old_param]: 9.61e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.47001e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 6.69999e-06
        [cse]: 1.405e-05
        [replace_applicator]: 5.81998e-06
[py_interpret_to_execute_after_opt_a]: 1.026e-05
[rewriter_after_opt_a]: 3.852e-05
[convert_after_rewriter]: 7.50998e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.0005111
[jit_opt_b]: 5.387e-05, [1]
    [Cycle 1]: 4.757e-05, [2]
        [frontend_op_eliminate]: 1.818e-05
        [inline_after_opt_a]: 1.797e-05
[cconv]: 2.266e-05
[loop_unroll]: 0.0004573
[jit_opt_after_cconv]: 0.00016313, [1]
    [Cycle 1]: 0.00015603, [11]
        [c_1]: 2.685e-05
        [parameter_eliminate]: 3.12002e-06
        [updatestate_depend_eliminate]: 5.94e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [cse]: 2.126e-05
        [call_graph_tuple_transform]: 2.435e-05
        [tuple_list_get_item_eliminator]: 6.43e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 7.35998e-06
[remove_dup_value]: 1.457e-05
[partial_unused_args_eliminate]: 2.61999e-06
[environ_conv]: 6.12001e-06
[add_recomputation]: 4.668e-05
[cse_after_recomputation]: 2.367e-05, [1]
    [Cycle 1]: 1.84e-05, [1]
        [cse]: 1.268e-05
[auto_monad_reorder]: 1.703e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 5.05001e-06
[opt_after_jit_grad]: 0.00046515
[symbol_engine_optimizer]: 7.576e-05, [1]
    [Cycle 1]: 6.923e-05, [6]
        [build]: 3.3e-06
        [elim_shapecalc]: 8.25e-06
        [elim_not_effective]: 1.371e-05
        [opt_reshape]: 6.53e-06
        [fold_const_symbol]: 9.54e-06
        [renormalize]: 4.60015e-07
[validate]: 3.55e-05
Sums
    bootstrap                            :     0.000266s :  3.06%
    type_inference                       :     0.004557s : 52.41%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000061s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.41%
    rewriter_before_opt_a                :     0.000086s :  0.99%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.75%
    jit_opt_a.loop_unroll                :     0.000042s :  0.49%
    jit_opt_a.a_1                        :     0.000740s :  8.51%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.09%
    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       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000018s :  0.21%
    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 :     0.000029s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000021s :  0.24%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000659s :  7.58%
    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      :     0.000021s :  0.24%
    jit_opt_a.cse                        :     0.000040s :  0.46%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000039s :  0.44%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000511s :  5.88%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.21%
    cconv                                :     0.000023s :  0.26%
    loop_unroll                          :     0.000457s :  5.26%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.31%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  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      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.08%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000047s :  0.54%
    cse_after_recomputation.cse          :     0.000013s :  0.15%
    auto_monad_reorder                   :     0.000017s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000465s :  5.35%
    symbol_engine_optimizer.build        :     0.000003s :  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  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000035s :  0.41%


Time group info:
------[substitution.]   0.000215    28
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.58% :     0.000006s :      4: substitution.graph_param_transform
81.99% :     0.000177s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000005s :      4: substitution.replace_old_param
 7.82% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004500     2
80.30% :     0.003614s :      1: type_inference.infer
19.70% :     0.000887s :      1: type_inference.specialize
------[replace.]   0.000082     8
56.28% :     0.000046s :      4: replace.inline
43.72% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     8
92.15% :     0.000174s :      4: match.inline
 7.85% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   912
 1.22% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     13: predicate.addn_check_dump
 1.26% :     0.000002s :     13: predicate.addn_zero_filter
 1.94% :     0.000003s :     13: predicate.arithmetic_simplify
 1.36% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     13: predicate.compare_switch_simplify
 1.15% :     0.000002s :     13: predicate.depend_value_elim
 1.15% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      4: predicate.elim_not_effective
 0.52% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     13: predicate.environ_get_eliminate
 1.10% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.99% :     0.000001s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 5.58% :     0.000008s :     29: predicate.inline
 0.96% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.13% :     0.000002s :      8: predicate.less_batch_normalization
 1.53% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.91% :     0.000003s :     21: predicate.load_eliminater
 1.37% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.26% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     13: predicate.merge_addn
 1.10% :     0.000001s :     13: predicate.minmaximum_grad
 1.43% :     0.000002s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.67% :     0.000004s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     13: predicate.reduce_eliminate
 1.56% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.99% :     0.000003s :     25: predicate.replace_applicator
 0.83% :     0.000001s :      8: predicate.replace_old_param
 0.36% :     0.000000s :      4: predicate.reset_defer_inline
 1.21% :     0.000002s :     13: predicate.reshape_eliminate
 1.16% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     13: predicate.same_eliminate
 0.64% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      8: predicate.special_op_eliminate
 0.99% :     0.000001s :      8: predicate.specialize_transform
 1.39% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.66% :     0.000004s :     21: predicate.switch_defer_inline
 2.23% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.83% :     0.000011s :     67: predicate.switch_simplify
 1.14% :     0.000002s :     13: predicate.tile_eliminate
 1.15% :     0.000002s :     13: predicate.transpose_eliminate
 1.44% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.23% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 2.02% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.16% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000584    12
47.45% :     0.000277s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.15% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.41% :     0.000271s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014051    72
 0.35% :     0.000050s :      1: add_recomputation
 0.46% :     0.000065s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 1.97% :     0.000277s :      1: bootstrap
 0.18% :     0.000025s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.17% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
36.74% :     0.005163s :      1: jit_opt_a
 1.18% :     0.000166s :      1: jit_opt_after_cconv
 0.40% :     0.000057s :      1: jit_opt_b
 3.31% :     0.000465s :      1: loop_unroll
 3.70% :     0.000520s :      1: mutable_eliminate
 7.04% :     0.000989s :     26: opt.transform.jit_opt_a
 0.44% :     0.000062s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000030s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.37% :     0.000473s :      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
 0.27% :     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
 2.53% :     0.000356s :      1: renormalize.infer
 2.09% :     0.000294s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.30% :     0.000042s :      1: rewriter_after_opt_a
 0.63% :     0.000089s :      1: rewriter_before_opt_a
 0.56% :     0.000079s :      1: symbol_engine_optimizer
32.50% :     0.004567s :      1: type_inference


TotalTime = 0.0121936, [30]
[bootstrap]: 0.0002423
[type_inference]: 0.00441242
[event_method]: 1.977e-05
[auto_monad]: 5.605e-05
[graph_reusing]: 4.70999e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.68e-05
[rewriter_before_opt_a]: 8.699e-05
[expand_dump_flag]: 2.53e-06
[jit_opt_a]: 0.00526702, [2]
    [Cycle 1]: 0.00177791, [27]
        [switch_simplify]: 6.013e-05
        [loop_unroll]: 3.647e-05
        [a_1]: 0.00064768
        [with_stream_mark]: 1.648e-05
        [recompute_prepare]: 8.63001e-06
        [updatestate_depend_eliminate]: 4.03001e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 3.43e-06
        [parameter_eliminate]: 1.67999e-06
        [specialize_transform]: 1.508e-05
        [updatestate_useless_node_eliminater]: 6.43e-06
        [accelerated_algorithm]: 6.69999e-06
        [meta_shard_fg_expand]: 2.42001e-06
        [get_grad_eliminate_]: 6.63e-06
        [merge_forward]: 3.98999e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.526e-05
        [j_node_and_user_rematch]: 9.42001e-06
        [meta_fg_expand]: 2.75002e-06
        [replace_old_param]: 1.101e-05
        [inline_without_move]: 6.26e-06
        [renormalize]: 0.00068366
        [add_forward_monad_depend]: 5.77999e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.253e-05
        [cse]: 2.565e-05
        [replace_applicator]: 1.213e-05
    [Cycle 2]: 0.00035556, [27]
        [switch_simplify]: 7.36999e-06
        [loop_unroll]: 5.95002e-06
        [a_1]: 0.0001148
        [with_stream_mark]: 8.55001e-06
        [recompute_prepare]: 5.89e-06
        [updatestate_depend_eliminate]: 2.97002e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 6.13002e-06
        [updatestate_useless_node_eliminater]: 5.91e-06
        [accelerated_algorithm]: 6.06998e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 5.42001e-06
        [merge_forward]: 2.62001e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.288e-05
        [j_node_and_user_rematch]: 8.67998e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 8.90001e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.11997e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 5.66e-06
        [cse]: 1.332e-05
        [replace_applicator]: 5.94999e-06
[py_interpret_to_execute_after_opt_a]: 1.014e-05
[rewriter_after_opt_a]: 3.361e-05
[convert_after_rewriter]: 7e-06
[order_py_execute_after_rewriter]: 5.42001e-06
[mutable_eliminate]: 0.00047745
[jit_opt_b]: 5.466e-05, [1]
    [Cycle 1]: 4.806e-05, [2]
        [frontend_op_eliminate]: 1.844e-05
        [inline_after_opt_a]: 1.807e-05
[cconv]: 2.316e-05
[loop_unroll]: 0.00044982
[jit_opt_after_cconv]: 0.00014938, [1]
    [Cycle 1]: 0.00014308, [11]
        [c_1]: 2.611e-05
        [parameter_eliminate]: 2.55002e-06
        [updatestate_depend_eliminate]: 5.49e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.054e-05
        [call_graph_tuple_transform]: 1.934e-05
        [tuple_list_get_item_eliminator]: 6.17001e-06
        [none_parameter_eliminate]: 1.28002e-06
        [renormalize]: 3.50003e-07
        [switch_simplify]: 6.02999e-06
[remove_dup_value]: 1.335e-05
[partial_unused_args_eliminate]: 2.28998e-06
[environ_conv]: 5.69e-06
[add_recomputation]: 4.768e-05
[cse_after_recomputation]: 2.388e-05, [1]
    [Cycle 1]: 1.826e-05, [1]
        [cse]: 1.237e-05
[auto_monad_reorder]: 1.714e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 4.83001e-06
[opt_after_jit_grad]: 0.00046633
[symbol_engine_optimizer]: 7.592e-05, [1]
    [Cycle 1]: 7.008e-05, [6]
        [build]: 3.06001e-06
        [elim_shapecalc]: 8.83001e-06
        [elim_not_effective]: 1.415e-05
        [opt_reshape]: 6.28e-06
        [fold_const_symbol]: 9.57999e-06
        [renormalize]: 3.00002e-07
[validate]: 3.368e-05
Sums
    bootstrap                            :     0.000242s :  2.85%
    type_inference                       :     0.004412s : 51.89%
    event_method                         :     0.000020s :  0.23%
    auto_monad                           :     0.000056s :  0.66%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.43%
    rewriter_before_opt_a                :     0.000087s :  1.02%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000067s :  0.79%
    jit_opt_a.loop_unroll                :     0.000042s :  0.50%
    jit_opt_a.a_1                        :     0.000762s :  8.97%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.29%
    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 :     0.000006s :  0.07%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000021s :  0.25%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.14%
    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             :     0.000005s :  0.06%
    jit_opt_a.replace_old_param          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000684s :  8.04%
    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                        :     0.000039s :  0.46%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000034s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000477s :  5.61%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.21%
    cconv                                :     0.000023s :  0.27%
    loop_unroll                          :     0.000450s :  5.29%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.31%
    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              :     0.000021s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  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 :  0.07%
    remove_dup_value                     :     0.000013s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000048s :  0.56%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000017s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000466s :  5.48%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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 :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.40%


Time group info:
------[substitution.]   0.000215    28
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000005s :      4: substitution.graph_param_transform
83.42% :     0.000179s :      4: substitution.inline
 1.32% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.20% :     0.000005s :      4: substitution.replace_old_param
 7.05% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004361     2
78.42% :     0.003420s :      1: type_inference.infer
21.58% :     0.000941s :      1: type_inference.specialize
------[replace.]   0.000086     8
58.25% :     0.000050s :      4: replace.inline
41.75% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     8
93.13% :     0.000176s :      4: match.inline
 6.87% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.25% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     13: predicate.addn_check_dump
 1.21% :     0.000002s :     13: predicate.addn_zero_filter
 2.00% :     0.000003s :     13: predicate.arithmetic_simplify
 1.39% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.14% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      4: predicate.elim_not_effective
 0.67% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     13: predicate.environ_get_eliminate
 1.08% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      4: predicate.fold_const_symbol
 0.99% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.24% :     0.000007s :     29: predicate.inline
 0.98% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      8: predicate.less_batch_normalization
 1.54% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.96% :     0.000003s :     21: predicate.load_eliminater
 1.24% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.17% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     13: predicate.merge_addn
 1.17% :     0.000002s :     13: predicate.minmaximum_grad
 1.44% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.61% :     0.000004s :     21: predicate.partial_eliminate
 1.22% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     13: predicate.reduce_eliminate
 1.65% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.92% :     0.000003s :     25: predicate.replace_applicator
 0.64% :     0.000001s :      8: predicate.replace_old_param
 0.32% :     0.000000s :      4: predicate.reset_defer_inline
 1.16% :     0.000002s :     13: predicate.reshape_eliminate
 1.16% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      8: predicate.special_op_eliminate
 1.01% :     0.000001s :      8: predicate.specialize_transform
 1.51% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.74% :     0.000004s :     21: predicate.switch_defer_inline
 2.18% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.75% :     0.000011s :     67: predicate.switch_simplify
 1.22% :     0.000002s :     13: predicate.tile_eliminate
 1.20% :     0.000002s :     13: predicate.transpose_eliminate
 1.51% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.86% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.15% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     13: predicate.value_based_eliminate
 0.42% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000605    12
42.65% :     0.000258s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.01% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.34% :     0.000311s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013967    72
 0.36% :     0.000050s :      1: add_recomputation
 0.42% :     0.000059s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 1.79% :     0.000251s :      1: bootstrap
 0.19% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.17% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
37.73% :     0.005270s :      1: jit_opt_a
 1.09% :     0.000152s :      1: jit_opt_after_cconv
 0.41% :     0.000058s :      1: jit_opt_b
 3.28% :     0.000458s :      1: loop_unroll
 3.48% :     0.000486s :      1: mutable_eliminate
 7.26% :     0.001014s :     26: opt.transform.jit_opt_a
 0.39% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000030s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.40% :     0.000474s :      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% :     0.000040s :      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
 2.77% :     0.000387s :      1: renormalize.infer
 2.07% :     0.000289s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000036s :      1: rewriter_after_opt_a
 0.64% :     0.000090s :      1: rewriter_before_opt_a
 0.56% :     0.000079s :      1: symbol_engine_optimizer
31.66% :     0.004422s :      1: type_inference


TotalTime = 0.0427259, [33]
[bootstrap]: 0.00028453
[type_inference]: 0.0234232
[event_method]: 1.983e-05
[auto_monad]: 0.00015776
[graph_reusing]: 7.2e-06
[pre_auto_parallel]: 3.4e-06
[py_interpret_to_execute]: 0.00012359
[rewriter_before_opt_a]: 9.929e-05
[expand_dump_flag]: 3.75e-06
[jit_opt_a]: 0.0063644, [2]
    [Cycle 1]: 0.00243566, [27]
        [switch_simplify]: 6.187e-05
        [loop_unroll]: 3.612e-05
        [a_1]: 0.00071437
        [with_stream_mark]: 2.023e-05
        [recompute_prepare]: 1.048e-05
        [updatestate_depend_eliminate]: 6.31998e-06
        [updatestate_assign_eliminate]: 5.48002e-06
        [updatestate_loads_eliminate]: 4.68001e-06
        [parameter_eliminate]: 2.33002e-06
        [specialize_transform]: 9.15999e-06
        [updatestate_useless_node_eliminater]: 1.163e-05
        [accelerated_algorithm]: 8.75999e-06
        [meta_shard_fg_expand]: 3.36001e-06
        [get_grad_eliminate_]: 8.36002e-06
        [merge_forward]: 5.56e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.003e-05
        [j_node_and_user_rematch]: 1.459e-05
        [meta_fg_expand]: 3.75e-06
        [replace_old_param]: 1.181e-05
        [inline_without_move]: 8.45001e-06
        [renormalize]: 0.00120339
        [add_forward_monad_depend]: 5.97001e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 2.128e-05
        [cse]: 3.952e-05
        [replace_applicator]: 1.569e-05
    [Cycle 2]: 0.000473, [27]
        [switch_simplify]: 1.02e-05
        [loop_unroll]: 7.96001e-06
        [a_1]: 0.00016771
        [with_stream_mark]: 1.229e-05
        [recompute_prepare]: 8.59e-06
        [updatestate_depend_eliminate]: 5.64e-06
        [updatestate_assign_eliminate]: 4.52998e-06
        [updatestate_loads_eliminate]: 4.18999e-06
        [parameter_eliminate]: 1.42999e-06
        [specialize_transform]: 8.03001e-06
        [updatestate_useless_node_eliminater]: 1.048e-05
        [accelerated_algorithm]: 8.07e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 7.48999e-06
        [merge_forward]: 4.57e-06
        [cell_reuse_recompute_pass]: 2.31e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.718e-05
        [j_node_and_user_rematch]: 1.264e-05
        [meta_fg_expand]: 2.91e-06
        [replace_old_param]: 1.046e-05
        [inline_without_move]: 7.38999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 9.89996e-07
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 1.026e-05
        [cse]: 2.188e-05
        [replace_applicator]: 7.93001e-06
[py_interpret_to_execute_after_opt_a]: 1.246e-05
[rewriter_after_opt_a]: 0.00029434
[convert_after_rewriter]: 1.118e-05
[order_py_execute_after_rewriter]: 7.33e-06
[mutable_eliminate]: 0.00059547
[jit_opt_b]: 6.729e-05, [1]
    [Cycle 1]: 6.063e-05, [2]
        [frontend_op_eliminate]: 2.441e-05
        [inline_after_opt_a]: 2.417e-05
[cconv]: 2.46e-05
[loop_unroll]: 0.00045282
[jit_opt_after_cconv]: 0.00019077, [1]
    [Cycle 1]: 0.00018397, [11]
        [c_1]: 4.375e-05
        [parameter_eliminate]: 2.17001e-06
        [updatestate_depend_eliminate]: 7.11001e-06
        [updatestate_assign_eliminate]: 4.48001e-06
        [updatestate_loads_eliminate]: 3.87998e-06
        [cse]: 2.949e-05
        [call_graph_tuple_transform]: 2.207e-05
        [tuple_list_get_item_eliminator]: 8.23999e-06
        [none_parameter_eliminate]: 1.91e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 8.60001e-06
[remove_dup_value]: 1.77e-05
[partial_unused_args_eliminate]: 2.78998e-06
[environ_conv]: 9.59999e-06
[add_recomputation]: 6.527e-05
[cse_after_recomputation]: 2.984e-05, [1]
    [Cycle 1]: 2.429e-05, [1]
        [cse]: 1.779e-05
[auto_monad_reorder]: 2.653e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 5.67001e-06
[opt_after_jit_grad]: 0.00050029
[symbol_engine_optimizer]: 9.817e-05, [1]
    [Cycle 1]: 9.19e-05, [6]
        [build]: 1.149e-05
        [elim_shapecalc]: 1.133e-05
        [elim_not_effective]: 1.778e-05
        [opt_reshape]: 8.83001e-06
        [fold_const_symbol]: 1.349e-05
        [renormalize]: 4.30009e-07
[validate]: 5.357e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00955717
[execute]: 6.59001e-06
Sums
    bootstrap                            :     0.000285s :  0.74%
    type_inference                       :     0.023423s : 60.61%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000158s :  0.41%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000124s :  0.32%
    rewriter_before_opt_a                :     0.000099s :  0.26%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.19%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000882s :  2.28%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001203s :  3.11%
    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      :     0.000032s :  0.08%
    jit_opt_a.cse                        :     0.000061s :  0.16%
    jit_opt_a.replace_applicator         :     0.000024s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000294s :  0.76%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000595s :  1.54%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000025s :  0.06%
    loop_unroll                          :     0.000453s :  1.17%
    jit_opt_after_cconv.c_1              :     0.000044s :  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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000018s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000010s :  0.02%
    add_recomputation                    :     0.000065s :  0.17%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000027s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000500s :  1.29%
    symbol_engine_optimizer.build        :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000054s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009557s : 24.73%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000279    47
 3.10% :     0.000009s :      2: substitution.depend_value_elim
 1.07% :     0.000003s :      4: substitution.elim_not_effective
 0.75% :     0.000002s :      4: substitution.fold_const_symbol
 2.38% :     0.000007s :      5: substitution.graph_param_transform
75.68% :     0.000211s :      4: substitution.inline
 1.80% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.75% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.69% :     0.000005s :      2: substitution.replace_old_param
 4.57% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 2.89% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.33% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023354     2
93.73% :     0.021890s :      1: type_inference.infer
 6.27% :     0.001464s :      1: type_inference.specialize
------[replace.]   0.000093     6
65.98% :     0.000062s :      4: replace.inline
34.02% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.68% :     0.000209s :      4: match.inline
 5.32% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1021
 1.30% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.11% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     15: predicate.addn_check_dump
 1.38% :     0.000002s :     15: predicate.addn_zero_filter
 2.11% :     0.000004s :     15: predicate.arithmetic_simplify
 1.33% :     0.000002s :     15: predicate.cast_eliminate
 0.47% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.41% :     0.000002s :     15: predicate.compare_switch_simplify
 1.40% :     0.000002s :     15: predicate.depend_value_elim
 1.14% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      5: predicate.elim_not_effective
 0.63% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     15: predicate.environ_get_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.09% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 5.14% :     0.000009s :     31: predicate.inline
 1.14% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :     10: predicate.less_batch_normalization
 1.35% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.05% :     0.000003s :     22: predicate.load_eliminater
 1.30% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.53% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     15: predicate.merge_addn
 1.21% :     0.000002s :     15: predicate.minmaximum_grad
 1.28% :     0.000002s :      5: predicate.mutable_eliminate
 0.60% :     0.000001s :      5: predicate.opt_reshape
 2.41% :     0.000004s :     22: predicate.partial_eliminate
 1.44% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.67% :     0.000003s :     15: predicate.reduce_eliminate
 1.49% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.75% :     0.000003s :     27: predicate.replace_applicator
 0.87% :     0.000001s :     10: predicate.replace_old_param
 0.31% :     0.000001s :      5: predicate.reset_defer_inline
 1.22% :     0.000002s :     15: predicate.reshape_eliminate
 1.23% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     15: predicate.same_eliminate
 0.54% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.13% :     0.000002s :     10: predicate.special_op_eliminate
 1.08% :     0.000002s :     10: predicate.specialize_transform
 1.49% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.30% :     0.000004s :     21: predicate.switch_defer_inline
 1.96% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.64% :     0.000011s :     68: predicate.switch_simplify
 1.26% :     0.000002s :     15: predicate.tile_eliminate
 1.36% :     0.000002s :     15: predicate.transpose_eliminate
 1.54% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.78% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000831    13
44.69% :     0.000371s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.88% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.42% :     0.000419s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045291    76
 0.15% :     0.000068s :      1: add_recomputation
 0.36% :     0.000163s :      1: auto_monad
 0.07% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000298s :      1: bootstrap
 0.06% :     0.000027s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.05% :     0.000024s :      1: event_method
 0.02% :     0.000010s :      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
14.06% :     0.006368s :      1: jit_opt_a
 0.43% :     0.000194s :      1: jit_opt_after_cconv
 0.15% :     0.000070s :      1: jit_opt_b
 1.02% :     0.000460s :      1: loop_unroll
 1.33% :     0.000604s :      1: mutable_eliminate
 2.62% :     0.001185s :     26: opt.transform.jit_opt_a
 0.17% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000048s :      4: opt.transform.symbol_engine_opt
 1.12% :     0.000508s :      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
 0.28% :     0.000127s :      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
 1.65% :     0.000749s :      1: renormalize.infer
 0.98% :     0.000445s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.66% :     0.000299s :      1: rewriter_after_opt_a
 0.23% :     0.000104s :      1: rewriter_before_opt_a
 0.22% :     0.000101s :      1: symbol_engine_optimizer
21.13% :     0.009571s :      1: task_emit
51.74% :     0.023436s :      1: type_inference
 0.17% :     0.000077s :      1: validate


TotalTime = 0.0126976, [30]
[bootstrap]: 0.00029737
[type_inference]: 0.0049434
[event_method]: 2.214e-05
[auto_monad]: 6.521e-05
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 4.802e-05
[rewriter_before_opt_a]: 8.967e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00515525, [2]
    [Cycle 1]: 0.00170571, [27]
        [switch_simplify]: 5.684e-05
        [loop_unroll]: 3.738e-05
        [a_1]: 0.0006507
        [with_stream_mark]: 1.531e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 3.34001e-06
        [updatestate_loads_eliminate]: 3.33e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 7.12002e-06
        [updatestate_useless_node_eliminater]: 6.26e-06
        [accelerated_algorithm]: 6.31e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.99999e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.508e-05
        [j_node_and_user_rematch]: 1.038e-05
        [meta_fg_expand]: 2.57001e-06
        [replace_old_param]: 1.054e-05
        [inline_without_move]: 6.06e-06
        [renormalize]: 0.00061046
        [add_forward_monad_depend]: 5.61e-06
        [auto_monad_grad]: 2.14e-06
        [auto_monad_eliminator]: 1.46e-05
        [cse]: 3.045e-05
        [replace_applicator]: 1.335e-05
    [Cycle 2]: 0.00036768, [27]
        [switch_simplify]: 6.53e-06
        [loop_unroll]: 5.96e-06
        [a_1]: 0.00011541
        [with_stream_mark]: 9.94999e-06
        [recompute_prepare]: 6.42001e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 2.78998e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [parameter_eliminate]: 1.38002e-06
        [specialize_transform]: 6.69001e-06
        [updatestate_useless_node_eliminater]: 5.96e-06
        [accelerated_algorithm]: 6.30002e-06
        [meta_shard_fg_expand]: 1.34998e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 2.80002e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.355e-05
        [j_node_and_user_rematch]: 8.82999e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 9.45001e-06
        [inline_without_move]: 5.61998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 6.40022e-07
        [auto_monad_eliminator]: 6.41e-06
        [cse]: 1.378e-05
        [replace_applicator]: 5.97999e-06
[py_interpret_to_execute_after_opt_a]: 9.26002e-06
[rewriter_after_opt_a]: 3.366e-05
[convert_after_rewriter]: 6.94001e-06
[order_py_execute_after_rewriter]: 5.27999e-06
[mutable_eliminate]: 0.00051957
[jit_opt_b]: 5.525e-05, [1]
    [Cycle 1]: 4.906e-05, [2]
        [frontend_op_eliminate]: 1.847e-05
        [inline_after_opt_a]: 1.745e-05
[cconv]: 2.458e-05
[loop_unroll]: 0.00041785
[jit_opt_after_cconv]: 0.00015187, [1]
    [Cycle 1]: 0.00014572, [11]
        [c_1]: 2.543e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.49e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.53998e-06
        [cse]: 2.157e-05
        [call_graph_tuple_transform]: 2.046e-05
        [tuple_list_get_item_eliminator]: 6.51e-06
        [none_parameter_eliminate]: 1.95001e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 6.29999e-06
[remove_dup_value]: 1.487e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.77999e-06
[add_recomputation]: 4.757e-05
[cse_after_recomputation]: 2.39e-05, [1]
    [Cycle 1]: 1.851e-05, [1]
        [cse]: 1.29e-05
[auto_monad_reorder]: 1.775e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.79e-06
[opt_after_jit_grad]: 0.0004468
[symbol_engine_optimizer]: 7.749e-05, [1]
    [Cycle 1]: 7.118e-05, [6]
        [build]: 3.08998e-06
        [elim_shapecalc]: 8.89e-06
        [elim_not_effective]: 1.403e-05
        [opt_reshape]: 6.44999e-06
        [fold_const_symbol]: 9.71998e-06
        [renormalize]: 4.59986e-07
[validate]: 3.539e-05
Sums
    bootstrap                            :     0.000297s :  3.29%
    type_inference                       :     0.004943s : 54.65%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000065s :  0.72%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000048s :  0.53%
    rewriter_before_opt_a                :     0.000090s :  0.99%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000063s :  0.70%
    jit_opt_a.loop_unroll                :     0.000043s :  0.48%
    jit_opt_a.a_1                        :     0.000766s :  8.47%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.16%
    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 :     0.000006s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000014s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000012s :  0.13%
    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 :     0.000029s :  0.32%
    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          :     0.000020s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000611s :  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      :     0.000021s :  0.23%
    jit_opt_a.cse                        :     0.000044s :  0.49%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000520s :  5.74%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000025s :  0.27%
    loop_unroll                          :     0.000418s :  4.62%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.28%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000022s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  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      :     0.000000s :  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                    :     0.000048s :  0.53%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.20%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000447s :  4.94%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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 :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000035s :  0.39%


Time group info:
------[substitution.]   0.000206    28
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 3.00% :     0.000006s :      4: substitution.graph_param_transform
80.99% :     0.000167s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.48% :     0.000005s :      4: substitution.replace_old_param
 8.23% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004884     2
80.08% :     0.003911s :      1: type_inference.infer
19.92% :     0.000973s :      1: type_inference.specialize
------[replace.]   0.000112     8
42.30% :     0.000047s :      4: replace.inline
57.70% :     0.000065s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     8
91.58% :     0.000164s :      4: match.inline
 8.42% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.34% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.11% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     13: predicate.addn_check_dump
 1.21% :     0.000002s :     13: predicate.addn_zero_filter
 2.12% :     0.000003s :     13: predicate.arithmetic_simplify
 1.19% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     13: predicate.compare_switch_simplify
 1.19% :     0.000002s :     13: predicate.depend_value_elim
 1.13% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.53% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     13: predicate.environ_get_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 1.01% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.55% :     0.000008s :     29: predicate.inline
 0.87% :     0.000001s :      8: predicate.inline_without_move
 0.42% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.12% :     0.000002s :      8: predicate.less_batch_normalization
 1.66% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.92% :     0.000003s :     21: predicate.load_eliminater
 1.28% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.35% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     13: predicate.merge_addn
 1.11% :     0.000002s :     13: predicate.minmaximum_grad
 1.43% :     0.000002s :      4: predicate.mutable_eliminate
 0.46% :     0.000001s :      4: predicate.opt_reshape
 2.62% :     0.000004s :     21: predicate.partial_eliminate
 1.15% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     13: predicate.reduce_eliminate
 1.63% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.02% :     0.000003s :     25: predicate.replace_applicator
 0.67% :     0.000001s :      8: predicate.replace_old_param
 0.37% :     0.000001s :      4: predicate.reset_defer_inline
 1.27% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      8: predicate.special_op_eliminate
 1.07% :     0.000001s :      8: predicate.specialize_transform
 1.41% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.53% :     0.000004s :     21: predicate.switch_defer_inline
 2.23% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.47% :     0.000010s :     67: predicate.switch_simplify
 1.26% :     0.000002s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.48% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.37% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.81% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     13: predicate.value_based_eliminate
 0.35% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000604    12
49.71% :     0.000300s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.86% :     0.000023s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.43% :     0.000280s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014398    72
 0.35% :     0.000050s :      1: add_recomputation
 0.48% :     0.000069s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 2.18% :     0.000314s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.18% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
35.83% :     0.005158s :      1: jit_opt_a
 1.08% :     0.000155s :      1: jit_opt_after_cconv
 0.40% :     0.000058s :      1: jit_opt_b
 2.96% :     0.000426s :      1: loop_unroll
 3.67% :     0.000528s :      1: mutable_eliminate
 7.01% :     0.001009s :     26: opt.transform.jit_opt_a
 0.38% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000030s :      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% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.15% :     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
 0.35% :     0.000051s :      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
 2.21% :     0.000317s :      1: renormalize.infer
 1.98% :     0.000285s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000040s :      1: rewriter_after_opt_a
 0.64% :     0.000093s :      1: rewriter_before_opt_a
 0.56% :     0.000080s :      1: symbol_engine_optimizer
34.41% :     0.004954s :      1: type_inference


TotalTime = 0.0164084, [30]
[bootstrap]: 0.0002702
[type_inference]: 0.00486102
[event_method]: 2.221e-05
[auto_monad]: 6.595e-05
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 3.953e-05
[rewriter_before_opt_a]: 9.03e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.00894541, [2]
    [Cycle 1]: 0.00177442, [27]
        [switch_simplify]: 5.994e-05
        [loop_unroll]: 3.705e-05
        [a_1]: 0.00064447
        [with_stream_mark]: 1.955e-05
        [recompute_prepare]: 8.70001e-06
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 3.00002e-06
        [parameter_eliminate]: 1.81003e-06
        [specialize_transform]: 7.17997e-06
        [updatestate_useless_node_eliminater]: 6.20002e-06
        [accelerated_algorithm]: 6.26e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 6.02999e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.6e-05
        [j_node_and_user_rematch]: 9.76998e-06
        [meta_fg_expand]: 2.71e-06
        [replace_old_param]: 1.054e-05
        [inline_without_move]: 6.19999e-06
        [renormalize]: 0.00067483
        [add_forward_monad_depend]: 5.47001e-06
        [auto_monad_grad]: 2.58998e-06
        [auto_monad_eliminator]: 1.607e-05
        [cse]: 3.251e-05
        [replace_applicator]: 1.358e-05
    [Cycle 2]: 0.00036336, [27]
        [switch_simplify]: 6.74001e-06
        [loop_unroll]: 6.19001e-06
        [a_1]: 0.0001154
        [with_stream_mark]: 9.36998e-06
        [recompute_prepare]: 6.06e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 5.91e-06
        [accelerated_algorithm]: 5.80002e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 3.11999e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.338e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 9.36e-06
        [inline_without_move]: 5.52999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 6.28e-06
        [cse]: 1.386e-05
        [replace_applicator]: 6.07001e-06
[py_interpret_to_execute_after_opt_a]: 1.449e-05
[rewriter_after_opt_a]: 3.647e-05
[convert_after_rewriter]: 7.66001e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00048948
[jit_opt_b]: 5.667e-05, [1]
    [Cycle 1]: 4.976e-05, [2]
        [frontend_op_eliminate]: 1.798e-05
        [inline_after_opt_a]: 1.956e-05
[cconv]: 2.377e-05
[loop_unroll]: 0.0004238
[jit_opt_after_cconv]: 0.00015379, [1]
    [Cycle 1]: 0.00014722, [11]
        [c_1]: 2.553e-05
        [parameter_eliminate]: 2.75997e-06
        [updatestate_depend_eliminate]: 6.19999e-06
        [updatestate_assign_eliminate]: 2.68003e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [cse]: 2.016e-05
        [call_graph_tuple_transform]: 2.108e-05
        [tuple_list_get_item_eliminator]: 6.51e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 6.49001e-06
[remove_dup_value]: 1.648e-05
[partial_unused_args_eliminate]: 2.59999e-06
[environ_conv]: 5.44e-06
[add_recomputation]: 5.075e-05
[cse_after_recomputation]: 2.406e-05, [1]
    [Cycle 1]: 1.859e-05, [1]
        [cse]: 1.272e-05
[auto_monad_reorder]: 1.692e-05
[get_jit_bprop_graph]: 1.49998e-06
[rewriter_after_jit_bprop_graph]: 4.70001e-06
[opt_after_jit_grad]: 0.00045637
[symbol_engine_optimizer]: 7.571e-05, [1]
    [Cycle 1]: 6.955e-05, [6]
        [build]: 3.17002e-06
        [elim_shapecalc]: 8.71002e-06
        [elim_not_effective]: 1.403e-05
        [opt_reshape]: 6.34001e-06
        [fold_const_symbol]: 9.42001e-06
        [renormalize]: 4.39992e-07
[validate]: 5.763e-05
Sums
    bootstrap                            :     0.000270s :  3.00%
    type_inference                       :     0.004861s : 53.91%
    event_method                         :     0.000022s :  0.25%
    auto_monad                           :     0.000066s :  0.73%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.44%
    rewriter_before_opt_a                :     0.000090s :  1.00%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000067s :  0.74%
    jit_opt_a.loop_unroll                :     0.000043s :  0.48%
    jit_opt_a.a_1                        :     0.000760s :  8.43%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000015s :  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 :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    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 :  0.13%
    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 :     0.000029s :  0.33%
    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          :     0.000020s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000675s :  7.49%
    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      :     0.000022s :  0.25%
    jit_opt_a.cse                        :     0.000046s :  0.51%
    jit_opt_a.replace_applicator         :     0.000020s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.16%
    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                    :     0.000489s :  5.43%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.22%
    cconv                                :     0.000024s :  0.26%
    loop_unroll                          :     0.000424s :  4.70%
    jit_opt_after_cconv.c_1              :     0.000026s :  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 :     0.000003s :  0.03%
    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 :     0.000021s :  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  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000051s :  0.56%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000017s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000456s :  5.06%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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 :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000058s :  0.64%


Time group info:
------[substitution.]   0.000220    28
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000006s :      4: substitution.graph_param_transform
82.69% :     0.000182s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000005s :      4: substitution.replace_old_param
 7.40% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004797     2
79.56% :     0.003816s :      1: type_inference.infer
20.44% :     0.000981s :      1: type_inference.specialize
------[replace.]   0.000088     8
58.73% :     0.000052s :      4: replace.inline
41.27% :     0.000036s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     8
92.59% :     0.000179s :      4: match.inline
 7.41% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.27% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     13: predicate.addn_check_dump
 1.25% :     0.000002s :     13: predicate.addn_zero_filter
 2.03% :     0.000003s :     13: predicate.arithmetic_simplify
 1.56% :     0.000002s :     13: predicate.cast_eliminate
 0.43% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     13: predicate.compare_switch_simplify
 1.12% :     0.000002s :     13: predicate.depend_value_elim
 1.15% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.94% :     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
 1.14% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     13: predicate.environ_get_eliminate
 1.22% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000001s :      8: predicate.get_grad_eliminate
 0.30% :     0.000000s :      4: predicate.graph_param_transform
 5.67% :     0.000008s :     29: predicate.inline
 0.96% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      8: predicate.less_batch_normalization
 1.61% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     21: predicate.load_eliminater
 1.22% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.25% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.84% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     13: predicate.merge_addn
 1.09% :     0.000002s :     13: predicate.minmaximum_grad
 1.64% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.53% :     0.000003s :     21: predicate.partial_eliminate
 1.19% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     13: predicate.reduce_eliminate
 1.62% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.93% :     0.000003s :     25: predicate.replace_applicator
 0.82% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.22% :     0.000002s :     13: predicate.reshape_eliminate
 1.17% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      8: predicate.special_op_eliminate
 1.03% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     21: predicate.switch_defer_inline
 2.15% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.94% :     0.000011s :     67: predicate.switch_simplify
 1.21% :     0.000002s :     13: predicate.tile_eliminate
 1.15% :     0.000002s :     13: predicate.transpose_eliminate
 1.50% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.35% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.86% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     13: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000627    12
47.46% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.76% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.78% :     0.000299s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018143    72
 0.30% :     0.000054s :      1: add_recomputation
 0.38% :     0.000070s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 1.56% :     0.000283s :      1: bootstrap
 0.15% :     0.000027s :      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
 0.15% :     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
49.32% :     0.008948s :      1: jit_opt_a
 0.86% :     0.000157s :      1: jit_opt_after_cconv
 0.33% :     0.000060s :      1: jit_opt_b
 2.38% :     0.000431s :      1: loop_unroll
 2.74% :     0.000498s :      1: mutable_eliminate
 5.54% :     0.001004s :     26: opt.transform.jit_opt_a
 0.31% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000031s :      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
 0.13% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.56% :     0.000464s :      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
 0.23% :     0.000042s :      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
 1.89% :     0.000343s :      1: renormalize.infer
 1.79% :     0.000324s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000040s :      1: rewriter_after_opt_a
 0.51% :     0.000093s :      1: rewriter_before_opt_a
 0.43% :     0.000078s :      1: symbol_engine_optimizer
26.95% :     0.004889s :      1: type_inference


TotalTime = 0.0442905, [33]
[bootstrap]: 0.00029537
[type_inference]: 0.023229
[event_method]: 1.993e-05
[auto_monad]: 0.00016089
[graph_reusing]: 6.19001e-06
[pre_auto_parallel]: 2.59999e-06
[py_interpret_to_execute]: 0.000123
[rewriter_before_opt_a]: 0.00013275
[expand_dump_flag]: 4.1e-06
[jit_opt_a]: 0.00671616, [2]
    [Cycle 1]: 0.00253787, [27]
        [switch_simplify]: 6.785e-05
        [loop_unroll]: 3.627e-05
        [a_1]: 0.00081073
        [with_stream_mark]: 2.581e-05
        [recompute_prepare]: 1.04e-05
        [updatestate_depend_eliminate]: 7.26999e-06
        [updatestate_assign_eliminate]: 6.44001e-06
        [updatestate_loads_eliminate]: 5.15999e-06
        [parameter_eliminate]: 2.12001e-06
        [specialize_transform]: 9.70002e-06
        [updatestate_useless_node_eliminater]: 1.191e-05
        [accelerated_algorithm]: 8.50999e-06
        [meta_shard_fg_expand]: 3.11001e-06
        [get_grad_eliminate_]: 7.83001e-06
        [merge_forward]: 6.04999e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.048e-05
        [j_node_and_user_rematch]: 1.384e-05
        [meta_fg_expand]: 3.75e-06
        [replace_old_param]: 1.171e-05
        [inline_without_move]: 8.18999e-06
        [renormalize]: 0.00116736
        [add_forward_monad_depend]: 6.24001e-06
        [auto_monad_grad]: 2.84999e-06
        [auto_monad_eliminator]: 2.252e-05
        [cse]: 4.421e-05
        [replace_applicator]: 1.768e-05
    [Cycle 2]: 0.00046691, [27]
        [switch_simplify]: 9.57001e-06
        [loop_unroll]: 8.1e-06
        [a_1]: 0.00016729
        [with_stream_mark]: 1.46e-05
        [recompute_prepare]: 7.86001e-06
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 4.29997e-06
        [updatestate_loads_eliminate]: 3.77998e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 7.83999e-06
        [updatestate_useless_node_eliminater]: 1.072e-05
        [accelerated_algorithm]: 7.83001e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 7.23999e-06
        [merge_forward]: 4.55001e-06
        [cell_reuse_recompute_pass]: 1.93002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.712e-05
        [j_node_and_user_rematch]: 1.246e-05
        [meta_fg_expand]: 2.99999e-06
        [replace_old_param]: 1.043e-05
        [inline_without_move]: 7.49002e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 9.42999e-06
        [cse]: 1.97e-05
        [replace_applicator]: 7.51999e-06
[py_interpret_to_execute_after_opt_a]: 1.546e-05
[rewriter_after_opt_a]: 0.00028819
[convert_after_rewriter]: 1.154e-05
[order_py_execute_after_rewriter]: 8.07e-06
[mutable_eliminate]: 0.00072842
[jit_opt_b]: 6.981e-05, [1]
    [Cycle 1]: 6.172e-05, [2]
        [frontend_op_eliminate]: 2.461e-05
        [inline_after_opt_a]: 2.513e-05
[cconv]: 3.19e-05
[loop_unroll]: 0.00048723
[jit_opt_after_cconv]: 0.00020451, [1]
    [Cycle 1]: 0.00019721, [11]
        [c_1]: 4.78e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 9.26998e-06
        [updatestate_assign_eliminate]: 4.67e-06
        [updatestate_loads_eliminate]: 4.15e-06
        [cse]: 3.317e-05
        [call_graph_tuple_transform]: 2.308e-05
        [tuple_list_get_item_eliminator]: 8.30999e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 5.49975e-07
        [switch_simplify]: 8.95001e-06
[remove_dup_value]: 1.922e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 1.134e-05
[add_recomputation]: 6.798e-05
[cse_after_recomputation]: 2.832e-05, [1]
    [Cycle 1]: 2.294e-05, [1]
        [cse]: 1.688e-05
[auto_monad_reorder]: 2.606e-05
[get_jit_bprop_graph]: 2.22999e-06
[rewriter_after_jit_bprop_graph]: 6.00002e-06
[opt_after_jit_grad]: 0.00052113
[symbol_engine_optimizer]: 0.00010865, [1]
    [Cycle 1]: 0.00010126, [6]
        [build]: 1.542e-05
        [elim_shapecalc]: 1.293e-05
        [elim_not_effective]: 1.926e-05
        [opt_reshape]: 8.72998e-06
        [fold_const_symbol]: 1.419e-05
        [renormalize]: 8.59989e-07
[validate]: 6.625e-05
[backend_pass]: 1.03001e-06
[task_emit]: 0.0105684
[execute]: 8.34002e-06
Sums
    bootstrap                            :     0.000295s :  0.74%
    type_inference                       :     0.023229s : 58.33%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000161s :  0.40%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000123s :  0.31%
    rewriter_before_opt_a                :     0.000133s :  0.33%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.19%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000978s :  2.46%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  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 :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  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 :     0.000038s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001167s :  2.93%
    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      :     0.000032s :  0.08%
    jit_opt_a.cse                        :     0.000064s :  0.16%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000288s :  0.72%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000728s :  1.83%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000032s :  0.08%
    loop_unroll                          :     0.000487s :  1.22%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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 :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000068s :  0.17%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000521s :  1.31%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000066s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010568s : 26.54%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000311    47
 3.05% :     0.000009s :      2: substitution.depend_value_elim
 0.97% :     0.000003s :      4: substitution.elim_not_effective
 0.72% :     0.000002s :      4: substitution.fold_const_symbol
 2.42% :     0.000008s :      5: substitution.graph_param_transform
76.88% :     0.000239s :      4: substitution.inline
 1.69% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.33% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.50% :     0.000005s :      2: substitution.replace_old_param
 4.26% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.13% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 3.05% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023157     2
93.86% :     0.021735s :      1: type_inference.infer
 6.14% :     0.001423s :      1: type_inference.specialize
------[replace.]   0.000099     6
67.16% :     0.000066s :      4: replace.inline
32.84% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000248     6
95.12% :     0.000236s :      4: match.inline
 4.88% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000174  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     15: predicate.addn_check_dump
 1.31% :     0.000002s :     15: predicate.addn_zero_filter
 2.17% :     0.000004s :     15: predicate.arithmetic_simplify
 1.37% :     0.000002s :     15: predicate.cast_eliminate
 0.61% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     15: predicate.compare_switch_simplify
 1.38% :     0.000002s :     15: predicate.depend_value_elim
 1.15% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      5: predicate.elim_not_effective
 0.61% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.09% :     0.000002s :     10: predicate.get_grad_eliminate
 0.26% :     0.000000s :      5: predicate.graph_param_transform
 5.59% :     0.000010s :     31: predicate.inline
 1.01% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :     10: predicate.less_batch_normalization
 1.39% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.86% :     0.000003s :     22: predicate.load_eliminater
 1.73% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.32% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.86% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000002s :     15: predicate.merge_addn
 1.20% :     0.000002s :     15: predicate.minmaximum_grad
 1.92% :     0.000003s :      5: predicate.mutable_eliminate
 0.54% :     0.000001s :      5: predicate.opt_reshape
 2.34% :     0.000004s :     22: predicate.partial_eliminate
 1.29% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.79% :     0.000003s :     15: predicate.reduce_eliminate
 1.47% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.78% :     0.000003s :     27: predicate.replace_applicator
 0.71% :     0.000001s :     10: predicate.replace_old_param
 0.35% :     0.000001s :      5: predicate.reset_defer_inline
 1.16% :     0.000002s :     15: predicate.reshape_eliminate
 1.30% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     15: predicate.same_eliminate
 0.54% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :     10: predicate.special_op_eliminate
 1.04% :     0.000002s :     10: predicate.specialize_transform
 1.44% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.12% :     0.000004s :     21: predicate.switch_defer_inline
 1.83% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.81% :     0.000012s :     68: predicate.switch_simplify
 1.20% :     0.000002s :     15: predicate.tile_eliminate
 1.19% :     0.000002s :     15: predicate.transpose_eliminate
 1.55% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.09% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 2.06% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.15% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000003s :     15: predicate.value_based_eliminate
 0.45% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000829    13
41.59% :     0.000345s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.72% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.69% :     0.000420s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046861    76
 0.15% :     0.000071s :      1: add_recomputation
 0.36% :     0.000167s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000314s :      1: bootstrap
 0.07% :     0.000035s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000024s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
14.34% :     0.006720s :      1: jit_opt_a
 0.44% :     0.000207s :      1: jit_opt_after_cconv
 0.16% :     0.000073s :      1: jit_opt_b
 1.06% :     0.000497s :      1: loop_unroll
 1.58% :     0.000739s :      1: mutable_eliminate
 2.74% :     0.001286s :     26: opt.transform.jit_opt_a
 0.18% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.13% :     0.000530s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000127s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.49% :     0.000698s :      1: renormalize.infer
 0.98% :     0.000460s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.62% :     0.000292s :      1: rewriter_after_opt_a
 0.30% :     0.000139s :      1: rewriter_before_opt_a
 0.24% :     0.000112s :      1: symbol_engine_optimizer
22.60% :     0.010588s :      1: task_emit
49.60% :     0.023244s :      1: type_inference
 0.20% :     0.000094s :      1: validate


TotalTime = 0.0114605, [30]
[bootstrap]: 0.00028812
[type_inference]: 0.00429072
[event_method]: 1.959e-05
[auto_monad]: 3.985e-05
[graph_reusing]: 3.48e-06
[pre_auto_parallel]: 1.69e-06
[py_interpret_to_execute]: 3.415e-05
[rewriter_before_opt_a]: 7.982e-05
[expand_dump_flag]: 1.89e-06
[jit_opt_a]: 0.00481614, [2]
    [Cycle 1]: 0.00154818, [27]
        [switch_simplify]: 5.74e-05
        [loop_unroll]: 3.687e-05
        [a_1]: 0.00058847
        [with_stream_mark]: 9.46998e-06
        [recompute_prepare]: 7.62002e-06
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 6.86999e-06
        [updatestate_useless_node_eliminater]: 6.04001e-06
        [accelerated_algorithm]: 6.41998e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 7.77e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.587e-05
        [j_node_and_user_rematch]: 9.36e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 9.77001e-06
        [inline_without_move]: 6.14001e-06
        [renormalize]: 0.00054695
        [add_forward_monad_depend]: 3.93999e-06
        [auto_monad_grad]: 1.10999e-06
        [auto_monad_eliminator]: 9.00999e-06
        [cse]: 1.865e-05
        [replace_applicator]: 1.065e-05
    [Cycle 2]: 0.00035099, [27]
        [switch_simplify]: 6.93e-06
        [loop_unroll]: 6.16e-06
        [a_1]: 0.00011463
        [with_stream_mark]: 7.3e-06
        [recompute_prepare]: 5.67999e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 6.06e-06
        [updatestate_useless_node_eliminater]: 5.81e-06
        [accelerated_algorithm]: 5.92999e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 5.38002e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.222e-05
        [j_node_and_user_rematch]: 8.50001e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.65001e-06
        [inline_without_move]: 5.71e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.64998e-06
        [cse]: 1.347e-05
        [replace_applicator]: 6.04999e-06
[py_interpret_to_execute_after_opt_a]: 8.48999e-06
[rewriter_after_opt_a]: 2.791e-05
[convert_after_rewriter]: 5.46998e-06
[order_py_execute_after_rewriter]: 5.07e-06
[mutable_eliminate]: 0.000444
[jit_opt_b]: 5.325e-05, [1]
    [Cycle 1]: 4.723e-05, [2]
        [frontend_op_eliminate]: 1.806e-05
        [inline_after_opt_a]: 1.817e-05
[cconv]: 1.398e-05
[loop_unroll]: 0.00039176
[jit_opt_after_cconv]: 0.00013814, [1]
    [Cycle 1]: 0.00013268, [11]
        [c_1]: 2.51e-05
        [parameter_eliminate]: 1.77001e-06
        [updatestate_depend_eliminate]: 4.63999e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 1.671e-05
        [call_graph_tuple_transform]: 1.729e-05
        [tuple_list_get_item_eliminator]: 6.01e-06
        [none_parameter_eliminate]: 7.59988e-07
        [renormalize]: 3.30008e-07
        [switch_simplify]: 6.36e-06
[remove_dup_value]: 7.56999e-06
[partial_unused_args_eliminate]: 1.48002e-06
[environ_conv]: 4.16001e-06
[add_recomputation]: 3.005e-05
[cse_after_recomputation]: 2.206e-05, [1]
    [Cycle 1]: 1.728e-05, [1]
        [cse]: 1.175e-05
[auto_monad_reorder]: 1.151e-05
[get_jit_bprop_graph]: 1.28002e-06
[rewriter_after_jit_bprop_graph]: 4.23001e-06
[opt_after_jit_grad]: 0.00047702
[symbol_engine_optimizer]: 7.2e-05, [1]
    [Cycle 1]: 6.629e-05, [6]
        [build]: 2.41e-06
        [elim_shapecalc]: 8.03001e-06
        [elim_not_effective]: 1.217e-05
        [opt_reshape]: 6.42001e-06
        [fold_const_symbol]: 9.44e-06
        [renormalize]: 3.9002e-07
[validate]: 2.632e-05
Sums
    bootstrap                            :     0.000288s :  3.60%
    type_inference                       :     0.004291s : 53.59%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000040s :  0.50%
    graph_reusing                        :     0.000003s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.43%
    rewriter_before_opt_a                :     0.000080s :  1.00%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.80%
    jit_opt_a.loop_unroll                :     0.000043s :  0.54%
    jit_opt_a.a_1                        :     0.000703s :  8.78%
    jit_opt_a.with_stream_mark           :     0.000017s :  0.21%
    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 :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.15%
    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_        :     0.000013s :  0.16%
    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 :     0.000028s :  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 :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.15%
    jit_opt_a.renormalize                :     0.000547s :  6.83%
    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 :  0.18%
    jit_opt_a.cse                        :     0.000032s :  0.40%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000028s :  0.35%
    convert_after_rewriter               :     0.000005s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000444s :  5.55%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.23%
    cconv                                :     0.000014s :  0.17%
    loop_unroll                          :     0.000392s :  4.89%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.31%
    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 :     0.000003s :  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 :     0.000017s :  0.22%
    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 :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000008s :  0.09%
    partial_unused_args_eliminate        :     0.000001s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000030s :  0.38%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000012s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000477s :  5.96%
    symbol_engine_optimizer.build        :     0.000002s :  0.03%
    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 :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.33%


Time group info:
------[substitution.]   0.000169    28
 0.78% :     0.000001s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 1.99% :     0.000003s :      4: substitution.graph_param_transform
83.83% :     0.000142s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.40% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000003s :      4: substitution.replace_old_param
 6.87% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004244     2
78.95% :     0.003351s :      1: type_inference.infer
21.05% :     0.000893s :      1: type_inference.specialize
------[replace.]   0.000082     8
56.60% :     0.000046s :      4: replace.inline
43.40% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000149     8
93.47% :     0.000139s :      4: match.inline
 6.53% :     0.000010s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   912
 1.32% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.11% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     13: predicate.addn_check_dump
 1.22% :     0.000002s :     13: predicate.addn_zero_filter
 1.90% :     0.000003s :     13: predicate.arithmetic_simplify
 1.36% :     0.000002s :     13: predicate.cast_eliminate
 0.41% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     13: predicate.compare_switch_simplify
 1.21% :     0.000002s :     13: predicate.depend_value_elim
 1.13% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      4: predicate.elim_not_effective
 0.54% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     13: predicate.environ_get_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 1.31% :     0.000002s :      8: predicate.get_grad_eliminate
 0.24% :     0.000000s :      4: predicate.graph_param_transform
 5.05% :     0.000007s :     29: predicate.inline
 0.96% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      8: predicate.less_batch_normalization
 1.76% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.11% :     0.000003s :     21: predicate.load_eliminater
 1.25% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.40% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.02% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     13: predicate.merge_addn
 1.11% :     0.000002s :     13: predicate.minmaximum_grad
 1.33% :     0.000002s :      4: predicate.mutable_eliminate
 0.49% :     0.000001s :      4: predicate.opt_reshape
 2.53% :     0.000003s :     21: predicate.partial_eliminate
 1.19% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     13: predicate.reduce_eliminate
 1.59% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.95% :     0.000003s :     25: predicate.replace_applicator
 0.64% :     0.000001s :      8: predicate.replace_old_param
 0.38% :     0.000001s :      4: predicate.reset_defer_inline
 1.21% :     0.000002s :     13: predicate.reshape_eliminate
 1.25% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     13: predicate.same_eliminate
 0.60% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.53% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.63% :     0.000004s :     21: predicate.switch_defer_inline
 2.17% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.90% :     0.000011s :     67: predicate.switch_simplify
 1.24% :     0.000002s :     13: predicate.tile_eliminate
 1.29% :     0.000002s :     13: predicate.transpose_eliminate
 1.47% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     13: 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.]   0.000550    12
42.96% :     0.000236s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.07% :     0.000022s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.97% :     0.000291s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013036    72
 0.25% :     0.000033s :      1: add_recomputation
 0.33% :     0.000043s :      1: auto_monad
 0.11% :     0.000014s :      1: auto_monad_reorder
 2.28% :     0.000298s :      1: bootstrap
 0.13% :     0.000017s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.17% :     0.000023s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
36.97% :     0.004819s :      1: jit_opt_a
 1.08% :     0.000141s :      1: jit_opt_after_cconv
 0.43% :     0.000056s :      1: jit_opt_b
 3.05% :     0.000398s :      1: loop_unroll
 3.46% :     0.000451s :      1: mutable_eliminate
 7.23% :     0.000942s :     26: opt.transform.jit_opt_a
 0.40% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000031s :      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
 0.18% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.71% :     0.000484s :      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% :     0.000037s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000010s :      1: remove_dup_value
 2.18% :     0.000284s :      1: renormalize.infer
 1.97% :     0.000257s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000031s :      1: rewriter_after_opt_a
 0.63% :     0.000083s :      1: rewriter_before_opt_a
 0.57% :     0.000075s :      1: symbol_engine_optimizer
32.98% :     0.004299s :      1: type_inference


TotalTime = 0.0121005, [30]
[bootstrap]: 0.00023974
[type_inference]: 0.00451101
[event_method]: 2.268e-05
[auto_monad]: 6.484e-05
[graph_reusing]: 4.4e-06
[pre_auto_parallel]: 1.99999e-06
[py_interpret_to_execute]: 3.924e-05
[rewriter_before_opt_a]: 8.569e-05
[expand_dump_flag]: 2.24999e-06
[jit_opt_a]: 0.00510854, [2]
    [Cycle 1]: 0.00166581, [27]
        [switch_simplify]: 5.355e-05
        [loop_unroll]: 3.609e-05
        [a_1]: 0.00060812
        [with_stream_mark]: 1.415e-05
        [recompute_prepare]: 8.73001e-06
        [updatestate_depend_eliminate]: 3.49001e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 6.87002e-06
        [updatestate_useless_node_eliminater]: 6.62002e-06
        [accelerated_algorithm]: 6.19001e-06
        [meta_shard_fg_expand]: 1.67999e-06
        [get_grad_eliminate_]: 5.94999e-06
        [merge_forward]: 2.97002e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.228e-05
        [j_node_and_user_rematch]: 9.71998e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 1.066e-05
        [inline_without_move]: 6.12999e-06
        [renormalize]: 0.00063081
        [add_forward_monad_depend]: 4.65999e-06
        [auto_monad_grad]: 1.86003e-06
        [auto_monad_eliminator]: 1.16e-05
        [cse]: 2.384e-05
        [replace_applicator]: 1.183e-05
    [Cycle 2]: 0.00035682, [27]
        [switch_simplify]: 6.84999e-06
        [loop_unroll]: 6.01e-06
        [a_1]: 0.0001162
        [with_stream_mark]: 9.04e-06
        [recompute_prepare]: 5.91e-06
        [updatestate_depend_eliminate]: 3.45003e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 5.56998e-06
        [updatestate_useless_node_eliminater]: 5.87999e-06
        [accelerated_algorithm]: 6.74001e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 5.42001e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.292e-05
        [j_node_and_user_rematch]: 8.60999e-06
        [meta_fg_expand]: 1.87999e-06
        [replace_old_param]: 9.17999e-06
        [inline_without_move]: 5.51998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.86e-06
        [cse]: 1.322e-05
        [replace_applicator]: 5.84999e-06
[py_interpret_to_execute_after_opt_a]: 1.004e-05
[rewriter_after_opt_a]: 2.963e-05
[convert_after_rewriter]: 5.85002e-06
[order_py_execute_after_rewriter]: 4.35999e-06
[mutable_eliminate]: 0.00049461
[jit_opt_b]: 5.449e-05, [1]
    [Cycle 1]: 4.846e-05, [2]
        [frontend_op_eliminate]: 1.901e-05
        [inline_after_opt_a]: 1.798e-05
[cconv]: 1.745e-05
[loop_unroll]: 0.00041915
[jit_opt_after_cconv]: 0.00015005, [1]
    [Cycle 1]: 0.00014371, [11]
        [c_1]: 2.609e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 5.59998e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [cse]: 1.982e-05
        [call_graph_tuple_transform]: 1.969e-05
        [tuple_list_get_item_eliminator]: 6.31998e-06
        [none_parameter_eliminate]: 9.39996e-07
        [renormalize]: 3.69997e-07
        [switch_simplify]: 6.61e-06
[remove_dup_value]: 1.038e-05
[partial_unused_args_eliminate]: 1.74e-06
[environ_conv]: 4.52e-06
[add_recomputation]: 3.668e-05
[cse_after_recomputation]: 2.262e-05, [1]
    [Cycle 1]: 1.792e-05, [1]
        [cse]: 1.239e-05
[auto_monad_reorder]: 1.284e-05
[get_jit_bprop_graph]: 9.60001e-06
[rewriter_after_jit_bprop_graph]: 4.96002e-06
[opt_after_jit_grad]: 0.00046602
[symbol_engine_optimizer]: 7.566e-05, [1]
    [Cycle 1]: 6.987e-05, [6]
        [build]: 2.97002e-06
        [elim_shapecalc]: 8.78001e-06
        [elim_not_effective]: 1.392e-05
        [opt_reshape]: 6.59999e-06
        [fold_const_symbol]: 9.43002e-06
        [renormalize]: 3.80009e-07
[validate]: 3.281e-05
Sums
    bootstrap                            :     0.000240s :  2.84%
    type_inference                       :     0.004511s : 53.38%
    event_method                         :     0.000023s :  0.27%
    auto_monad                           :     0.000065s :  0.77%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.46%
    rewriter_before_opt_a                :     0.000086s :  1.01%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000060s :  0.71%
    jit_opt_a.loop_unroll                :     0.000042s :  0.50%
    jit_opt_a.a_1                        :     0.000724s :  8.57%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.27%
    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 :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.15%
    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              :     0.000006s :  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 :  0.30%
    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          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000631s :  7.47%
    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      :     0.000017s :  0.21%
    jit_opt_a.cse                        :     0.000037s :  0.44%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000030s :  0.35%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000495s :  5.85%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.21%
    cconv                                :     0.000017s :  0.21%
    loop_unroll                          :     0.000419s :  4.96%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.31%
    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              :     0.000020s :  0.23%
    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 :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.08%
    remove_dup_value                     :     0.000010s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000037s :  0.43%
    cse_after_recomputation.cse          :     0.000012s :  0.15%
    auto_monad_reorder                   :     0.000013s :  0.15%
    get_jit_bprop_graph                  :     0.000010s :  0.11%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000466s :  5.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  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  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.39%


Time group info:
------[substitution.]   0.000186    28
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000005s :      4: substitution.graph_param_transform
81.98% :     0.000153s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.45% :     0.000005s :      4: substitution.replace_old_param
 7.54% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004455     2
78.27% :     0.003487s :      1: type_inference.infer
21.73% :     0.000968s :      1: type_inference.specialize
------[replace.]   0.000083     8
55.25% :     0.000046s :      4: replace.inline
44.75% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     8
92.50% :     0.000150s :      4: match.inline
 7.50% :     0.000012s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   912
 1.26% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     13: predicate.addn_check_dump
 1.34% :     0.000002s :     13: predicate.addn_zero_filter
 2.02% :     0.000003s :     13: predicate.arithmetic_simplify
 1.32% :     0.000002s :     13: predicate.cast_eliminate
 0.53% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.19% :     0.000002s :     13: predicate.compare_switch_simplify
 1.17% :     0.000002s :     13: predicate.depend_value_elim
 1.14% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.54% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     13: predicate.environ_get_eliminate
 1.18% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      4: predicate.fold_const_symbol
 1.03% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.42% :     0.000008s :     29: predicate.inline
 0.96% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      8: predicate.less_batch_normalization
 1.60% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     21: predicate.load_eliminater
 1.36% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.20% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     13: predicate.merge_addn
 1.14% :     0.000002s :     13: predicate.minmaximum_grad
 1.42% :     0.000002s :      4: predicate.mutable_eliminate
 0.51% :     0.000001s :      4: predicate.opt_reshape
 2.53% :     0.000004s :     21: predicate.partial_eliminate
 1.19% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.89% :     0.000003s :     13: predicate.reduce_eliminate
 1.56% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.00% :     0.000003s :     25: predicate.replace_applicator
 0.65% :     0.000001s :      8: predicate.replace_old_param
 0.32% :     0.000000s :      4: predicate.reset_defer_inline
 1.13% :     0.000002s :     13: predicate.reshape_eliminate
 1.14% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     13: predicate.same_eliminate
 0.64% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      8: predicate.special_op_eliminate
 0.97% :     0.000001s :      8: predicate.specialize_transform
 1.32% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.63% :     0.000004s :     21: predicate.switch_defer_inline
 2.22% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.94% :     0.000011s :     67: predicate.switch_simplify
 1.17% :     0.000002s :     13: predicate.tile_eliminate
 1.23% :     0.000002s :     13: predicate.transpose_eliminate
 1.56% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000584    12
44.94% :     0.000262s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.93% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.13% :     0.000293s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013777    72
 0.28% :     0.000039s :      1: add_recomputation
 0.50% :     0.000069s :      1: auto_monad
 0.11% :     0.000015s :      1: auto_monad_reorder
 1.82% :     0.000251s :      1: bootstrap
 0.15% :     0.000020s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.20% :     0.000027s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.09% :     0.000012s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
37.10% :     0.005111s :      1: jit_opt_a
 1.11% :     0.000153s :      1: jit_opt_after_cconv
 0.41% :     0.000057s :      1: jit_opt_b
 3.10% :     0.000427s :      1: loop_unroll
 3.65% :     0.000503s :      1: mutable_eliminate
 6.96% :     0.000959s :     26: opt.transform.jit_opt_a
 0.40% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000031s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.44% :     0.000474s :      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
 0.31% :     0.000042s :      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
 2.50% :     0.000344s :      1: renormalize.infer
 2.03% :     0.000280s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000032s :      1: rewriter_after_opt_a
 0.64% :     0.000089s :      1: rewriter_before_opt_a
 0.57% :     0.000078s :      1: symbol_engine_optimizer
32.84% :     0.004524s :      1: type_inference


TotalTime = 0.0131568, [30]
[bootstrap]: 0.00032214
[type_inference]: 0.00487619
[event_method]: 2.188e-05
[auto_monad]: 6.701e-05
[graph_reusing]: 5.21998e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 4.194e-05
[rewriter_before_opt_a]: 8.456e-05
[expand_dump_flag]: 1.99e-06
[jit_opt_a]: 0.00554857, [2]
    [Cycle 1]: 0.00179428, [27]
        [switch_simplify]: 5.98e-05
        [loop_unroll]: 3.666e-05
        [a_1]: 0.00065724
        [with_stream_mark]: 1.396e-05
        [recompute_prepare]: 9.10999e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 2.75002e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 6.98998e-06
        [updatestate_useless_node_eliminater]: 6.28002e-06
        [accelerated_algorithm]: 6.42001e-06
        [meta_shard_fg_expand]: 1.58002e-06
        [get_grad_eliminate_]: 6.07999e-06
        [merge_forward]: 2.87002e-06
        [cell_reuse_recompute_pass]: 8.10018e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.225e-05
        [j_node_and_user_rematch]: 1.01e-05
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 1.044e-05
        [inline_without_move]: 6.12001e-06
        [renormalize]: 0.00070828
        [add_forward_monad_depend]: 4.88001e-06
        [auto_monad_grad]: 1.63002e-06
        [auto_monad_eliminator]: 1.185e-05
        [cse]: 2.244e-05
        [replace_applicator]: 1.365e-05
    [Cycle 2]: 0.00036712, [27]
        [switch_simplify]: 7.26001e-06
        [loop_unroll]: 6.11e-06
        [a_1]: 0.00011866
        [with_stream_mark]: 9.26998e-06
        [recompute_prepare]: 6.22001e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 6.09999e-06
        [updatestate_useless_node_eliminater]: 5.87999e-06
        [accelerated_algorithm]: 6.17999e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 5.55001e-06
        [merge_forward]: 3.06999e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.309e-05
        [j_node_and_user_rematch]: 9.00999e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 9.49999e-06
        [inline_without_move]: 5.46998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 6.05002e-06
        [cse]: 1.379e-05
        [replace_applicator]: 6.02999e-06
[py_interpret_to_execute_after_opt_a]: 1.071e-05
[rewriter_after_opt_a]: 3.068e-05
[convert_after_rewriter]: 5.74e-06
[order_py_execute_after_rewriter]: 4.31002e-06
[mutable_eliminate]: 0.00058286
[jit_opt_b]: 5.868e-05, [1]
    [Cycle 1]: 5.101e-05, [2]
        [frontend_op_eliminate]: 2.092e-05
        [inline_after_opt_a]: 1.82e-05
[cconv]: 1.739e-05
[loop_unroll]: 0.00045006
[jit_opt_after_cconv]: 0.00016411, [1]
    [Cycle 1]: 0.00015703, [11]
        [c_1]: 2.608e-05
        [parameter_eliminate]: 2.89001e-06
        [updatestate_depend_eliminate]: 1.172e-05
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [cse]: 2.38e-05
        [call_graph_tuple_transform]: 2.04e-05
        [tuple_list_get_item_eliminator]: 6.69999e-06
        [none_parameter_eliminate]: 1.22e-06
        [renormalize]: 3.7998e-07
        [switch_simplify]: 6.69001e-06
[remove_dup_value]: 9.76e-06
[partial_unused_args_eliminate]: 1.94999e-06
[environ_conv]: 4.90001e-06
[add_recomputation]: 3.517e-05
[cse_after_recomputation]: 2.485e-05, [1]
    [Cycle 1]: 1.932e-05, [1]
        [cse]: 1.33e-05
[auto_monad_reorder]: 1.31e-05
[get_jit_bprop_graph]: 1.52999e-06
[rewriter_after_jit_bprop_graph]: 5.64e-06
[opt_after_jit_grad]: 0.00048454
[symbol_engine_optimizer]: 7.95e-05, [1]
    [Cycle 1]: 7.343e-05, [6]
        [build]: 2.96001e-06
        [elim_shapecalc]: 9.33002e-06
        [elim_not_effective]: 1.386e-05
        [opt_reshape]: 7.6e-06
        [fold_const_symbol]: 1.05e-05
        [renormalize]: 4.89992e-07
[validate]: 3.299e-05
Sums
    bootstrap                            :     0.000322s :  3.50%
    type_inference                       :     0.004876s : 53.05%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000067s :  0.73%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000042s :  0.46%
    rewriter_before_opt_a                :     0.000085s :  0.92%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.73%
    jit_opt_a.loop_unroll                :     0.000043s :  0.47%
    jit_opt_a.a_1                        :     0.000776s :  8.44%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.25%
    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        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.13%
    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 :  0.28%
    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          :     0.000020s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000708s :  7.71%
    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                        :     0.000036s :  0.39%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000031s :  0.33%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000583s :  6.34%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000017s :  0.19%
    loop_unroll                          :     0.000450s :  4.90%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.13%
    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 :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  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  :     0.000007s :  0.07%
    remove_dup_value                     :     0.000010s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000035s :  0.38%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000013s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000485s :  5.27%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000033s :  0.36%


Time group info:
------[substitution.]   0.000218    28
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.26% :     0.000005s :      4: substitution.graph_param_transform
82.92% :     0.000180s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.79% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      4: substitution.replace_old_param
 8.25% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004813     2
79.85% :     0.003844s :      1: type_inference.infer
20.15% :     0.000970s :      1: type_inference.specialize
------[replace.]   0.000090     8
56.58% :     0.000051s :      4: replace.inline
43.42% :     0.000039s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     8
91.84% :     0.000177s :      4: match.inline
 8.16% :     0.000016s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000145   912
 1.29% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.10% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     13: predicate.addn_check_dump
 1.48% :     0.000002s :     13: predicate.addn_zero_filter
 1.96% :     0.000003s :     13: predicate.arithmetic_simplify
 1.28% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.03% :     0.000002s :     13: predicate.compare_switch_simplify
 1.18% :     0.000002s :     13: predicate.depend_value_elim
 1.25% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      4: predicate.elim_not_effective
 0.55% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     13: predicate.environ_get_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.88% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 5.03% :     0.000007s :     29: predicate.inline
 0.88% :     0.000001s :      8: predicate.inline_without_move
 0.41% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.16% :     0.000002s :      8: predicate.less_batch_normalization
 1.67% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.03% :     0.000003s :     21: predicate.load_eliminater
 1.53% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.15% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     13: predicate.merge_addn
 1.03% :     0.000001s :     13: predicate.minmaximum_grad
 1.63% :     0.000002s :      4: predicate.mutable_eliminate
 0.66% :     0.000001s :      4: predicate.opt_reshape
 2.41% :     0.000003s :     21: predicate.partial_eliminate
 1.15% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.78% :     0.000003s :     13: predicate.reduce_eliminate
 1.84% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.94% :     0.000003s :     25: predicate.replace_applicator
 0.63% :     0.000001s :      8: predicate.replace_old_param
 0.32% :     0.000000s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.18% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     13: predicate.same_eliminate
 0.59% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      8: predicate.special_op_eliminate
 0.93% :     0.000001s :      8: predicate.specialize_transform
 1.49% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.57% :     0.000004s :     21: predicate.switch_defer_inline
 2.04% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.36% :     0.000011s :     67: predicate.switch_simplify
 1.18% :     0.000002s :     13: predicate.tile_eliminate
 1.21% :     0.000002s :     13: predicate.transpose_eliminate
 1.59% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.48% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     13: predicate.value_based_eliminate
 0.34% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000623    12
45.08% :     0.000281s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.59% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.34% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014977    72
 0.25% :     0.000038s :      1: add_recomputation
 0.47% :     0.000070s :      1: auto_monad
 0.10% :     0.000016s :      1: auto_monad_reorder
 2.25% :     0.000337s :      1: bootstrap
 0.13% :     0.000020s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000027s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.17% :     0.000026s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.07% :     0.005552s :      1: jit_opt_a
 1.12% :     0.000167s :      1: jit_opt_after_cconv
 0.41% :     0.000062s :      1: jit_opt_b
 3.06% :     0.000458s :      1: loop_unroll
 3.96% :     0.000592s :      1: mutable_eliminate
 6.82% :     0.001021s :     26: opt.transform.jit_opt_a
 0.37% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000033s :      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
 0.16% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.29% :     0.000492s :      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% :     0.000045s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000012s :      1: remove_dup_value
 2.47% :     0.000369s :      1: renormalize.infer
 2.22% :     0.000332s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000034s :      1: rewriter_after_opt_a
 0.59% :     0.000088s :      1: rewriter_before_opt_a
 0.55% :     0.000082s :      1: symbol_engine_optimizer
32.64% :     0.004889s :      1: type_inference


TotalTime = 0.0129815, [30]
[bootstrap]: 0.00025292
[type_inference]: 0.00456479
[event_method]: 2.119e-05
[auto_monad]: 6.481e-05
[graph_reusing]: 5.04e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 3.975e-05
[rewriter_before_opt_a]: 9.783e-05
[expand_dump_flag]: 3.56999e-06
[jit_opt_a]: 0.00563005, [2]
    [Cycle 1]: 0.00195738, [27]
        [switch_simplify]: 6.106e-05
        [loop_unroll]: 3.708e-05
        [a_1]: 0.0006784
        [with_stream_mark]: 1.831e-05
        [recompute_prepare]: 7.71999e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 3.50003e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 1.76998e-06
        [specialize_transform]: 7.13998e-06
        [updatestate_useless_node_eliminater]: 6.29001e-06
        [accelerated_algorithm]: 6.44001e-06
        [meta_shard_fg_expand]: 2.74001e-06
        [get_grad_eliminate_]: 5.92001e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.598e-05
        [j_node_and_user_rematch]: 1.757e-05
        [meta_fg_expand]: 2.78e-06
        [replace_old_param]: 1.168e-05
        [inline_without_move]: 6.04999e-06
        [renormalize]: 0.0008222
        [add_forward_monad_depend]: 6.41e-06
        [auto_monad_grad]: 2.36e-06
        [auto_monad_eliminator]: 1.459e-05
        [cse]: 2.546e-05
        [replace_applicator]: 1.255e-05
    [Cycle 2]: 0.00036447, [27]
        [switch_simplify]: 6.76e-06
        [loop_unroll]: 6.01e-06
        [a_1]: 0.00011633
        [with_stream_mark]: 9.99999e-06
        [recompute_prepare]: 5.59998e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 5.72999e-06
        [updatestate_useless_node_eliminater]: 6.04001e-06
        [accelerated_algorithm]: 6.06998e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 8.68001e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 9.90002e-06
        [inline_without_move]: 5.67999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.16002e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 6.71e-06
        [cse]: 1.363e-05
        [replace_applicator]: 5.94e-06
[py_interpret_to_execute_after_opt_a]: 1.191e-05
[rewriter_after_opt_a]: 3.605e-05
[convert_after_rewriter]: 7.51001e-06
[order_py_execute_after_rewriter]: 5.60001e-06
[mutable_eliminate]: 0.00056954
[jit_opt_b]: 5.951e-05, [1]
    [Cycle 1]: 5.19e-05, [2]
        [frontend_op_eliminate]: 2.044e-05
        [inline_after_opt_a]: 1.955e-05
[cconv]: 2.517e-05
[loop_unroll]: 0.00050276
[jit_opt_after_cconv]: 0.00016153, [1]
    [Cycle 1]: 0.0001548, [11]
        [c_1]: 2.554e-05
        [parameter_eliminate]: 2.83e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.72001e-06
        [cse]: 2.262e-05
        [call_graph_tuple_transform]: 2.203e-05
        [tuple_list_get_item_eliminator]: 6.51999e-06
        [none_parameter_eliminate]: 1.82999e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 6.65998e-06
[remove_dup_value]: 1.626e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 6.21998e-06
[add_recomputation]: 4.742e-05
[cse_after_recomputation]: 2.32e-05, [1]
    [Cycle 1]: 1.764e-05, [1]
        [cse]: 1.162e-05
[auto_monad_reorder]: 1.883e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 5.65001e-06
[opt_after_jit_grad]: 0.00051403
[symbol_engine_optimizer]: 7.849e-05, [1]
    [Cycle 1]: 7.219e-05, [6]
        [build]: 3.29001e-06
        [elim_shapecalc]: 8.92e-06
        [elim_not_effective]: 1.429e-05
        [opt_reshape]: 6.41e-06
        [fold_const_symbol]: 9.62999e-06
        [renormalize]: 7.89994e-07
[validate]: 3.877e-05
Sums
    bootstrap                            :     0.000253s :  2.78%
    type_inference                       :     0.004565s : 50.19%
    event_method                         :     0.000021s :  0.23%
    auto_monad                           :     0.000065s :  0.71%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.44%
    rewriter_before_opt_a                :     0.000098s :  1.08%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000068s :  0.75%
    jit_opt_a.loop_unroll                :     0.000043s :  0.47%
    jit_opt_a.a_1                        :     0.000795s :  8.74%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    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 :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.14%
    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              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.29%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000022s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000822s :  9.04%
    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      :     0.000021s :  0.23%
    jit_opt_a.cse                        :     0.000039s :  0.43%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000036s :  0.40%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000570s :  6.26%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.21%
    cconv                                :     0.000025s :  0.28%
    loop_unroll                          :     0.000503s :  5.53%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.28%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000023s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.24%
    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 :  0.07%
    remove_dup_value                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000047s :  0.52%
    cse_after_recomputation.cse          :     0.000012s :  0.13%
    auto_monad_reorder                   :     0.000019s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000514s :  5.65%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  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 :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.43%


Time group info:
------[substitution.]   0.000245    28
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000006s :      4: substitution.graph_param_transform
83.75% :     0.000205s :      4: substitution.inline
 1.27% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000005s :      4: substitution.replace_old_param
 7.08% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004505     2
77.91% :     0.003510s :      1: type_inference.infer
22.09% :     0.000995s :      1: type_inference.specialize
------[replace.]   0.000091     8
58.62% :     0.000054s :      4: replace.inline
41.38% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000217     8
92.97% :     0.000202s :      4: match.inline
 7.03% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000145   912
 1.44% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.20% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.04% :     0.000002s :     13: predicate.addn_check_dump
 1.31% :     0.000002s :     13: predicate.addn_zero_filter
 2.15% :     0.000003s :     13: predicate.arithmetic_simplify
 1.51% :     0.000002s :     13: predicate.cast_eliminate
 0.50% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     13: predicate.compare_switch_simplify
 1.13% :     0.000002s :     13: predicate.depend_value_elim
 1.20% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.48% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      4: predicate.elim_not_effective
 0.55% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.06% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     13: predicate.environ_get_eliminate
 1.06% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      4: predicate.fold_const_symbol
 1.02% :     0.000001s :      8: predicate.get_grad_eliminate
 0.30% :     0.000000s :      4: predicate.graph_param_transform
 5.48% :     0.000008s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.05% :     0.000002s :      8: predicate.less_batch_normalization
 1.60% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.87% :     0.000003s :     21: predicate.load_eliminater
 1.15% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.14% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     13: predicate.merge_addn
 1.07% :     0.000002s :     13: predicate.minmaximum_grad
 1.75% :     0.000003s :      4: predicate.mutable_eliminate
 0.47% :     0.000001s :      4: predicate.opt_reshape
 2.52% :     0.000004s :     21: predicate.partial_eliminate
 1.22% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.86% :     0.000003s :     13: predicate.reduce_eliminate
 1.60% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.83% :     0.000003s :     25: predicate.replace_applicator
 0.73% :     0.000001s :      8: predicate.replace_old_param
 0.31% :     0.000000s :      4: predicate.reset_defer_inline
 1.27% :     0.000002s :     13: predicate.reshape_eliminate
 1.13% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     13: predicate.same_eliminate
 0.57% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.19% :     0.000002s :      8: predicate.special_op_eliminate
 0.93% :     0.000001s :      8: predicate.specialize_transform
 1.35% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.60% :     0.000004s :     21: predicate.switch_defer_inline
 2.18% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.66% :     0.000011s :     67: predicate.switch_simplify
 1.28% :     0.000002s :     13: predicate.tile_eliminate
 1.18% :     0.000002s :     13: predicate.transpose_eliminate
 1.44% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.73% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     13: predicate.value_based_eliminate
 0.37% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000630    12
43.47% :     0.000274s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.50% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.03% :     0.000315s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014928    72
 0.34% :     0.000050s :      1: add_recomputation
 0.46% :     0.000069s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 1.77% :     0.000264s :      1: bootstrap
 0.19% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.17% :     0.000025s :      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
37.74% :     0.005633s :      1: jit_opt_a
 1.10% :     0.000165s :      1: jit_opt_after_cconv
 0.42% :     0.000062s :      1: jit_opt_b
 3.43% :     0.000512s :      1: loop_unroll
 3.87% :     0.000578s :      1: mutable_eliminate
 7.01% :     0.001047s :     26: opt.transform.jit_opt_a
 0.38% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000033s :      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% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.49% :     0.000522s :      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% :     0.000043s :      1: py_interpret_to_execute
 0.10% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000019s :      1: remove_dup_value
 3.05% :     0.000455s :      1: renormalize.infer
 2.41% :     0.000359s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000039s :      1: rewriter_after_opt_a
 0.68% :     0.000101s :      1: rewriter_before_opt_a
 0.54% :     0.000081s :      1: symbol_engine_optimizer
30.65% :     0.004576s :      1: type_inference


TotalTime = 0.0143232, [30]
[bootstrap]: 0.00031422
[type_inference]: 0.00527621
[event_method]: 2.259e-05
[auto_monad]: 6.691e-05
[graph_reusing]: 6.32001e-06
[pre_auto_parallel]: 3.22002e-06
[py_interpret_to_execute]: 4.479e-05
[rewriter_before_opt_a]: 9.186e-05
[expand_dump_flag]: 3.31999e-06
[jit_opt_a]: 0.00618393, [2]
    [Cycle 1]: 0.00202313, [27]
        [switch_simplify]: 6.63e-05
        [loop_unroll]: 3.629e-05
        [a_1]: 0.00068549
        [with_stream_mark]: 2.145e-05
        [recompute_prepare]: 8.50001e-06
        [updatestate_depend_eliminate]: 4.85001e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 3.06999e-06
        [parameter_eliminate]: 2.10002e-06
        [specialize_transform]: 7.55998e-06
        [updatestate_useless_node_eliminater]: 6.57002e-06
        [accelerated_algorithm]: 7.03e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 6.07001e-06
        [merge_forward]: 3.75998e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.573e-05
        [j_node_and_user_rematch]: 9.87001e-06
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 1.196e-05
        [inline_without_move]: 5.87999e-06
        [renormalize]: 0.00085667
        [add_forward_monad_depend]: 7.12997e-06
        [auto_monad_grad]: 2.84001e-06
        [auto_monad_eliminator]: 1.577e-05
        [cse]: 3.085e-05
        [replace_applicator]: 1.37e-05
    [Cycle 2]: 0.00038205, [27]
        [switch_simplify]: 7.06999e-06
        [loop_unroll]: 6.20002e-06
        [a_1]: 0.00012686
        [with_stream_mark]: 1.249e-05
        [recompute_prepare]: 6.34001e-06
        [updatestate_depend_eliminate]: 3.53e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.58002e-06
        [specialize_transform]: 5.89e-06
        [updatestate_useless_node_eliminater]: 5.73002e-06
        [accelerated_algorithm]: 6.33e-06
        [meta_shard_fg_expand]: 1.53002e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.25998e-06
        [cell_reuse_recompute_pass]: 1.73997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.405e-05
        [j_node_and_user_rematch]: 9.04e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 9.42001e-06
        [inline_without_move]: 5.84999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 7.41001e-06
        [cse]: 1.426e-05
        [replace_applicator]: 5.71e-06
[py_interpret_to_execute_after_opt_a]: 1.412e-05
[rewriter_after_opt_a]: 3.835e-05
[convert_after_rewriter]: 7.09001e-06
[order_py_execute_after_rewriter]: 5.35001e-06
[mutable_eliminate]: 0.00060891
[jit_opt_b]: 5.65e-05, [1]
    [Cycle 1]: 4.956e-05, [2]
        [frontend_op_eliminate]: 1.956e-05
        [inline_after_opt_a]: 1.792e-05
[cconv]: 2.503e-05
[loop_unroll]: 0.00046149
[jit_opt_after_cconv]: 0.0001536, [1]
    [Cycle 1]: 0.00014692, [11]
        [c_1]: 2.525e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 6.30997e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.43998e-06
        [cse]: 2.286e-05
        [call_graph_tuple_transform]: 2.164e-05
        [tuple_list_get_item_eliminator]: 6.14001e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 6.02999e-06
[remove_dup_value]: 1.513e-05
[partial_unused_args_eliminate]: 2.50002e-06
[environ_conv]: 5.91e-06
[add_recomputation]: 4.858e-05
[cse_after_recomputation]: 2.312e-05, [1]
    [Cycle 1]: 1.821e-05, [1]
        [cse]: 1.243e-05
[auto_monad_reorder]: 1.759e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 3.099e-05
[opt_after_jit_grad]: 0.00050627
[symbol_engine_optimizer]: 7.802e-05, [1]
    [Cycle 1]: 7.103e-05, [6]
        [build]: 3.88999e-06
        [elim_shapecalc]: 9.50001e-06
        [elim_not_effective]: 1.442e-05
        [opt_reshape]: 6.81001e-06
        [fold_const_symbol]: 9.27001e-06
        [renormalize]: 4.2998e-07
[validate]: 4.008e-05
Sums
    bootstrap                            :     0.000314s :  3.16%
    type_inference                       :     0.005276s : 52.98%
    event_method                         :     0.000023s :  0.23%
    auto_monad                           :     0.000067s :  0.67%
    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                :     0.000092s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000073s :  0.74%
    jit_opt_a.loop_unroll                :     0.000042s :  0.43%
    jit_opt_a.a_1                        :     0.000812s :  8.16%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.34%
    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       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.12%
    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    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000021s :  0.21%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000857s :  8.60%
    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      :     0.000023s :  0.23%
    jit_opt_a.cse                        :     0.000045s :  0.45%
    jit_opt_a.replace_applicator         :     0.000019s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.14%
    rewriter_after_opt_a                 :     0.000038s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000609s :  6.11%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.18%
    cconv                                :     0.000025s :  0.25%
    loop_unroll                          :     0.000461s :  4.63%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.25%
    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              :     0.000023s :  0.23%
    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  :     0.000006s :  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                    :     0.000049s :  0.49%
    cse_after_recomputation.cse          :     0.000012s :  0.12%
    auto_monad_reorder                   :     0.000018s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000031s :  0.31%
    opt_after_jit_grad                   :     0.000506s :  5.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.40%


Time group info:
------[substitution.]   0.000253    28
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000006s :      4: substitution.graph_param_transform
84.41% :     0.000213s :      4: substitution.inline
 1.20% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.78% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.36% :     0.000006s :      4: substitution.replace_old_param
 6.37% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005200     2
79.35% :     0.004126s :      1: type_inference.infer
20.65% :     0.001074s :      1: type_inference.specialize
------[replace.]   0.000089     8
56.57% :     0.000050s :      4: replace.inline
43.43% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000224     8
93.75% :     0.000210s :      4: match.inline
 6.25% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000144   912
 1.13% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.08% :     0.000002s :     13: predicate.addn_check_dump
 1.23% :     0.000002s :     13: predicate.addn_zero_filter
 1.91% :     0.000003s :     13: predicate.arithmetic_simplify
 1.34% :     0.000002s :     13: predicate.cast_eliminate
 0.40% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     13: predicate.compare_switch_simplify
 1.10% :     0.000002s :     13: predicate.depend_value_elim
 1.07% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.17% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.12% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      4: predicate.elim_not_effective
 0.65% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     13: predicate.environ_get_eliminate
 1.06% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 0.89% :     0.000001s :      8: predicate.get_grad_eliminate
 0.29% :     0.000000s :      4: predicate.graph_param_transform
 5.79% :     0.000008s :     29: predicate.inline
 1.09% :     0.000002s :      8: predicate.inline_without_move
 0.45% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.50% :     0.000002s :      8: predicate.less_batch_normalization
 1.54% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     21: predicate.load_eliminater
 1.28% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.10% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     13: predicate.merge_addn
 1.04% :     0.000002s :     13: predicate.minmaximum_grad
 1.85% :     0.000003s :      4: predicate.mutable_eliminate
 0.74% :     0.000001s :      4: predicate.opt_reshape
 2.62% :     0.000004s :     21: predicate.partial_eliminate
 1.25% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     13: predicate.reduce_eliminate
 1.51% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.05% :     0.000003s :     25: predicate.replace_applicator
 0.63% :     0.000001s :      8: predicate.replace_old_param
 0.32% :     0.000000s :      4: predicate.reset_defer_inline
 1.14% :     0.000002s :     13: predicate.reshape_eliminate
 1.19% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.37% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.69% :     0.000004s :     21: predicate.switch_defer_inline
 2.09% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.31% :     0.000011s :     67: predicate.switch_simplify
 1.17% :     0.000002s :     13: predicate.tile_eliminate
 1.14% :     0.000002s :     13: predicate.transpose_eliminate
 1.58% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 2.05% :     0.000003s :     13: predicate.value_based_eliminate
 0.42% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.49% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000695    12
43.99% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.94% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.07% :     0.000341s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016342    72
 0.32% :     0.000052s :      1: add_recomputation
 0.43% :     0.000070s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 2.00% :     0.000327s :      1: bootstrap
 0.17% :     0.000028s :      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
 0.17% :     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
37.86% :     0.006187s :      1: jit_opt_a
 0.96% :     0.000156s :      1: jit_opt_after_cconv
 0.36% :     0.000059s :      1: jit_opt_b
 2.87% :     0.000470s :      1: loop_unroll
 3.77% :     0.000617s :      1: mutable_eliminate
 6.53% :     0.001067s :     26: opt.transform.jit_opt_a
 0.34% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000031s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.27% :     0.000044s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.15% :     0.000514s :      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% :     0.000048s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.98% :     0.000487s :      1: renormalize.infer
 2.21% :     0.000361s :      1: renormalize.specialize
 0.21% :     0.000034s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000041s :      1: rewriter_after_opt_a
 0.58% :     0.000095s :      1: rewriter_before_opt_a
 0.49% :     0.000081s :      1: symbol_engine_optimizer
32.38% :     0.005291s :      1: type_inference


TotalTime = 0.0461852, [33]
[bootstrap]: 0.00032957
[type_inference]: 0.0232821
[event_method]: 2.037e-05
[auto_monad]: 0.00015929
[graph_reusing]: 6.67002e-06
[pre_auto_parallel]: 3.29001e-06
[py_interpret_to_execute]: 0.00012801
[rewriter_before_opt_a]: 9.868e-05
[expand_dump_flag]: 3.9e-06
[jit_opt_a]: 0.00662204, [2]
    [Cycle 1]: 0.0024998, [27]
        [switch_simplify]: 6.651e-05
        [loop_unroll]: 3.595e-05
        [a_1]: 0.00073479
        [with_stream_mark]: 2.346e-05
        [recompute_prepare]: 1.102e-05
        [updatestate_depend_eliminate]: 5.97001e-06
        [updatestate_assign_eliminate]: 5.54e-06
        [updatestate_loads_eliminate]: 4.99e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 9.07999e-06
        [updatestate_useless_node_eliminater]: 1.138e-05
        [accelerated_algorithm]: 9.08002e-06
        [meta_shard_fg_expand]: 3.04999e-06
        [get_grad_eliminate_]: 8.69e-06
        [merge_forward]: 5.66998e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.022e-05
        [j_node_and_user_rematch]: 1.437e-05
        [meta_fg_expand]: 3.63999e-06
        [replace_old_param]: 1.138e-05
        [inline_without_move]: 8.41002e-06
        [renormalize]: 0.00124145
        [add_forward_monad_depend]: 5.74999e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 2.038e-05
        [cse]: 3.965e-05
        [replace_applicator]: 1.616e-05
    [Cycle 2]: 0.00045964, [27]
        [switch_simplify]: 9.07001e-06
        [loop_unroll]: 7.88999e-06
        [a_1]: 0.00016524
        [with_stream_mark]: 1.187e-05
        [recompute_prepare]: 7.93001e-06
        [updatestate_depend_eliminate]: 4.99998e-06
        [updatestate_assign_eliminate]: 4.66002e-06
        [updatestate_loads_eliminate]: 4.06001e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 7.92003e-06
        [updatestate_useless_node_eliminater]: 1.088e-05
        [accelerated_algorithm]: 7.6e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 7.41999e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.63e-05
        [j_node_and_user_rematch]: 1.23e-05
        [meta_fg_expand]: 2.94999e-06
        [replace_old_param]: 9.84999e-06
        [inline_without_move]: 7.41999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 9.56e-06
        [cse]: 1.996e-05
        [replace_applicator]: 7.6e-06
[py_interpret_to_execute_after_opt_a]: 1.338e-05
[rewriter_after_opt_a]: 0.00029144
[convert_after_rewriter]: 1.069e-05
[order_py_execute_after_rewriter]: 6.76e-06
[mutable_eliminate]: 0.00062176
[jit_opt_b]: 6.728e-05, [1]
    [Cycle 1]: 6.025e-05, [2]
        [frontend_op_eliminate]: 2.437e-05
        [inline_after_opt_a]: 2.379e-05
[cconv]: 2.619e-05
[loop_unroll]: 0.00045539
[jit_opt_after_cconv]: 0.00019314, [1]
    [Cycle 1]: 0.00018599, [11]
        [c_1]: 4.643e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 8.12e-06
        [updatestate_assign_eliminate]: 4.45e-06
        [updatestate_loads_eliminate]: 3.97e-06
        [cse]: 2.822e-05
        [call_graph_tuple_transform]: 2.131e-05
        [tuple_list_get_item_eliminator]: 8.25e-06
        [none_parameter_eliminate]: 1.94e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 8.37998e-06
[remove_dup_value]: 1.894e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 1.255e-05
[add_recomputation]: 6.867e-05
[cse_after_recomputation]: 2.943e-05, [1]
    [Cycle 1]: 2.38e-05, [1]
        [cse]: 1.781e-05
[auto_monad_reorder]: 2.401e-05
[get_jit_bprop_graph]: 2.63e-06
[rewriter_after_jit_bprop_graph]: 4.70999e-06
[opt_after_jit_grad]: 0.0005526
[symbol_engine_optimizer]: 0.0001066, [1]
    [Cycle 1]: 9.948e-05, [6]
        [build]: 1.349e-05
        [elim_shapecalc]: 1.177e-05
        [elim_not_effective]: 2.083e-05
        [opt_reshape]: 8.92e-06
        [fold_const_symbol]: 1.39e-05
        [renormalize]: 3.59985e-07
[validate]: 5.986e-05
[backend_pass]: 1.03001e-06
[task_emit]: 0.0127271
[execute]: 8.1e-06
Sums
    bootstrap                            :     0.000330s :  0.79%
    type_inference                       :     0.023282s : 55.59%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000159s :  0.38%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000128s :  0.31%
    rewriter_before_opt_a                :     0.000099s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000076s :  0.18%
    jit_opt_a.loop_unroll                :     0.000044s :  0.10%
    jit_opt_a.a_1                        :     0.000900s :  2.15%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    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    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001242s :  2.96%
    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      :     0.000030s :  0.07%
    jit_opt_a.cse                        :     0.000060s :  0.14%
    jit_opt_a.replace_applicator         :     0.000024s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000291s :  0.70%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000622s :  1.48%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000026s :  0.06%
    loop_unroll                          :     0.000455s :  1.09%
    jit_opt_after_cconv.c_1              :     0.000046s :  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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.05%
    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                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000069s :  0.16%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000024s :  0.06%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000553s :  1.32%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000060s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012727s : 30.39%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000301    47
 2.91% :     0.000009s :      2: substitution.depend_value_elim
 1.07% :     0.000003s :      4: substitution.elim_not_effective
 0.68% :     0.000002s :      4: substitution.fold_const_symbol
 2.11% :     0.000006s :      5: substitution.graph_param_transform
77.67% :     0.000234s :      4: substitution.inline
 1.51% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.45% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.39% :     0.000004s :      2: substitution.replace_old_param
 4.26% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.11% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.84% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023208     2
93.47% :     0.021693s :      1: type_inference.infer
 6.53% :     0.001515s :      1: type_inference.specialize
------[replace.]   0.000092     6
65.61% :     0.000060s :      4: replace.inline
34.39% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000243     6
95.18% :     0.000231s :      4: match.inline
 4.82% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000172  1021
 1.46% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     15: predicate.addn_check_dump
 1.24% :     0.000002s :     15: predicate.addn_zero_filter
 2.15% :     0.000004s :     15: predicate.arithmetic_simplify
 1.34% :     0.000002s :     15: predicate.cast_eliminate
 0.48% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.19% :     0.000002s :     15: predicate.compare_switch_simplify
 1.45% :     0.000003s :     15: predicate.depend_value_elim
 1.19% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.93% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.67% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.42% :     0.000002s :     15: predicate.environ_get_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.15% :     0.000002s :     10: predicate.get_grad_eliminate
 0.26% :     0.000000s :      5: predicate.graph_param_transform
 5.25% :     0.000009s :     31: predicate.inline
 1.08% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :     10: predicate.less_batch_normalization
 1.41% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.92% :     0.000003s :     22: predicate.load_eliminater
 1.38% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.35% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.12% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     15: predicate.merge_addn
 1.20% :     0.000002s :     15: predicate.minmaximum_grad
 1.40% :     0.000002s :      5: predicate.mutable_eliminate
 0.54% :     0.000001s :      5: predicate.opt_reshape
 2.37% :     0.000004s :     22: predicate.partial_eliminate
 1.19% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.71% :     0.000003s :     15: predicate.reduce_eliminate
 1.40% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.73% :     0.000003s :     27: predicate.replace_applicator
 0.67% :     0.000001s :     10: predicate.replace_old_param
 0.41% :     0.000001s :      5: predicate.reset_defer_inline
 1.28% :     0.000002s :     15: predicate.reshape_eliminate
 1.34% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     15: predicate.same_eliminate
 0.55% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.05% :     0.000002s :     10: predicate.special_op_eliminate
 1.05% :     0.000002s :     10: predicate.specialize_transform
 1.43% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.31% :     0.000004s :     21: predicate.switch_defer_inline
 1.89% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.63% :     0.000011s :     68: predicate.switch_simplify
 1.20% :     0.000002s :     15: predicate.tile_eliminate
 1.20% :     0.000002s :     15: predicate.transpose_eliminate
 1.55% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.67% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.38% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000897    13
45.74% :     0.000410s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.74% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.52% :     0.000426s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048817    76
 0.15% :     0.000072s :      1: add_recomputation
 0.34% :     0.000165s :      1: auto_monad
 0.05% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.72% :     0.000350s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.06% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.05% :     0.000025s :      1: event_method
 0.02% :     0.000012s :      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
13.57% :     0.006625s :      1: jit_opt_a
 0.40% :     0.000196s :      1: jit_opt_after_cconv
 0.14% :     0.000070s :      1: jit_opt_b
 0.95% :     0.000466s :      1: loop_unroll
 1.29% :     0.000631s :      1: mutable_eliminate
 2.47% :     0.001205s :     26: opt.transform.jit_opt_a
 0.16% :     0.000081s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      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% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000562s :      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
 0.27% :     0.000132s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.55% :     0.000758s :      1: renormalize.infer
 0.97% :     0.000474s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000296s :      1: rewriter_after_opt_a
 0.21% :     0.000103s :      1: rewriter_before_opt_a
 0.22% :     0.000109s :      1: symbol_engine_optimizer
26.09% :     0.012739s :      1: task_emit
47.73% :     0.023298s :      1: type_inference
 0.17% :     0.000083s :      1: validate


TotalTime = 0.0140105, [30]
[bootstrap]: 0.00030519
[type_inference]: 0.00535994
[event_method]: 2.319e-05
[auto_monad]: 6.84e-05
[graph_reusing]: 5.89e-06
[pre_auto_parallel]: 2.69001e-06
[py_interpret_to_execute]: 4.242e-05
[rewriter_before_opt_a]: 9.352e-05
[expand_dump_flag]: 3.37997e-06
[jit_opt_a]: 0.00574829, [2]
    [Cycle 1]: 0.00191597, [27]
        [switch_simplify]: 6.486e-05
        [loop_unroll]: 3.743e-05
        [a_1]: 0.00064963
        [with_stream_mark]: 2.27e-05
        [recompute_prepare]: 9.97999e-06
        [updatestate_depend_eliminate]: 4.87e-06
        [updatestate_assign_eliminate]: 3.48e-06
        [updatestate_loads_eliminate]: 2.79999e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 8.35999e-06
        [updatestate_useless_node_eliminater]: 6.18002e-06
        [accelerated_algorithm]: 6.72002e-06
        [meta_shard_fg_expand]: 2.66e-06
        [get_grad_eliminate_]: 6.16e-06
        [merge_forward]: 4.35e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.831e-05
        [j_node_and_user_rematch]: 1.071e-05
        [meta_fg_expand]: 2.76999e-06
        [replace_old_param]: 1.218e-05
        [inline_without_move]: 6.30002e-06
        [renormalize]: 0.00076737
        [add_forward_monad_depend]: 6.96999e-06
        [auto_monad_grad]: 2.48e-06
        [auto_monad_eliminator]: 1.642e-05
        [cse]: 3.325e-05
        [replace_applicator]: 1.545e-05
    [Cycle 2]: 0.00038495, [27]
        [switch_simplify]: 6.86999e-06
        [loop_unroll]: 5.77999e-06
        [a_1]: 0.00011622
        [with_stream_mark]: 1.306e-05
        [recompute_prepare]: 6.42001e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 6.37001e-06
        [accelerated_algorithm]: 5.86998e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.62002e-06
        [cell_reuse_recompute_pass]: 2.11998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.592e-05
        [j_node_and_user_rematch]: 9.17001e-06
        [meta_fg_expand]: 1.95001e-06
        [replace_old_param]: 9.78998e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.49998e-06
        [auto_monad_grad]: 1.54e-06
        [auto_monad_eliminator]: 7.32997e-06
        [cse]: 1.471e-05
        [replace_applicator]: 7.13998e-06
[py_interpret_to_execute_after_opt_a]: 1.303e-05
[rewriter_after_opt_a]: 3.825e-05
[convert_after_rewriter]: 7.38999e-06
[order_py_execute_after_rewriter]: 4.98001e-06
[mutable_eliminate]: 0.00058777
[jit_opt_b]: 5.763e-05, [1]
    [Cycle 1]: 5.04e-05, [2]
        [frontend_op_eliminate]: 1.978e-05
        [inline_after_opt_a]: 1.865e-05
[cconv]: 2.619e-05
[loop_unroll]: 0.00049749
[jit_opt_after_cconv]: 0.0001644, [1]
    [Cycle 1]: 0.00015679, [11]
        [c_1]: 2.74e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 6.29001e-06
        [updatestate_assign_eliminate]: 3.08998e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.569e-05
        [call_graph_tuple_transform]: 2.243e-05
        [tuple_list_get_item_eliminator]: 6.80002e-06
        [none_parameter_eliminate]: 2.05002e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 6.23e-06
[remove_dup_value]: 1.607e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 6.13998e-06
[add_recomputation]: 5.199e-05
[cse_after_recomputation]: 2.658e-05, [1]
    [Cycle 1]: 1.991e-05, [1]
        [cse]: 1.366e-05
[auto_monad_reorder]: 1.832e-05
[get_jit_bprop_graph]: 2.05002e-06
[rewriter_after_jit_bprop_graph]: 7e-06
[opt_after_jit_grad]: 0.00051702
[symbol_engine_optimizer]: 8.279e-05, [1]
    [Cycle 1]: 7.48e-05, [6]
        [build]: 4.69002e-06
        [elim_shapecalc]: 9.49999e-06
        [elim_not_effective]: 1.446e-05
        [opt_reshape]: 6.93e-06
        [fold_const_symbol]: 9.84001e-06
        [renormalize]: 5.29981e-07
[validate]: 4.155e-05
Sums
    bootstrap                            :     0.000305s :  3.07%
    type_inference                       :     0.005360s : 53.95%
    event_method                         :     0.000023s :  0.23%
    auto_monad                           :     0.000068s :  0.69%
    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                :     0.000094s :  0.94%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000072s :  0.72%
    jit_opt_a.loop_unroll                :     0.000043s :  0.43%
    jit_opt_a.a_1                        :     0.000766s :  7.71%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.36%
    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 :     0.000006s :  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       :     0.000014s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.13%
    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_        :     0.000012s :  0.12%
    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 :  0.34%
    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          :     0.000022s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000767s :  7.72%
    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 :  0.24%
    jit_opt_a.cse                        :     0.000048s :  0.48%
    jit_opt_a.replace_applicator         :     0.000023s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.13%
    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                    :     0.000588s :  5.92%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.19%
    cconv                                :     0.000026s :  0.26%
    loop_unroll                          :     0.000497s :  5.01%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.28%
    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              :     0.000026s :  0.26%
    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  :     0.000006s :  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                    :     0.000052s :  0.52%
    cse_after_recomputation.cse          :     0.000014s :  0.14%
    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                   :     0.000517s :  5.20%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.42%


Time group info:
------[substitution.]   0.000225    28
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.63% :     0.000006s :      4: substitution.graph_param_transform
81.71% :     0.000184s :      4: substitution.inline
 1.76% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.53% :     0.000006s :      4: substitution.replace_old_param
 7.72% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005289     2
80.15% :     0.004240s :      1: type_inference.infer
19.85% :     0.001050s :      1: type_inference.specialize
------[replace.]   0.000090     8
57.11% :     0.000051s :      4: replace.inline
42.89% :     0.000038s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     8
92.13% :     0.000181s :      4: match.inline
 7.87% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000145   912
 1.17% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     13: predicate.addn_check_dump
 1.26% :     0.000002s :     13: predicate.addn_zero_filter
 1.93% :     0.000003s :     13: predicate.arithmetic_simplify
 1.31% :     0.000002s :     13: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.06% :     0.000002s :     13: predicate.compare_switch_simplify
 1.11% :     0.000002s :     13: predicate.depend_value_elim
 1.06% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     13: predicate.dict_set_item_eliminator
 1.15% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      4: predicate.elim_not_effective
 0.79% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.07% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     13: predicate.environ_get_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 0.86% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.64% :     0.000008s :     29: predicate.inline
 1.07% :     0.000002s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.12% :     0.000002s :      8: predicate.less_batch_normalization
 1.55% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     21: predicate.load_eliminater
 1.42% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.46% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.85% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     13: predicate.merge_addn
 1.13% :     0.000002s :     13: predicate.minmaximum_grad
 1.46% :     0.000002s :      4: predicate.mutable_eliminate
 0.46% :     0.000001s :      4: predicate.opt_reshape
 2.47% :     0.000004s :     21: predicate.partial_eliminate
 1.11% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     13: predicate.reduce_eliminate
 1.53% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.11% :     0.000003s :     25: predicate.replace_applicator
 0.90% :     0.000001s :      8: predicate.replace_old_param
 0.47% :     0.000001s :      4: predicate.reset_defer_inline
 1.08% :     0.000002s :     13: predicate.reshape_eliminate
 1.14% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.14% :     0.000002s :     13: predicate.same_eliminate
 0.61% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.15% :     0.000002s :      8: predicate.special_op_eliminate
 1.09% :     0.000002s :      8: predicate.specialize_transform
 1.33% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.49% :     0.000004s :     21: predicate.switch_defer_inline
 2.11% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.96% :     0.000012s :     67: predicate.switch_simplify
 1.06% :     0.000002s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.39% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.35% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.78% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.18% :     0.000005s :     29: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     13: predicate.value_based_eliminate
 0.32% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000675    12
49.83% :     0.000336s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.42% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.75% :     0.000309s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015885    72
 0.35% :     0.000055s :      1: add_recomputation
 0.45% :     0.000072s :      1: auto_monad
 0.14% :     0.000022s :      1: auto_monad_reorder
 2.02% :     0.000321s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000029s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.17% :     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
36.21% :     0.005752s :      1: jit_opt_a
 1.06% :     0.000168s :      1: jit_opt_after_cconv
 0.38% :     0.000061s :      1: jit_opt_b
 3.19% :     0.000507s :      1: loop_unroll
 3.76% :     0.000597s :      1: mutable_eliminate
 6.48% :     0.001029s :     26: opt.transform.jit_opt_a
 0.37% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000032s :      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
 0.18% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.31% :     0.000527s :      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% :     0.000045s :      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
 2.51% :     0.000399s :      1: renormalize.infer
 2.26% :     0.000359s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000042s :      1: rewriter_after_opt_a
 0.61% :     0.000097s :      1: rewriter_before_opt_a
 0.54% :     0.000086s :      1: symbol_engine_optimizer
33.84% :     0.005376s :      1: type_inference


TotalTime = 0.0147574, [30]
[bootstrap]: 0.00031966
[type_inference]: 0.00537802
[event_method]: 2.307e-05
[auto_monad]: 6.781e-05
[graph_reusing]: 6.09001e-06
[pre_auto_parallel]: 2.79999e-06
[py_interpret_to_execute]: 4.847e-05
[rewriter_before_opt_a]: 9.067e-05
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.0065455, [2]
    [Cycle 1]: 0.00220864, [27]
        [switch_simplify]: 6.871e-05
        [loop_unroll]: 3.693e-05
        [a_1]: 0.00072349
        [with_stream_mark]: 2.111e-05
        [recompute_prepare]: 8.87e-06
        [updatestate_depend_eliminate]: 4.03001e-06
        [updatestate_assign_eliminate]: 3.41999e-06
        [updatestate_loads_eliminate]: 3.11999e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 8e-06
        [updatestate_useless_node_eliminater]: 6.73e-06
        [accelerated_algorithm]: 6.73e-06
        [meta_shard_fg_expand]: 2.43998e-06
        [get_grad_eliminate_]: 6.26e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 1.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.645e-05
        [j_node_and_user_rematch]: 1.026e-05
        [meta_fg_expand]: 2.93998e-06
        [replace_old_param]: 1.124e-05
        [inline_without_move]: 6.04999e-06
        [renormalize]: 0.00095887
        [add_forward_monad_depend]: 6.92002e-06
        [auto_monad_grad]: 3.14999e-06
        [auto_monad_eliminator]: 1.714e-05
        [cse]: 3.651e-05
        [replace_applicator]: 1.786e-05
    [Cycle 2]: 0.0004417, [27]
        [switch_simplify]: 7.78001e-06
        [loop_unroll]: 6.69001e-06
        [a_1]: 0.00014527
        [with_stream_mark]: 1.33e-05
        [recompute_prepare]: 6.51e-06
        [updatestate_depend_eliminate]: 3.88999e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.94999e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 6.91001e-06
        [updatestate_useless_node_eliminater]: 6.61e-06
        [accelerated_algorithm]: 7.16001e-06
        [meta_shard_fg_expand]: 1.65001e-06
        [get_grad_eliminate_]: 6.18998e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 3.13e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.818e-05
        [j_node_and_user_rematch]: 1.022e-05
        [meta_fg_expand]: 3.19001e-06
        [replace_old_param]: 1.162e-05
        [inline_without_move]: 6.32001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.46002e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 7.70998e-06
        [cse]: 1.637e-05
        [replace_applicator]: 7.06999e-06
[py_interpret_to_execute_after_opt_a]: 1.585e-05
[rewriter_after_opt_a]: 4.054e-05
[convert_after_rewriter]: 1.279e-05
[order_py_execute_after_rewriter]: 5.72999e-06
[mutable_eliminate]: 0.00061006
[jit_opt_b]: 6.51e-05, [1]
    [Cycle 1]: 5.649e-05, [2]
        [frontend_op_eliminate]: 2.149e-05
        [inline_after_opt_a]: 2.115e-05
[cconv]: 2.796e-05
[loop_unroll]: 0.00045356
[jit_opt_after_cconv]: 0.00015735, [1]
    [Cycle 1]: 0.00015089, [11]
        [c_1]: 2.619e-05
        [parameter_eliminate]: 2.78e-06
        [updatestate_depend_eliminate]: 5.77999e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.65997e-06
        [cse]: 2.197e-05
        [call_graph_tuple_transform]: 2.156e-05
        [tuple_list_get_item_eliminator]: 6.63e-06
        [none_parameter_eliminate]: 1.93002e-06
        [renormalize]: 3.9002e-07
        [switch_simplify]: 6.63e-06
[remove_dup_value]: 1.399e-05
[partial_unused_args_eliminate]: 2.45002e-06
[environ_conv]: 5.62999e-06
[add_recomputation]: 4.752e-05
[cse_after_recomputation]: 2.468e-05, [1]
    [Cycle 1]: 1.878e-05, [1]
        [cse]: 1.313e-05
[auto_monad_reorder]: 1.771e-05
[get_jit_bprop_graph]: 1.97001e-06
[rewriter_after_jit_bprop_graph]: 5.14998e-06
[opt_after_jit_grad]: 0.00047216
[symbol_engine_optimizer]: 7.899e-05, [1]
    [Cycle 1]: 7.287e-05, [6]
        [build]: 3.58999e-06
        [elim_shapecalc]: 9.00999e-06
        [elim_not_effective]: 1.45e-05
        [opt_reshape]: 6.76999e-06
        [fold_const_symbol]: 9.77999e-06
        [renormalize]: 5.3001e-07
[validate]: 3.916e-05
Sums
    bootstrap                            :     0.000320s :  3.13%
    type_inference                       :     0.005378s : 52.64%
    event_method                         :     0.000023s :  0.23%
    auto_monad                           :     0.000068s :  0.66%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000048s :  0.47%
    rewriter_before_opt_a                :     0.000091s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000076s :  0.75%
    jit_opt_a.loop_unroll                :     0.000044s :  0.43%
    jit_opt_a.a_1                        :     0.000869s :  8.50%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.34%
    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 :     0.000006s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000015s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  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 :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.06%
    jit_opt_a.replace_old_param          :     0.000023s :  0.22%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000959s :  9.39%
    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      :     0.000025s :  0.24%
    jit_opt_a.cse                        :     0.000053s :  0.52%
    jit_opt_a.replace_applicator         :     0.000025s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.16%
    rewriter_after_opt_a                 :     0.000041s :  0.40%
    convert_after_rewriter               :     0.000013s :  0.13%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000610s :  5.97%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.21%
    cconv                                :     0.000028s :  0.27%
    loop_unroll                          :     0.000454s :  4.44%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.26%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000022s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  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  :     0.000007s :  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                    :     0.000048s :  0.47%
    cse_after_recomputation.cse          :     0.000013s :  0.13%
    auto_monad_reorder                   :     0.000018s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000472s :  4.62%
    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 :  0.14%
    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                             :     0.000039s :  0.38%


Time group info:
------[substitution.]   0.000289    28
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000002s :      2: substitution.fold_const_symbol
 2.02% :     0.000006s :      4: substitution.graph_param_transform
85.69% :     0.000247s :      4: substitution.inline
 1.24% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.75% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000006s :      4: substitution.replace_old_param
 6.02% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005308     2
80.12% :     0.004253s :      1: type_inference.infer
19.88% :     0.001055s :      1: type_inference.specialize
------[replace.]   0.000094     8
56.73% :     0.000054s :      4: replace.inline
43.27% :     0.000041s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000260     8
94.08% :     0.000244s :      4: match.inline
 5.92% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168   912
 1.05% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.90% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     13: predicate.addn_check_dump
 1.24% :     0.000002s :     13: predicate.addn_zero_filter
 1.96% :     0.000003s :     13: predicate.arithmetic_simplify
 1.14% :     0.000002s :     13: predicate.cast_eliminate
 0.47% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.05% :     0.000002s :     13: predicate.compare_switch_simplify
 0.98% :     0.000002s :     13: predicate.depend_value_elim
 0.95% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.05% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      4: predicate.elim_not_effective
 0.48% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000002s :     13: predicate.environ_add_const_eliminate
 0.92% :     0.000002s :     13: predicate.environ_get_add_eliminate
 0.87% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     13: predicate.environ_get_eliminate
 0.96% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000001s :      8: predicate.get_grad_eliminate
 0.26% :     0.000000s :      4: predicate.graph_param_transform
 4.95% :     0.000008s :     29: predicate.inline
 0.87% :     0.000001s :      8: predicate.inline_without_move
 0.37% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      8: predicate.less_batch_normalization
 1.48% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.73% :     0.000003s :     21: predicate.load_eliminater
 1.20% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.69% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.71% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000002s :     13: predicate.merge_addn
 1.07% :     0.000002s :     13: predicate.minmaximum_grad
 1.49% :     0.000002s :      4: predicate.mutable_eliminate
 0.44% :     0.000001s :      4: predicate.opt_reshape
 2.15% :     0.000004s :     21: predicate.partial_eliminate
 1.01% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     13: predicate.reduce_eliminate
 1.39% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.72% :     0.000003s :     25: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.29% :     0.000000s :      4: predicate.reset_defer_inline
11.13% :     0.000019s :     13: predicate.reshape_eliminate
 1.05% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.01% :     0.000002s :     13: predicate.same_eliminate
 0.52% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.96% :     0.000002s :      8: predicate.special_op_eliminate
 0.92% :     0.000002s :      8: predicate.specialize_transform
 1.53% :     0.000003s :     13: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.30% :     0.000004s :     21: predicate.switch_defer_inline
 1.98% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.63% :     0.000011s :     67: predicate.switch_simplify
 1.18% :     0.000002s :     13: predicate.tile_eliminate
 1.08% :     0.000002s :     13: predicate.transpose_eliminate
 1.40% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000007s :     25: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.30% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.54% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000005s :     29: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000003s :     13: predicate.value_based_eliminate
 0.33% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000755    12
43.64% :     0.000330s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.68% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.68% :     0.000375s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016940    72
 0.30% :     0.000051s :      1: add_recomputation
 0.42% :     0.000072s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 1.98% :     0.000336s :      1: bootstrap
 0.18% :     0.000031s :      1: cconv
 0.09% :     0.000015s :      1: convert_after_rewriter
 0.16% :     0.000027s :      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
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
38.66% :     0.006550s :      1: jit_opt_a
 0.95% :     0.000160s :      1: jit_opt_after_cconv
 0.40% :     0.000068s :      1: jit_opt_b
 2.72% :     0.000461s :      1: loop_unroll
 3.66% :     0.000619s :      1: mutable_eliminate
 6.74% :     0.001141s :     26: opt.transform.jit_opt_a
 0.34% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000035s :      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
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.83% :     0.000480s :      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% :     0.000052s :      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
 2.94% :     0.000498s :      1: renormalize.infer
 2.66% :     0.000451s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000043s :      1: rewriter_after_opt_a
 0.55% :     0.000094s :      1: rewriter_before_opt_a
 0.48% :     0.000082s :      1: symbol_engine_optimizer
31.84% :     0.005394s :      1: type_inference


TotalTime = 0.0502539, [33]
[bootstrap]: 0.00029226
[type_inference]: 0.0278642
[event_method]: 1.964e-05
[auto_monad]: 0.00015527
[graph_reusing]: 6.44001e-06
[pre_auto_parallel]: 2.84001e-06
[py_interpret_to_execute]: 0.00012593
[rewriter_before_opt_a]: 9.865e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.00639065, [2]
    [Cycle 1]: 0.00246026, [27]
        [switch_simplify]: 6.351e-05
        [loop_unroll]: 3.608e-05
        [a_1]: 0.00071565
        [with_stream_mark]: 2.051e-05
        [recompute_prepare]: 1.083e-05
        [updatestate_depend_eliminate]: 6.89999e-06
        [updatestate_assign_eliminate]: 5.81e-06
        [updatestate_loads_eliminate]: 5.02e-06
        [parameter_eliminate]: 2.56998e-06
        [specialize_transform]: 9.79e-06
        [updatestate_useless_node_eliminater]: 1.155e-05
        [accelerated_algorithm]: 8.65999e-06
        [meta_shard_fg_expand]: 3.13998e-06
        [get_grad_eliminate_]: 8.31002e-06
        [merge_forward]: 5.14998e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.085e-05
        [j_node_and_user_rematch]: 1.39e-05
        [meta_fg_expand]: 3.73999e-06
        [replace_old_param]: 1.27e-05
        [inline_without_move]: 8.42e-06
        [renormalize]: 0.00121056
        [add_forward_monad_depend]: 6.51e-06
        [auto_monad_grad]: 2.17999e-06
        [auto_monad_eliminator]: 2.381e-05
        [cse]: 4.132e-05
        [replace_applicator]: 1.823e-05
    [Cycle 2]: 0.000478, [27]
        [switch_simplify]: 9.14998e-06
        [loop_unroll]: 8.62998e-06
        [a_1]: 0.00017142
        [with_stream_mark]: 1.277e-05
        [recompute_prepare]: 8.74998e-06
        [updatestate_depend_eliminate]: 5.25001e-06
        [updatestate_assign_eliminate]: 5.25001e-06
        [updatestate_loads_eliminate]: 4.02e-06
        [parameter_eliminate]: 1.30001e-06
        [specialize_transform]: 8.29998e-06
        [updatestate_useless_node_eliminater]: 1.107e-05
        [accelerated_algorithm]: 7.88999e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 7.20998e-06
        [merge_forward]: 4.94e-06
        [cell_reuse_recompute_pass]: 2.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.701e-05
        [j_node_and_user_rematch]: 1.28e-05
        [meta_fg_expand]: 2.95002e-06
        [replace_old_param]: 1.165e-05
        [inline_without_move]: 7.80998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 9.60019e-07
        [auto_monad_eliminator]: 9.54e-06
        [cse]: 2.068e-05
        [replace_applicator]: 7.66999e-06
[py_interpret_to_execute_after_opt_a]: 1.348e-05
[rewriter_after_opt_a]: 0.00030673
[convert_after_rewriter]: 1.146e-05
[order_py_execute_after_rewriter]: 8.65001e-06
[mutable_eliminate]: 0.00061862
[jit_opt_b]: 7.142e-05, [1]
    [Cycle 1]: 6.401e-05, [2]
        [frontend_op_eliminate]: 2.641e-05
        [inline_after_opt_a]: 2.503e-05
[cconv]: 3.037e-05
[loop_unroll]: 0.0004525
[jit_opt_after_cconv]: 0.000194, [1]
    [Cycle 1]: 0.00018744, [11]
        [c_1]: 4.438e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 8.79e-06
        [updatestate_assign_eliminate]: 4.47e-06
        [updatestate_loads_eliminate]: 4.03001e-06
        [cse]: 2.971e-05
        [call_graph_tuple_transform]: 2.17e-05
        [tuple_list_get_item_eliminator]: 8.33001e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 3.80009e-07
        [switch_simplify]: 9.69999e-06
[remove_dup_value]: 2.07e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 1.15e-05
[add_recomputation]: 6.487e-05
[cse_after_recomputation]: 2.967e-05, [1]
    [Cycle 1]: 2.362e-05, [1]
        [cse]: 1.786e-05
[auto_monad_reorder]: 2.41e-05
[get_jit_bprop_graph]: 1.86998e-06
[rewriter_after_jit_bprop_graph]: 4.87998e-06
[opt_after_jit_grad]: 0.00049776
[symbol_engine_optimizer]: 0.00010419, [1]
    [Cycle 1]: 9.759e-05, [6]
        [build]: 1.383e-05
        [elim_shapecalc]: 1.151e-05
        [elim_not_effective]: 1.934e-05
        [opt_reshape]: 9.11998e-06
        [fold_const_symbol]: 1.411e-05
        [renormalize]: 4.19997e-07
[validate]: 5.88e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.0125388
[execute]: 8.43001e-06
Sums
    bootstrap                            :     0.000292s :  0.63%
    type_inference                       :     0.027864s : 60.37%
    event_method                         :     0.000020s :  0.04%
    auto_monad                           :     0.000155s :  0.34%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000126s :  0.27%
    rewriter_before_opt_a                :     0.000099s :  0.21%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.16%
    jit_opt_a.loop_unroll                :     0.000045s :  0.10%
    jit_opt_a.a_1                        :     0.000887s :  1.92%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  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 :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.03%
    jit_opt_a.merge_forward              :     0.000010s :  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 :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001211s :  2.62%
    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      :     0.000033s :  0.07%
    jit_opt_a.cse                        :     0.000062s :  0.13%
    jit_opt_a.replace_applicator         :     0.000026s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000307s :  0.66%
    convert_after_rewriter               :     0.000011s :  0.02%
    order_py_execute_after_rewriter      :     0.000009s :  0.02%
    mutable_eliminate                    :     0.000619s :  1.34%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.05%
    cconv                                :     0.000030s :  0.07%
    loop_unroll                          :     0.000453s :  0.98%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000030s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.05%
    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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000021s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000012s :  0.02%
    add_recomputation                    :     0.000065s :  0.14%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000024s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000498s :  1.08%
    symbol_engine_optimizer.build        :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000059s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012539s : 27.16%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000275    47
 3.40% :     0.000009s :      2: substitution.depend_value_elim
 1.07% :     0.000003s :      4: substitution.elim_not_effective
 0.75% :     0.000002s :      4: substitution.fold_const_symbol
 2.31% :     0.000006s :      5: substitution.graph_param_transform
75.00% :     0.000206s :      4: substitution.inline
 1.74% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.63% :     0.000007s :      8: substitution.remove_not_recompute_node
 2.35% :     0.000006s :      2: substitution.replace_old_param
 4.42% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.08% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.24% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.027796     2
94.56% :     0.026283s :      1: type_inference.infer
 5.44% :     0.001513s :      1: type_inference.specialize
------[replace.]   0.000097     6
66.23% :     0.000065s :      4: replace.inline
33.77% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
94.89% :     0.000203s :      4: match.inline
 5.11% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000176  1021
 1.40% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.48% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.31% :     0.000002s :     15: predicate.addn_zero_filter
 1.95% :     0.000003s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.47% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     15: predicate.compare_switch_simplify
 1.52% :     0.000003s :     15: predicate.depend_value_elim
 1.22% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.95% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.31% :     0.000001s :      5: predicate.elim_not_effective
 0.67% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     15: predicate.environ_get_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.02% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 5.02% :     0.000009s :     31: predicate.inline
 1.00% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.44% :     0.000003s :     10: predicate.less_batch_normalization
 1.33% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     22: predicate.load_eliminater
 1.49% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.41% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.05% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     15: predicate.merge_addn
 1.14% :     0.000002s :     15: predicate.minmaximum_grad
 1.75% :     0.000003s :      5: predicate.mutable_eliminate
 0.52% :     0.000001s :      5: predicate.opt_reshape
 2.34% :     0.000004s :     22: predicate.partial_eliminate
 1.49% :     0.000003s :     15: predicate.print_const_string_wrapper
 2.03% :     0.000004s :     15: predicate.reduce_eliminate
 1.45% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.75% :     0.000003s :     27: predicate.replace_applicator
 0.79% :     0.000001s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.38% :     0.000002s :     15: predicate.reshape_eliminate
 1.26% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.09% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.18% :     0.000002s :     10: predicate.special_op_eliminate
 1.04% :     0.000002s :     10: predicate.specialize_transform
 1.36% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.21% :     0.000004s :     21: predicate.switch_defer_inline
 1.85% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.55% :     0.000012s :     68: predicate.switch_simplify
 1.20% :     0.000002s :     15: predicate.tile_eliminate
 1.47% :     0.000003s :     15: predicate.transpose_eliminate
 1.50% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.81% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.43% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000826    13
43.25% :     0.000357s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.86% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.89% :     0.000429s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.052850    76
 0.13% :     0.000068s :      1: add_recomputation
 0.30% :     0.000160s :      1: auto_monad
 0.05% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.58% :     0.000306s :      1: bootstrap
 0.06% :     0.000033s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.06% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.04% :     0.000023s :      1: event_method
 0.02% :     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
12.10% :     0.006394s :      1: jit_opt_a
 0.37% :     0.000197s :      1: jit_opt_after_cconv
 0.14% :     0.000074s :      1: jit_opt_b
 0.87% :     0.000461s :      1: loop_unroll
 1.19% :     0.000628s :      1: mutable_eliminate
 2.26% :     0.001196s :     26: opt.transform.jit_opt_a
 0.15% :     0.000080s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000044s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000051s :      4: opt.transform.symbol_engine_opt
 0.96% :     0.000506s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000130s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000023s :      1: remove_dup_value
 1.30% :     0.000687s :      1: renormalize.infer
 0.97% :     0.000514s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.59% :     0.000312s :      1: rewriter_after_opt_a
 0.20% :     0.000103s :      1: rewriter_before_opt_a
 0.20% :     0.000107s :      1: symbol_engine_optimizer
23.76% :     0.012556s :      1: task_emit
52.75% :     0.027876s :      1: type_inference
 0.16% :     0.000085s :      1: validate


TotalTime = 0.012517, [30]
[bootstrap]: 0.00026239
[type_inference]: 0.00451324
[event_method]: 2.036e-05
[auto_monad]: 5.618e-05
[graph_reusing]: 4.88001e-06
[pre_auto_parallel]: 1.63002e-06
[py_interpret_to_execute]: 4.251e-05
[rewriter_before_opt_a]: 8.502e-05
[expand_dump_flag]: 2.93e-06
[jit_opt_a]: 0.00532348, [2]
    [Cycle 1]: 0.00180876, [27]
        [switch_simplify]: 5.677e-05
        [loop_unroll]: 3.674e-05
        [a_1]: 0.00062935
        [with_stream_mark]: 1.498e-05
        [recompute_prepare]: 8.45001e-06
        [updatestate_depend_eliminate]: 3.40998e-06
        [updatestate_assign_eliminate]: 3.01999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 7.5e-06
        [updatestate_useless_node_eliminater]: 6.25002e-06
        [accelerated_algorithm]: 6.43998e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 6.27001e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.407e-05
        [j_node_and_user_rematch]: 1.055e-05
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 1.091e-05
        [inline_without_move]: 6.20002e-06
        [renormalize]: 0.00074702
        [add_forward_monad_depend]: 5.50001e-06
        [auto_monad_grad]: 2.14e-06
        [auto_monad_eliminator]: 1.473e-05
        [cse]: 2.352e-05
        [replace_applicator]: 1.341e-05
    [Cycle 2]: 0.00037052, [27]
        [switch_simplify]: 6.89999e-06
        [loop_unroll]: 6.01e-06
        [a_1]: 0.00011724
        [with_stream_mark]: 9.22999e-06
        [recompute_prepare]: 5.84e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 6.07001e-06
        [updatestate_useless_node_eliminater]: 6.04999e-06
        [accelerated_algorithm]: 6.03998e-06
        [meta_shard_fg_expand]: 1.63002e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 1.67999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.393e-05
        [j_node_and_user_rematch]: 9.15001e-06
        [meta_fg_expand]: 1.91998e-06
        [replace_old_param]: 9.63997e-06
        [inline_without_move]: 5.79999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 6.81001e-06
        [cse]: 1.396e-05
        [replace_applicator]: 6.17999e-06
[py_interpret_to_execute_after_opt_a]: 1.191e-05
[rewriter_after_opt_a]: 3.316e-05
[convert_after_rewriter]: 7.65e-06
[order_py_execute_after_rewriter]: 5.37001e-06
[mutable_eliminate]: 0.00053724
[jit_opt_b]: 5.62e-05, [1]
    [Cycle 1]: 4.941e-05, [2]
        [frontend_op_eliminate]: 1.93e-05
        [inline_after_opt_a]: 1.791e-05
[cconv]: 2.442e-05
[loop_unroll]: 0.00045541
[jit_opt_after_cconv]: 0.00016495, [1]
    [Cycle 1]: 0.00015757, [11]
        [c_1]: 2.634e-05
        [parameter_eliminate]: 3.01001e-06
        [updatestate_depend_eliminate]: 7.35998e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 2.62e-05
        [call_graph_tuple_transform]: 2.291e-05
        [tuple_list_get_item_eliminator]: 6.30997e-06
        [none_parameter_eliminate]: 1.77999e-06
        [renormalize]: 5.40022e-07
        [switch_simplify]: 6.06998e-06
[remove_dup_value]: 1.609e-05
[partial_unused_args_eliminate]: 2.76e-06
[environ_conv]: 6.12999e-06
[add_recomputation]: 5.059e-05
[cse_after_recomputation]: 2.649e-05, [1]
    [Cycle 1]: 2.139e-05, [1]
        [cse]: 1.437e-05
[auto_monad_reorder]: 1.735e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 5.62999e-06
[opt_after_jit_grad]: 0.00048945
[symbol_engine_optimizer]: 8.317e-05, [1]
    [Cycle 1]: 7.671e-05, [6]
        [build]: 3.82998e-06
        [elim_shapecalc]: 8.95999e-06
        [elim_not_effective]: 1.531e-05
        [opt_reshape]: 7.35e-06
        [fold_const_symbol]: 1.046e-05
        [renormalize]: 6.50005e-07
[validate]: 3.883e-05
Sums
    bootstrap                            :     0.000262s :  2.98%
    type_inference                       :     0.004513s : 51.34%
    event_method                         :     0.000020s :  0.23%
    auto_monad                           :     0.000056s :  0.64%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.48%
    rewriter_before_opt_a                :     0.000085s :  0.97%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000064s :  0.72%
    jit_opt_a.loop_unroll                :     0.000043s :  0.49%
    jit_opt_a.a_1                        :     0.000747s :  8.49%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.28%
    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 :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000014s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000012s :  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 :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000021s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000747s :  8.50%
    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      :     0.000022s :  0.25%
    jit_opt_a.cse                        :     0.000037s :  0.43%
    jit_opt_a.replace_applicator         :     0.000020s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.14%
    rewriter_after_opt_a                 :     0.000033s :  0.38%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000537s :  6.11%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000024s :  0.28%
    loop_unroll                          :     0.000455s :  5.18%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.30%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000026s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.26%
    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                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000051s :  0.58%
    cse_after_recomputation.cse          :     0.000014s :  0.16%
    auto_monad_reorder                   :     0.000017s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000489s :  5.57%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    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 :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.44%


Time group info:
------[substitution.]   0.000212    28
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000002s :      2: substitution.fold_const_symbol
 3.08% :     0.000007s :      4: substitution.graph_param_transform
82.24% :     0.000175s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.42% :     0.000005s :      4: substitution.replace_old_param
 6.96% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004460     2
78.70% :     0.003510s :      1: type_inference.infer
21.30% :     0.000950s :      1: type_inference.specialize
------[replace.]   0.000082     8
56.89% :     0.000047s :      4: replace.inline
43.11% :     0.000035s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     8
93.06% :     0.000172s :      4: match.inline
 6.94% :     0.000013s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   912
 1.21% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.10% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.08% :     0.000002s :     13: predicate.addn_check_dump
 1.31% :     0.000002s :     13: predicate.addn_zero_filter
 1.93% :     0.000003s :     13: predicate.arithmetic_simplify
 1.31% :     0.000002s :     13: predicate.cast_eliminate
 0.49% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     13: predicate.compare_switch_simplify
 1.14% :     0.000002s :     13: predicate.depend_value_elim
 1.19% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      4: predicate.elim_not_effective
 0.56% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     13: predicate.environ_get_eliminate
 1.08% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      4: predicate.fold_const_symbol
 1.13% :     0.000002s :      8: predicate.get_grad_eliminate
 0.36% :     0.000001s :      4: predicate.graph_param_transform
 5.43% :     0.000008s :     29: predicate.inline
 1.03% :     0.000001s :      8: predicate.inline_without_move
 0.45% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :      8: predicate.less_batch_normalization
 1.63% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     21: predicate.load_eliminater
 1.61% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.22% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.96% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000002s :     13: predicate.merge_addn
 1.13% :     0.000002s :     13: predicate.minmaximum_grad
 1.38% :     0.000002s :      4: predicate.mutable_eliminate
 0.49% :     0.000001s :      4: predicate.opt_reshape
 2.61% :     0.000004s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     13: predicate.reduce_eliminate
 1.57% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.89% :     0.000003s :     25: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.58% :     0.000001s :      4: predicate.reset_defer_inline
 1.27% :     0.000002s :     13: predicate.reshape_eliminate
 1.36% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     13: predicate.same_eliminate
 0.62% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.53% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.52% :     0.000004s :     21: predicate.switch_defer_inline
 2.09% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.67% :     0.000011s :     67: predicate.switch_simplify
 1.26% :     0.000002s :     13: predicate.tile_eliminate
 1.34% :     0.000002s :     13: predicate.transpose_eliminate
 1.43% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.81% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     13: predicate.value_based_eliminate
 0.44% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000617    12
44.11% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.58% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.31% :     0.000310s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014334    72
 0.37% :     0.000053s :      1: add_recomputation
 0.42% :     0.000060s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 1.92% :     0.000275s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.20% :     0.000029s :      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
 0.05% :     0.000007s :      1: graph_reusing
37.16% :     0.005327s :      1: jit_opt_a
 1.18% :     0.000169s :      1: jit_opt_after_cconv
 0.41% :     0.000059s :      1: jit_opt_b
 3.24% :     0.000464s :      1: loop_unroll
 3.81% :     0.000546s :      1: mutable_eliminate
 6.91% :     0.000990s :     26: opt.transform.jit_opt_a
 0.40% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000031s :      4: opt.transform.jit_opt_b
 0.10% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.47% :     0.000498s :      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
 0.32% :     0.000046s :      1: py_interpret_to_execute
 0.10% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000019s :      1: remove_dup_value
 2.99% :     0.000428s :      1: renormalize.infer
 2.17% :     0.000311s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000036s :      1: rewriter_after_opt_a
 0.61% :     0.000088s :      1: rewriter_before_opt_a
 0.60% :     0.000086s :      1: symbol_engine_optimizer
31.56% :     0.004523s :      1: type_inference


TotalTime = 0.0130111, [30]
[bootstrap]: 0.00026548
[type_inference]: 0.00472648
[event_method]: 2.199e-05
[auto_monad]: 6.393e-05
[graph_reusing]: 5.61e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 4.671e-05
[rewriter_before_opt_a]: 8.905e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00553065, [2]
    [Cycle 1]: 0.00186411, [27]
        [switch_simplify]: 6.38e-05
        [loop_unroll]: 3.664e-05
        [a_1]: 0.00064729
        [with_stream_mark]: 1.786e-05
        [recompute_prepare]: 8.90999e-06
        [updatestate_depend_eliminate]: 3.81999e-06
        [updatestate_assign_eliminate]: 3.63e-06
        [updatestate_loads_eliminate]: 2.93998e-06
        [parameter_eliminate]: 2.22999e-06
        [specialize_transform]: 7.36001e-06
        [updatestate_useless_node_eliminater]: 6.34001e-06
        [accelerated_algorithm]: 6.04999e-06
        [meta_shard_fg_expand]: 2.26998e-06
        [get_grad_eliminate_]: 6.35002e-06
        [merge_forward]: 4.48999e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.532e-05
        [j_node_and_user_rematch]: 9.81e-06
        [meta_fg_expand]: 2.45002e-06
        [replace_old_param]: 1.09e-05
        [inline_without_move]: 6.57002e-06
        [renormalize]: 0.00073154
        [add_forward_monad_depend]: 5.61e-06
        [auto_monad_grad]: 2.67001e-06
        [auto_monad_eliminator]: 1.574e-05
        [cse]: 3.212e-05
        [replace_applicator]: 1.331e-05
    [Cycle 2]: 0.00036607, [27]
        [switch_simplify]: 6.94999e-06
        [loop_unroll]: 6.14001e-06
        [a_1]: 0.00011647
        [with_stream_mark]: 9.19e-06
        [recompute_prepare]: 5.94999e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.79e-06
        [updatestate_useless_node_eliminater]: 5.96998e-06
        [accelerated_algorithm]: 6.19999e-06
        [meta_shard_fg_expand]: 1.52999e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 3.02002e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.37e-05
        [j_node_and_user_rematch]: 8.60999e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 9.77001e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.34003e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 6.02999e-06
        [cse]: 1.291e-05
        [replace_applicator]: 5.86e-06
[py_interpret_to_execute_after_opt_a]: 1.096e-05
[rewriter_after_opt_a]: 3.502e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00055728
[jit_opt_b]: 5.993e-05, [1]
    [Cycle 1]: 5.249e-05, [2]
        [frontend_op_eliminate]: 2.104e-05
        [inline_after_opt_a]: 1.917e-05
[cconv]: 2.648e-05
[loop_unroll]: 0.00046508
[jit_opt_after_cconv]: 0.00017083, [1]
    [Cycle 1]: 0.00016441, [11]
        [c_1]: 2.558e-05
        [parameter_eliminate]: 2.93998e-06
        [updatestate_depend_eliminate]: 6.39999e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.82002e-06
        [cse]: 2.696e-05
        [call_graph_tuple_transform]: 2.259e-05
        [tuple_list_get_item_eliminator]: 6.21e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 9.00007e-07
        [switch_simplify]: 6.77002e-06
[remove_dup_value]: 1.48e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 6.24999e-06
[add_recomputation]: 5.009e-05
[cse_after_recomputation]: 2.474e-05, [1]
    [Cycle 1]: 1.905e-05, [1]
        [cse]: 1.279e-05
[auto_monad_reorder]: 1.778e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 5.79e-06
[opt_after_jit_grad]: 0.00049647
[symbol_engine_optimizer]: 8.369e-05, [1]
    [Cycle 1]: 7.689e-05, [6]
        [build]: 4.01001e-06
        [elim_shapecalc]: 8.38001e-06
        [elim_not_effective]: 1.534e-05
        [opt_reshape]: 6.87002e-06
        [fold_const_symbol]: 1.131e-05
        [renormalize]: 8.10018e-07
[validate]: 4.011e-05
Sums
    bootstrap                            :     0.000265s :  2.92%
    type_inference                       :     0.004726s : 52.00%
    event_method                         :     0.000022s :  0.24%
    auto_monad                           :     0.000064s :  0.70%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.51%
    rewriter_before_opt_a                :     0.000089s :  0.98%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000071s :  0.78%
    jit_opt_a.loop_unroll                :     0.000043s :  0.47%
    jit_opt_a.a_1                        :     0.000764s :  8.40%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.16%
    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 :     0.000005s :  0.06%
    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 :     0.000012s :  0.14%
    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_        :     0.000012s :  0.13%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  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 :  0.05%
    jit_opt_a.replace_old_param          :     0.000021s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000732s :  8.05%
    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      :     0.000022s :  0.24%
    jit_opt_a.cse                        :     0.000045s :  0.50%
    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 :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000557s :  6.13%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.21%
    cconv                                :     0.000026s :  0.29%
    loop_unroll                          :     0.000465s :  5.12%
    jit_opt_after_cconv.c_1              :     0.000026s :  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 :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000027s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  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      :     0.000001s :  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 :  0.07%
    add_recomputation                    :     0.000050s :  0.55%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000018s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000496s :  5.46%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    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 :     0.000011s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.44%


Time group info:
------[substitution.]   0.000222    28
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.82% :     0.000006s :      4: substitution.graph_param_transform
81.94% :     0.000182s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.28% :     0.000005s :      4: substitution.replace_old_param
 7.72% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004664     2
79.17% :     0.003693s :      1: type_inference.infer
20.83% :     0.000972s :      1: type_inference.specialize
------[replace.]   0.000088     8
54.50% :     0.000048s :      4: replace.inline
45.50% :     0.000040s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000194     8
92.28% :     0.000179s :      4: match.inline
 7.72% :     0.000015s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   912
 1.29% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     13: predicate.addn_check_dump
 1.25% :     0.000002s :     13: predicate.addn_zero_filter
 1.88% :     0.000003s :     13: predicate.arithmetic_simplify
 1.33% :     0.000002s :     13: predicate.cast_eliminate
 0.48% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     13: predicate.compare_switch_simplify
 1.11% :     0.000002s :     13: predicate.depend_value_elim
 1.15% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      4: predicate.elim_not_effective
 0.55% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     13: predicate.environ_get_eliminate
 1.20% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 1.00% :     0.000001s :      8: predicate.get_grad_eliminate
 0.24% :     0.000000s :      4: predicate.graph_param_transform
 5.48% :     0.000008s :     29: predicate.inline
 0.95% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 0.97% :     0.000001s :      8: predicate.less_batch_normalization
 1.55% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.02% :     0.000003s :     21: predicate.load_eliminater
 1.50% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.14% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     13: predicate.merge_addn
 1.12% :     0.000002s :     13: predicate.minmaximum_grad
 1.60% :     0.000002s :      4: predicate.mutable_eliminate
 0.48% :     0.000001s :      4: predicate.opt_reshape
 2.65% :     0.000004s :     21: predicate.partial_eliminate
 1.15% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     13: predicate.reduce_eliminate
 1.57% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.94% :     0.000003s :     25: predicate.replace_applicator
 0.80% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000001s :      4: predicate.reset_defer_inline
 1.21% :     0.000002s :     13: predicate.reshape_eliminate
 1.20% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     13: predicate.same_eliminate
 0.64% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      8: predicate.special_op_eliminate
 0.90% :     0.000001s :      8: predicate.specialize_transform
 1.37% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.46% :     0.000004s :     21: predicate.switch_defer_inline
 2.09% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.83% :     0.000011s :     67: predicate.switch_simplify
 1.22% :     0.000002s :     13: predicate.tile_eliminate
 1.17% :     0.000002s :     13: predicate.transpose_eliminate
 1.55% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.80% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     13: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000661    12
47.82% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.48% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.70% :     0.000315s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014836    72
 0.36% :     0.000053s :      1: add_recomputation
 0.45% :     0.000067s :      1: auto_monad
 0.14% :     0.000020s :      1: auto_monad_reorder
 1.87% :     0.000278s :      1: bootstrap
 0.20% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000027s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.17% :     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
37.30% :     0.005534s :      1: jit_opt_a
 1.17% :     0.000174s :      1: jit_opt_after_cconv
 0.42% :     0.000063s :      1: jit_opt_b
 3.19% :     0.000474s :      1: loop_unroll
 3.83% :     0.000568s :      1: mutable_eliminate
 6.82% :     0.001012s :     26: opt.transform.jit_opt_a
 0.39% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000033s :      4: opt.transform.jit_opt_b
 0.10% :     0.000015s :      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
 0.26% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.41% :     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% :     0.000005s :      1: pre_auto_parallel
 0.34% :     0.000050s :      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
 2.57% :     0.000382s :      1: renormalize.infer
 2.30% :     0.000342s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000038s :      1: rewriter_after_opt_a
 0.62% :     0.000092s :      1: rewriter_before_opt_a
 0.59% :     0.000087s :      1: symbol_engine_optimizer
31.94% :     0.004739s :      1: type_inference


TotalTime = 0.0416014, [33]
[bootstrap]: 0.00029418
[type_inference]: 0.0219848
[event_method]: 2.083e-05
[auto_monad]: 0.0001525
[graph_reusing]: 6.59001e-06
[pre_auto_parallel]: 2.89999e-06
[py_interpret_to_execute]: 0.00011877
[rewriter_before_opt_a]: 9.652e-05
[expand_dump_flag]: 3.4e-06
[jit_opt_a]: 0.00604044, [2]
    [Cycle 1]: 0.00213241, [27]
        [switch_simplify]: 6.166e-05
        [loop_unroll]: 3.574e-05
        [a_1]: 0.00066062
        [with_stream_mark]: 1.381e-05
        [recompute_prepare]: 1.039e-05
        [updatestate_depend_eliminate]: 5.67001e-06
        [updatestate_assign_eliminate]: 4.84e-06
        [updatestate_loads_eliminate]: 4.02e-06
        [parameter_eliminate]: 8.30012e-07
        [specialize_transform]: 8.82e-06
        [updatestate_useless_node_eliminater]: 1.11e-05
        [accelerated_algorithm]: 8.94e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 7.95e-06
        [merge_forward]: 3.87998e-06
        [cell_reuse_recompute_pass]: 1.60001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.573e-05
        [j_node_and_user_rematch]: 1.332e-05
        [meta_fg_expand]: 3.14001e-06
        [replace_old_param]: 1.081e-05
        [inline_without_move]: 8.13001e-06
        [renormalize]: 0.0010138
        [add_forward_monad_depend]: 4.33001e-06
        [auto_monad_grad]: 1.24998e-06
        [auto_monad_eliminator]: 1.608e-05
        [cse]: 2.852e-05
        [replace_applicator]: 1.456e-05
    [Cycle 2]: 0.00046105, [27]
        [switch_simplify]: 9.29e-06
        [loop_unroll]: 7.91001e-06
        [a_1]: 0.00016647
        [with_stream_mark]: 1.117e-05
        [recompute_prepare]: 8.26002e-06
        [updatestate_depend_eliminate]: 5.00001e-06
        [updatestate_assign_eliminate]: 4.58999e-06
        [updatestate_loads_eliminate]: 4.17e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 8.38999e-06
        [updatestate_useless_node_eliminater]: 1.062e-05
        [accelerated_algorithm]: 7.80998e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 7.44002e-06
        [merge_forward]: 4.14002e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.587e-05
        [j_node_and_user_rematch]: 1.225e-05
        [meta_fg_expand]: 2.68e-06
        [replace_old_param]: 1.007e-05
        [inline_without_move]: 7.46001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 1.016e-05
        [cse]: 1.998e-05
        [replace_applicator]: 7.6e-06
[py_interpret_to_execute_after_opt_a]: 1.107e-05
[rewriter_after_opt_a]: 0.00027399
[convert_after_rewriter]: 9.77999e-06
[order_py_execute_after_rewriter]: 6.45002e-06
[mutable_eliminate]: 0.00056391
[jit_opt_b]: 6.576e-05, [1]
    [Cycle 1]: 5.946e-05, [2]
        [frontend_op_eliminate]: 2.377e-05
        [inline_after_opt_a]: 2.415e-05
[cconv]: 1.774e-05
[loop_unroll]: 0.00044134
[jit_opt_after_cconv]: 0.00018424, [1]
    [Cycle 1]: 0.00017813, [11]
        [c_1]: 4.359e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 6.82002e-06
        [updatestate_assign_eliminate]: 4.13001e-06
        [updatestate_loads_eliminate]: 4.03999e-06
        [cse]: 2.565e-05
        [call_graph_tuple_transform]: 2.101e-05
        [tuple_list_get_item_eliminator]: 8.49002e-06
        [none_parameter_eliminate]: 7.89994e-07
        [renormalize]: 3.80009e-07
        [switch_simplify]: 8.72e-06
[remove_dup_value]: 1.221e-05
[partial_unused_args_eliminate]: 1.34998e-06
[environ_conv]: 8.70001e-06
[add_recomputation]: 4.97e-05
[cse_after_recomputation]: 3.008e-05, [1]
    [Cycle 1]: 2.456e-05, [1]
        [cse]: 1.812e-05
[auto_monad_reorder]: 2.015e-05
[get_jit_bprop_graph]: 1.32e-06
[rewriter_after_jit_bprop_graph]: 5.61e-06
[opt_after_jit_grad]: 0.00051047
[symbol_engine_optimizer]: 0.00010882, [1]
    [Cycle 1]: 0.00010287, [6]
        [build]: 1.241e-05
        [elim_shapecalc]: 1.278e-05
        [elim_not_effective]: 2.172e-05
        [opt_reshape]: 1.036e-05
        [fold_const_symbol]: 1.436e-05
        [renormalize]: 3.80009e-07
[validate]: 4.926e-05
[backend_pass]: 7.79983e-07
[task_emit]: 0.0102867
[execute]: 8.28001e-06
Sums
    bootstrap                            :     0.000294s :  0.78%
    type_inference                       :     0.021985s : 58.56%
    event_method                         :     0.000021s :  0.06%
    auto_monad                           :     0.000152s :  0.41%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000119s :  0.32%
    rewriter_before_opt_a                :     0.000097s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.19%
    jit_opt_a.loop_unroll                :     0.000044s :  0.12%
    jit_opt_a.a_1                        :     0.000827s :  2.20%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  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 :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001014s :  2.70%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.07%
    jit_opt_a.cse                        :     0.000048s :  0.13%
    jit_opt_a.replace_applicator         :     0.000022s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000274s :  0.73%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000564s :  1.50%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000018s :  0.05%
    loop_unroll                          :     0.000441s :  1.18%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.12%
    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              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000012s :  0.03%
    partial_unused_args_eliminate        :     0.000001s :  0.00%
    environ_conv                         :     0.000009s :  0.02%
    add_recomputation                    :     0.000050s :  0.13%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000020s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000510s :  1.36%
    symbol_engine_optimizer.build        :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  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  :     0.000000s :  0.00%
    validate                             :     0.000049s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010287s : 27.40%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000231    47
 3.40% :     0.000008s :      2: substitution.depend_value_elim
 1.21% :     0.000003s :      4: substitution.elim_not_effective
 0.74% :     0.000002s :      4: substitution.fold_const_symbol
 2.32% :     0.000005s :      5: substitution.graph_param_transform
75.49% :     0.000174s :      4: substitution.inline
 1.83% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.75% :     0.000006s :      8: substitution.remove_not_recompute_node
 1.52% :     0.000003s :      2: substitution.replace_old_param
 3.38% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
 3.73% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.63% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021919     2
93.22% :     0.020433s :      1: type_inference.infer
 6.78% :     0.001486s :      1: type_inference.specialize
------[replace.]   0.000086     6
64.77% :     0.000056s :      4: replace.inline
35.23% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     6
96.24% :     0.000171s :      4: match.inline
 3.76% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1021
 1.49% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.40% :     0.000002s :     15: predicate.addn_zero_filter
 2.09% :     0.000004s :     15: predicate.arithmetic_simplify
 1.66% :     0.000003s :     15: predicate.cast_eliminate
 0.54% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     15: predicate.compare_switch_simplify
 1.42% :     0.000002s :     15: predicate.depend_value_elim
 1.15% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.49% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      5: predicate.elim_not_effective
 0.78% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.31% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     15: predicate.environ_get_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.29% :     0.000000s :      5: predicate.graph_param_transform
 4.85% :     0.000008s :     31: predicate.inline
 1.10% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.57% :     0.000003s :     10: predicate.less_batch_normalization
 1.57% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     22: predicate.load_eliminater
 1.24% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.42% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.11% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 1.47% :     0.000002s :      5: predicate.mutable_eliminate
 0.70% :     0.000001s :      5: predicate.opt_reshape
 2.35% :     0.000004s :     22: predicate.partial_eliminate
 1.32% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.65% :     0.000003s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.65% :     0.000003s :     27: predicate.replace_applicator
 0.57% :     0.000001s :     10: predicate.replace_old_param
 0.32% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000002s :     15: predicate.reshape_eliminate
 1.35% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.11% :     0.000002s :     10: predicate.special_op_eliminate
 1.07% :     0.000002s :     10: predicate.specialize_transform
 1.38% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.32% :     0.000004s :     21: predicate.switch_defer_inline
 1.96% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.62% :     0.000011s :     68: predicate.switch_simplify
 1.28% :     0.000002s :     15: predicate.tile_eliminate
 1.25% :     0.000002s :     15: predicate.transpose_eliminate
 1.53% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.81% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.21% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     15: predicate.value_based_eliminate
 0.46% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000781    13
43.99% :     0.000343s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.61% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.40% :     0.000401s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043927    76
 0.12% :     0.000053s :      1: add_recomputation
 0.36% :     0.000157s :      1: auto_monad
 0.05% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.70% :     0.000307s :      1: bootstrap
 0.05% :     0.000021s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000011s :      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% :     0.000009s :      1: graph_reusing
13.76% :     0.006044s :      1: jit_opt_a
 0.43% :     0.000187s :      1: jit_opt_after_cconv
 0.16% :     0.000068s :      1: jit_opt_b
 1.02% :     0.000448s :      1: loop_unroll
 1.30% :     0.000572s :      1: mutable_eliminate
 2.55% :     0.001122s :     26: opt.transform.jit_opt_a
 0.18% :     0.000078s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      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
 0.13% :     0.000055s :      4: opt.transform.symbol_engine_opt
 1.18% :     0.000519s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000123s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000015s :      1: remove_dup_value
 1.38% :     0.000606s :      1: renormalize.infer
 0.91% :     0.000402s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000279s :      1: rewriter_after_opt_a
 0.23% :     0.000101s :      1: rewriter_before_opt_a
 0.25% :     0.000112s :      1: symbol_engine_optimizer
23.46% :     0.010306s :      1: task_emit
50.08% :     0.021997s :      1: type_inference
 0.16% :     0.000070s :      1: validate


TotalTime = 0.0134168, [30]
[bootstrap]: 0.00028189
[type_inference]: 0.00491478
[event_method]: 2.227e-05
[auto_monad]: 6.825e-05
[graph_reusing]: 5.61998e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 5.211e-05
[rewriter_before_opt_a]: 0.00010124
[expand_dump_flag]: 3.88001e-06
[jit_opt_a]: 0.00563847, [2]
    [Cycle 1]: 0.00195262, [27]
        [switch_simplify]: 6.224e-05
        [loop_unroll]: 4.2e-05
        [a_1]: 0.00070323
        [with_stream_mark]: 1.878e-05
        [recompute_prepare]: 1.045e-05
        [updatestate_depend_eliminate]: 4.60001e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 8.63001e-06
        [updatestate_useless_node_eliminater]: 6.99001e-06
        [accelerated_algorithm]: 7.42998e-06
        [meta_shard_fg_expand]: 2.22999e-06
        [get_grad_eliminate_]: 6.57002e-06
        [merge_forward]: 4.35e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.694e-05
        [j_node_and_user_rematch]: 1.119e-05
        [meta_fg_expand]: 2.58e-06
        [replace_old_param]: 1.169e-05
        [inline_without_move]: 6.68998e-06
        [renormalize]: 0.00075025
        [add_forward_monad_depend]: 6.01e-06
        [auto_monad_grad]: 2.80002e-06
        [auto_monad_eliminator]: 1.593e-05
        [cse]: 3.471e-05
        [replace_applicator]: 1.478e-05
    [Cycle 2]: 0.00037956, [27]
        [switch_simplify]: 7.01001e-06
        [loop_unroll]: 6.26e-06
        [a_1]: 0.00011829
        [with_stream_mark]: 1.053e-05
        [recompute_prepare]: 6.43e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 5.97999e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 5.90002e-06
        [merge_forward]: 3.03998e-06
        [cell_reuse_recompute_pass]: 2.17001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.56e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 1.083e-05
        [inline_without_move]: 5.56e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 6.57002e-06
        [cse]: 1.399e-05
        [replace_applicator]: 5.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.407e-05
[rewriter_after_opt_a]: 3.815e-05
[convert_after_rewriter]: 7.39002e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00062069
[jit_opt_b]: 6.043e-05, [1]
    [Cycle 1]: 5.215e-05, [2]
        [frontend_op_eliminate]: 2.028e-05
        [inline_after_opt_a]: 1.867e-05
[cconv]: 2.585e-05
[loop_unroll]: 0.00045965
[jit_opt_after_cconv]: 0.00016448, [1]
    [Cycle 1]: 0.00015743, [11]
        [c_1]: 2.6e-05
        [parameter_eliminate]: 2.86e-06
        [updatestate_depend_eliminate]: 5.84999e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [cse]: 2.576e-05
        [call_graph_tuple_transform]: 2.174e-05
        [tuple_list_get_item_eliminator]: 6.42001e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 5.29981e-07
        [switch_simplify]: 7.15e-06
[remove_dup_value]: 1.603e-05
[partial_unused_args_eliminate]: 2.64999e-06
[environ_conv]: 6.21998e-06
[add_recomputation]: 5.09e-05
[cse_after_recomputation]: 2.662e-05, [1]
    [Cycle 1]: 2.07e-05, [1]
        [cse]: 1.482e-05
[auto_monad_reorder]: 1.863e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 5.74e-06
[opt_after_jit_grad]: 0.00049375
[symbol_engine_optimizer]: 8.18e-05, [1]
    [Cycle 1]: 7.364e-05, [6]
        [build]: 3.85998e-06
        [elim_shapecalc]: 8.70999e-06
        [elim_not_effective]: 1.473e-05
        [opt_reshape]: 6.53998e-06
        [fold_const_symbol]: 9.75002e-06
        [renormalize]: 5.19998e-07
[validate]: 3.944e-05
Sums
    bootstrap                            :     0.000282s :  2.97%
    type_inference                       :     0.004915s : 51.84%
    event_method                         :     0.000022s :  0.23%
    auto_monad                           :     0.000068s :  0.72%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000052s :  0.55%
    rewriter_before_opt_a                :     0.000101s :  1.07%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000069s :  0.73%
    jit_opt_a.loop_unroll                :     0.000048s :  0.51%
    jit_opt_a.a_1                        :     0.000822s :  8.66%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.09%
    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       :     0.000015s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.13%
    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 :     0.000033s :  0.34%
    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          :     0.000023s :  0.24%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000750s :  7.91%
    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      :     0.000023s :  0.24%
    jit_opt_a.cse                        :     0.000049s :  0.51%
    jit_opt_a.replace_applicator         :     0.000021s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.15%
    rewriter_after_opt_a                 :     0.000038s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000621s :  6.55%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.20%
    cconv                                :     0.000026s :  0.27%
    loop_unroll                          :     0.000460s :  4.85%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.27%
    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 :  0.03%
    jit_opt_after_cconv.cse              :     0.000026s :  0.27%
    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 :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.08%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000051s :  0.54%
    cse_after_recomputation.cse          :     0.000015s :  0.16%
    auto_monad_reorder                   :     0.000019s :  0.20%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000494s :  5.21%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    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                             :     0.000039s :  0.42%


Time group info:
------[substitution.]   0.000230    28
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.76% :     0.000006s :      4: substitution.graph_param_transform
81.33% :     0.000187s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.50% :     0.000006s :      4: substitution.replace_old_param
 8.34% :     0.000019s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004847     2
78.66% :     0.003813s :      1: type_inference.infer
21.34% :     0.001034s :      1: type_inference.specialize
------[replace.]   0.000094     8
54.65% :     0.000052s :      4: replace.inline
45.35% :     0.000043s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000201     8
91.50% :     0.000184s :      4: match.inline
 8.50% :     0.000017s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000152   912
 1.13% :     0.000002s :     13: predicate.accumulaten_eliminater
 0.99% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.05% :     0.000002s :     13: predicate.addn_check_dump
 1.44% :     0.000002s :     13: predicate.addn_zero_filter
 2.21% :     0.000003s :     13: predicate.arithmetic_simplify
 1.44% :     0.000002s :     13: predicate.cast_eliminate
 0.39% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.02% :     0.000002s :     13: predicate.compare_switch_simplify
 1.13% :     0.000002s :     13: predicate.depend_value_elim
 1.11% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      4: predicate.elim_not_effective
 0.43% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     13: predicate.environ_get_eliminate
 1.15% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      4: predicate.fold_const_symbol
 0.97% :     0.000001s :      8: predicate.get_grad_eliminate
 0.34% :     0.000001s :      4: predicate.graph_param_transform
 5.42% :     0.000008s :     29: predicate.inline
 0.84% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.08% :     0.000002s :      8: predicate.less_batch_normalization
 1.59% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.90% :     0.000003s :     21: predicate.load_eliminater
 1.06% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.57% :     0.000007s :     42: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     13: predicate.merge_addn
 1.08% :     0.000002s :     13: predicate.minmaximum_grad
 1.43% :     0.000002s :      4: predicate.mutable_eliminate
 0.41% :     0.000001s :      4: predicate.opt_reshape
 2.40% :     0.000004s :     21: predicate.partial_eliminate
 1.13% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.83% :     0.000003s :     13: predicate.reduce_eliminate
 1.61% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.87% :     0.000003s :     25: predicate.replace_applicator
 0.86% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000001s :      4: predicate.reset_defer_inline
 1.38% :     0.000002s :     13: predicate.reshape_eliminate
 1.10% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     13: predicate.same_eliminate
 0.63% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      8: predicate.special_op_eliminate
 0.98% :     0.000001s :      8: predicate.specialize_transform
 1.31% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.62% :     0.000004s :     21: predicate.switch_defer_inline
 2.57% :     0.000004s :     21: predicate.switch_layer_defer_inline
 7.57% :     0.000011s :     67: predicate.switch_simplify
 1.26% :     0.000002s :     13: predicate.tile_eliminate
 1.17% :     0.000002s :     13: predicate.transpose_eliminate
 1.34% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 4.25% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000003s :     13: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.80% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000005s :     29: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     13: predicate.value_based_eliminate
 0.35% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000647    12
46.79% :     0.000303s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.82% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.39% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015336    72
 0.35% :     0.000054s :      1: add_recomputation
 0.47% :     0.000072s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 1.93% :     0.000295s :      1: bootstrap
 0.19% :     0.000029s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.19% :     0.000029s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.17% :     0.000027s :      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
36.79% :     0.005642s :      1: jit_opt_a
 1.09% :     0.000167s :      1: jit_opt_after_cconv
 0.50% :     0.000076s :      1: jit_opt_b
 3.05% :     0.000467s :      1: loop_unroll
 4.11% :     0.000630s :      1: mutable_eliminate
 7.09% :     0.001088s :     26: opt.transform.jit_opt_a
 0.38% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000032s :      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
 0.17% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.28% :     0.000503s :      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% :     0.000056s :      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
 2.62% :     0.000402s :      1: renormalize.infer
 2.22% :     0.000340s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000042s :      1: rewriter_after_opt_a
 0.68% :     0.000105s :      1: rewriter_before_opt_a
 0.55% :     0.000085s :      1: symbol_engine_optimizer
32.13% :     0.004927s :      1: type_inference


TotalTime = 0.0131912, [30]
[bootstrap]: 0.00030954
[type_inference]: 0.00520727
[event_method]: 2.252e-05
[auto_monad]: 6.591e-05
[graph_reusing]: 5.62001e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 4.495e-05
[rewriter_before_opt_a]: 9.011e-05
[expand_dump_flag]: 3.43999e-06
[jit_opt_a]: 0.00525735, [2]
    [Cycle 1]: 0.00169475, [27]
        [switch_simplify]: 5.646e-05
        [loop_unroll]: 3.738e-05
        [a_1]: 0.00063114
        [with_stream_mark]: 1.578e-05
        [recompute_prepare]: 8.07998e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 3.49001e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [parameter_eliminate]: 2.32999e-06
        [specialize_transform]: 6.75998e-06
        [updatestate_useless_node_eliminater]: 6.01e-06
        [accelerated_algorithm]: 6.38e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 6.19001e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.522e-05
        [j_node_and_user_rematch]: 9.86e-06
        [meta_fg_expand]: 2.59001e-06
        [replace_old_param]: 1.139e-05
        [inline_without_move]: 6.24999e-06
        [renormalize]: 0.00061396
        [add_forward_monad_depend]: 6.15002e-06
        [auto_monad_grad]: 1.88997e-06
        [auto_monad_eliminator]: 1.542e-05
        [cse]: 3.4e-05
        [replace_applicator]: 1.339e-05
    [Cycle 2]: 0.00037419, [27]
        [switch_simplify]: 6.99001e-06
        [loop_unroll]: 5.90002e-06
        [a_1]: 0.00011518
        [with_stream_mark]: 1.056e-05
        [recompute_prepare]: 6.56e-06
        [updatestate_depend_eliminate]: 3.30998e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.89e-06
        [accelerated_algorithm]: 6.43e-06
        [meta_shard_fg_expand]: 1.37999e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.309e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 1.87999e-06
        [replace_old_param]: 1.026e-05
        [inline_without_move]: 5.69e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 9.10019e-07
        [auto_monad_eliminator]: 7.2e-06
        [cse]: 1.682e-05
        [replace_applicator]: 6.28998e-06
[py_interpret_to_execute_after_opt_a]: 1.096e-05
[rewriter_after_opt_a]: 3.632e-05
[convert_after_rewriter]: 7.51001e-06
[order_py_execute_after_rewriter]: 5.69e-06
[mutable_eliminate]: 0.00052238
[jit_opt_b]: 5.505e-05, [1]
    [Cycle 1]: 4.778e-05, [2]
        [frontend_op_eliminate]: 1.838e-05
        [inline_after_opt_a]: 1.77e-05
[cconv]: 2.568e-05
[loop_unroll]: 0.00046766
[jit_opt_after_cconv]: 0.00016284, [1]
    [Cycle 1]: 0.00015541, [11]
        [c_1]: 2.636e-05
        [parameter_eliminate]: 2.57001e-06
        [updatestate_depend_eliminate]: 5.97001e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [cse]: 2.443e-05
        [call_graph_tuple_transform]: 2.135e-05
        [tuple_list_get_item_eliminator]: 6.30002e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 6.62002e-06
[remove_dup_value]: 1.637e-05
[partial_unused_args_eliminate]: 2.76e-06
[environ_conv]: 5.76e-06
[add_recomputation]: 5.159e-05
[cse_after_recomputation]: 2.483e-05, [1]
    [Cycle 1]: 1.907e-05, [1]
        [cse]: 1.305e-05
[auto_monad_reorder]: 1.918e-05
[get_jit_bprop_graph]: 1.66e-06
[rewriter_after_jit_bprop_graph]: 5.94999e-06
[opt_after_jit_grad]: 0.00046612
[symbol_engine_optimizer]: 7.549e-05, [1]
    [Cycle 1]: 6.952e-05, [6]
        [build]: 3.45e-06
        [elim_shapecalc]: 8.57e-06
        [elim_not_effective]: 1.408e-05
        [opt_reshape]: 6.29999e-06
        [fold_const_symbol]: 9.48997e-06
        [renormalize]: 4.40021e-07
[validate]: 3.971e-05
Sums
    bootstrap                            :     0.000310s :  3.29%
    type_inference                       :     0.005207s : 55.33%
    event_method                         :     0.000023s :  0.24%
    auto_monad                           :     0.000066s :  0.70%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.48%
    rewriter_before_opt_a                :     0.000090s :  0.96%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.67%
    jit_opt_a.loop_unroll                :     0.000043s :  0.46%
    jit_opt_a.a_1                        :     0.000746s :  7.93%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000015s :  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 :     0.000005s :  0.06%
    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 :     0.000012s :  0.13%
    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_        :     0.000012s :  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 :     0.000028s :  0.30%
    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          :     0.000022s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000614s :  6.52%
    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      :     0.000023s :  0.24%
    jit_opt_a.cse                        :     0.000051s :  0.54%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000036s :  0.39%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000522s :  5.55%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.19%
    cconv                                :     0.000026s :  0.27%
    loop_unroll                          :     0.000468s :  4.97%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.28%
    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 :  0.03%
    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 :     0.000006s :  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  :     0.000007s :  0.07%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000052s :  0.55%
    cse_after_recomputation.cse          :     0.000013s :  0.14%
    auto_monad_reorder                   :     0.000019s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000466s :  4.95%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.42%


Time group info:
------[substitution.]   0.000212    28
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.91% :     0.000006s :      4: substitution.graph_param_transform
80.77% :     0.000171s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000005s :      4: substitution.remove_not_recompute_node
 3.04% :     0.000006s :      4: substitution.replace_old_param
 7.77% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005138     2
80.59% :     0.004141s :      1: type_inference.infer
19.41% :     0.000997s :      1: type_inference.specialize
------[replace.]   0.000087     8
56.68% :     0.000049s :      4: replace.inline
43.32% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     8
92.12% :     0.000168s :      4: match.inline
 7.88% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.41% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.13% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     13: predicate.addn_check_dump
 1.25% :     0.000002s :     13: predicate.addn_zero_filter
 1.80% :     0.000002s :     13: predicate.arithmetic_simplify
 1.26% :     0.000002s :     13: predicate.cast_eliminate
 0.46% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     13: predicate.compare_switch_simplify
 1.12% :     0.000002s :     13: predicate.depend_value_elim
 1.12% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.17% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      4: predicate.elim_not_effective
 0.66% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     13: predicate.environ_get_eliminate
 1.17% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.95% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.68% :     0.000008s :     29: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      8: predicate.less_batch_normalization
 1.56% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.96% :     0.000003s :     21: predicate.load_eliminater
 1.41% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.45% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     13: predicate.merge_addn
 1.14% :     0.000002s :     13: predicate.minmaximum_grad
 1.46% :     0.000002s :      4: predicate.mutable_eliminate
 0.49% :     0.000001s :      4: predicate.opt_reshape
 2.67% :     0.000004s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     13: predicate.reduce_eliminate
 1.54% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.96% :     0.000003s :     25: predicate.replace_applicator
 0.86% :     0.000001s :      8: predicate.replace_old_param
 0.30% :     0.000000s :      4: predicate.reset_defer_inline
 1.17% :     0.000002s :     13: predicate.reshape_eliminate
 1.20% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     13: predicate.same_eliminate
 0.62% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      8: predicate.special_op_eliminate
 0.97% :     0.000001s :      8: predicate.specialize_transform
 1.40% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.56% :     0.000004s :     21: predicate.switch_defer_inline
 2.12% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.74% :     0.000011s :     67: predicate.switch_simplify
 1.41% :     0.000002s :     13: predicate.tile_eliminate
 1.22% :     0.000002s :     13: predicate.transpose_eliminate
 1.56% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.81% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     13: predicate.value_based_eliminate
 0.33% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000646    12
50.93% :     0.000329s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.33% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.74% :     0.000289s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014867    72
 0.37% :     0.000054s :      1: add_recomputation
 0.47% :     0.000070s :      1: auto_monad
 0.15% :     0.000022s :      1: auto_monad_reorder
 2.19% :     0.000325s :      1: bootstrap
 0.19% :     0.000028s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000027s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.18% :     0.000027s :      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
35.38% :     0.005260s :      1: jit_opt_a
 1.12% :     0.000166s :      1: jit_opt_after_cconv
 0.39% :     0.000058s :      1: jit_opt_b
 3.20% :     0.000476s :      1: loop_unroll
 3.58% :     0.000532s :      1: mutable_eliminate
 6.66% :     0.000990s :     26: opt.transform.jit_opt_a
 0.39% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000030s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.19% :     0.000474s :      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
 0.33% :     0.000048s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000019s :      1: remove_dup_value
 2.13% :     0.000317s :      1: renormalize.infer
 1.94% :     0.000288s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000040s :      1: rewriter_after_opt_a
 0.63% :     0.000093s :      1: rewriter_before_opt_a
 0.53% :     0.000078s :      1: symbol_engine_optimizer
35.12% :     0.005222s :      1: type_inference


TotalTime = 0.0448196, [33]
[bootstrap]: 0.00028819
[type_inference]: 0.023236
[event_method]: 2.135e-05
[auto_monad]: 0.00015671
[graph_reusing]: 6.43e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 0.00011843
[rewriter_before_opt_a]: 0.0001014
[expand_dump_flag]: 3.31999e-06
[jit_opt_a]: 0.00626185, [2]
    [Cycle 1]: 0.00242591, [27]
        [switch_simplify]: 6.504e-05
        [loop_unroll]: 3.661e-05
        [a_1]: 0.00071169
        [with_stream_mark]: 7.146e-05
        [recompute_prepare]: 1.173e-05
        [updatestate_depend_eliminate]: 7.01001e-06
        [updatestate_assign_eliminate]: 5.69e-06
        [updatestate_loads_eliminate]: 5.29998e-06
        [parameter_eliminate]: 2.12001e-06
        [specialize_transform]: 9.66e-06
        [updatestate_useless_node_eliminater]: 1.19e-05
        [accelerated_algorithm]: 8.87e-06
        [meta_shard_fg_expand]: 2.94001e-06
        [get_grad_eliminate_]: 7.85e-06
        [merge_forward]: 5.65001e-06
        [cell_reuse_recompute_pass]: 1.50001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.054e-05
        [j_node_and_user_rematch]: 1.392e-05
        [meta_fg_expand]: 3.74002e-06
        [replace_old_param]: 1.213e-05
        [inline_without_move]: 7.91001e-06
        [renormalize]: 0.00113027
        [add_forward_monad_depend]: 6.17001e-06
        [auto_monad_grad]: 2.87002e-06
        [auto_monad_eliminator]: 2.077e-05
        [cse]: 4.057e-05
        [replace_applicator]: 1.723e-05
    [Cycle 2]: 0.00046969, [27]
        [switch_simplify]: 8.94998e-06
        [loop_unroll]: 7.97e-06
        [a_1]: 0.0001681
        [with_stream_mark]: 1.175e-05
        [recompute_prepare]: 8.15e-06
        [updatestate_depend_eliminate]: 4.95001e-06
        [updatestate_assign_eliminate]: 4.63999e-06
        [updatestate_loads_eliminate]: 3.85e-06
        [parameter_eliminate]: 1.30001e-06
        [specialize_transform]: 8.07e-06
        [updatestate_useless_node_eliminater]: 1.141e-05
        [accelerated_algorithm]: 7.93999e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 7.46999e-06
        [merge_forward]: 4.67e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.709e-05
        [j_node_and_user_rematch]: 1.432e-05
        [meta_fg_expand]: 2.79999e-06
        [replace_old_param]: 1.037e-05
        [inline_without_move]: 7.38e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.19998e-06
        [auto_monad_grad]: 9.80013e-07
        [auto_monad_eliminator]: 1.038e-05
        [cse]: 2.037e-05
        [replace_applicator]: 7.93001e-06
[py_interpret_to_execute_after_opt_a]: 1.393e-05
[rewriter_after_opt_a]: 0.00028573
[convert_after_rewriter]: 1.174e-05
[order_py_execute_after_rewriter]: 8.02e-06
[mutable_eliminate]: 0.00069537
[jit_opt_b]: 6.926e-05, [1]
    [Cycle 1]: 6.176e-05, [2]
        [frontend_op_eliminate]: 2.462e-05
        [inline_after_opt_a]: 2.468e-05
[cconv]: 2.904e-05
[loop_unroll]: 0.00047126
[jit_opt_after_cconv]: 0.00020756, [1]
    [Cycle 1]: 0.00020136, [11]
        [c_1]: 4.735e-05
        [parameter_eliminate]: 3.36999e-06
        [updatestate_depend_eliminate]: 9.91998e-06
        [updatestate_assign_eliminate]: 4.52e-06
        [updatestate_loads_eliminate]: 4.37998e-06
        [cse]: 3.532e-05
        [call_graph_tuple_transform]: 2.331e-05
        [tuple_list_get_item_eliminator]: 8.27e-06
        [none_parameter_eliminate]: 1.61002e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 9.46998e-06
[remove_dup_value]: 1.994e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 1.289e-05
[add_recomputation]: 6.746e-05
[cse_after_recomputation]: 6.222e-05, [1]
    [Cycle 1]: 5.632e-05, [1]
        [cse]: 4.934e-05
[auto_monad_reorder]: 2.78e-05
[get_jit_bprop_graph]: 2.09999e-06
[rewriter_after_jit_bprop_graph]: 6.15002e-06
[opt_after_jit_grad]: 0.00051867
[symbol_engine_optimizer]: 0.00011058, [1]
    [Cycle 1]: 0.00010387, [6]
        [build]: 1.678e-05
        [elim_shapecalc]: 1.234e-05
        [elim_not_effective]: 1.944e-05
        [opt_reshape]: 9.42001e-06
        [fold_const_symbol]: 1.436e-05
        [renormalize]: 4.59986e-07
[validate]: 6.049e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.0116492
[execute]: 9.60001e-06
Sums
    bootstrap                            :     0.000288s :  0.71%
    type_inference                       :     0.023236s : 57.03%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000157s :  0.38%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000118s :  0.29%
    rewriter_before_opt_a                :     0.000101s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.18%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000880s :  2.16%
    jit_opt_a.with_stream_mark           :     0.000083s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001130s :  2.77%
    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      :     0.000031s :  0.08%
    jit_opt_a.cse                        :     0.000061s :  0.15%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000286s :  0.70%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000695s :  1.71%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000029s :  0.07%
    loop_unroll                          :     0.000471s :  1.16%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  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              :     0.000035s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000067s :  0.17%
    cse_after_recomputation.cse          :     0.000049s :  0.12%
    auto_monad_reorder                   :     0.000028s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000519s :  1.27%
    symbol_engine_optimizer.build        :     0.000017s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000060s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011649s : 28.59%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000280    47
 3.36% :     0.000009s :      2: substitution.depend_value_elim
 1.04% :     0.000003s :      4: substitution.elim_not_effective
 0.81% :     0.000002s :      4: substitution.fold_const_symbol
 2.42% :     0.000007s :      5: substitution.graph_param_transform
74.40% :     0.000208s :      4: substitution.inline
 1.83% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.79% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.92% :     0.000005s :      2: substitution.replace_old_param
 4.57% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.68% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 3.17% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023162     2
93.55% :     0.021668s :      1: type_inference.infer
 6.45% :     0.001494s :      1: type_inference.specialize
------[replace.]   0.000091     6
67.80% :     0.000062s :      4: replace.inline
32.20% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000217     6
94.65% :     0.000205s :      4: match.inline
 5.35% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000174  1021
 1.52% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.29% :     0.000002s :     15: predicate.addn_zero_filter
 2.16% :     0.000004s :     15: predicate.arithmetic_simplify
 1.25% :     0.000002s :     15: predicate.cast_eliminate
 0.49% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.29% :     0.000002s :     15: predicate.depend_value_elim
 1.18% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.09% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.84% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     15: predicate.environ_get_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.06% :     0.000002s :     10: predicate.get_grad_eliminate
 0.48% :     0.000001s :      5: predicate.graph_param_transform
 5.34% :     0.000009s :     31: predicate.inline
 0.94% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :     10: predicate.less_batch_normalization
 1.47% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     22: predicate.load_eliminater
 1.30% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.29% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     15: predicate.merge_addn
 1.14% :     0.000002s :     15: predicate.minmaximum_grad
 1.92% :     0.000003s :      5: predicate.mutable_eliminate
 0.47% :     0.000001s :      5: predicate.opt_reshape
 2.29% :     0.000004s :     22: predicate.partial_eliminate
 1.29% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.60% :     0.000003s :     15: predicate.reduce_eliminate
 1.56% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.74% :     0.000003s :     27: predicate.replace_applicator
 0.66% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.28% :     0.000002s :     15: predicate.reshape_eliminate
 1.31% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.07% :     0.000002s :     10: predicate.special_op_eliminate
 1.09% :     0.000002s :     10: predicate.specialize_transform
 1.38% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.59% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.20% :     0.000004s :     21: predicate.switch_defer_inline
 1.90% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.79% :     0.000012s :     68: predicate.switch_simplify
 1.21% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000002s :     15: predicate.transpose_eliminate
 1.56% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.17% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000003s :     15: predicate.value_based_eliminate
 0.50% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.83% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000863    13
45.44% :     0.000392s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.04% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.52% :     0.000427s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047330    76
 0.15% :     0.000071s :      1: add_recomputation
 0.34% :     0.000162s :      1: auto_monad
 0.06% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.72% :     0.000343s :      1: bootstrap
 0.07% :     0.000032s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.14% :     0.000064s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.06% :     0.000026s :      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% :     0.000009s :      1: graph_reusing
13.24% :     0.006265s :      1: jit_opt_a
 0.44% :     0.000210s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 1.01% :     0.000480s :      1: loop_unroll
 1.49% :     0.000705s :      1: mutable_eliminate
 2.51% :     0.001189s :     26: opt.transform.jit_opt_a
 0.18% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.12% :     0.000528s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000122s :      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
 1.41% :     0.000667s :      1: renormalize.infer
 0.96% :     0.000454s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000290s :      1: rewriter_after_opt_a
 0.22% :     0.000106s :      1: rewriter_before_opt_a
 0.24% :     0.000113s :      1: symbol_engine_optimizer
24.66% :     0.011673s :      1: task_emit
49.13% :     0.023254s :      1: type_inference
 0.18% :     0.000086s :      1: validate


TotalTime = 0.0122574, [30]
[bootstrap]: 0.00032328
[type_inference]: 0.0045875
[event_method]: 2.069e-05
[auto_monad]: 4.764e-05
[graph_reusing]: 3.66001e-06
[pre_auto_parallel]: 1.72999e-06
[py_interpret_to_execute]: 4.235e-05
[rewriter_before_opt_a]: 8.323e-05
[expand_dump_flag]: 2.71e-06
[jit_opt_a]: 0.00513545, [2]
    [Cycle 1]: 0.00171092, [27]
        [switch_simplify]: 5.566e-05
        [loop_unroll]: 3.673e-05
        [a_1]: 0.00062874
        [with_stream_mark]: 1.461e-05
        [recompute_prepare]: 8.25999e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.28998e-06
        [parameter_eliminate]: 1.29998e-06
        [specialize_transform]: 6.75002e-06
        [updatestate_useless_node_eliminater]: 6.06e-06
        [accelerated_algorithm]: 6.29001e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 5.92999e-06
        [merge_forward]: 3.28e-06
        [cell_reuse_recompute_pass]: 9.40025e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.218e-05
        [j_node_and_user_rematch]: 9.84999e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 1.055e-05
        [inline_without_move]: 6.71e-06
        [renormalize]: 0.00066228
        [add_forward_monad_depend]: 4.42e-06
        [auto_monad_grad]: 1.75001e-06
        [auto_monad_eliminator]: 1.205e-05
        [cse]: 2.303e-05
        [replace_applicator]: 1.258e-05
    [Cycle 2]: 0.00036741, [27]
        [switch_simplify]: 7.06001e-06
        [loop_unroll]: 6.19001e-06
        [a_1]: 0.00011626
        [with_stream_mark]: 9.67001e-06
        [recompute_prepare]: 6.06998e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 6.06003e-06
        [updatestate_useless_node_eliminater]: 5.80002e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 5.99e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.348e-05
        [j_node_and_user_rematch]: 9.05999e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 9.67001e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24003e-06
        [auto_monad_grad]: 9.79984e-07
        [auto_monad_eliminator]: 6.13998e-06
        [cse]: 1.332e-05
        [replace_applicator]: 6.39999e-06
[py_interpret_to_execute_after_opt_a]: 1.066e-05
[rewriter_after_opt_a]: 3.021e-05
[convert_after_rewriter]: 6.28e-06
[order_py_execute_after_rewriter]: 4.55001e-06
[mutable_eliminate]: 0.00047389
[jit_opt_b]: 5.623e-05, [1]
    [Cycle 1]: 4.899e-05, [2]
        [frontend_op_eliminate]: 1.88e-05
        [inline_after_opt_a]: 1.766e-05
[cconv]: 1.711e-05
[loop_unroll]: 0.00042068
[jit_opt_after_cconv]: 0.00016266, [1]
    [Cycle 1]: 0.00015636, [11]
        [c_1]: 2.593e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 8.98002e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.69e-05
        [call_graph_tuple_transform]: 1.998e-05
        [tuple_list_get_item_eliminator]: 6.71999e-06
        [none_parameter_eliminate]: 1.34998e-06
        [renormalize]: 3.59985e-07
        [switch_simplify]: 6.64999e-06
[remove_dup_value]: 1.167e-05
[partial_unused_args_eliminate]: 1.86e-06
[environ_conv]: 5.92999e-06
[add_recomputation]: 3.39e-05
[cse_after_recomputation]: 2.394e-05, [1]
    [Cycle 1]: 1.832e-05, [1]
        [cse]: 1.24e-05
[auto_monad_reorder]: 1.233e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.48999e-06
[opt_after_jit_grad]: 0.0004476
[symbol_engine_optimizer]: 9.252e-05, [1]
    [Cycle 1]: 8.621e-05, [6]
        [build]: 2.71e-06
        [elim_shapecalc]: 9.39e-06
        [elim_not_effective]: 1.385e-05
        [opt_reshape]: 6.58e-06
        [fold_const_symbol]: 2.026e-05
        [renormalize]: 4.80009e-07
[validate]: 3.197e-05
Sums
    bootstrap                            :     0.000323s :  3.75%
    type_inference                       :     0.004588s : 53.17%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000048s :  0.55%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.49%
    rewriter_before_opt_a                :     0.000083s :  0.96%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000063s :  0.73%
    jit_opt_a.loop_unroll                :     0.000043s :  0.50%
    jit_opt_a.a_1                        :     0.000745s :  8.64%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000014s :  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 :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  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 :  0.30%
    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          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.14%
    jit_opt_a.renormalize                :     0.000662s :  7.68%
    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 :  0.21%
    jit_opt_a.cse                        :     0.000036s :  0.42%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000030s :  0.35%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000474s :  5.49%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000017s :  0.20%
    loop_unroll                          :     0.000421s :  4.88%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.30%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    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 :  0.31%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.08%
    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  :     0.000007s :  0.08%
    remove_dup_value                     :     0.000012s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000034s :  0.39%
    cse_after_recomputation.cse          :     0.000012s :  0.14%
    auto_monad_reorder                   :     0.000012s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000448s :  5.19%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000020s :  0.23%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000032s :  0.37%


Time group info:
------[substitution.]   0.000202    28
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.05% :     0.000004s :      4: substitution.graph_param_transform
83.54% :     0.000169s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.53% :     0.000005s :      4: substitution.replace_old_param
 6.92% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004533     2
79.33% :     0.003596s :      1: type_inference.infer
20.67% :     0.000937s :      1: type_inference.specialize
------[replace.]   0.000084     8
56.05% :     0.000047s :      4: replace.inline
43.95% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     8
93.24% :     0.000166s :      4: match.inline
 6.76% :     0.000012s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   912
 1.19% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     13: predicate.addn_check_dump
 1.28% :     0.000002s :     13: predicate.addn_zero_filter
 1.90% :     0.000003s :     13: predicate.arithmetic_simplify
 1.32% :     0.000002s :     13: predicate.cast_eliminate
 0.49% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     13: predicate.compare_switch_simplify
 1.24% :     0.000002s :     13: predicate.depend_value_elim
 1.16% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      4: predicate.elim_not_effective
 0.61% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     13: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     13: predicate.environ_get_eliminate
 1.12% :     0.000002s :     13: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 0.92% :     0.000001s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 5.11% :     0.000007s :     29: predicate.inline
 0.87% :     0.000001s :      8: predicate.inline_without_move
 0.49% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      8: predicate.less_batch_normalization
 1.56% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.09% :     0.000003s :     21: predicate.load_eliminater
 1.63% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.26% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     17: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     13: predicate.merge_addn
 1.15% :     0.000002s :     13: predicate.minmaximum_grad
 1.30% :     0.000002s :      4: predicate.mutable_eliminate
 0.50% :     0.000001s :      4: predicate.opt_reshape
 2.49% :     0.000003s :     21: predicate.partial_eliminate
 1.18% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     13: predicate.reduce_eliminate
 1.65% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.96% :     0.000003s :     25: predicate.replace_applicator
 0.64% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000000s :      4: predicate.reset_defer_inline
 1.40% :     0.000002s :     13: predicate.reshape_eliminate
 1.21% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     13: predicate.same_eliminate
 0.68% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.38% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.98% :     0.000011s :     67: predicate.switch_simplify
 1.40% :     0.000002s :     13: predicate.tile_eliminate
 1.19% :     0.000002s :     13: predicate.transpose_eliminate
 1.61% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     25: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.21% :     0.000004s :     29: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     13: predicate.value_based_eliminate
 0.41% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
42.87% :     0.000268s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.27% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.85% :     0.000324s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013992    72
 0.26% :     0.000037s :      1: add_recomputation
 0.37% :     0.000051s :      1: auto_monad
 0.11% :     0.000015s :      1: auto_monad_reorder
 2.41% :     0.000337s :      1: bootstrap
 0.14% :     0.000020s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.17% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
36.73% :     0.005139s :      1: jit_opt_a
 1.18% :     0.000166s :      1: jit_opt_after_cconv
 0.42% :     0.000059s :      1: jit_opt_b
 3.06% :     0.000429s :      1: loop_unroll
 3.45% :     0.000482s :      1: mutable_eliminate
 7.04% :     0.000986s :     26: opt.transform.jit_opt_a
 0.40% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000030s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.26% :     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
 0.33% :     0.000046s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000014s :      1: remove_dup_value
 2.50% :     0.000350s :      1: renormalize.infer
 2.18% :     0.000305s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000033s :      1: rewriter_after_opt_a
 0.62% :     0.000086s :      1: rewriter_before_opt_a
 0.68% :     0.000095s :      1: symbol_engine_optimizer
32.87% :     0.004598s :      1: type_inference


TotalTime = 0.0123607, [30]
[bootstrap]: 0.00026128
[type_inference]: 0.00458828
[event_method]: 2.075e-05
[auto_monad]: 6.175e-05
[graph_reusing]: 5.04e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 4.643e-05
[rewriter_before_opt_a]: 8.62e-05
[expand_dump_flag]: 3.03998e-06
[jit_opt_a]: 0.00526397, [2]
    [Cycle 1]: 0.00176095, [27]
        [switch_simplify]: 5.765e-05
        [loop_unroll]: 3.724e-05
        [a_1]: 0.00064172
        [with_stream_mark]: 1.56e-05
        [recompute_prepare]: 8.31002e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 1.56002e-06
        [specialize_transform]: 7.2e-06
        [updatestate_useless_node_eliminater]: 6.32001e-06
        [accelerated_algorithm]: 6.23e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 5.85002e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.417e-05
        [j_node_and_user_rematch]: 9.72999e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 1.094e-05
        [inline_without_move]: 6.11e-06
        [renormalize]: 0.00070443
        [add_forward_monad_depend]: 5.17999e-06
        [auto_monad_grad]: 2.24999e-06
        [auto_monad_eliminator]: 1.368e-05
        [cse]: 2.072e-05
        [replace_applicator]: 1.202e-05
    [Cycle 2]: 0.00036021, [27]
        [switch_simplify]: 6.47001e-06
        [loop_unroll]: 5.87001e-06
        [a_1]: 0.00011558
        [with_stream_mark]: 8.64e-06
        [recompute_prepare]: 5.87999e-06
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.92999e-06
        [accelerated_algorithm]: 6.24999e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.33e-05
        [j_node_and_user_rematch]: 9.25001e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 9.07001e-06
        [inline_without_move]: 5.52999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 5.89e-06
        [cse]: 1.331e-05
        [replace_applicator]: 6.10002e-06
[py_interpret_to_execute_after_opt_a]: 1.105e-05
[rewriter_after_opt_a]: 3.342e-05
[convert_after_rewriter]: 7.38e-06
[order_py_execute_after_rewriter]: 5.62001e-06
[mutable_eliminate]: 0.00047702
[jit_opt_b]: 5.564e-05, [1]
    [Cycle 1]: 4.904e-05, [2]
        [frontend_op_eliminate]: 1.904e-05
        [inline_after_opt_a]: 1.878e-05
[cconv]: 2.563e-05
[loop_unroll]: 0.00041688
[jit_opt_after_cconv]: 0.00015389, [1]
    [Cycle 1]: 0.00014712, [11]
        [c_1]: 2.549e-05
        [parameter_eliminate]: 2.84001e-06
        [updatestate_depend_eliminate]: 5.79999e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 2.102e-05
        [call_graph_tuple_transform]: 2.065e-05
        [tuple_list_get_item_eliminator]: 6.45997e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 6.29999e-06
[remove_dup_value]: 1.441e-05
[partial_unused_args_eliminate]: 2.96999e-06
[environ_conv]: 6.24001e-06
[add_recomputation]: 4.433e-05
[cse_after_recomputation]: 2.436e-05, [1]
    [Cycle 1]: 1.894e-05, [1]
        [cse]: 1.306e-05
[auto_monad_reorder]: 1.817e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 5.05999e-06
[opt_after_jit_grad]: 0.00044901
[symbol_engine_optimizer]: 7.674e-05, [1]
    [Cycle 1]: 7.025e-05, [6]
        [build]: 3.92998e-06
        [elim_shapecalc]: 9.11998e-06
        [elim_not_effective]: 1.321e-05
        [opt_reshape]: 6.40002e-06
        [fold_const_symbol]: 9.71e-06
        [renormalize]: 4.39992e-07
[validate]: 3.671e-05
Sums
    bootstrap                            :     0.000261s :  3.01%
    type_inference                       :     0.004588s : 52.90%
    event_method                         :     0.000021s :  0.24%
    auto_monad                           :     0.000062s :  0.71%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000046s :  0.54%
    rewriter_before_opt_a                :     0.000086s :  0.99%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000064s :  0.74%
    jit_opt_a.loop_unroll                :     0.000043s :  0.50%
    jit_opt_a.a_1                        :     0.000757s :  8.73%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.28%
    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 :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.14%
    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_        :     0.000011s :  0.13%
    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 :  0.32%
    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          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.000705s :  8.12%
    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      :     0.000020s :  0.23%
    jit_opt_a.cse                        :     0.000034s :  0.39%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000033s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000477s :  5.50%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.22%
    cconv                                :     0.000026s :  0.30%
    loop_unroll                          :     0.000417s :  4.81%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.29%
    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 :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  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  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000044s :  0.51%
    cse_after_recomputation.cse          :     0.000013s :  0.15%
    auto_monad_reorder                   :     0.000018s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000449s :  5.18%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    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 :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000037s :  0.42%


Time group info:
------[substitution.]   0.000226    28
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000002s :      2: substitution.fold_const_symbol
 2.57% :     0.000006s :      4: substitution.graph_param_transform
83.46% :     0.000189s :      4: substitution.inline
 1.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000005s :      4: substitution.replace_old_param
 7.09% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004533     2
79.29% :     0.003594s :      1: type_inference.infer
20.71% :     0.000939s :      1: type_inference.specialize
------[replace.]   0.000085     8
56.78% :     0.000048s :      4: replace.inline
43.22% :     0.000037s :      4: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     8
93.07% :     0.000185s :      4: match.inline
 6.93% :     0.000014s :      4: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   912
 1.17% :     0.000002s :     13: predicate.accumulaten_eliminater
 1.15% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     13: predicate.addn_check_dump
 1.28% :     0.000002s :     13: predicate.addn_zero_filter
 1.95% :     0.000003s :     13: predicate.arithmetic_simplify
 1.32% :     0.000002s :     13: predicate.cast_eliminate
 0.46% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     13: predicate.compare_switch_simplify
 1.37% :     0.000002s :     13: predicate.depend_value_elim
 1.11% :     0.000002s :     13: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     13: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     13: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      4: predicate.elim_not_effective
 0.58% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     13: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     13: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     13: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     13: predicate.environ_get_eliminate
 1.06% :     0.000001s :     13: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      4: predicate.fold_const_symbol
 0.91% :     0.000001s :      8: predicate.get_grad_eliminate
 0.23% :     0.000000s :      4: predicate.graph_param_transform
 5.58% :     0.000008s :     29: predicate.inline
 0.99% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.11% :     0.000002s :      8: predicate.less_batch_normalization
 1.59% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.99% :     0.000003s :     21: predicate.load_eliminater
 1.29% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.29% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     17: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     13: predicate.merge_addn
 1.21% :     0.000002s :     13: predicate.minmaximum_grad
 1.32% :     0.000002s :      4: predicate.mutable_eliminate
 0.45% :     0.000001s :      4: predicate.opt_reshape
 2.47% :     0.000003s :     21: predicate.partial_eliminate
 1.27% :     0.000002s :     13: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     13: predicate.reduce_eliminate
 1.59% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      8: predicate.remove_not_recompute_node
 2.01% :     0.000003s :     25: predicate.replace_applicator
 0.82% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.20% :     0.000002s :     13: predicate.reshape_eliminate
 1.21% :     0.000002s :     13: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     13: predicate.same_eliminate
 0.62% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      8: predicate.special_op_eliminate
 0.96% :     0.000001s :      8: predicate.specialize_transform
 1.40% :     0.000002s :     13: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     13: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.65% :     0.000004s :     21: predicate.switch_defer_inline
 2.19% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.61% :     0.000011s :     67: predicate.switch_simplify
 1.35% :     0.000002s :     13: predicate.tile_eliminate
 1.55% :     0.000002s :     13: predicate.transpose_eliminate
 1.46% :     0.000002s :     13: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     13: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000006s :     25: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     13: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.86% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.29% :     0.000005s :     29: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     13: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000635    12
42.47% :     0.000270s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.01% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.52% :     0.000327s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014144    72
 0.33% :     0.000047s :      1: add_recomputation
 0.46% :     0.000065s :      1: auto_monad
 0.15% :     0.000021s :      1: auto_monad_reorder
 1.91% :     0.000270s :      1: bootstrap
 0.20% :     0.000028s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.19% :     0.000027s :      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% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
37.24% :     0.005267s :      1: jit_opt_a
 1.11% :     0.000157s :      1: jit_opt_after_cconv
 0.41% :     0.000058s :      1: jit_opt_b
 3.01% :     0.000426s :      1: loop_unroll
 3.43% :     0.000485s :      1: mutable_eliminate
 7.06% :     0.000998s :     26: opt.transform.jit_opt_a
 0.39% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000031s :      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
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.23% :     0.000457s :      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% :     0.000004s :      1: pre_auto_parallel
 0.35% :     0.000050s :      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
 2.66% :     0.000376s :      1: renormalize.infer
 2.27% :     0.000321s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000040s :      1: rewriter_after_opt_a
 0.63% :     0.000089s :      1: rewriter_before_opt_a
 0.56% :     0.000079s :      1: symbol_engine_optimizer
32.50% :     0.004596s :      1: type_inference


TotalTime = 0.0466746, [33]
[bootstrap]: 0.00030498
[type_inference]: 0.0231794
[event_method]: 2.385e-05
[auto_monad]: 0.00017398
[graph_reusing]: 5.69e-06
[pre_auto_parallel]: 3.42002e-06
[py_interpret_to_execute]: 0.00012333
[rewriter_before_opt_a]: 0.00010027
[expand_dump_flag]: 3.76001e-06
[jit_opt_a]: 0.00705506, [2]
    [Cycle 1]: 0.002642, [27]
        [switch_simplify]: 6.482e-05
        [loop_unroll]: 3.878e-05
        [a_1]: 0.00077869
        [with_stream_mark]: 2.304e-05
        [recompute_prepare]: 1.174e-05
        [updatestate_depend_eliminate]: 5.65001e-06
        [updatestate_assign_eliminate]: 5.05001e-06
        [updatestate_loads_eliminate]: 4.45e-06
        [parameter_eliminate]: 1.47999e-06
        [specialize_transform]: 9.72999e-06
        [updatestate_useless_node_eliminater]: 1.287e-05
        [accelerated_algorithm]: 8.86002e-06
        [meta_shard_fg_expand]: 2.79999e-06
        [get_grad_eliminate_]: 8.48999e-06
        [merge_forward]: 5.90002e-06
        [cell_reuse_recompute_pass]: 1.81998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.888e-05
        [j_node_and_user_rematch]: 1.403e-05
        [meta_fg_expand]: 3.48e-06
        [replace_old_param]: 1.336e-05
        [inline_without_move]: 8.11002e-06
        [renormalize]: 0.00132542
        [add_forward_monad_depend]: 6.90998e-06
        [auto_monad_grad]: 3.06999e-06
        [auto_monad_eliminator]: 2.382e-05
        [cse]: 3.787e-05
        [replace_applicator]: 1.815e-05
    [Cycle 2]: 0.00048023, [27]
        [switch_simplify]: 9.15001e-06
        [loop_unroll]: 8.12e-06
        [a_1]: 0.00017139
        [with_stream_mark]: 1.526e-05
        [recompute_prepare]: 8.24002e-06
        [updatestate_depend_eliminate]: 5.42999e-06
        [updatestate_assign_eliminate]: 4.57e-06
        [updatestate_loads_eliminate]: 4.02e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 8.43999e-06
        [updatestate_useless_node_eliminater]: 1.018e-05
        [accelerated_algorithm]: 8.32e-06
        [meta_shard_fg_expand]: 2.01003e-06
        [get_grad_eliminate_]: 7.78001e-06
        [merge_forward]: 4.60001e-06
        [cell_reuse_recompute_pass]: 2.37001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.806e-05
        [j_node_and_user_rematch]: 1.279e-05
        [meta_fg_expand]: 2.98e-06
        [replace_old_param]: 1.093e-05
        [inline_without_move]: 7.71001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.46998e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 9.79e-06
        [cse]: 2.125e-05
        [replace_applicator]: 7.61999e-06
[py_interpret_to_execute_after_opt_a]: 1.475e-05
[rewriter_after_opt_a]: 0.00032392
[convert_after_rewriter]: 1.182e-05
[order_py_execute_after_rewriter]: 8.28999e-06
[mutable_eliminate]: 0.00088582
[jit_opt_b]: 7.678e-05, [1]
    [Cycle 1]: 6.679e-05, [2]
        [frontend_op_eliminate]: 2.641e-05
        [inline_after_opt_a]: 2.751e-05
[cconv]: 3.751e-05
[loop_unroll]: 0.00059462
[jit_opt_after_cconv]: 0.00023802, [1]
    [Cycle 1]: 0.00022998, [11]
        [c_1]: 5.837e-05
        [parameter_eliminate]: 5.39e-06
        [updatestate_depend_eliminate]: 9.54e-06
        [updatestate_assign_eliminate]: 4.88001e-06
        [updatestate_loads_eliminate]: 4.44002e-06
        [cse]: 4.671e-05
        [call_graph_tuple_transform]: 2.546e-05
        [tuple_list_get_item_eliminator]: 8.55999e-06
        [none_parameter_eliminate]: 1.28002e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 9.46e-06
[remove_dup_value]: 2.216e-05
[partial_unused_args_eliminate]: 2.30002e-06
[environ_conv]: 1.635e-05
[add_recomputation]: 7.452e-05
[cse_after_recomputation]: 3.677e-05, [1]
    [Cycle 1]: 2.898e-05, [1]
        [cse]: 2.114e-05
[auto_monad_reorder]: 2.782e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 8.90999e-06
[opt_after_jit_grad]: 0.00066964
[symbol_engine_optimizer]: 0.00011579, [1]
    [Cycle 1]: 0.00010795, [6]
        [build]: 1.774e-05
        [elim_shapecalc]: 1.28e-05
        [elim_not_effective]: 2.165e-05
        [opt_reshape]: 1.005e-05
        [fold_const_symbol]: 1.461e-05
        [renormalize]: 4.89992e-07
[validate]: 6.394e-05
[backend_pass]: 1.05001e-06
[task_emit]: 0.0121873
[execute]: 1.029e-05
Sums
    bootstrap                            :     0.000305s :  0.73%
    type_inference                       :     0.023179s : 55.15%
    event_method                         :     0.000024s :  0.06%
    auto_monad                           :     0.000174s :  0.41%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000123s :  0.29%
    rewriter_before_opt_a                :     0.000100s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.18%
    jit_opt_a.loop_unroll                :     0.000047s :  0.11%
    jit_opt_a.a_1                        :     0.000950s :  2.26%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000024s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001325s :  3.15%
    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      :     0.000034s :  0.08%
    jit_opt_a.cse                        :     0.000059s :  0.14%
    jit_opt_a.replace_applicator         :     0.000026s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000324s :  0.77%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000886s :  2.11%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000028s :  0.07%
    cconv                                :     0.000038s :  0.09%
    loop_unroll                          :     0.000595s :  1.41%
    jit_opt_after_cconv.c_1              :     0.000058s :  0.14%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000047s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  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                    :     0.000075s :  0.18%
    cse_after_recomputation.cse          :     0.000021s :  0.05%
    auto_monad_reorder                   :     0.000028s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000670s :  1.59%
    symbol_engine_optimizer.build        :     0.000018s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000064s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012187s : 29.00%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000337    47
 3.18% :     0.000011s :      2: substitution.depend_value_elim
 1.00% :     0.000003s :      4: substitution.elim_not_effective
 0.74% :     0.000002s :      4: substitution.fold_const_symbol
 2.26% :     0.000008s :      5: substitution.graph_param_transform
76.22% :     0.000257s :      4: substitution.inline
 1.47% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.24% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.94% :     0.000007s :      2: substitution.replace_old_param
 3.76% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 4.12% :     0.000014s :      3: substitution.updatestate_pure_node_eliminater
 3.07% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023083     2
93.26% :     0.021528s :      1: type_inference.infer
 6.74% :     0.001555s :      1: type_inference.specialize
------[replace.]   0.000101     6
67.39% :     0.000068s :      4: replace.inline
32.61% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     6
95.63% :     0.000254s :      4: match.inline
 4.37% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000184  1021
 1.18% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.44% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.07% :     0.000002s :     15: predicate.addn_check_dump
 1.34% :     0.000002s :     15: predicate.addn_zero_filter
 2.11% :     0.000004s :     15: predicate.arithmetic_simplify
 1.22% :     0.000002s :     15: predicate.cast_eliminate
 0.45% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.06% :     0.000002s :     15: predicate.compare_switch_simplify
 1.34% :     0.000002s :     15: predicate.depend_value_elim
 1.10% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.13% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.09% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.27% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      5: predicate.elim_not_effective
 0.65% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.29% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     15: predicate.environ_get_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      5: predicate.fold_const_symbol
 1.08% :     0.000002s :     10: predicate.get_grad_eliminate
 0.33% :     0.000001s :      5: predicate.graph_param_transform
 5.27% :     0.000010s :     31: predicate.inline
 1.05% :     0.000002s :     10: predicate.inline_without_move
 0.40% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.45% :     0.000003s :     10: predicate.less_batch_normalization
 1.42% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.94% :     0.000004s :     22: predicate.load_eliminater
 1.92% :     0.000004s :      5: predicate.loop_unroll_after_grad
 3.69% :     0.000007s :     42: predicate.loop_unroll_before_grad
 2.03% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000002s :     15: predicate.merge_addn
 1.07% :     0.000002s :     15: predicate.minmaximum_grad
 1.98% :     0.000004s :      5: predicate.mutable_eliminate
 0.52% :     0.000001s :      5: predicate.opt_reshape
 2.12% :     0.000004s :     22: predicate.partial_eliminate
 1.21% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.59% :     0.000003s :     15: predicate.reduce_eliminate
 1.36% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.59% :     0.000003s :     27: predicate.replace_applicator
 0.82% :     0.000002s :     10: predicate.replace_old_param
 0.70% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000002s :     15: predicate.reshape_eliminate
 1.28% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.20% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.56% :     0.000003s :     15: predicate.same_eliminate
 0.52% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.01% :     0.000002s :     10: predicate.special_op_eliminate
 0.97% :     0.000002s :     10: predicate.specialize_transform
 1.39% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.74% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.38% :     0.000004s :     21: predicate.switch_defer_inline
 1.75% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.38% :     0.000012s :     68: predicate.switch_simplify
 1.15% :     0.000002s :     15: predicate.tile_eliminate
 1.23% :     0.000002s :     15: predicate.transpose_eliminate
 1.39% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.56% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.70% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.45% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000003s :     15: predicate.value_based_eliminate
 0.63% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000910    13
41.87% :     0.000381s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.96% :     0.000063s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.17% :     0.000466s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049487    76
 0.16% :     0.000078s :      1: add_recomputation
 0.36% :     0.000180s :      1: auto_monad
 0.06% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.65% :     0.000323s :      1: bootstrap
 0.08% :     0.000041s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.08% :     0.000039s :      1: cse_after_recomputation
 0.04% :     0.000019s :      1: environ_conv
 0.06% :     0.000029s :      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% :     0.000008s :      1: graph_reusing
14.26% :     0.007058s :      1: jit_opt_a
 0.49% :     0.000241s :      1: jit_opt_after_cconv
 0.16% :     0.000081s :      1: jit_opt_b
 1.25% :     0.000619s :      1: loop_unroll
 1.81% :     0.000898s :      1: mutable_eliminate
 2.56% :     0.001265s :     26: opt.transform.jit_opt_a
 0.20% :     0.000097s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000046s :      4: opt.transform.jit_opt_b
 0.04% :     0.000022s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000055s :      4: opt.transform.symbol_engine_opt
 1.38% :     0.000682s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000127s :      1: py_interpret_to_execute
 0.03% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000025s :      1: remove_dup_value
 1.64% :     0.000812s :      1: renormalize.infer
 1.02% :     0.000503s :      1: renormalize.specialize
 0.02% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.66% :     0.000329s :      1: rewriter_after_opt_a
 0.21% :     0.000105s :      1: rewriter_before_opt_a
 0.24% :     0.000119s :      1: symbol_engine_optimizer
24.66% :     0.012204s :      1: task_emit
46.88% :     0.023201s :      1: type_inference
 0.18% :     0.000091s :      1: validate


TotalTime = 0.0513266, [33]
[bootstrap]: 0.00042062
[type_inference]: 0.0277007
[event_method]: 2.277e-05
[auto_monad]: 0.00017619
[graph_reusing]: 6.14001e-06
[pre_auto_parallel]: 3.90998e-06
[py_interpret_to_execute]: 0.00015886
[rewriter_before_opt_a]: 0.00010452
[expand_dump_flag]: 3.69002e-06
[jit_opt_a]: 0.00771652, [2]
    [Cycle 1]: 0.00285251, [27]
        [switch_simplify]: 7.53e-05
        [loop_unroll]: 3.744e-05
        [a_1]: 0.00084102
        [with_stream_mark]: 2.694e-05
        [recompute_prepare]: 1.161e-05
        [updatestate_depend_eliminate]: 7.18e-06
        [updatestate_assign_eliminate]: 5.49e-06
        [updatestate_loads_eliminate]: 4.92999e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 9.51e-06
        [updatestate_useless_node_eliminater]: 1.478e-05
        [accelerated_algorithm]: 1.027e-05
        [meta_shard_fg_expand]: 3.7e-06
        [get_grad_eliminate_]: 9.20999e-06
        [merge_forward]: 5.72999e-06
        [cell_reuse_recompute_pass]: 2.20002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.193e-05
        [j_node_and_user_rematch]: 1.402e-05
        [meta_fg_expand]: 3.48999e-06
        [replace_old_param]: 1.36e-05
        [inline_without_move]: 8.97e-06
        [renormalize]: 0.00140523
        [add_forward_monad_depend]: 9.27001e-06
        [auto_monad_grad]: 2.67001e-06
        [auto_monad_eliminator]: 2.614e-05
        [cse]: 4.382e-05
        [replace_applicator]: 2.424e-05
    [Cycle 2]: 0.00050508, [27]
        [switch_simplify]: 9.60001e-06
        [loop_unroll]: 9.13002e-06
        [a_1]: 0.00017978
        [with_stream_mark]: 1.555e-05
        [recompute_prepare]: 8.73001e-06
        [updatestate_depend_eliminate]: 5.75001e-06
        [updatestate_assign_eliminate]: 5.79e-06
        [updatestate_loads_eliminate]: 4.17998e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 8.30999e-06
        [updatestate_useless_node_eliminater]: 1.089e-05
        [accelerated_algorithm]: 8.45001e-06
        [meta_shard_fg_expand]: 2.41998e-06
        [get_grad_eliminate_]: 7.69002e-06
        [merge_forward]: 4.65001e-06
        [cell_reuse_recompute_pass]: 3.75998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.834e-05
        [j_node_and_user_rematch]: 1.405e-05
        [meta_fg_expand]: 2.99999e-06
        [replace_old_param]: 1.137e-05
        [inline_without_move]: 7.73001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.36002e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 1.161e-05
        [cse]: 2.28e-05
        [replace_applicator]: 8.25e-06
[py_interpret_to_execute_after_opt_a]: 1.695e-05
[rewriter_after_opt_a]: 0.00030571
[convert_after_rewriter]: 1.207e-05
[order_py_execute_after_rewriter]: 8.27e-06
[mutable_eliminate]: 0.00087077
[jit_opt_b]: 7.726e-05, [1]
    [Cycle 1]: 6.806e-05, [2]
        [frontend_op_eliminate]: 2.667e-05
        [inline_after_opt_a]: 2.743e-05
[cconv]: 3.577e-05
[loop_unroll]: 0.0006488
[jit_opt_after_cconv]: 0.00025528, [1]
    [Cycle 1]: 0.00023563, [11]
        [c_1]: 5.738e-05
        [parameter_eliminate]: 6.43998e-06
        [updatestate_depend_eliminate]: 9.78002e-06
        [updatestate_assign_eliminate]: 4.89e-06
        [updatestate_loads_eliminate]: 5.22e-06
        [cse]: 4.569e-05
        [call_graph_tuple_transform]: 2.521e-05
        [tuple_list_get_item_eliminator]: 9.09998e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 4.20026e-07
        [switch_simplify]: 1.211e-05
[remove_dup_value]: 2.235e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 1.592e-05
[add_recomputation]: 7.625e-05
[cse_after_recomputation]: 3.423e-05, [1]
    [Cycle 1]: 2.617e-05, [1]
        [cse]: 1.874e-05
[auto_monad_reorder]: 2.631e-05
[get_jit_bprop_graph]: 2.18998e-06
[rewriter_after_jit_bprop_graph]: 1.005e-05
[opt_after_jit_grad]: 0.00067283
[symbol_engine_optimizer]: 0.00011697, [1]
    [Cycle 1]: 0.00010929, [6]
        [build]: 1.844e-05
        [elim_shapecalc]: 1.249e-05
        [elim_not_effective]: 2.183e-05
        [opt_reshape]: 9.54e-06
        [fold_const_symbol]: 1.506e-05
        [renormalize]: 9.00007e-07
[validate]: 7.346e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.0114356
[execute]: 1.065e-05
Sums
    bootstrap                            :     0.000421s :  0.91%
    type_inference                       :     0.027701s : 59.96%
    event_method                         :     0.000023s :  0.05%
    auto_monad                           :     0.000176s :  0.38%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000159s :  0.34%
    rewriter_before_opt_a                :     0.000105s :  0.23%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000085s :  0.18%
    jit_opt_a.loop_unroll                :     0.000047s :  0.10%
    jit_opt_a.a_1                        :     0.001021s :  2.21%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  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 :     0.000026s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000025s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001405s :  3.04%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000038s :  0.08%
    jit_opt_a.cse                        :     0.000067s :  0.14%
    jit_opt_a.replace_applicator         :     0.000032s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.04%
    rewriter_after_opt_a                 :     0.000306s :  0.66%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000871s :  1.88%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.06%
    cconv                                :     0.000036s :  0.08%
    loop_unroll                          :     0.000649s :  1.40%
    jit_opt_after_cconv.c_1              :     0.000057s :  0.12%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000046s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000012s :  0.03%
    remove_dup_value                     :     0.000022s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000016s :  0.03%
    add_recomputation                    :     0.000076s :  0.17%
    cse_after_recomputation.cse          :     0.000019s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.02%
    opt_after_jit_grad                   :     0.000673s :  1.46%
    symbol_engine_optimizer.build        :     0.000018s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000073s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011436s : 24.75%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000350    47
 3.25% :     0.000011s :      2: substitution.depend_value_elim
 0.84% :     0.000003s :      4: substitution.elim_not_effective
 0.72% :     0.000003s :      4: substitution.fold_const_symbol
 1.98% :     0.000007s :      5: substitution.graph_param_transform
76.60% :     0.000268s :      4: substitution.inline
 1.54% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.47% :     0.000009s :      8: substitution.remove_not_recompute_node
 1.84% :     0.000006s :      2: substitution.replace_old_param
 4.04% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.51% :     0.000012s :      3: substitution.updatestate_pure_node_eliminater
 3.21% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.027618     2
93.97% :     0.025953s :      1: type_inference.infer
 6.03% :     0.001666s :      1: type_inference.specialize
------[replace.]   0.000143     6
46.42% :     0.000066s :      4: replace.inline
53.58% :     0.000076s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000278     6
95.30% :     0.000265s :      4: match.inline
 4.70% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000187  1021
 1.85% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     15: predicate.addn_check_dump
 1.35% :     0.000003s :     15: predicate.addn_zero_filter
 1.70% :     0.000003s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.49% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     15: predicate.compare_switch_simplify
 1.49% :     0.000003s :     15: predicate.depend_value_elim
 1.10% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.09% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.98% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.44% :     0.000003s :     15: predicate.environ_get_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      5: predicate.fold_const_symbol
 1.17% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000001s :      5: predicate.graph_param_transform
 5.42% :     0.000010s :     31: predicate.inline
 0.92% :     0.000002s :     10: predicate.inline_without_move
 0.38% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.58% :     0.000003s :     10: predicate.less_batch_normalization
 1.59% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.73% :     0.000003s :     22: predicate.load_eliminater
 1.77% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.38% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 2.49% :     0.000005s :      5: predicate.mutable_eliminate
 0.55% :     0.000001s :      5: predicate.opt_reshape
 2.15% :     0.000004s :     22: predicate.partial_eliminate
 1.23% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.73% :     0.000003s :     15: predicate.reduce_eliminate
 1.36% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.77% :     0.000003s :     27: predicate.replace_applicator
 0.72% :     0.000001s :     10: predicate.replace_old_param
 0.31% :     0.000001s :      5: predicate.reset_defer_inline
 1.25% :     0.000002s :     15: predicate.reshape_eliminate
 1.36% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.95% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.93% :     0.000002s :     10: predicate.special_op_eliminate
 1.02% :     0.000002s :     10: predicate.specialize_transform
 1.46% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.09% :     0.000004s :     21: predicate.switch_defer_inline
 1.72% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.45% :     0.000012s :     68: predicate.switch_simplify
 1.29% :     0.000002s :     15: predicate.tile_eliminate
 1.21% :     0.000002s :     15: predicate.transpose_eliminate
 1.54% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 4.24% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.77% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000003s :     15: predicate.value_based_eliminate
 0.39% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.84% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001064    13
48.76% :     0.000519s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.31% :     0.000067s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.92% :     0.000478s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.054309    76
 0.15% :     0.000079s :      1: add_recomputation
 0.34% :     0.000182s :      1: auto_monad
 0.05% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.83% :     0.000449s :      1: bootstrap
 0.07% :     0.000038s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000036s :      1: cse_after_recomputation
 0.03% :     0.000018s :      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% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
14.22% :     0.007720s :      1: jit_opt_a
 0.48% :     0.000259s :      1: jit_opt_after_cconv
 0.15% :     0.000081s :      1: jit_opt_b
 1.22% :     0.000661s :      1: loop_unroll
 1.63% :     0.000884s :      1: mutable_eliminate
 2.50% :     0.001359s :     26: opt.transform.jit_opt_a
 0.18% :     0.000099s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000046s :      4: opt.transform.jit_opt_b
 0.04% :     0.000023s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000028s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000055s :      4: opt.transform.symbol_engine_opt
 1.26% :     0.000684s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.30% :     0.000163s :      1: py_interpret_to_execute
 0.04% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000025s :      1: remove_dup_value
 1.53% :     0.000832s :      1: renormalize.infer
 1.04% :     0.000562s :      1: renormalize.specialize
 0.02% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.57% :     0.000311s :      1: rewriter_after_opt_a
 0.20% :     0.000109s :      1: rewriter_before_opt_a
 0.22% :     0.000120s :      1: symbol_engine_optimizer
21.10% :     0.011457s :      1: task_emit
51.05% :     0.027722s :      1: type_inference
 0.19% :     0.000103s :      1: validate


TotalTime = 0.0471888, [33]
[bootstrap]: 0.00029611
[type_inference]: 0.0239754
[event_method]: 2.406e-05
[auto_monad]: 0.00017843
[graph_reusing]: 6.81001e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 0.00016313
[rewriter_before_opt_a]: 0.00010675
[expand_dump_flag]: 3.61999e-06
[jit_opt_a]: 0.00746428, [2]
    [Cycle 1]: 0.00274643, [27]
        [switch_simplify]: 6.383e-05
        [loop_unroll]: 3.71e-05
        [a_1]: 0.00083493
        [with_stream_mark]: 2.322e-05
        [recompute_prepare]: 1.128e-05
        [updatestate_depend_eliminate]: 5.91e-06
        [updatestate_assign_eliminate]: 5.17e-06
        [updatestate_loads_eliminate]: 4.32e-06
        [parameter_eliminate]: 1.49998e-06
        [specialize_transform]: 1.013e-05
        [updatestate_useless_node_eliminater]: 1.291e-05
        [accelerated_algorithm]: 8.60001e-06
        [meta_shard_fg_expand]: 3.02002e-06
        [get_grad_eliminate_]: 8.37e-06
        [merge_forward]: 4.87e-06
        [cell_reuse_recompute_pass]: 1.66002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.956e-05
        [j_node_and_user_rematch]: 1.354e-05
        [meta_fg_expand]: 3.86999e-06
        [replace_old_param]: 1.223e-05
        [inline_without_move]: 8.52e-06
        [renormalize]: 0.00134705
        [add_forward_monad_depend]: 7.85e-06
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 2.625e-05
        [cse]: 3.824e-05
        [replace_applicator]: 2.086e-05
    [Cycle 2]: 0.00049964, [27]
        [switch_simplify]: 9.29e-06
        [loop_unroll]: 7.97998e-06
        [a_1]: 0.00017498
        [with_stream_mark]: 1.549e-05
        [recompute_prepare]: 7.68999e-06
        [updatestate_depend_eliminate]: 5.44e-06
        [updatestate_assign_eliminate]: 5.14998e-06
        [updatestate_loads_eliminate]: 4.60001e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 8.03001e-06
        [updatestate_useless_node_eliminater]: 1.084e-05
        [accelerated_algorithm]: 8.3e-06
        [meta_shard_fg_expand]: 2.85002e-06
        [get_grad_eliminate_]: 7.68001e-06
        [merge_forward]: 4.77998e-06
        [cell_reuse_recompute_pass]: 2.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.974e-05
        [j_node_and_user_rematch]: 1.375e-05
        [meta_fg_expand]: 3.31001e-06
        [replace_old_param]: 1.216e-05
        [inline_without_move]: 7.88001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.52999e-06
        [auto_monad_grad]: 1.19998e-06
        [auto_monad_eliminator]: 1.13e-05
        [cse]: 2.348e-05
        [replace_applicator]: 8.17e-06
[py_interpret_to_execute_after_opt_a]: 1.922e-05
[rewriter_after_opt_a]: 0.00031184
[convert_after_rewriter]: 1.342e-05
[order_py_execute_after_rewriter]: 7.8e-06
[mutable_eliminate]: 0.00094746
[jit_opt_b]: 8.209e-05, [1]
    [Cycle 1]: 7.384e-05, [2]
        [frontend_op_eliminate]: 3.057e-05
        [inline_after_opt_a]: 2.821e-05
[cconv]: 3.377e-05
[loop_unroll]: 0.00055684
[jit_opt_after_cconv]: 0.00022305, [1]
    [Cycle 1]: 0.00021566, [11]
        [c_1]: 4.981e-05
        [parameter_eliminate]: 4.41002e-06
        [updatestate_depend_eliminate]: 9.10001e-06
        [updatestate_assign_eliminate]: 5.86e-06
        [updatestate_loads_eliminate]: 4.48999e-06
        [cse]: 4.439e-05
        [call_graph_tuple_transform]: 2.433e-05
        [tuple_list_get_item_eliminator]: 8.43001e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 9.62001e-06
[remove_dup_value]: 2.255e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 1.294e-05
[add_recomputation]: 7.051e-05
[cse_after_recomputation]: 3.11e-05, [1]
    [Cycle 1]: 2.458e-05, [1]
        [cse]: 1.869e-05
[auto_monad_reorder]: 2.607e-05
[get_jit_bprop_graph]: 2.36998e-06
[rewriter_after_jit_bprop_graph]: 7.56001e-06
[opt_after_jit_grad]: 0.00056734
[symbol_engine_optimizer]: 0.00011287, [1]
    [Cycle 1]: 0.0001059, [6]
        [build]: 1.87e-05
        [elim_shapecalc]: 1.268e-05
        [elim_not_effective]: 1.899e-05
        [opt_reshape]: 1.017e-05
        [fold_const_symbol]: 1.423e-05
        [renormalize]: 5.00004e-07
[validate]: 6.578e-05
[backend_pass]: 1.05999e-06
[task_emit]: 0.0115755
[execute]: 8.81002e-06
Sums
    bootstrap                            :     0.000296s :  0.70%
    type_inference                       :     0.023975s : 56.76%
    event_method                         :     0.000024s :  0.06%
    auto_monad                           :     0.000178s :  0.42%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000163s :  0.39%
    rewriter_before_opt_a                :     0.000107s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.001010s :  2.39%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000024s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001347s :  3.19%
    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      :     0.000038s :  0.09%
    jit_opt_a.cse                        :     0.000062s :  0.15%
    jit_opt_a.replace_applicator         :     0.000029s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.05%
    rewriter_after_opt_a                 :     0.000312s :  0.74%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000947s :  2.24%
    jit_opt_b.frontend_op_eliminate      :     0.000031s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000028s :  0.07%
    cconv                                :     0.000034s :  0.08%
    loop_unroll                          :     0.000557s :  1.32%
    jit_opt_after_cconv.c_1              :     0.000050s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000044s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000023s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000071s :  0.17%
    cse_after_recomputation.cse          :     0.000019s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000567s :  1.34%
    symbol_engine_optimizer.build        :     0.000019s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000066s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011576s : 27.40%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000330    47
 3.28% :     0.000011s :      2: substitution.depend_value_elim
 0.93% :     0.000003s :      4: substitution.elim_not_effective
 0.68% :     0.000002s :      4: substitution.fold_const_symbol
 2.24% :     0.000007s :      5: substitution.graph_param_transform
75.95% :     0.000251s :      4: substitution.inline
 1.59% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.76% :     0.000009s :      8: substitution.remove_not_recompute_node
 1.76% :     0.000006s :      2: substitution.replace_old_param
 4.15% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.30% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.38% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023890     2
93.21% :     0.022269s :      1: type_inference.infer
 6.79% :     0.001622s :      1: type_inference.specialize
------[replace.]   0.000102     6
63.92% :     0.000065s :      4: replace.inline
36.08% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000260     6
95.20% :     0.000248s :      4: match.inline
 4.80% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000187  1021
 1.27% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.54% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.37% :     0.000003s :     15: predicate.addn_check_dump
 1.30% :     0.000002s :     15: predicate.addn_zero_filter
 2.04% :     0.000004s :     15: predicate.arithmetic_simplify
 1.63% :     0.000003s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.44% :     0.000003s :     15: predicate.depend_value_elim
 1.21% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.11% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.84% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.78% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     15: predicate.environ_get_eliminate
 1.03% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 4.82% :     0.000009s :     31: predicate.inline
 0.94% :     0.000002s :     10: predicate.inline_without_move
 0.40% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :     10: predicate.less_batch_normalization
 1.42% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.82% :     0.000003s :     22: predicate.load_eliminater
 1.51% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.13% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.97% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 1.98% :     0.000004s :      5: predicate.mutable_eliminate
 0.86% :     0.000002s :      5: predicate.opt_reshape
 2.36% :     0.000004s :     22: predicate.partial_eliminate
 1.24% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.86% :     0.000003s :     15: predicate.reduce_eliminate
 1.50% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.77% :     0.000003s :     27: predicate.replace_applicator
 0.72% :     0.000001s :     10: predicate.replace_old_param
 0.44% :     0.000001s :      5: predicate.reset_defer_inline
 1.42% :     0.000003s :     15: predicate.reshape_eliminate
 1.12% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     15: predicate.same_eliminate
 0.52% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.24% :     0.000002s :     10: predicate.special_op_eliminate
 1.00% :     0.000002s :     10: predicate.specialize_transform
 1.45% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.75% :     0.000005s :     21: predicate.switch_defer_inline
 1.92% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.18% :     0.000012s :     68: predicate.switch_simplify
 1.33% :     0.000002s :     15: predicate.tile_eliminate
 1.21% :     0.000002s :     15: predicate.transpose_eliminate
 1.43% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 2.03% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.66% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.23% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 1.04% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000928    13
43.18% :     0.000401s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.60% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.22% :     0.000466s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050070    76
 0.15% :     0.000074s :      1: add_recomputation
 0.37% :     0.000185s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.63% :     0.000317s :      1: bootstrap
 0.07% :     0.000036s :      1: cconv
 0.03% :     0.000017s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.06% :     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% :     0.000010s :      1: graph_reusing
14.92% :     0.007468s :      1: jit_opt_a
 0.45% :     0.000226s :      1: jit_opt_after_cconv
 0.17% :     0.000085s :      1: jit_opt_b
 1.13% :     0.000567s :      1: loop_unroll
 1.92% :     0.000962s :      1: mutable_eliminate
 2.65% :     0.001329s :     26: opt.transform.jit_opt_a
 0.18% :     0.000089s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000049s :      4: opt.transform.jit_opt_b
 0.04% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000577s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.34% :     0.000168s :      1: py_interpret_to_execute
 0.05% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000025s :      1: remove_dup_value
 1.60% :     0.000800s :      1: renormalize.infer
 1.07% :     0.000536s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.64% :     0.000318s :      1: rewriter_after_opt_a
 0.22% :     0.000111s :      1: rewriter_before_opt_a
 0.23% :     0.000116s :      1: symbol_engine_optimizer
23.16% :     0.011595s :      1: task_emit
47.92% :     0.023994s :      1: type_inference
 0.19% :     0.000095s :      1: validate


TotalTime = 0.0463109, [33]
[bootstrap]: 0.00030148
[type_inference]: 0.0249332
[event_method]: 2.149e-05
[auto_monad]: 0.00015982
[graph_reusing]: 6.71999e-06
[pre_auto_parallel]: 3.01999e-06
[py_interpret_to_execute]: 0.00012363
[rewriter_before_opt_a]: 9.717e-05
[expand_dump_flag]: 3.44001e-06
[jit_opt_a]: 0.00643945, [2]
    [Cycle 1]: 0.00244026, [27]
        [switch_simplify]: 6.421e-05
        [loop_unroll]: 3.649e-05
        [a_1]: 0.00073987
        [with_stream_mark]: 2.596e-05
        [recompute_prepare]: 1.252e-05
        [updatestate_depend_eliminate]: 6.76999e-06
        [updatestate_assign_eliminate]: 6.07001e-06
        [updatestate_loads_eliminate]: 4.79998e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 1.055e-05
        [updatestate_useless_node_eliminater]: 1.377e-05
        [accelerated_algorithm]: 8.38001e-06
        [meta_shard_fg_expand]: 3.16999e-06
        [get_grad_eliminate_]: 8.18999e-06
        [merge_forward]: 5.91998e-06
        [cell_reuse_recompute_pass]: 2.18998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.045e-05
        [j_node_and_user_rematch]: 1.477e-05
        [meta_fg_expand]: 3.53e-06
        [replace_old_param]: 1.225e-05
        [inline_without_move]: 8.24002e-06
        [renormalize]: 0.00114408
        [add_forward_monad_depend]: 6.14001e-06
        [auto_monad_grad]: 2.42001e-06
        [auto_monad_eliminator]: 2.126e-05
        [cse]: 4.125e-05
        [replace_applicator]: 1.835e-05
    [Cycle 2]: 0.00048995, [27]
        [switch_simplify]: 9.15999e-06
        [loop_unroll]: 8.22e-06
        [a_1]: 0.00018479
        [with_stream_mark]: 1.387e-05
        [recompute_prepare]: 8.40001e-06
        [updatestate_depend_eliminate]: 4.94998e-06
        [updatestate_assign_eliminate]: 4.55999e-06
        [updatestate_loads_eliminate]: 3.85e-06
        [parameter_eliminate]: 1.46998e-06
        [specialize_transform]: 8.22e-06
        [updatestate_useless_node_eliminater]: 1.064e-05
        [accelerated_algorithm]: 8.18999e-06
        [meta_shard_fg_expand]: 2.53998e-06
        [get_grad_eliminate_]: 7.55998e-06
        [merge_forward]: 4.30999e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.678e-05
        [j_node_and_user_rematch]: 1.256e-05
        [meta_fg_expand]: 2.67001e-06
        [replace_old_param]: 1.063e-05
        [inline_without_move]: 7.63001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.42999e-06
        [auto_monad_grad]: 1.02998e-06
        [auto_monad_eliminator]: 1.015e-05
        [cse]: 2.111e-05
        [replace_applicator]: 8.02e-06
[py_interpret_to_execute_after_opt_a]: 1.421e-05
[rewriter_after_opt_a]: 0.00028489
[convert_after_rewriter]: 1.092e-05
[order_py_execute_after_rewriter]: 7.57002e-06
[mutable_eliminate]: 0.00063677
[jit_opt_b]: 6.96e-05, [1]
    [Cycle 1]: 6.22e-05, [2]
        [frontend_op_eliminate]: 2.457e-05
        [inline_after_opt_a]: 2.508e-05
[cconv]: 2.657e-05
[loop_unroll]: 0.00045978
[jit_opt_after_cconv]: 0.0001989, [1]
    [Cycle 1]: 0.00019211, [11]
        [c_1]: 4.618e-05
        [parameter_eliminate]: 2.77002e-06
        [updatestate_depend_eliminate]: 7.63001e-06
        [updatestate_assign_eliminate]: 4.35e-06
        [updatestate_loads_eliminate]: 3.95e-06
        [cse]: 3.155e-05
        [call_graph_tuple_transform]: 2.358e-05
        [tuple_list_get_item_eliminator]: 8.53001e-06
        [none_parameter_eliminate]: 1.89e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 8.67e-06
[remove_dup_value]: 1.922e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 1.06e-05
[add_recomputation]: 6.647e-05
[cse_after_recomputation]: 2.976e-05, [1]
    [Cycle 1]: 2.393e-05, [1]
        [cse]: 1.724e-05
[auto_monad_reorder]: 2.62e-05
[get_jit_bprop_graph]: 1.96003e-06
[rewriter_after_jit_bprop_graph]: 6.83e-06
[opt_after_jit_grad]: 0.00051064
[symbol_engine_optimizer]: 0.00010745, [1]
    [Cycle 1]: 0.00010051, [6]
        [build]: 1.548e-05
        [elim_shapecalc]: 1.234e-05
        [elim_not_effective]: 1.907e-05
        [opt_reshape]: 9.39e-06
        [fold_const_symbol]: 1.395e-05
        [renormalize]: 6.50005e-07
[validate]: 5.617e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.011433
[execute]: 7.73999e-06
Sums
    bootstrap                            :     0.000301s :  0.72%
    type_inference                       :     0.024933s : 59.18%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000160s :  0.38%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000124s :  0.29%
    rewriter_before_opt_a                :     0.000097s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000925s :  2.19%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    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        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001144s :  2.72%
    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      :     0.000031s :  0.07%
    jit_opt_a.cse                        :     0.000062s :  0.15%
    jit_opt_a.replace_applicator         :     0.000026s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000285s :  0.68%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000637s :  1.51%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000027s :  0.06%
    loop_unroll                          :     0.000460s :  1.09%
    jit_opt_after_cconv.c_1              :     0.000046s :  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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000066s :  0.16%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000511s :  1.21%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000056s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011433s : 27.14%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000299    47
 3.24% :     0.000010s :      2: substitution.depend_value_elim
 0.94% :     0.000003s :      4: substitution.elim_not_effective
 0.71% :     0.000002s :      4: substitution.fold_const_symbol
 2.35% :     0.000007s :      5: substitution.graph_param_transform
75.75% :     0.000227s :      4: substitution.inline
 1.87% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.48% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.91% :     0.000006s :      2: substitution.replace_old_param
 4.40% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.12% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.24% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024859     2
94.11% :     0.023396s :      1: type_inference.infer
 5.89% :     0.001463s :      1: type_inference.specialize
------[replace.]   0.000099     6
65.17% :     0.000064s :      4: replace.inline
34.83% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000236     6
94.92% :     0.000224s :      4: match.inline
 5.08% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1021
 1.47% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     15: predicate.addn_check_dump
 1.19% :     0.000002s :     15: predicate.addn_zero_filter
 1.96% :     0.000003s :     15: predicate.arithmetic_simplify
 1.51% :     0.000003s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.83% :     0.000003s :     15: predicate.depend_value_elim
 1.16% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.68% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     15: predicate.environ_get_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.29% :     0.000000s :      5: predicate.graph_param_transform
 5.36% :     0.000009s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :     10: predicate.less_batch_normalization
 1.59% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     22: predicate.load_eliminater
 1.34% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.39% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     15: predicate.merge_addn
 1.09% :     0.000002s :     15: predicate.minmaximum_grad
 1.49% :     0.000003s :      5: predicate.mutable_eliminate
 0.57% :     0.000001s :      5: predicate.opt_reshape
 2.40% :     0.000004s :     22: predicate.partial_eliminate
 1.20% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.85% :     0.000003s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.69% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.83% :     0.000003s :     27: predicate.replace_applicator
 0.56% :     0.000001s :     10: predicate.replace_old_param
 0.39% :     0.000001s :      5: predicate.reset_defer_inline
 1.35% :     0.000002s :     15: predicate.reshape_eliminate
 1.25% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     15: predicate.same_eliminate
 0.60% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.09% :     0.000002s :     10: predicate.special_op_eliminate
 1.23% :     0.000002s :     10: predicate.specialize_transform
 1.48% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.23% :     0.000004s :     21: predicate.switch_defer_inline
 1.86% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.60% :     0.000011s :     68: predicate.switch_simplify
 1.30% :     0.000002s :     15: predicate.tile_eliminate
 1.31% :     0.000002s :     15: predicate.transpose_eliminate
 1.48% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.50% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000861    13
46.01% :     0.000396s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.84% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.15% :     0.000415s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048879    76
 0.14% :     0.000069s :      1: add_recomputation
 0.34% :     0.000166s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000322s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.05% :     0.000025s :      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
13.18% :     0.006443s :      1: jit_opt_a
 0.41% :     0.000202s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 0.96% :     0.000468s :      1: loop_unroll
 1.32% :     0.000646s :      1: mutable_eliminate
 2.53% :     0.001239s :     26: opt.transform.jit_opt_a
 0.17% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      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
 1.06% :     0.000519s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000128s :      1: py_interpret_to_execute
 0.03% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000022s :      1: remove_dup_value
 1.40% :     0.000685s :      1: renormalize.infer
 0.92% :     0.000449s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.59% :     0.000290s :      1: rewriter_after_opt_a
 0.21% :     0.000102s :      1: rewriter_before_opt_a
 0.23% :     0.000110s :      1: symbol_engine_optimizer
23.43% :     0.011453s :      1: task_emit
51.04% :     0.024947s :      1: type_inference
 0.16% :     0.000080s :      1: validate


TotalTime = 0.0444305, [33]
[bootstrap]: 0.00033872
[type_inference]: 0.0229582
[event_method]: 2.061e-05
[auto_monad]: 0.00016509
[graph_reusing]: 6.46e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 0.00013715
[rewriter_before_opt_a]: 9.638e-05
[expand_dump_flag]: 3.56001e-06
[jit_opt_a]: 0.00675441, [2]
    [Cycle 1]: 0.00249093, [27]
        [switch_simplify]: 7.061e-05
        [loop_unroll]: 3.661e-05
        [a_1]: 0.00077045
        [with_stream_mark]: 2.474e-05
        [recompute_prepare]: 1.296e-05
        [updatestate_depend_eliminate]: 6.51999e-06
        [updatestate_assign_eliminate]: 5.81e-06
        [updatestate_loads_eliminate]: 4.84e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 9.66e-06
        [updatestate_useless_node_eliminater]: 1.174e-05
        [accelerated_algorithm]: 1.003e-05
        [meta_shard_fg_expand]: 2.93998e-06
        [get_grad_eliminate_]: 8.57e-06
        [merge_forward]: 5.39e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.063e-05
        [j_node_and_user_rematch]: 1.403e-05
        [meta_fg_expand]: 3.65e-06
        [replace_old_param]: 1.152e-05
        [inline_without_move]: 8.23001e-06
        [renormalize]: 0.00116419
        [add_forward_monad_depend]: 8.54002e-06
        [auto_monad_grad]: 2.81999e-06
        [auto_monad_eliminator]: 2.119e-05
        [cse]: 3.33e-05
        [replace_applicator]: 2.147e-05
    [Cycle 2]: 0.00048642, [27]
        [switch_simplify]: 8.69e-06
        [loop_unroll]: 9.12001e-06
        [a_1]: 0.00017117
        [with_stream_mark]: 1.61e-05
        [recompute_prepare]: 8.54e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 5.17e-06
        [updatestate_loads_eliminate]: 4.43999e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 8.49002e-06
        [updatestate_useless_node_eliminater]: 1.096e-05
        [accelerated_algorithm]: 7.9e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 7.56999e-06
        [merge_forward]: 4.53999e-06
        [cell_reuse_recompute_pass]: 3.11999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.789e-05
        [j_node_and_user_rematch]: 1.305e-05
        [meta_fg_expand]: 3.11001e-06
        [replace_old_param]: 1.047e-05
        [inline_without_move]: 7.4e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.67001e-06
        [auto_monad_grad]: 1.20001e-06
        [auto_monad_eliminator]: 1.107e-05
        [cse]: 2.166e-05
        [replace_applicator]: 8.24998e-06
[py_interpret_to_execute_after_opt_a]: 1.493e-05
[rewriter_after_opt_a]: 0.00031307
[convert_after_rewriter]: 1.226e-05
[order_py_execute_after_rewriter]: 7.55e-06
[mutable_eliminate]: 0.00077945
[jit_opt_b]: 7.017e-05, [1]
    [Cycle 1]: 6.163e-05, [2]
        [frontend_op_eliminate]: 2.494e-05
        [inline_after_opt_a]: 2.419e-05
[cconv]: 3.104e-05
[loop_unroll]: 0.00049942
[jit_opt_after_cconv]: 0.00021491, [1]
    [Cycle 1]: 0.00020737, [11]
        [c_1]: 5.099e-05
        [parameter_eliminate]: 3.36001e-06
        [updatestate_depend_eliminate]: 8.96998e-06
        [updatestate_assign_eliminate]: 5.12e-06
        [updatestate_loads_eliminate]: 4.75001e-06
        [cse]: 3.698e-05
        [call_graph_tuple_transform]: 2.43e-05
        [tuple_list_get_item_eliminator]: 8.40001e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 1.22e-06
        [switch_simplify]: 9.18002e-06
[remove_dup_value]: 2.118e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 1.049e-05
[add_recomputation]: 6.811e-05
[cse_after_recomputation]: 3.142e-05, [1]
    [Cycle 1]: 2.533e-05, [1]
        [cse]: 1.88e-05
[auto_monad_reorder]: 2.757e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 7.26999e-06
[opt_after_jit_grad]: 0.00054939
[symbol_engine_optimizer]: 0.00010758, [1]
    [Cycle 1]: 0.0001008, [6]
        [build]: 1.398e-05
        [elim_shapecalc]: 1.238e-05
        [elim_not_effective]: 1.942e-05
        [opt_reshape]: 9.33002e-06
        [fold_const_symbol]: 1.405e-05
        [renormalize]: 7.30011e-07
[validate]: 6.268e-05
[backend_pass]: 1.24e-06
[task_emit]: 0.0108664
[execute]: 8.50001e-06
Sums
    bootstrap                            :     0.000339s :  0.85%
    type_inference                       :     0.022958s : 57.44%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000165s :  0.41%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000137s :  0.34%
    rewriter_before_opt_a                :     0.000096s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000079s :  0.20%
    jit_opt_a.loop_unroll                :     0.000046s :  0.11%
    jit_opt_a.a_1                        :     0.000942s :  2.36%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001164s :  2.91%
    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      :     0.000032s :  0.08%
    jit_opt_a.cse                        :     0.000055s :  0.14%
    jit_opt_a.replace_applicator         :     0.000030s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000313s :  0.78%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000779s :  1.95%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000031s :  0.08%
    loop_unroll                          :     0.000499s :  1.25%
    jit_opt_after_cconv.c_1              :     0.000051s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000037s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000068s :  0.17%
    cse_after_recomputation.cse          :     0.000019s :  0.05%
    auto_monad_reorder                   :     0.000028s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000549s :  1.37%
    symbol_engine_optimizer.build        :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000063s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010866s : 27.19%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000322    47
 3.16% :     0.000010s :      2: substitution.depend_value_elim
 0.90% :     0.000003s :      4: substitution.elim_not_effective
 0.66% :     0.000002s :      4: substitution.fold_const_symbol
 2.19% :     0.000007s :      5: substitution.graph_param_transform
76.15% :     0.000245s :      4: substitution.inline
 1.58% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.42% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.58% :     0.000005s :      2: substitution.replace_old_param
 4.49% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.72% :     0.000012s :      3: substitution.updatestate_pure_node_eliminater
 3.15% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022883     2
93.24% :     0.021337s :      1: type_inference.infer
 6.76% :     0.001546s :      1: type_inference.specialize
------[replace.]   0.000097     6
66.42% :     0.000064s :      4: replace.inline
33.58% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     6
94.78% :     0.000242s :      4: match.inline
 5.22% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000179  1021
 1.42% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.25% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     15: predicate.addn_check_dump
 1.31% :     0.000002s :     15: predicate.addn_zero_filter
 2.00% :     0.000004s :     15: predicate.arithmetic_simplify
 1.59% :     0.000003s :     15: predicate.cast_eliminate
 0.45% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     15: predicate.compare_switch_simplify
 1.50% :     0.000003s :     15: predicate.depend_value_elim
 1.12% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.02% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.58% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.22% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     15: predicate.environ_get_eliminate
 1.26% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.18% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 4.80% :     0.000009s :     31: predicate.inline
 0.97% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.89% :     0.000003s :     10: predicate.less_batch_normalization
 1.51% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000004s :     22: predicate.load_eliminater
 1.36% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.34% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.01% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 1.69% :     0.000003s :      5: predicate.mutable_eliminate
 0.75% :     0.000001s :      5: predicate.opt_reshape
 2.46% :     0.000004s :     22: predicate.partial_eliminate
 1.16% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.63% :     0.000003s :     15: predicate.reduce_eliminate
 1.56% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.73% :     0.000003s :     27: predicate.replace_applicator
 0.51% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.68% :     0.000003s :     15: predicate.reshape_eliminate
 1.24% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.13% :     0.000002s :     15: predicate.same_eliminate
 0.54% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.29% :     0.000002s :     10: predicate.special_op_eliminate
 1.10% :     0.000002s :     10: predicate.specialize_transform
 1.39% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.17% :     0.000004s :     21: predicate.switch_defer_inline
 1.74% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.40% :     0.000011s :     68: predicate.switch_simplify
 1.37% :     0.000002s :     15: predicate.tile_eliminate
 1.32% :     0.000002s :     15: predicate.transpose_eliminate
 1.47% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 2.02% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.34% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.74% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.43% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000889    13
45.42% :     0.000404s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.35% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.23% :     0.000429s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047057    76
 0.15% :     0.000072s :      1: add_recomputation
 0.36% :     0.000170s :      1: auto_monad
 0.07% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.77% :     0.000361s :      1: bootstrap
 0.07% :     0.000034s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.07% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.05% :     0.000024s :      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% :     0.000009s :      1: graph_reusing
14.36% :     0.006758s :      1: jit_opt_a
 0.46% :     0.000218s :      1: jit_opt_after_cconv
 0.15% :     0.000073s :      1: jit_opt_b
 1.08% :     0.000509s :      1: loop_unroll
 1.67% :     0.000788s :      1: mutable_eliminate
 2.69% :     0.001264s :     26: opt.transform.jit_opt_a
 0.19% :     0.000089s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000560s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000141s :      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
 1.45% :     0.000681s :      1: renormalize.infer
 1.01% :     0.000473s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.68% :     0.000319s :      1: rewriter_after_opt_a
 0.21% :     0.000101s :      1: rewriter_before_opt_a
 0.23% :     0.000110s :      1: symbol_engine_optimizer
23.13% :     0.010884s :      1: task_emit
48.82% :     0.022972s :      1: type_inference
 0.19% :     0.000088s :      1: validate


TotalTime = 0.0435721, [33]
[bootstrap]: 0.0002784
[type_inference]: 0.0230528
[event_method]: 2.226e-05
[auto_monad]: 0.00015946
[graph_reusing]: 6.54001e-06
[pre_auto_parallel]: 3.48999e-06
[py_interpret_to_execute]: 0.00012118
[rewriter_before_opt_a]: 9.918e-05
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.00641011, [2]
    [Cycle 1]: 0.00240742, [27]
        [switch_simplify]: 6.428e-05
        [loop_unroll]: 3.656e-05
        [a_1]: 0.00072483
        [with_stream_mark]: 2.409e-05
        [recompute_prepare]: 1.006e-05
        [updatestate_depend_eliminate]: 6.34999e-06
        [updatestate_assign_eliminate]: 5.97999e-06
        [updatestate_loads_eliminate]: 5.25001e-06
        [parameter_eliminate]: 1.99999e-06
        [specialize_transform]: 9.35001e-06
        [updatestate_useless_node_eliminater]: 1.134e-05
        [accelerated_algorithm]: 8.35999e-06
        [meta_shard_fg_expand]: 2.93e-06
        [get_grad_eliminate_]: 7.93001e-06
        [merge_forward]: 5.57001e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.974e-05
        [j_node_and_user_rematch]: 1.351e-05
        [meta_fg_expand]: 3.43e-06
        [replace_old_param]: 1.294e-05
        [inline_without_move]: 7.87e-06
        [renormalize]: 0.00114765
        [add_forward_monad_depend]: 6.69999e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 2.057e-05
        [cse]: 4.418e-05
        [replace_applicator]: 1.785e-05
    [Cycle 2]: 0.00056072, [27]
        [switch_simplify]: 8.94e-06
        [loop_unroll]: 7.68999e-06
        [a_1]: 0.00022635
        [with_stream_mark]: 1.956e-05
        [recompute_prepare]: 9.79e-06
        [updatestate_depend_eliminate]: 5.79e-06
        [updatestate_assign_eliminate]: 4.60001e-06
        [updatestate_loads_eliminate]: 3.97e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 8.37e-06
        [updatestate_useless_node_eliminater]: 1.227e-05
        [accelerated_algorithm]: 8.72998e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 8.1e-06
        [merge_forward]: 5.12e-06
        [cell_reuse_recompute_pass]: 2.38998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.734e-05
        [j_node_and_user_rematch]: 1.331e-05
        [meta_fg_expand]: 2.83998e-06
        [replace_old_param]: 1.186e-05
        [inline_without_move]: 7.6e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 2.02999e-06
        [auto_monad_eliminator]: 1.138e-05
        [cse]: 2.501e-05
        [replace_applicator]: 8.40001e-06
[py_interpret_to_execute_after_opt_a]: 1.503e-05
[rewriter_after_opt_a]: 0.00030864
[convert_after_rewriter]: 1.386e-05
[order_py_execute_after_rewriter]: 8.02e-06
[mutable_eliminate]: 0.00069733
[jit_opt_b]: 7.543e-05, [1]
    [Cycle 1]: 6.77e-05, [2]
        [frontend_op_eliminate]: 2.764e-05
        [inline_after_opt_a]: 2.659e-05
[cconv]: 3.047e-05
[loop_unroll]: 0.00049798
[jit_opt_after_cconv]: 0.00020605, [1]
    [Cycle 1]: 0.00019898, [11]
        [c_1]: 4.846e-05
        [parameter_eliminate]: 3.15998e-06
        [updatestate_depend_eliminate]: 9.20999e-06
        [updatestate_assign_eliminate]: 4.90001e-06
        [updatestate_loads_eliminate]: 4.23001e-06
        [cse]: 3.124e-05
        [call_graph_tuple_transform]: 2.304e-05
        [tuple_list_get_item_eliminator]: 8.52e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 9.71e-06
[remove_dup_value]: 1.981e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 1.145e-05
[add_recomputation]: 7.048e-05
[cse_after_recomputation]: 3.293e-05, [1]
    [Cycle 1]: 2.641e-05, [1]
        [cse]: 2.011e-05
[auto_monad_reorder]: 2.778e-05
[get_jit_bprop_graph]: 2.22001e-06
[rewriter_after_jit_bprop_graph]: 6.38e-06
[opt_after_jit_grad]: 0.00054177
[symbol_engine_optimizer]: 0.00011073, [1]
    [Cycle 1]: 0.00010349, [6]
        [build]: 1.274e-05
        [elim_shapecalc]: 1.267e-05
        [elim_not_effective]: 2.003e-05
        [opt_reshape]: 1.035e-05
        [fold_const_symbol]: 1.529e-05
        [renormalize]: 4.89992e-07
[validate]: 5.837e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.0104302
[execute]: 9.41e-06
Sums
    bootstrap                            :     0.000278s :  0.71%
    type_inference                       :     0.023053s : 58.44%
    event_method                         :     0.000022s :  0.06%
    auto_monad                           :     0.000159s :  0.40%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000121s :  0.31%
    rewriter_before_opt_a                :     0.000099s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.19%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000951s :  2.41%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    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  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001148s :  2.91%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000032s :  0.08%
    jit_opt_a.cse                        :     0.000069s :  0.18%
    jit_opt_a.replace_applicator         :     0.000026s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000309s :  0.78%
    convert_after_rewriter               :     0.000014s :  0.04%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000697s :  1.77%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.07%
    cconv                                :     0.000030s :  0.08%
    loop_unroll                          :     0.000498s :  1.26%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000031s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000070s :  0.18%
    cse_after_recomputation.cse          :     0.000020s :  0.05%
    auto_monad_reorder                   :     0.000028s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000542s :  1.37%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000058s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010430s : 26.44%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000299    47
 3.01% :     0.000009s :      2: substitution.depend_value_elim
 0.97% :     0.000003s :      4: substitution.elim_not_effective
 0.70% :     0.000002s :      4: substitution.fold_const_symbol
 2.23% :     0.000007s :      5: substitution.graph_param_transform
75.64% :     0.000226s :      4: substitution.inline
 1.65% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.50% :     0.000007s :      8: substitution.remove_not_recompute_node
 2.20% :     0.000007s :      2: substitution.replace_old_param
 4.40% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.67% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.03% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022977     2
93.54% :     0.021493s :      1: type_inference.infer
 6.46% :     0.001484s :      1: type_inference.specialize
------[replace.]   0.000091     6
65.77% :     0.000060s :      4: replace.inline
34.23% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000235     6
94.87% :     0.000223s :      4: match.inline
 5.13% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000177  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.50% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.23% :     0.000002s :     15: predicate.addn_zero_filter
 1.85% :     0.000003s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.85% :     0.000002s :      5: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     15: predicate.compare_switch_simplify
 1.35% :     0.000002s :     15: predicate.depend_value_elim
 1.15% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.97% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.56% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     15: predicate.environ_get_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.29% :     0.000001s :      5: predicate.fold_const_symbol
 1.19% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 5.33% :     0.000009s :     31: predicate.inline
 1.02% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :     10: predicate.less_batch_normalization
 1.52% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.88% :     0.000003s :     22: predicate.load_eliminater
 1.59% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.22% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.84% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.09% :     0.000002s :     15: predicate.minmaximum_grad
 1.99% :     0.000004s :      5: predicate.mutable_eliminate
 0.84% :     0.000001s :      5: predicate.opt_reshape
 2.53% :     0.000004s :     22: predicate.partial_eliminate
 1.23% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.60% :     0.000003s :     15: predicate.reduce_eliminate
 1.43% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.73% :     0.000003s :     27: predicate.replace_applicator
 0.71% :     0.000001s :     10: predicate.replace_old_param
 0.37% :     0.000001s :      5: predicate.reset_defer_inline
 1.32% :     0.000002s :     15: predicate.reshape_eliminate
 1.27% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     15: predicate.same_eliminate
 0.76% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.36% :     0.000002s :     10: predicate.special_op_eliminate
 1.13% :     0.000002s :     10: predicate.specialize_transform
 1.39% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.41% :     0.000004s :     21: predicate.switch_defer_inline
 1.87% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.39% :     0.000011s :     68: predicate.switch_simplify
 1.20% :     0.000002s :     15: predicate.tile_eliminate
 1.23% :     0.000002s :     15: predicate.transpose_eliminate
 1.54% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.78% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.27% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000003s :     15: predicate.value_based_eliminate
 0.47% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000885    13
42.30% :     0.000374s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.98% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.72% :     0.000458s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046191    76
 0.16% :     0.000074s :      1: add_recomputation
 0.36% :     0.000164s :      1: auto_monad
 0.07% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.63% :     0.000293s :      1: bootstrap
 0.07% :     0.000034s :      1: cconv
 0.04% :     0.000017s :      1: convert_after_rewriter
 0.08% :     0.000035s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.06% :     0.000027s :      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% :     0.000009s :      1: graph_reusing
13.89% :     0.006414s :      1: jit_opt_a
 0.45% :     0.000209s :      1: jit_opt_after_cconv
 0.17% :     0.000079s :      1: jit_opt_b
 1.10% :     0.000508s :      1: loop_unroll
 1.53% :     0.000708s :      1: mutable_eliminate
 2.73% :     0.001262s :     26: opt.transform.jit_opt_a
 0.19% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000047s :      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
 0.08% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000054s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000551s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000125s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000023s :      1: remove_dup_value
 1.47% :     0.000681s :      1: renormalize.infer
 0.99% :     0.000459s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.68% :     0.000315s :      1: rewriter_after_opt_a
 0.22% :     0.000103s :      1: rewriter_before_opt_a
 0.25% :     0.000114s :      1: symbol_engine_optimizer
22.62% :     0.010449s :      1: task_emit
49.94% :     0.023069s :      1: type_inference
 0.18% :     0.000084s :      1: validate


TotalTime = 0.0417846, [33]
[bootstrap]: 0.00031204
[type_inference]: 0.0219007
[event_method]: 1.965e-05
[auto_monad]: 0.00015697
[graph_reusing]: 6.55002e-06
[pre_auto_parallel]: 2.34001e-06
[py_interpret_to_execute]: 0.00012345
[rewriter_before_opt_a]: 9.544e-05
[expand_dump_flag]: 4.01001e-06
[jit_opt_a]: 0.00622299, [2]
    [Cycle 1]: 0.00240743, [27]
        [switch_simplify]: 6.336e-05
        [loop_unroll]: 3.67e-05
        [a_1]: 0.00074315
        [with_stream_mark]: 2.343e-05
        [recompute_prepare]: 1.078e-05
        [updatestate_depend_eliminate]: 6.41998e-06
        [updatestate_assign_eliminate]: 5.42001e-06
        [updatestate_loads_eliminate]: 4.57e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 9.52001e-06
        [updatestate_useless_node_eliminater]: 1.189e-05
        [accelerated_algorithm]: 8.47e-06
        [meta_shard_fg_expand]: 2.83e-06
        [get_grad_eliminate_]: 8.32e-06
        [merge_forward]: 4.89e-06
        [cell_reuse_recompute_pass]: 1.61998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.892e-05
        [j_node_and_user_rematch]: 1.423e-05
        [meta_fg_expand]: 3.91001e-06
        [replace_old_param]: 1.187e-05
        [inline_without_move]: 7.97998e-06
        [renormalize]: 0.00114017
        [add_forward_monad_depend]: 5.99e-06
        [auto_monad_grad]: 2.86e-06
        [auto_monad_eliminator]: 2.071e-05
        [cse]: 4.026e-05
        [replace_applicator]: 1.729e-05
    [Cycle 2]: 0.00046974, [27]
        [switch_simplify]: 9.17001e-06
        [loop_unroll]: 8.32998e-06
        [a_1]: 0.00016904
        [with_stream_mark]: 1.159e-05
        [recompute_prepare]: 8.25e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 4.4e-06
        [updatestate_loads_eliminate]: 4.22998e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 8.02998e-06
        [updatestate_useless_node_eliminater]: 1.052e-05
        [accelerated_algorithm]: 7.82e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 7.51999e-06
        [merge_forward]: 4.35e-06
        [cell_reuse_recompute_pass]: 2.07999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.614e-05
        [j_node_and_user_rematch]: 1.267e-05
        [meta_fg_expand]: 3.09001e-06
        [replace_old_param]: 1.029e-05
        [inline_without_move]: 7.35998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 1.029e-05
        [cse]: 2.098e-05
        [replace_applicator]: 7.98001e-06
[py_interpret_to_execute_after_opt_a]: 1.308e-05
[rewriter_after_opt_a]: 0.00027883
[convert_after_rewriter]: 1.225e-05
[order_py_execute_after_rewriter]: 7.71999e-06
[mutable_eliminate]: 0.0007164
[jit_opt_b]: 6.9e-05, [1]
    [Cycle 1]: 6.114e-05, [2]
        [frontend_op_eliminate]: 2.418e-05
        [inline_after_opt_a]: 2.549e-05
[cconv]: 2.642e-05
[loop_unroll]: 0.00045868
[jit_opt_after_cconv]: 0.00019968, [1]
    [Cycle 1]: 0.00019303, [11]
        [c_1]: 4.727e-05
        [parameter_eliminate]: 2.84999e-06
        [updatestate_depend_eliminate]: 8.31002e-06
        [updatestate_assign_eliminate]: 4.52e-06
        [updatestate_loads_eliminate]: 4.13999e-06
        [cse]: 3.089e-05
        [call_graph_tuple_transform]: 2.279e-05
        [tuple_list_get_item_eliminator]: 8.51002e-06
        [none_parameter_eliminate]: 1.58002e-06
        [renormalize]: 7.30011e-07
        [switch_simplify]: 9.32999e-06
[remove_dup_value]: 1.861e-05
[partial_unused_args_eliminate]: 2.03997e-06
[environ_conv]: 1.139e-05
[add_recomputation]: 6.789e-05
[cse_after_recomputation]: 2.921e-05, [1]
    [Cycle 1]: 2.327e-05, [1]
        [cse]: 1.738e-05
[auto_monad_reorder]: 2.657e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 6.44999e-06
[opt_after_jit_grad]: 0.00049765
[symbol_engine_optimizer]: 0.0001009, [1]
    [Cycle 1]: 9.41e-05, [6]
        [build]: 1.062e-05
        [elim_shapecalc]: 1.208e-05
        [elim_not_effective]: 1.83e-05
        [opt_reshape]: 8.95001e-06
        [fold_const_symbol]: 1.407e-05
        [renormalize]: 6.19999e-07
[validate]: 5.347e-05
[backend_pass]: 1.27e-06
[task_emit]: 0.010115
[execute]: 8.31002e-06
Sums
    bootstrap                            :     0.000312s :  0.83%
    type_inference                       :     0.021901s : 57.94%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000157s :  0.42%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000123s :  0.33%
    rewriter_before_opt_a                :     0.000095s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000912s :  2.41%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    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 :     0.000035s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000015s :  0.04%
    jit_opt_a.renormalize                :     0.001140s :  3.02%
    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      :     0.000031s :  0.08%
    jit_opt_a.cse                        :     0.000061s :  0.16%
    jit_opt_a.replace_applicator         :     0.000025s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000279s :  0.74%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000716s :  1.90%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.07%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000459s :  1.21%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.13%
    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              :     0.000031s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000068s :  0.18%
    cse_after_recomputation.cse          :     0.000017s :  0.05%
    auto_monad_reorder                   :     0.000027s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000498s :  1.32%
    symbol_engine_optimizer.build        :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000053s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010115s : 26.76%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000286    47
 3.01% :     0.000009s :      2: substitution.depend_value_elim
 0.94% :     0.000003s :      4: substitution.elim_not_effective
 0.80% :     0.000002s :      4: substitution.fold_const_symbol
 2.52% :     0.000007s :      5: substitution.graph_param_transform
75.62% :     0.000216s :      4: substitution.inline
 1.76% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.40% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.66% :     0.000005s :      2: substitution.replace_old_param
 4.64% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.42% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 3.23% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021829     2
93.28% :     0.020362s :      1: type_inference.infer
 6.72% :     0.001467s :      1: type_inference.specialize
------[replace.]   0.000100     6
68.42% :     0.000068s :      4: replace.inline
31.58% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000225     6
94.63% :     0.000213s :      4: match.inline
 5.37% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000177  1021
 1.35% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.42% :     0.000003s :     15: predicate.addn_check_dump
 1.22% :     0.000002s :     15: predicate.addn_zero_filter
 1.99% :     0.000004s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.48% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     15: predicate.compare_switch_simplify
 1.41% :     0.000002s :     15: predicate.depend_value_elim
 1.18% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      5: predicate.elim_not_effective
 0.58% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.22% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     15: predicate.environ_get_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.08% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 5.56% :     0.000010s :     31: predicate.inline
 1.04% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :     10: predicate.less_batch_normalization
 1.45% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000004s :     22: predicate.load_eliminater
 1.44% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.49% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.11% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000002s :     15: predicate.merge_addn
 1.22% :     0.000002s :     15: predicate.minmaximum_grad
 1.44% :     0.000003s :      5: predicate.mutable_eliminate
 0.47% :     0.000001s :      5: predicate.opt_reshape
 2.60% :     0.000005s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.77% :     0.000003s :     15: predicate.reduce_eliminate
 1.47% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.80% :     0.000003s :     27: predicate.replace_applicator
 0.80% :     0.000001s :     10: predicate.replace_old_param
 0.46% :     0.000001s :      5: predicate.reset_defer_inline
 1.18% :     0.000002s :     15: predicate.reshape_eliminate
 1.35% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.89% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     15: predicate.same_eliminate
 0.55% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.08% :     0.000002s :     10: predicate.special_op_eliminate
 1.08% :     0.000002s :     10: predicate.specialize_transform
 1.43% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.29% :     0.000004s :     21: predicate.switch_defer_inline
 2.23% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.67% :     0.000012s :     68: predicate.switch_simplify
 1.28% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.48% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 2.11% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.23% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000800    13
45.30% :     0.000362s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.29% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.42% :     0.000387s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044330    76
 0.16% :     0.000071s :      1: add_recomputation
 0.37% :     0.000162s :      1: auto_monad
 0.07% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.74% :     0.000328s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000023s :      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
14.05% :     0.006226s :      1: jit_opt_a
 0.46% :     0.000203s :      1: jit_opt_after_cconv
 0.16% :     0.000072s :      1: jit_opt_b
 1.05% :     0.000467s :      1: loop_unroll
 1.64% :     0.000727s :      1: mutable_eliminate
 2.74% :     0.001215s :     26: opt.transform.jit_opt_a
 0.19% :     0.000084s :      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% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.14% :     0.000507s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000127s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.55% :     0.000689s :      1: renormalize.infer
 1.00% :     0.000443s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.64% :     0.000283s :      1: rewriter_after_opt_a
 0.22% :     0.000099s :      1: rewriter_before_opt_a
 0.23% :     0.000104s :      1: symbol_engine_optimizer
22.85% :     0.010128s :      1: task_emit
49.43% :     0.021914s :      1: type_inference
 0.18% :     0.000078s :      1: validate


TotalTime = 0.0440062, [33]
[bootstrap]: 0.00028829
[type_inference]: 0.0218843
[event_method]: 1.907e-05
[auto_monad]: 0.00014777
[graph_reusing]: 5.96e-06
[pre_auto_parallel]: 2.89001e-06
[py_interpret_to_execute]: 0.00012812
[rewriter_before_opt_a]: 9.542e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00615105, [2]
    [Cycle 1]: 0.00229087, [27]
        [switch_simplify]: 5.496e-05
        [loop_unroll]: 3.658e-05
        [a_1]: 0.00070143
        [with_stream_mark]: 1.803e-05
        [recompute_prepare]: 1.074e-05
        [updatestate_depend_eliminate]: 5.64998e-06
        [updatestate_assign_eliminate]: 5.14e-06
        [updatestate_loads_eliminate]: 4.62e-06
        [parameter_eliminate]: 1.36998e-06
        [specialize_transform]: 9.52001e-06
        [updatestate_useless_node_eliminater]: 1.132e-05
        [accelerated_algorithm]: 8.64003e-06
        [meta_shard_fg_expand]: 2.83e-06
        [get_grad_eliminate_]: 8.75999e-06
        [merge_forward]: 4.62998e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.92e-05
        [j_node_and_user_rematch]: 1.345e-05
        [meta_fg_expand]: 3.77002e-06
        [replace_old_param]: 1.226e-05
        [inline_without_move]: 8.31002e-06
        [renormalize]: 0.00110074
        [add_forward_monad_depend]: 5.44998e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.992e-05
        [cse]: 3.489e-05
        [replace_applicator]: 1.768e-05
    [Cycle 2]: 0.00047593, [27]
        [switch_simplify]: 8.82999e-06
        [loop_unroll]: 7.85e-06
        [a_1]: 0.00016442
        [with_stream_mark]: 1.088e-05
        [recompute_prepare]: 8.38001e-06
        [updatestate_depend_eliminate]: 4.95001e-06
        [updatestate_assign_eliminate]: 4.91002e-06
        [updatestate_loads_eliminate]: 4.13001e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 7.85e-06
        [updatestate_useless_node_eliminater]: 1.051e-05
        [accelerated_algorithm]: 7.58999e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 7.74002e-06
        [merge_forward]: 4.31002e-06
        [cell_reuse_recompute_pass]: 1.92001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.63e-05
        [j_node_and_user_rematch]: 1.319e-05
        [meta_fg_expand]: 2.88e-06
        [replace_old_param]: 1.031e-05
        [inline_without_move]: 7.43e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.67001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 1.008e-05
        [cse]: 2.104e-05
        [replace_applicator]: 7.64002e-06
[py_interpret_to_execute_after_opt_a]: 1.316e-05
[rewriter_after_opt_a]: 0.00027983
[convert_after_rewriter]: 1.022e-05
[order_py_execute_after_rewriter]: 7.69002e-06
[mutable_eliminate]: 0.00065012
[jit_opt_b]: 8.034e-05, [1]
    [Cycle 1]: 7.351e-05, [2]
        [frontend_op_eliminate]: 3.526e-05
        [inline_after_opt_a]: 2.6e-05
[cconv]: 3.569e-05
[loop_unroll]: 0.00046495
[jit_opt_after_cconv]: 0.00018959, [1]
    [Cycle 1]: 0.0001828, [11]
        [c_1]: 4.419e-05
        [parameter_eliminate]: 2.70002e-06
        [updatestate_depend_eliminate]: 7.33e-06
        [updatestate_assign_eliminate]: 4.63001e-06
        [updatestate_loads_eliminate]: 4.33999e-06
        [cse]: 2.734e-05
        [call_graph_tuple_transform]: 2.224e-05
        [tuple_list_get_item_eliminator]: 8.35001e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 5.29981e-07
        [switch_simplify]: 8.62e-06
[remove_dup_value]: 1.773e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 1.181e-05
[add_recomputation]: 5.922e-05
[cse_after_recomputation]: 2.899e-05, [1]
    [Cycle 1]: 2.33e-05, [1]
        [cse]: 1.709e-05
[auto_monad_reorder]: 2.442e-05
[get_jit_bprop_graph]: 2.12001e-06
[rewriter_after_jit_bprop_graph]: 5.39e-06
[opt_after_jit_grad]: 0.00050853
[symbol_engine_optimizer]: 0.00010246, [1]
    [Cycle 1]: 9.584e-05, [6]
        [build]: 1.201e-05
        [elim_shapecalc]: 1.226e-05
        [elim_not_effective]: 1.924e-05
        [opt_reshape]: 8.84003e-06
        [fold_const_symbol]: 1.35e-05
        [renormalize]: 5.29981e-07
[validate]: 5.488e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.0125008
[execute]: 8.53001e-06
Sums
    bootstrap                            :     0.000288s :  0.72%
    type_inference                       :     0.021884s : 54.73%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000148s :  0.37%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000128s :  0.32%
    rewriter_before_opt_a                :     0.000095s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.16%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000866s :  2.17%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001101s :  2.75%
    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      :     0.000030s :  0.08%
    jit_opt_a.cse                        :     0.000056s :  0.14%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000280s :  0.70%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000650s :  1.63%
    jit_opt_b.frontend_op_eliminate      :     0.000035s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.07%
    cconv                                :     0.000036s :  0.09%
    loop_unroll                          :     0.000465s :  1.16%
    jit_opt_after_cconv.c_1              :     0.000044s :  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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000018s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000059s :  0.15%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000024s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000509s :  1.27%
    symbol_engine_optimizer.build        :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000055s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012501s : 31.27%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000269    47
 3.13% :     0.000008s :      2: substitution.depend_value_elim
 1.10% :     0.000003s :      4: substitution.elim_not_effective
 0.72% :     0.000002s :      4: substitution.fold_const_symbol
 2.43% :     0.000007s :      5: substitution.graph_param_transform
75.42% :     0.000203s :      4: substitution.inline
 1.80% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.68% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.73% :     0.000005s :      2: substitution.replace_old_param
 4.80% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.06% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 3.13% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.021819     2
93.08% :     0.020308s :      1: type_inference.infer
 6.92% :     0.001511s :      1: type_inference.specialize
------[replace.]   0.000092     6
65.96% :     0.000061s :      4: replace.inline
34.04% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.49% :     0.000200s :      4: match.inline
 5.51% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.36% :     0.000002s :     15: predicate.addn_zero_filter
 2.28% :     0.000004s :     15: predicate.arithmetic_simplify
 1.52% :     0.000003s :     15: predicate.cast_eliminate
 0.53% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.28% :     0.000002s :     15: predicate.compare_switch_simplify
 1.35% :     0.000002s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.60% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     15: predicate.environ_get_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.20% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 4.81% :     0.000008s :     31: predicate.inline
 1.06% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :     10: predicate.less_batch_normalization
 1.61% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.03% :     0.000003s :     22: predicate.load_eliminater
 1.38% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.61% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.34% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     15: predicate.merge_addn
 1.13% :     0.000002s :     15: predicate.minmaximum_grad
 1.26% :     0.000002s :      5: predicate.mutable_eliminate
 0.51% :     0.000001s :      5: predicate.opt_reshape
 2.36% :     0.000004s :     22: predicate.partial_eliminate
 1.19% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.80% :     0.000003s :     15: predicate.reduce_eliminate
 1.57% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.79% :     0.000003s :     27: predicate.replace_applicator
 0.84% :     0.000001s :     10: predicate.replace_old_param
 0.36% :     0.000001s :      5: predicate.reset_defer_inline
 1.22% :     0.000002s :     15: predicate.reshape_eliminate
 1.32% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.50% :     0.000003s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.03% :     0.000002s :     10: predicate.special_op_eliminate
 1.18% :     0.000002s :     10: predicate.specialize_transform
 1.54% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.07% :     0.000004s :     21: predicate.switch_defer_inline
 2.00% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.37% :     0.000011s :     68: predicate.switch_simplify
 1.28% :     0.000002s :     15: predicate.tile_eliminate
 1.24% :     0.000002s :     15: predicate.transpose_eliminate
 1.73% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000844    13
46.80% :     0.000395s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.08% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.12% :     0.000406s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046467    76
 0.13% :     0.000062s :      1: add_recomputation
 0.33% :     0.000153s :      1: auto_monad
 0.06% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000307s :      1: bootstrap
 0.08% :     0.000039s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000023s :      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% :     0.000009s :      1: graph_reusing
13.25% :     0.006155s :      1: jit_opt_a
 0.41% :     0.000192s :      1: jit_opt_after_cconv
 0.18% :     0.000083s :      1: jit_opt_b
 1.02% :     0.000473s :      1: loop_unroll
 1.42% :     0.000659s :      1: mutable_eliminate
 2.50% :     0.001164s :     26: opt.transform.jit_opt_a
 0.17% :     0.000080s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000054s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.11% :     0.000517s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000132s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000020s :      1: remove_dup_value
 1.40% :     0.000653s :      1: renormalize.infer
 0.95% :     0.000440s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000284s :      1: rewriter_after_opt_a
 0.21% :     0.000099s :      1: rewriter_before_opt_a
 0.23% :     0.000105s :      1: symbol_engine_optimizer
26.94% :     0.012518s :      1: task_emit
47.12% :     0.021895s :      1: type_inference
 0.17% :     0.000077s :      1: validate


TotalTime = 0.0462371, [33]
[bootstrap]: 0.00029254
[type_inference]: 0.0232696
[event_method]: 2.287e-05
[auto_monad]: 0.00018859
[graph_reusing]: 6.70002e-06
[pre_auto_parallel]: 3.06999e-06
[py_interpret_to_execute]: 0.00013345
[rewriter_before_opt_a]: 0.00011141
[expand_dump_flag]: 3.86001e-06
[jit_opt_a]: 0.00642714, [2]
    [Cycle 1]: 0.00240571, [27]
        [switch_simplify]: 6.336e-05
        [loop_unroll]: 3.678e-05
        [a_1]: 0.00073169
        [with_stream_mark]: 2.052e-05
        [recompute_prepare]: 1.179e-05
        [updatestate_depend_eliminate]: 6.21998e-06
        [updatestate_assign_eliminate]: 5.72999e-06
        [updatestate_loads_eliminate]: 4.80999e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 9.44e-06
        [updatestate_useless_node_eliminater]: 1.114e-05
        [accelerated_algorithm]: 8.70001e-06
        [meta_shard_fg_expand]: 3.28e-06
        [get_grad_eliminate_]: 8.31002e-06
        [merge_forward]: 5.60001e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.893e-05
        [j_node_and_user_rematch]: 1.385e-05
        [meta_fg_expand]: 3.78001e-06
        [replace_old_param]: 1.16e-05
        [inline_without_move]: 7.9e-06
        [renormalize]: 0.00114729
        [add_forward_monad_depend]: 6.11e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 2.187e-05
        [cse]: 4.076e-05
        [replace_applicator]: 1.723e-05
    [Cycle 2]: 0.00046766, [27]
        [switch_simplify]: 8.98002e-06
        [loop_unroll]: 7.69997e-06
        [a_1]: 0.00016579
        [with_stream_mark]: 1.251e-05
        [recompute_prepare]: 8.02e-06
        [updatestate_depend_eliminate]: 4.84e-06
        [updatestate_assign_eliminate]: 4.78001e-06
        [updatestate_loads_eliminate]: 4.05e-06
        [parameter_eliminate]: 1.49998e-06
        [specialize_transform]: 7.88999e-06
        [updatestate_useless_node_eliminater]: 1.075e-05
        [accelerated_algorithm]: 7.95e-06
        [meta_shard_fg_expand]: 1.98997e-06
        [get_grad_eliminate_]: 7.5e-06
        [merge_forward]: 4.68999e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.706e-05
        [j_node_and_user_rematch]: 1.278e-05
        [meta_fg_expand]: 2.73e-06
        [replace_old_param]: 1.004e-05
        [inline_without_move]: 7.70998e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 9.90025e-07
        [auto_monad_eliminator]: 1e-05
        [cse]: 2.146e-05
        [replace_applicator]: 7.76001e-06
[py_interpret_to_execute_after_opt_a]: 1.218e-05
[rewriter_after_opt_a]: 0.000281
[convert_after_rewriter]: 1.183e-05
[order_py_execute_after_rewriter]: 7.35e-06
[mutable_eliminate]: 0.00061789
[jit_opt_b]: 6.954e-05, [1]
    [Cycle 1]: 6.2e-05, [2]
        [frontend_op_eliminate]: 2.496e-05
        [inline_after_opt_a]: 2.488e-05
[cconv]: 2.664e-05
[loop_unroll]: 0.00046026
[jit_opt_after_cconv]: 0.00019673, [1]
    [Cycle 1]: 0.00019, [11]
        [c_1]: 4.485e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 8.79998e-06
        [updatestate_assign_eliminate]: 4.48999e-06
        [updatestate_loads_eliminate]: 4e-06
        [cse]: 2.929e-05
        [call_graph_tuple_transform]: 2.307e-05
        [tuple_list_get_item_eliminator]: 8.77999e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 9.12001e-06
[remove_dup_value]: 1.918e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 1.134e-05
[add_recomputation]: 0.00011128
[cse_after_recomputation]: 3.182e-05, [1]
    [Cycle 1]: 2.521e-05, [1]
        [cse]: 1.891e-05
[auto_monad_reorder]: 2.615e-05
[get_jit_bprop_graph]: 2.56e-06
[rewriter_after_jit_bprop_graph]: 5.60001e-06
[opt_after_jit_grad]: 0.00051476
[symbol_engine_optimizer]: 0.00010656, [1]
    [Cycle 1]: 9.936e-05, [6]
        [build]: 1.412e-05
        [elim_shapecalc]: 1.224e-05
        [elim_not_effective]: 1.885e-05
        [opt_reshape]: 9.24998e-06
        [fold_const_symbol]: 1.433e-05
        [renormalize]: 5.8001e-07
[validate]: 6.02e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.0129482
[execute]: 7.95e-06
Sums
    bootstrap                            :     0.000293s :  0.70%
    type_inference                       :     0.023270s : 55.39%
    event_method                         :     0.000023s :  0.05%
    auto_monad                           :     0.000189s :  0.45%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000133s :  0.32%
    rewriter_before_opt_a                :     0.000111s :  0.27%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.17%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000897s :  2.14%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001147s :  2.73%
    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      :     0.000032s :  0.08%
    jit_opt_a.cse                        :     0.000062s :  0.15%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000281s :  0.67%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000618s :  1.47%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000027s :  0.06%
    loop_unroll                          :     0.000460s :  1.10%
    jit_opt_after_cconv.c_1              :     0.000045s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000029s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000111s :  0.26%
    cse_after_recomputation.cse          :     0.000019s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000515s :  1.23%
    symbol_engine_optimizer.build        :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000060s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012948s : 30.82%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000280    47
 3.39% :     0.000009s :      2: substitution.depend_value_elim
 1.01% :     0.000003s :      4: substitution.elim_not_effective
 0.73% :     0.000002s :      4: substitution.fold_const_symbol
 2.47% :     0.000007s :      5: substitution.graph_param_transform
75.29% :     0.000211s :      4: substitution.inline
 1.84% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.40% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.59% :     0.000004s :      2: substitution.replace_old_param
 4.77% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.05% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.46% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023193     2
93.32% :     0.021644s :      1: type_inference.infer
 6.68% :     0.001549s :      1: type_inference.specialize
------[replace.]   0.000097     6
65.98% :     0.000064s :      4: replace.inline
34.02% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.47% :     0.000208s :      4: match.inline
 5.53% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000174  1021
 1.46% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     15: predicate.addn_check_dump
 1.42% :     0.000002s :     15: predicate.addn_zero_filter
 2.19% :     0.000004s :     15: predicate.arithmetic_simplify
 1.75% :     0.000003s :     15: predicate.cast_eliminate
 0.73% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     15: predicate.compare_switch_simplify
 1.36% :     0.000002s :     15: predicate.depend_value_elim
 1.13% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.01% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.76% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.44% :     0.000003s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.09% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 5.23% :     0.000009s :     31: predicate.inline
 1.04% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :     10: predicate.less_batch_normalization
 1.51% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     22: predicate.load_eliminater
 1.23% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.33% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     15: predicate.merge_addn
 1.13% :     0.000002s :     15: predicate.minmaximum_grad
 1.57% :     0.000003s :      5: predicate.mutable_eliminate
 0.51% :     0.000001s :      5: predicate.opt_reshape
 2.39% :     0.000004s :     22: predicate.partial_eliminate
 1.26% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.68% :     0.000003s :     15: predicate.reduce_eliminate
 1.40% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.74% :     0.000003s :     27: predicate.replace_applicator
 0.80% :     0.000001s :     10: predicate.replace_old_param
 0.40% :     0.000001s :      5: predicate.reset_defer_inline
 1.17% :     0.000002s :     15: predicate.reshape_eliminate
 1.19% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.89% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.02% :     0.000002s :     10: predicate.special_op_eliminate
 1.09% :     0.000002s :     10: predicate.specialize_transform
 1.64% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.14% :     0.000004s :     21: predicate.switch_defer_inline
 1.85% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.67% :     0.000012s :     68: predicate.switch_simplify
 1.31% :     0.000002s :     15: predicate.tile_eliminate
 1.52% :     0.000003s :     15: predicate.transpose_eliminate
 1.42% :     0.000002s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.79% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.25% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     15: predicate.value_based_eliminate
 0.42% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000838    13
43.18% :     0.000362s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.06% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.75% :     0.000434s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048772    76
 0.24% :     0.000115s :      1: add_recomputation
 0.40% :     0.000195s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.63% :     0.000307s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.06% :     0.000028s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
13.18% :     0.006430s :      1: jit_opt_a
 0.41% :     0.000200s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 0.96% :     0.000469s :      1: loop_unroll
 1.29% :     0.000627s :      1: mutable_eliminate
 2.46% :     0.001200s :     26: opt.transform.jit_opt_a
 0.17% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.07% :     0.000524s :      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
 0.28% :     0.000139s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000022s :      1: remove_dup_value
 1.41% :     0.000690s :      1: renormalize.infer
 0.92% :     0.000449s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.59% :     0.000286s :      1: rewriter_after_opt_a
 0.24% :     0.000116s :      1: rewriter_before_opt_a
 0.22% :     0.000109s :      1: symbol_engine_optimizer
26.58% :     0.012965s :      1: task_emit
47.77% :     0.023300s :      1: type_inference
 0.18% :     0.000085s :      1: validate


TotalTime = 0.044633, [33]
[bootstrap]: 0.0003076
[type_inference]: 0.0226463
[event_method]: 2.215e-05
[auto_monad]: 0.00016348
[graph_reusing]: 6.93e-06
[pre_auto_parallel]: 3.4e-06
[py_interpret_to_execute]: 0.00012899
[rewriter_before_opt_a]: 0.00011286
[expand_dump_flag]: 4.73001e-06
[jit_opt_a]: 0.00670204, [2]
    [Cycle 1]: 0.00263478, [27]
        [switch_simplify]: 6.83e-05
        [loop_unroll]: 3.656e-05
        [a_1]: 0.00082827
        [with_stream_mark]: 2.604e-05
        [recompute_prepare]: 1.087e-05
        [updatestate_depend_eliminate]: 6.36998e-06
        [updatestate_assign_eliminate]: 6.01e-06
        [updatestate_loads_eliminate]: 4.56002e-06
        [parameter_eliminate]: 2.17001e-06
        [specialize_transform]: 9.59999e-06
        [updatestate_useless_node_eliminater]: 1.155e-05
        [accelerated_algorithm]: 8.55999e-06
        [meta_shard_fg_expand]: 3.04001e-06
        [get_grad_eliminate_]: 8.17e-06
        [merge_forward]: 5.86998e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.926e-05
        [j_node_and_user_rematch]: 1.384e-05
        [meta_fg_expand]: 3.71999e-06
        [replace_old_param]: 1.24e-05
        [inline_without_move]: 8.22e-06
        [renormalize]: 0.00125488
        [add_forward_monad_depend]: 6.84999e-06
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 2.213e-05
        [cse]: 4.358e-05
        [replace_applicator]: 1.636e-05
    [Cycle 2]: 0.00046344, [27]
        [switch_simplify]: 9.09998e-06
        [loop_unroll]: 7.83001e-06
        [a_1]: 0.00016606
        [with_stream_mark]: 1.348e-05
        [recompute_prepare]: 8.41002e-06
        [updatestate_depend_eliminate]: 5.17e-06
        [updatestate_assign_eliminate]: 4.58001e-06
        [updatestate_loads_eliminate]: 4.2e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 8.00999e-06
        [updatestate_useless_node_eliminater]: 1.058e-05
        [accelerated_algorithm]: 7.62998e-06
        [meta_shard_fg_expand]: 1.58002e-06
        [get_grad_eliminate_]: 7.35e-06
        [merge_forward]: 4.17e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.673e-05
        [j_node_and_user_rematch]: 1.237e-05
        [meta_fg_expand]: 2.93998e-06
        [replace_old_param]: 9.79e-06
        [inline_without_move]: 7.55e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 1.004e-05
        [cse]: 1.978e-05
        [replace_applicator]: 7.94002e-06
[py_interpret_to_execute_after_opt_a]: 1.37e-05
[rewriter_after_opt_a]: 0.00028989
[convert_after_rewriter]: 1.161e-05
[order_py_execute_after_rewriter]: 7.23999e-06
[mutable_eliminate]: 0.00067187
[jit_opt_b]: 6.908e-05, [1]
    [Cycle 1]: 6.147e-05, [2]
        [frontend_op_eliminate]: 2.478e-05
        [inline_after_opt_a]: 2.477e-05
[cconv]: 2.878e-05
[loop_unroll]: 0.00050747
[jit_opt_after_cconv]: 0.00020465, [1]
    [Cycle 1]: 0.00019709, [11]
        [c_1]: 4.65e-05
        [parameter_eliminate]: 2.88003e-06
        [updatestate_depend_eliminate]: 9.22999e-06
        [updatestate_assign_eliminate]: 4.28001e-06
        [updatestate_loads_eliminate]: 4.50999e-06
        [cse]: 3.233e-05
        [call_graph_tuple_transform]: 2.423e-05
        [tuple_list_get_item_eliminator]: 8.17e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 1.03e-05
[remove_dup_value]: 1.973e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 1.181e-05
[add_recomputation]: 7.032e-05
[cse_after_recomputation]: 2.961e-05, [1]
    [Cycle 1]: 2.354e-05, [1]
        [cse]: 1.734e-05
[auto_monad_reorder]: 2.69e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 6.37001e-06
[opt_after_jit_grad]: 0.00051092
[symbol_engine_optimizer]: 0.00010938, [1]
    [Cycle 1]: 0.0001023, [6]
        [build]: 1.54e-05
        [elim_shapecalc]: 1.303e-05
        [elim_not_effective]: 1.938e-05
        [opt_reshape]: 9.56e-06
        [fold_const_symbol]: 1.443e-05
        [renormalize]: 4.50003e-07
[validate]: 6.042e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.0116271
[execute]: 9.54e-06
Sums
    bootstrap                            :     0.000308s :  0.76%
    type_inference                       :     0.022646s : 56.12%
    event_method                         :     0.000022s :  0.05%
    auto_monad                           :     0.000163s :  0.41%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000129s :  0.32%
    rewriter_before_opt_a                :     0.000113s :  0.28%
    expand_dump_flag                     :     0.000005s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.19%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000994s :  2.46%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001255s :  3.11%
    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      :     0.000032s :  0.08%
    jit_opt_a.cse                        :     0.000063s :  0.16%
    jit_opt_a.replace_applicator         :     0.000024s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000290s :  0.72%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000672s :  1.67%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000029s :  0.07%
    loop_unroll                          :     0.000507s :  1.26%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000010s :  0.03%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000070s :  0.17%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000027s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000511s :  1.27%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000060s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011627s : 28.82%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000327    47
 3.02% :     0.000010s :      2: substitution.depend_value_elim
 0.88% :     0.000003s :      4: substitution.elim_not_effective
 0.71% :     0.000002s :      4: substitution.fold_const_symbol
 2.01% :     0.000007s :      5: substitution.graph_param_transform
78.31% :     0.000256s :      4: substitution.inline
 1.49% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.21% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.42% :     0.000005s :      2: substitution.replace_old_param
 4.24% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 2.84% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.88% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022569     2
93.13% :     0.021019s :      1: type_inference.infer
 6.87% :     0.001550s :      1: type_inference.specialize
------[replace.]   0.000096     6
67.67% :     0.000065s :      4: replace.inline
32.33% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     6
95.18% :     0.000253s :      4: match.inline
 4.82% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000172  1021
 1.31% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.24% :     0.000002s :     15: predicate.addn_check_dump
 1.42% :     0.000002s :     15: predicate.addn_zero_filter
 2.16% :     0.000004s :     15: predicate.arithmetic_simplify
 1.37% :     0.000002s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.29% :     0.000002s :     15: predicate.depend_value_elim
 1.12% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.51% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.93% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.67% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     15: predicate.environ_get_eliminate
 1.10% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.08% :     0.000002s :     10: predicate.get_grad_eliminate
 0.31% :     0.000001s :      5: predicate.graph_param_transform
 5.43% :     0.000009s :     31: predicate.inline
 0.99% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.16% :     0.000002s :     10: predicate.less_batch_normalization
 1.44% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.82% :     0.000003s :     22: predicate.load_eliminater
 1.54% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.38% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     15: predicate.merge_addn
 1.16% :     0.000002s :     15: predicate.minmaximum_grad
 1.63% :     0.000003s :      5: predicate.mutable_eliminate
 0.56% :     0.000001s :      5: predicate.opt_reshape
 2.31% :     0.000004s :     22: predicate.partial_eliminate
 1.25% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.95% :     0.000003s :     15: predicate.reduce_eliminate
 1.43% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.64% :     0.000003s :     27: predicate.replace_applicator
 0.53% :     0.000001s :     10: predicate.replace_old_param
 0.30% :     0.000001s :      5: predicate.reset_defer_inline
 1.43% :     0.000002s :     15: predicate.reshape_eliminate
 1.38% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.15% :     0.000002s :     10: predicate.special_op_eliminate
 1.06% :     0.000002s :     10: predicate.specialize_transform
 1.71% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.10% :     0.000004s :     21: predicate.switch_defer_inline
 1.94% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.76% :     0.000012s :     68: predicate.switch_simplify
 1.18% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000002s :     15: predicate.transpose_eliminate
 1.51% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 2.12% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.20% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     15: predicate.value_based_eliminate
 0.42% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000891    13
42.82% :     0.000381s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.01% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.17% :     0.000447s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047379    76
 0.16% :     0.000074s :      1: add_recomputation
 0.36% :     0.000169s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000325s :      1: bootstrap
 0.07% :     0.000031s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.06% :     0.000027s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.02% :     0.000008s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
14.15% :     0.006706s :      1: jit_opt_a
 0.44% :     0.000208s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 1.09% :     0.000517s :      1: loop_unroll
 1.44% :     0.000681s :      1: mutable_eliminate
 2.75% :     0.001301s :     26: opt.transform.jit_opt_a
 0.18% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.10% :     0.000520s :      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% :     0.000006s :      1: pre_auto_parallel
 0.28% :     0.000133s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.62% :     0.000768s :      1: renormalize.infer
 1.01% :     0.000477s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.62% :     0.000295s :      1: rewriter_after_opt_a
 0.25% :     0.000119s :      1: rewriter_before_opt_a
 0.24% :     0.000112s :      1: symbol_engine_optimizer
24.58% :     0.011647s :      1: task_emit
47.84% :     0.022664s :      1: type_inference
 0.19% :     0.000088s :      1: validate


TotalTime = 0.0463025, [33]
[bootstrap]: 0.00031289
[type_inference]: 0.0242443
[event_method]: 2.399e-05
[auto_monad]: 0.00016363
[graph_reusing]: 6.52001e-06
[pre_auto_parallel]: 3.2e-06
[py_interpret_to_execute]: 0.0001344
[rewriter_before_opt_a]: 9.979e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.006515, [2]
    [Cycle 1]: 0.00240899, [27]
        [switch_simplify]: 6.497e-05
        [loop_unroll]: 3.619e-05
        [a_1]: 0.00072633
        [with_stream_mark]: 2.224e-05
        [recompute_prepare]: 1.15e-05
        [updatestate_depend_eliminate]: 6.48e-06
        [updatestate_assign_eliminate]: 5.56e-06
        [updatestate_loads_eliminate]: 4.46002e-06
        [parameter_eliminate]: 2.79999e-06
        [specialize_transform]: 9.34998e-06
        [updatestate_useless_node_eliminater]: 1.133e-05
        [accelerated_algorithm]: 8.11002e-06
        [meta_shard_fg_expand]: 2.97002e-06
        [get_grad_eliminate_]: 8.05e-06
        [merge_forward]: 5.30001e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.947e-05
        [j_node_and_user_rematch]: 1.333e-05
        [meta_fg_expand]: 3.61999e-06
        [replace_old_param]: 1.124e-05
        [inline_without_move]: 8.36002e-06
        [renormalize]: 0.00113964
        [add_forward_monad_depend]: 1.193e-05
        [auto_monad_grad]: 2.98998e-06
        [auto_monad_eliminator]: 2.296e-05
        [cse]: 4.122e-05
        [replace_applicator]: 1.731e-05
    [Cycle 2]: 0.0004651, [27]
        [switch_simplify]: 9.17001e-06
        [loop_unroll]: 8.22e-06
        [a_1]: 0.00016521
        [with_stream_mark]: 1.326e-05
        [recompute_prepare]: 7.92e-06
        [updatestate_depend_eliminate]: 4.92e-06
        [updatestate_assign_eliminate]: 4.79998e-06
        [updatestate_loads_eliminate]: 3.98999e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 7.86001e-06
        [updatestate_useless_node_eliminater]: 1.028e-05
        [accelerated_algorithm]: 7.91001e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 7.28999e-06
        [merge_forward]: 4.2e-06
        [cell_reuse_recompute_pass]: 1.75001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.645e-05
        [j_node_and_user_rematch]: 1.3e-05
        [meta_fg_expand]: 2.96001e-06
        [replace_old_param]: 1.004e-05
        [inline_without_move]: 7.38e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 1.16002e-06
        [auto_monad_eliminator]: 9.73002e-06
        [cse]: 2.062e-05
        [replace_applicator]: 7.58999e-06
[py_interpret_to_execute_after_opt_a]: 1.44e-05
[rewriter_after_opt_a]: 0.00028861
[convert_after_rewriter]: 1.186e-05
[order_py_execute_after_rewriter]: 7.80998e-06
[mutable_eliminate]: 0.00123045
[jit_opt_b]: 6.842e-05, [1]
    [Cycle 1]: 6.078e-05, [2]
        [frontend_op_eliminate]: 2.425e-05
        [inline_after_opt_a]: 2.456e-05
[cconv]: 3.011e-05
[loop_unroll]: 0.00049245
[jit_opt_after_cconv]: 0.00020748, [1]
    [Cycle 1]: 0.00020099, [11]
        [c_1]: 4.706e-05
        [parameter_eliminate]: 3.59002e-06
        [updatestate_depend_eliminate]: 9.94001e-06
        [updatestate_assign_eliminate]: 4.62998e-06
        [updatestate_loads_eliminate]: 4.02002e-06
        [cse]: 3.431e-05
        [call_graph_tuple_transform]: 2.367e-05
        [tuple_list_get_item_eliminator]: 8.70999e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 9.32001e-06
[remove_dup_value]: 2.05e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 1.279e-05
[add_recomputation]: 7.022e-05
[cse_after_recomputation]: 3.106e-05, [1]
    [Cycle 1]: 2.517e-05, [1]
        [cse]: 1.918e-05
[auto_monad_reorder]: 2.593e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 5.97001e-06
[opt_after_jit_grad]: 0.00052271
[symbol_engine_optimizer]: 0.0001068, [1]
    [Cycle 1]: 0.00010018, [6]
        [build]: 1.551e-05
        [elim_shapecalc]: 1.222e-05
        [elim_not_effective]: 1.947e-05
        [opt_reshape]: 8.50999e-06
        [fold_const_symbol]: 1.417e-05
        [renormalize]: 3.60014e-07
[validate]: 6.131e-05
[backend_pass]: 1.05999e-06
[task_emit]: 0.0113283
[execute]: 7.65998e-06
Sums
    bootstrap                            :     0.000313s :  0.75%
    type_inference                       :     0.024244s : 57.75%
    event_method                         :     0.000024s :  0.06%
    auto_monad                           :     0.000164s :  0.39%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000134s :  0.32%
    rewriter_before_opt_a                :     0.000100s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.18%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000892s :  2.12%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001140s :  2.71%
    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      :     0.000033s :  0.08%
    jit_opt_a.cse                        :     0.000062s :  0.15%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000289s :  0.69%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.001230s :  2.93%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000030s :  0.07%
    loop_unroll                          :     0.000492s :  1.17%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.11%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000070s :  0.17%
    cse_after_recomputation.cse          :     0.000019s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000523s :  1.25%
    symbol_engine_optimizer.build        :     0.000016s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000061s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011328s : 26.98%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000294    47
 3.08% :     0.000009s :      2: substitution.depend_value_elim
 0.97% :     0.000003s :      4: substitution.elim_not_effective
 0.71% :     0.000002s :      4: substitution.fold_const_symbol
 2.41% :     0.000007s :      5: substitution.graph_param_transform
77.07% :     0.000227s :      4: substitution.inline
 1.62% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.38% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.45% :     0.000004s :      2: substitution.replace_old_param
 4.46% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 2.95% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.90% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024161     2
93.63% :     0.022623s :      1: type_inference.infer
 6.37% :     0.001538s :      1: type_inference.specialize
------[replace.]   0.000092     6
66.96% :     0.000062s :      4: replace.inline
33.04% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000236     6
94.91% :     0.000224s :      4: match.inline
 5.09% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1021
 1.33% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.15% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     15: predicate.addn_check_dump
 1.20% :     0.000002s :     15: predicate.addn_zero_filter
 1.93% :     0.000003s :     15: predicate.arithmetic_simplify
 1.39% :     0.000002s :     15: predicate.cast_eliminate
 0.63% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     15: predicate.compare_switch_simplify
 1.44% :     0.000002s :     15: predicate.depend_value_elim
 1.13% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.95% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      5: predicate.elim_not_effective
 0.66% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     15: predicate.environ_get_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.04% :     0.000002s :     10: predicate.get_grad_eliminate
 0.44% :     0.000001s :      5: predicate.graph_param_transform
 5.31% :     0.000009s :     31: predicate.inline
 1.02% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.16% :     0.000002s :     10: predicate.less_batch_normalization
 1.53% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.88% :     0.000003s :     22: predicate.load_eliminater
 1.48% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.50% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.12% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.79% :     0.000003s :      5: predicate.mutable_eliminate
 0.49% :     0.000001s :      5: predicate.opt_reshape
 2.42% :     0.000004s :     22: predicate.partial_eliminate
 1.22% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     15: predicate.reduce_eliminate
 1.45% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.74% :     0.000003s :     27: predicate.replace_applicator
 0.59% :     0.000001s :     10: predicate.replace_old_param
 0.50% :     0.000001s :      5: predicate.reset_defer_inline
 1.18% :     0.000002s :     15: predicate.reshape_eliminate
 1.20% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.12% :     0.000002s :     10: predicate.special_op_eliminate
 1.12% :     0.000002s :     10: predicate.specialize_transform
 1.46% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.16% :     0.000004s :     21: predicate.switch_defer_inline
 1.90% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.68% :     0.000011s :     68: predicate.switch_simplify
 1.28% :     0.000002s :     15: predicate.tile_eliminate
 1.24% :     0.000002s :     15: predicate.transpose_eliminate
 1.52% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.25% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.82% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.16% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000003s :     15: predicate.value_based_eliminate
 0.45% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000921    13
47.68% :     0.000439s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.39% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.93% :     0.000423s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048830    76
 0.15% :     0.000073s :      1: add_recomputation
 0.35% :     0.000169s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.68% :     0.000333s :      1: bootstrap
 0.07% :     0.000033s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.06% :     0.000029s :      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% :     0.000009s :      1: graph_reusing
13.35% :     0.006519s :      1: jit_opt_a
 0.43% :     0.000211s :      1: jit_opt_after_cconv
 0.15% :     0.000071s :      1: jit_opt_b
 1.03% :     0.000502s :      1: loop_unroll
 2.55% :     0.001243s :      1: mutable_eliminate
 2.45% :     0.001194s :     26: opt.transform.jit_opt_a
 0.17% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.09% :     0.000532s :      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
 0.28% :     0.000139s :      1: py_interpret_to_execute
 0.03% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000023s :      1: remove_dup_value
 1.38% :     0.000674s :      1: renormalize.infer
 0.94% :     0.000457s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.60% :     0.000293s :      1: rewriter_after_opt_a
 0.21% :     0.000105s :      1: rewriter_before_opt_a
 0.22% :     0.000109s :      1: symbol_engine_optimizer
23.24% :     0.011347s :      1: task_emit
49.69% :     0.024264s :      1: type_inference
 0.18% :     0.000087s :      1: validate


TotalTime = 0.0423056, [33]
[bootstrap]: 0.00031201
[type_inference]: 0.0222197
[event_method]: 2.113e-05
[auto_monad]: 0.00015782
[graph_reusing]: 6.64001e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 0.00012334
[rewriter_before_opt_a]: 9.831e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00634381, [2]
    [Cycle 1]: 0.00240612, [27]
        [switch_simplify]: 6.132e-05
        [loop_unroll]: 3.706e-05
        [a_1]: 0.00070754
        [with_stream_mark]: 2.166e-05
        [recompute_prepare]: 1.063e-05
        [updatestate_depend_eliminate]: 5.644e-05
        [updatestate_assign_eliminate]: 6.56e-06
        [updatestate_loads_eliminate]: 5.23002e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 9.83002e-06
        [updatestate_useless_node_eliminater]: 1.262e-05
        [accelerated_algorithm]: 8.54e-06
        [meta_shard_fg_expand]: 3.26999e-06
        [get_grad_eliminate_]: 8.63001e-06
        [merge_forward]: 5.29e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.947e-05
        [j_node_and_user_rematch]: 1.378e-05
        [meta_fg_expand]: 3.65e-06
        [replace_old_param]: 1.209e-05
        [inline_without_move]: 8.38001e-06
        [renormalize]: 0.00111643
        [add_forward_monad_depend]: 5.79999e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 2.149e-05
        [cse]: 4.243e-05
        [replace_applicator]: 1.588e-05
    [Cycle 2]: 0.00046637, [27]
        [switch_simplify]: 9.34998e-06
        [loop_unroll]: 8.82999e-06
        [a_1]: 0.00016719
        [with_stream_mark]: 1.167e-05
        [recompute_prepare]: 8.23001e-06
        [updatestate_depend_eliminate]: 4.70999e-06
        [updatestate_assign_eliminate]: 4.46002e-06
        [updatestate_loads_eliminate]: 3.89002e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 8.35001e-06
        [updatestate_useless_node_eliminater]: 1.1e-05
        [accelerated_algorithm]: 8.04002e-06
        [meta_shard_fg_expand]: 1.79e-06
        [get_grad_eliminate_]: 7.43999e-06
        [merge_forward]: 3.94002e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.611e-05
        [j_node_and_user_rematch]: 1.244e-05
        [meta_fg_expand]: 2.88e-06
        [replace_old_param]: 1.03e-05
        [inline_without_move]: 7.75998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 8.90024e-07
        [auto_monad_eliminator]: 1.041e-05
        [cse]: 2.118e-05
        [replace_applicator]: 7.90998e-06
[py_interpret_to_execute_after_opt_a]: 1.275e-05
[rewriter_after_opt_a]: 0.00028276
[convert_after_rewriter]: 1.123e-05
[order_py_execute_after_rewriter]: 7.16001e-06
[mutable_eliminate]: 0.0005609
[jit_opt_b]: 6.818e-05, [1]
    [Cycle 1]: 6.128e-05, [2]
        [frontend_op_eliminate]: 2.392e-05
        [inline_after_opt_a]: 2.505e-05
[cconv]: 2.599e-05
[loop_unroll]: 0.00044907
[jit_opt_after_cconv]: 0.00019392, [1]
    [Cycle 1]: 0.00018769, [11]
        [c_1]: 4.581e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 7.72998e-06
        [updatestate_assign_eliminate]: 4.42e-06
        [updatestate_loads_eliminate]: 4.28999e-06
        [cse]: 2.85e-05
        [call_graph_tuple_transform]: 2.242e-05
        [tuple_list_get_item_eliminator]: 8.72e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 8.87e-06
[remove_dup_value]: 1.961e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 9.99001e-06
[add_recomputation]: 6.646e-05
[cse_after_recomputation]: 2.994e-05, [1]
    [Cycle 1]: 2.395e-05, [1]
        [cse]: 1.819e-05
[auto_monad_reorder]: 2.648e-05
[get_jit_bprop_graph]: 1.83002e-06
[rewriter_after_jit_bprop_graph]: 4.63001e-06
[opt_after_jit_grad]: 0.00052056
[symbol_engine_optimizer]: 0.00010552, [1]
    [Cycle 1]: 9.889e-05, [6]
        [build]: 1.349e-05
        [elim_shapecalc]: 1.163e-05
        [elim_not_effective]: 2.06e-05
        [opt_reshape]: 8.86002e-06
        [fold_const_symbol]: 1.368e-05
        [renormalize]: 5.19998e-07
[validate]: 5.623e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.0103207
[execute]: 9.49e-06
Sums
    bootstrap                            :     0.000312s :  0.82%
    type_inference                       :     0.022220s : 58.20%
    event_method                         :     0.000021s :  0.06%
    auto_monad                           :     0.000158s :  0.41%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000123s :  0.32%
    rewriter_before_opt_a                :     0.000098s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.19%
    jit_opt_a.loop_unroll                :     0.000046s :  0.12%
    jit_opt_a.a_1                        :     0.000875s :  2.29%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000061s :  0.16%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001117s :  2.92%
    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      :     0.000032s :  0.08%
    jit_opt_a.cse                        :     0.000064s :  0.17%
    jit_opt_a.replace_applicator         :     0.000024s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000283s :  0.74%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000561s :  1.47%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.07%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000449s :  1.18%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.12%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000066s :  0.17%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000521s :  1.36%
    symbol_engine_optimizer.build        :     0.000013s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000056s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010321s : 27.03%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000275    47
 3.17% :     0.000009s :      2: substitution.depend_value_elim
 1.13% :     0.000003s :      4: substitution.elim_not_effective
 0.75% :     0.000002s :      4: substitution.fold_const_symbol
 2.34% :     0.000006s :      5: substitution.graph_param_transform
76.22% :     0.000210s :      4: substitution.inline
 1.65% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.40% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.64% :     0.000005s :      2: substitution.replace_old_param
 4.50% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.31% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.89% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022140     2
93.40% :     0.020678s :      1: type_inference.infer
 6.60% :     0.001462s :      1: type_inference.specialize
------[replace.]   0.000089     6
66.81% :     0.000059s :      4: replace.inline
33.19% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
94.80% :     0.000207s :      4: match.inline
 5.20% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1021
 1.40% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     15: predicate.addn_check_dump
 1.34% :     0.000002s :     15: predicate.addn_zero_filter
 2.32% :     0.000004s :     15: predicate.arithmetic_simplify
 1.36% :     0.000002s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     15: predicate.compare_switch_simplify
 1.31% :     0.000002s :     15: predicate.depend_value_elim
 1.26% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.91% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.55% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     15: predicate.environ_get_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.11% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 5.59% :     0.000009s :     31: predicate.inline
 1.06% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :     10: predicate.less_batch_normalization
 1.56% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.87% :     0.000003s :     22: predicate.load_eliminater
 1.56% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.48% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 1.50% :     0.000003s :      5: predicate.mutable_eliminate
 0.48% :     0.000001s :      5: predicate.opt_reshape
 2.42% :     0.000004s :     22: predicate.partial_eliminate
 1.21% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.73% :     0.000003s :     15: predicate.reduce_eliminate
 1.46% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.73% :     0.000003s :     27: predicate.replace_applicator
 0.74% :     0.000001s :     10: predicate.replace_old_param
 0.35% :     0.000001s :      5: predicate.reset_defer_inline
 1.25% :     0.000002s :     15: predicate.reshape_eliminate
 1.27% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     15: predicate.same_eliminate
 0.60% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.08% :     0.000002s :     10: predicate.special_op_eliminate
 1.21% :     0.000002s :     10: predicate.specialize_transform
 1.43% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.16% :     0.000004s :     21: predicate.switch_defer_inline
 1.86% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.78% :     0.000012s :     68: predicate.switch_simplify
 1.22% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000002s :     15: predicate.transpose_eliminate
 1.53% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.78% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.47% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     15: predicate.value_based_eliminate
 0.42% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000870    13
46.39% :     0.000404s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.62% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.99% :     0.000418s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044784    76
 0.16% :     0.000070s :      1: add_recomputation
 0.36% :     0.000163s :      1: auto_monad
 0.07% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.74% :     0.000331s :      1: bootstrap
 0.06% :     0.000029s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000012s :      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% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
14.17% :     0.006347s :      1: jit_opt_a
 0.44% :     0.000197s :      1: jit_opt_after_cconv
 0.16% :     0.000071s :      1: jit_opt_b
 1.02% :     0.000457s :      1: loop_unroll
 1.27% :     0.000570s :      1: mutable_eliminate
 2.63% :     0.001177s :     26: opt.transform.jit_opt_a
 0.18% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.18% :     0.000529s :      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
 0.28% :     0.000127s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.52% :     0.000681s :      1: renormalize.infer
 0.96% :     0.000428s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.64% :     0.000287s :      1: rewriter_after_opt_a
 0.23% :     0.000103s :      1: rewriter_before_opt_a
 0.24% :     0.000109s :      1: symbol_engine_optimizer
23.09% :     0.010339s :      1: task_emit
49.65% :     0.022236s :      1: type_inference
 0.18% :     0.000080s :      1: validate


TotalTime = 0.0474501, [33]
[bootstrap]: 0.00034907
[type_inference]: 0.0243663
[event_method]: 5.246e-05
[auto_monad]: 0.00016313
[graph_reusing]: 6.23e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 0.00013106
[rewriter_before_opt_a]: 9.788e-05
[expand_dump_flag]: 3.68e-06
[jit_opt_a]: 0.00650039, [2]
    [Cycle 1]: 0.00256743, [27]
        [switch_simplify]: 5.81e-05
        [loop_unroll]: 3.773e-05
        [a_1]: 0.00073728
        [with_stream_mark]: 2.241e-05
        [recompute_prepare]: 1.108e-05
        [updatestate_depend_eliminate]: 5.85002e-06
        [updatestate_assign_eliminate]: 4.74e-06
        [updatestate_loads_eliminate]: 4.34002e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 9.57001e-06
        [updatestate_useless_node_eliminater]: 1.161e-05
        [accelerated_algorithm]: 8.39998e-06
        [meta_shard_fg_expand]: 2.99001e-06
        [get_grad_eliminate_]: 8.40999e-06
        [merge_forward]: 5.66e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.935e-05
        [j_node_and_user_rematch]: 1.383e-05
        [meta_fg_expand]: 3.68e-06
        [replace_old_param]: 1.261e-05
        [inline_without_move]: 7.97e-06
        [renormalize]: 0.00124663
        [add_forward_monad_depend]: 6.17999e-06
        [auto_monad_grad]: 2.47001e-06
        [auto_monad_eliminator]: 2.259e-05
        [cse]: 3.881e-05
        [replace_applicator]: 1.642e-05
    [Cycle 2]: 0.00047557, [27]
        [switch_simplify]: 9.14998e-06
        [loop_unroll]: 8.69e-06
        [a_1]: 0.00017156
        [with_stream_mark]: 1.343e-05
        [recompute_prepare]: 7.75998e-06
        [updatestate_depend_eliminate]: 5.27001e-06
        [updatestate_assign_eliminate]: 4.67e-06
        [updatestate_loads_eliminate]: 3.95e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 8.13001e-06
        [updatestate_useless_node_eliminater]: 1.077e-05
        [accelerated_algorithm]: 8.20999e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 8e-06
        [merge_forward]: 4.14002e-06
        [cell_reuse_recompute_pass]: 1.84998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.689e-05
        [j_node_and_user_rematch]: 1.332e-05
        [meta_fg_expand]: 2.76999e-06
        [replace_old_param]: 1.037e-05
        [inline_without_move]: 7.95998e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 9.72999e-06
        [cse]: 2.076e-05
        [replace_applicator]: 8.37e-06
[py_interpret_to_execute_after_opt_a]: 1.357e-05
[rewriter_after_opt_a]: 0.00028517
[convert_after_rewriter]: 1.268e-05
[order_py_execute_after_rewriter]: 6.46e-06
[mutable_eliminate]: 0.00062949
[jit_opt_b]: 0.00012872, [1]
    [Cycle 1]: 0.00012152, [2]
        [frontend_op_eliminate]: 2.307e-05
        [inline_after_opt_a]: 8.538e-05
[cconv]: 2.772e-05
[loop_unroll]: 0.00046055
[jit_opt_after_cconv]: 0.0002046, [1]
    [Cycle 1]: 0.00019703, [11]
        [c_1]: 5.024e-05
        [parameter_eliminate]: 3.01999e-06
        [updatestate_depend_eliminate]: 8.67998e-06
        [updatestate_assign_eliminate]: 5.46e-06
        [updatestate_loads_eliminate]: 4.2e-06
        [cse]: 2.8e-05
        [call_graph_tuple_transform]: 2.329e-05
        [tuple_list_get_item_eliminator]: 9.20001e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 8.55999e-06
[remove_dup_value]: 1.852e-05
[partial_unused_args_eliminate]: 2.00002e-06
[environ_conv]: 1.065e-05
[add_recomputation]: 6.326e-05
[cse_after_recomputation]: 3.011e-05, [1]
    [Cycle 1]: 2.431e-05, [1]
        [cse]: 1.84e-05
[auto_monad_reorder]: 2.772e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 5.96998e-06
[opt_after_jit_grad]: 0.00050824
[symbol_engine_optimizer]: 0.0001043, [1]
    [Cycle 1]: 9.769e-05, [6]
        [build]: 1.422e-05
        [elim_shapecalc]: 1.21e-05
        [elim_not_effective]: 1.871e-05
        [opt_reshape]: 9.00999e-06
        [fold_const_symbol]: 1.38e-05
        [renormalize]: 6.69999e-07
[validate]: 5.547e-05
[backend_pass]: 1.04998e-06
[task_emit]: 0.0129093
[execute]: 1.05e-05
Sums
    bootstrap                            :     0.000349s :  0.81%
    type_inference                       :     0.024366s : 56.35%
    event_method                         :     0.000052s :  0.12%
    auto_monad                           :     0.000163s :  0.38%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000131s :  0.30%
    rewriter_before_opt_a                :     0.000098s :  0.23%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000067s :  0.16%
    jit_opt_a.loop_unroll                :     0.000046s :  0.11%
    jit_opt_a.a_1                        :     0.000909s :  2.10%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    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        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001247s :  2.88%
    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      :     0.000032s :  0.07%
    jit_opt_a.cse                        :     0.000060s :  0.14%
    jit_opt_a.replace_applicator         :     0.000025s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000285s :  0.66%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000629s :  1.46%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000085s :  0.20%
    cconv                                :     0.000028s :  0.06%
    loop_unroll                          :     0.000461s :  1.07%
    jit_opt_after_cconv.c_1              :     0.000050s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000028s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000019s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000011s :  0.02%
    add_recomputation                    :     0.000063s :  0.15%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000028s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000508s :  1.18%
    symbol_engine_optimizer.build        :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000055s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012909s : 29.85%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000294    47
 3.11% :     0.000009s :      2: substitution.depend_value_elim
 0.93% :     0.000003s :      4: substitution.elim_not_effective
 0.71% :     0.000002s :      4: substitution.fold_const_symbol
 2.24% :     0.000007s :      5: substitution.graph_param_transform
76.77% :     0.000226s :      4: substitution.inline
 1.66% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.41% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.79% :     0.000005s :      2: substitution.replace_old_param
 4.38% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.06% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.94% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024292     2
93.59% :     0.022736s :      1: type_inference.infer
 6.41% :     0.001557s :      1: type_inference.specialize
------[replace.]   0.000093     6
67.26% :     0.000063s :      4: replace.inline
32.74% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000234     6
95.02% :     0.000223s :      4: match.inline
 4.98% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1021
 1.56% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     15: predicate.addn_check_dump
 1.30% :     0.000002s :     15: predicate.addn_zero_filter
 1.98% :     0.000003s :     15: predicate.arithmetic_simplify
 1.66% :     0.000003s :     15: predicate.cast_eliminate
 0.53% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.40% :     0.000002s :     15: predicate.depend_value_elim
 1.22% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.09% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.61% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     15: predicate.environ_get_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.14% :     0.000002s :     10: predicate.get_grad_eliminate
 0.32% :     0.000001s :      5: predicate.graph_param_transform
 5.12% :     0.000009s :     31: predicate.inline
 1.10% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :     10: predicate.less_batch_normalization
 1.46% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.92% :     0.000003s :     22: predicate.load_eliminater
 1.34% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.51% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.12% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.15% :     0.000002s :     15: predicate.minmaximum_grad
 1.64% :     0.000003s :      5: predicate.mutable_eliminate
 0.63% :     0.000001s :      5: predicate.opt_reshape
 2.27% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.76% :     0.000003s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.66% :     0.000003s :     27: predicate.replace_applicator
 0.75% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.20% :     0.000002s :     15: predicate.reshape_eliminate
 1.45% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     15: predicate.same_eliminate
 0.63% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.00% :     0.000002s :     10: predicate.special_op_eliminate
 1.08% :     0.000002s :     10: predicate.specialize_transform
 1.42% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.26% :     0.000004s :     21: predicate.switch_defer_inline
 1.92% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.81% :     0.000012s :     68: predicate.switch_simplify
 1.32% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000002s :     15: predicate.transpose_eliminate
 1.54% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.87% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.34% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000914    13
44.34% :     0.000405s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.02% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.63% :     0.000453s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050157    76
 0.13% :     0.000067s :      1: add_recomputation
 0.34% :     0.000168s :      1: auto_monad
 0.06% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.73% :     0.000368s :      1: bootstrap
 0.06% :     0.000031s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.06% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.12% :     0.000058s :      1: event_method
 0.03% :     0.000016s :      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
12.97% :     0.006504s :      1: jit_opt_a
 0.43% :     0.000216s :      1: jit_opt_after_cconv
 0.26% :     0.000132s :      1: jit_opt_b
 0.97% :     0.000484s :      1: loop_unroll
 1.27% :     0.000639s :      1: mutable_eliminate
 2.42% :     0.001212s :     26: opt.transform.jit_opt_a
 0.17% :     0.000087s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000101s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.03% :     0.000517s :      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
 0.27% :     0.000135s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.54% :     0.000774s :      1: renormalize.infer
 0.92% :     0.000462s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.58% :     0.000290s :      1: rewriter_after_opt_a
 0.20% :     0.000102s :      1: rewriter_before_opt_a
 0.21% :     0.000107s :      1: symbol_engine_optimizer
25.78% :     0.012929s :      1: task_emit
48.61% :     0.024381s :      1: type_inference
 0.16% :     0.000079s :      1: validate


TotalTime = 0.0509512, [33]
[bootstrap]: 0.00034635
[type_inference]: 0.0262879
[event_method]: 2.177e-05
[auto_monad]: 0.00016234
[graph_reusing]: 6.40997e-06
[pre_auto_parallel]: 3.23e-06
[py_interpret_to_execute]: 0.0001299
[rewriter_before_opt_a]: 9.839e-05
[expand_dump_flag]: 3.85e-06
[jit_opt_a]: 0.0065655, [2]
    [Cycle 1]: 0.00247888, [27]
        [switch_simplify]: 6.965e-05
        [loop_unroll]: 3.701e-05
        [a_1]: 0.0007302
        [with_stream_mark]: 2.395e-05
        [recompute_prepare]: 1.062e-05
        [updatestate_depend_eliminate]: 7.04001e-06
        [updatestate_assign_eliminate]: 5.97999e-06
        [updatestate_loads_eliminate]: 4.87e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 9.57999e-06
        [updatestate_useless_node_eliminater]: 1.135e-05
        [accelerated_algorithm]: 8.97e-06
        [meta_shard_fg_expand]: 3.09999e-06
        [get_grad_eliminate_]: 8.43001e-06
        [merge_forward]: 6.03998e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.179e-05
        [j_node_and_user_rematch]: 1.369e-05
        [meta_fg_expand]: 3.4e-06
        [replace_old_param]: 1.311e-05
        [inline_without_move]: 8.3e-06
        [renormalize]: 0.0011816
        [add_forward_monad_depend]: 6.76e-06
        [auto_monad_grad]: 2.88e-06
        [auto_monad_eliminator]: 2.332e-05
        [cse]: 4.657e-05
        [replace_applicator]: 2.052e-05
    [Cycle 2]: 0.00047929, [27]
        [switch_simplify]: 1.04e-05
        [loop_unroll]: 8.05999e-06
        [a_1]: 0.00016953
        [with_stream_mark]: 1.262e-05
        [recompute_prepare]: 8.27e-06
        [updatestate_depend_eliminate]: 5.02999e-06
        [updatestate_assign_eliminate]: 4.98001e-06
        [updatestate_loads_eliminate]: 4.23001e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 8.42998e-06
        [updatestate_useless_node_eliminater]: 1.078e-05
        [accelerated_algorithm]: 8.05e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 7.65e-06
        [merge_forward]: 4.97e-06
        [cell_reuse_recompute_pass]: 2.63e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.714e-05
        [j_node_and_user_rematch]: 1.256e-05
        [meta_fg_expand]: 2.64001e-06
        [replace_old_param]: 1.069e-05
        [inline_without_move]: 7.6e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.57001e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 1.038e-05
        [cse]: 2.107e-05
        [replace_applicator]: 7.76001e-06
[py_interpret_to_execute_after_opt_a]: 1.323e-05
[rewriter_after_opt_a]: 0.00065929
[convert_after_rewriter]: 3.33e-05
[order_py_execute_after_rewriter]: 1.347e-05
[mutable_eliminate]: 0.00138756
[jit_opt_b]: 0.0001136, [1]
    [Cycle 1]: 0.00010123, [2]
        [frontend_op_eliminate]: 4.291e-05
        [inline_after_opt_a]: 3.986e-05
[cconv]: 5.102e-05
[loop_unroll]: 0.0007749
[jit_opt_after_cconv]: 0.00030568, [1]
    [Cycle 1]: 0.00029758, [11]
        [c_1]: 7.571e-05
        [parameter_eliminate]: 5.72001e-06
        [updatestate_depend_eliminate]: 1.473e-05
        [updatestate_assign_eliminate]: 6.67002e-06
        [updatestate_loads_eliminate]: 5.57999e-06
        [cse]: 6.451e-05
        [call_graph_tuple_transform]: 3.203e-05
        [tuple_list_get_item_eliminator]: 1.062e-05
        [none_parameter_eliminate]: 1.74998e-06
        [renormalize]: 1.00999e-06
        [switch_simplify]: 1.589e-05
[remove_dup_value]: 3.36e-05
[partial_unused_args_eliminate]: 2.43002e-06
[environ_conv]: 2.016e-05
[add_recomputation]: 9.747e-05
[cse_after_recomputation]: 4.991e-05, [1]
    [Cycle 1]: 4.011e-05, [1]
        [cse]: 3.02e-05
[auto_monad_reorder]: 3.532e-05
[get_jit_bprop_graph]: 3.61999e-06
[rewriter_after_jit_bprop_graph]: 1.39e-05
[opt_after_jit_grad]: 0.00071347
[symbol_engine_optimizer]: 0.00013006, [1]
    [Cycle 1]: 0.00012153, [6]
        [build]: 2.145e-05
        [elim_shapecalc]: 1.5e-05
        [elim_not_effective]: 2.347e-05
        [opt_reshape]: 1.112e-05
        [fold_const_symbol]: 1.643e-05
        [renormalize]: 7.39994e-07
[validate]: 9.32e-05
[backend_pass]: 1.23002e-06
[task_emit]: 0.0124295
[execute]: 1.062e-05
Sums
    bootstrap                            :     0.000346s :  0.74%
    type_inference                       :     0.026288s : 56.50%
    event_method                         :     0.000022s :  0.05%
    auto_monad                           :     0.000162s :  0.35%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000130s :  0.28%
    rewriter_before_opt_a                :     0.000098s :  0.21%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.10%
    jit_opt_a.a_1                        :     0.000900s :  1.93%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.03%
    jit_opt_a.merge_forward              :     0.000011s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.03%
    jit_opt_a.renormalize                :     0.001182s :  2.54%
    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      :     0.000034s :  0.07%
    jit_opt_a.cse                        :     0.000068s :  0.15%
    jit_opt_a.replace_applicator         :     0.000028s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000659s :  1.42%
    convert_after_rewriter               :     0.000033s :  0.07%
    order_py_execute_after_rewriter      :     0.000013s :  0.03%
    mutable_eliminate                    :     0.001388s :  2.98%
    jit_opt_b.frontend_op_eliminate      :     0.000043s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000040s :  0.09%
    cconv                                :     0.000051s :  0.11%
    loop_unroll                          :     0.000775s :  1.67%
    jit_opt_after_cconv.c_1              :     0.000076s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000015s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.cse              :     0.000065s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000032s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000011s :  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  :     0.000016s :  0.03%
    remove_dup_value                     :     0.000034s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000020s :  0.04%
    add_recomputation                    :     0.000097s :  0.21%
    cse_after_recomputation.cse          :     0.000030s :  0.06%
    auto_monad_reorder                   :     0.000035s :  0.08%
    get_jit_bprop_graph                  :     0.000004s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000014s :  0.03%
    opt_after_jit_grad                   :     0.000713s :  1.53%
    symbol_engine_optimizer.build        :     0.000021s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000015s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000011s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000093s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012429s : 26.71%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000307    47
 2.99% :     0.000009s :      2: substitution.depend_value_elim
 1.22% :     0.000004s :      4: substitution.elim_not_effective
 0.93% :     0.000003s :      4: substitution.fold_const_symbol
 2.78% :     0.000009s :      5: substitution.graph_param_transform
73.13% :     0.000224s :      4: substitution.inline
 1.54% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.64% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.87% :     0.000006s :      2: substitution.replace_old_param
 4.26% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 4.91% :     0.000015s :      3: substitution.updatestate_pure_node_eliminater
 3.72% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026212     2
93.91% :     0.024615s :      1: type_inference.infer
 6.09% :     0.001597s :      1: type_inference.specialize
------[replace.]   0.000095     6
67.81% :     0.000064s :      4: replace.inline
32.19% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000233     6
94.93% :     0.000221s :      4: match.inline
 5.07% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000193  1021
 1.13% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.86% :     0.000004s :      5: predicate.ad_related_special_op_eliminate
 0.99% :     0.000002s :     15: predicate.addn_check_dump
 1.21% :     0.000002s :     15: predicate.addn_zero_filter
 1.69% :     0.000003s :     15: predicate.arithmetic_simplify
 1.14% :     0.000002s :     15: predicate.cast_eliminate
 0.80% :     0.000002s :      5: predicate.check_bprop_eliminate
 0.98% :     0.000002s :     15: predicate.compare_switch_simplify
 1.35% :     0.000003s :     15: predicate.depend_value_elim
 1.05% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.13% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.13% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.38% :     0.000003s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.75% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.05% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.05% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.04% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.12% :     0.000002s :     15: predicate.environ_get_eliminate
 0.99% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 0.93% :     0.000002s :     10: predicate.get_grad_eliminate
 0.36% :     0.000001s :      5: predicate.graph_param_transform
 4.72% :     0.000009s :     31: predicate.inline
 0.99% :     0.000002s :     10: predicate.inline_without_move
 0.38% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :     10: predicate.less_batch_normalization
 1.34% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     22: predicate.load_eliminater
 3.03% :     0.000006s :      5: predicate.loop_unroll_after_grad
 3.22% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.42% :     0.000005s :     20: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000002s :     15: predicate.merge_addn
 1.02% :     0.000002s :     15: predicate.minmaximum_grad
 3.35% :     0.000006s :      5: predicate.mutable_eliminate
 0.62% :     0.000001s :      5: predicate.opt_reshape
 2.01% :     0.000004s :     22: predicate.partial_eliminate
 1.07% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.61% :     0.000003s :     15: predicate.reduce_eliminate
 1.28% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.52% :     0.000003s :     27: predicate.replace_applicator
 0.63% :     0.000001s :     10: predicate.replace_old_param
 0.67% :     0.000001s :      5: predicate.reset_defer_inline
 1.08% :     0.000002s :     15: predicate.reshape_eliminate
 1.18% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.57% :     0.000003s :      5: predicate.row_tensor_eliminate
 1.31% :     0.000003s :     15: predicate.same_eliminate
 0.50% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.96% :     0.000004s :     10: predicate.special_op_eliminate
 1.12% :     0.000002s :     10: predicate.specialize_transform
 1.38% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.94% :     0.000004s :     21: predicate.switch_defer_inline
 1.70% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.58% :     0.000013s :     68: predicate.switch_simplify
 1.20% :     0.000002s :     15: predicate.tile_eliminate
 1.11% :     0.000002s :     15: predicate.transpose_eliminate
 1.50% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.31% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.22% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.64% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.34% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000003s :     15: predicate.value_based_eliminate
 0.38% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.85% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002805    13
82.43% :     0.002312s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 1.49% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
16.09% :     0.000451s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.053634    76
 0.19% :     0.000103s :      1: add_recomputation
 0.31% :     0.000168s :      1: auto_monad
 0.07% :     0.000039s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000368s :      1: bootstrap
 0.10% :     0.000054s :      1: cconv
 0.07% :     0.000039s :      1: convert_after_rewriter
 0.10% :     0.000054s :      1: cse_after_recomputation
 0.04% :     0.000023s :      1: environ_conv
 0.05% :     0.000026s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000007s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
12.25% :     0.006569s :      1: jit_opt_a
 0.58% :     0.000309s :      1: jit_opt_after_cconv
 0.22% :     0.000118s :      1: jit_opt_b
 1.47% :     0.000788s :      1: loop_unroll
 2.62% :     0.001407s :      1: mutable_eliminate
 2.27% :     0.001218s :     26: opt.transform.jit_opt_a
 0.24% :     0.000127s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000071s :      4: opt.transform.jit_opt_b
 0.05% :     0.000029s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000047s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000042s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000061s :      4: opt.transform.symbol_engine_opt
 1.36% :     0.000727s :      1: opt_after_jit_grad
 0.03% :     0.000016s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000134s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000036s :      1: remove_dup_value
 1.30% :     0.000695s :      1: renormalize.infer
 0.89% :     0.000478s :      1: renormalize.specialize
 0.03% :     0.000017s :      1: rewriter_after_jit_bprop_graph
 1.27% :     0.000681s :      1: rewriter_after_opt_a
 0.19% :     0.000102s :      1: rewriter_before_opt_a
 0.25% :     0.000133s :      1: symbol_engine_optimizer
23.21% :     0.012451s :      1: task_emit
49.05% :     0.026305s :      1: type_inference
 0.24% :     0.000128s :      1: validate


TotalTime = 0.0567085, [33]
[bootstrap]: 0.00045679
[type_inference]: 0.0309319
[event_method]: 2.242e-05
[auto_monad]: 0.00017796
[graph_reusing]: 6.54001e-06
[pre_auto_parallel]: 3.36999e-06
[py_interpret_to_execute]: 0.00017724
[rewriter_before_opt_a]: 0.00010349
[expand_dump_flag]: 3.7e-06
[jit_opt_a]: 0.00857295, [2]
    [Cycle 1]: 0.00311052, [27]
        [switch_simplify]: 7.185e-05
        [loop_unroll]: 3.825e-05
        [a_1]: 0.00085328
        [with_stream_mark]: 2.516e-05
        [recompute_prepare]: 1.199e-05
        [updatestate_depend_eliminate]: 6.39001e-06
        [updatestate_assign_eliminate]: 5.84999e-06
        [updatestate_loads_eliminate]: 4.90001e-06
        [parameter_eliminate]: 1.69998e-06
        [specialize_transform]: 1.108e-05
        [updatestate_useless_node_eliminater]: 1.347e-05
        [accelerated_algorithm]: 9.29e-06
        [meta_shard_fg_expand]: 3.73001e-06
        [get_grad_eliminate_]: 8.79e-06
        [merge_forward]: 6.01e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.085e-05
        [j_node_and_user_rematch]: 1.473e-05
        [meta_fg_expand]: 3.6e-06
        [replace_old_param]: 1.369e-05
        [inline_without_move]: 8.24998e-06
        [renormalize]: 0.00167152
        [add_forward_monad_depend]: 8.15999e-06
        [auto_monad_grad]: 3.58999e-06
        [auto_monad_eliminator]: 2.498e-05
        [cse]: 4.579e-05
        [replace_applicator]: 2.015e-05
    [Cycle 2]: 0.00058559, [27]
        [switch_simplify]: 9.65002e-06
        [loop_unroll]: 8.92e-06
        [a_1]: 0.00019067
        [with_stream_mark]: 2.805e-05
        [recompute_prepare]: 9.12999e-06
        [updatestate_depend_eliminate]: 5.73002e-06
        [updatestate_assign_eliminate]: 5.56e-06
        [updatestate_loads_eliminate]: 4.13999e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 9.14e-06
        [updatestate_useless_node_eliminater]: 1.195e-05
        [accelerated_algorithm]: 9.12999e-06
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 7.97e-06
        [merge_forward]: 5.49998e-06
        [cell_reuse_recompute_pass]: 4.83001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.79e-05
        [j_node_and_user_rematch]: 1.289e-05
        [meta_fg_expand]: 3.17002e-06
        [replace_old_param]: 1.118e-05
        [inline_without_move]: 7.88001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 1.25999e-06
        [auto_monad_eliminator]: 1.13e-05
        [cse]: 2.017e-05
        [replace_applicator]: 8.32e-06
[py_interpret_to_execute_after_opt_a]: 1.822e-05
[rewriter_after_opt_a]: 0.00034622
[convert_after_rewriter]: 1.39e-05
[order_py_execute_after_rewriter]: 8.38999e-06
[mutable_eliminate]: 0.0009838
[jit_opt_b]: 7.623e-05, [1]
    [Cycle 1]: 6.681e-05, [2]
        [frontend_op_eliminate]: 2.731e-05
        [inline_after_opt_a]: 2.655e-05
[cconv]: 3.208e-05
[loop_unroll]: 0.00047761
[jit_opt_after_cconv]: 0.00021059, [1]
    [Cycle 1]: 0.00020342, [11]
        [c_1]: 4.995e-05
        [parameter_eliminate]: 3.35e-06
        [updatestate_depend_eliminate]: 8.72e-06
        [updatestate_assign_eliminate]: 4.67998e-06
        [updatestate_loads_eliminate]: 4.37e-06
        [cse]: 3.166e-05
        [call_graph_tuple_transform]: 2.588e-05
        [tuple_list_get_item_eliminator]: 8.51002e-06
        [none_parameter_eliminate]: 1.61998e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 9.34e-06
[remove_dup_value]: 2.065e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 1.84e-05
[add_recomputation]: 7.552e-05
[cse_after_recomputation]: 3.193e-05, [1]
    [Cycle 1]: 2.505e-05, [1]
        [cse]: 1.839e-05
[auto_monad_reorder]: 2.699e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 8.52998e-06
[opt_after_jit_grad]: 0.00053048
[symbol_engine_optimizer]: 0.00012731, [1]
    [Cycle 1]: 0.00012002, [6]
        [build]: 1.654e-05
        [elim_shapecalc]: 1.189e-05
        [elim_not_effective]: 2.59e-05
        [opt_reshape]: 1.056e-05
        [fold_const_symbol]: 2.112e-05
        [renormalize]: 3.30008e-07
[validate]: 8.32e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.0128822
[execute]: 1.017e-05
Sums
    bootstrap                            :     0.000457s :  0.89%
    type_inference                       :     0.030932s : 60.59%
    event_method                         :     0.000022s :  0.04%
    auto_monad                           :     0.000178s :  0.35%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000177s :  0.35%
    rewriter_before_opt_a                :     0.000103s :  0.20%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000082s :  0.16%
    jit_opt_a.loop_unroll                :     0.000047s :  0.09%
    jit_opt_a.a_1                        :     0.001044s :  2.04%
    jit_opt_a.with_stream_mark           :     0.000053s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000020s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.03%
    jit_opt_a.merge_forward              :     0.000012s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000025s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.03%
    jit_opt_a.renormalize                :     0.001672s :  3.27%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000036s :  0.07%
    jit_opt_a.cse                        :     0.000066s :  0.13%
    jit_opt_a.replace_applicator         :     0.000028s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.04%
    rewriter_after_opt_a                 :     0.000346s :  0.68%
    convert_after_rewriter               :     0.000014s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000984s :  1.93%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.05%
    cconv                                :     0.000032s :  0.06%
    loop_unroll                          :     0.000478s :  0.94%
    jit_opt_after_cconv.c_1              :     0.000050s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000032s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000021s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000018s :  0.04%
    add_recomputation                    :     0.000076s :  0.15%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000027s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000530s :  1.04%
    symbol_engine_optimizer.build        :     0.000017s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000026s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000011s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000021s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000083s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012882s : 25.23%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000379    47
 2.90% :     0.000011s :      2: substitution.depend_value_elim
 1.01% :     0.000004s :      4: substitution.elim_not_effective
 1.46% :     0.000006s :      4: substitution.fold_const_symbol
 2.05% :     0.000008s :      5: substitution.graph_param_transform
78.15% :     0.000296s :      4: substitution.inline
 1.43% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.01% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.58% :     0.000006s :      2: substitution.replace_old_param
 3.65% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 2.84% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 2.93% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.030832     2
92.78% :     0.028606s :      1: type_inference.infer
 7.22% :     0.002226s :      1: type_inference.specialize
------[replace.]   0.000112     6
64.55% :     0.000072s :      4: replace.inline
35.45% :     0.000040s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000306     6
95.85% :     0.000293s :      4: match.inline
 4.15% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000192  1021
 1.51% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.12% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.03% :     0.000002s :     15: predicate.addn_check_dump
 1.91% :     0.000004s :     15: predicate.addn_zero_filter
 2.41% :     0.000005s :     15: predicate.arithmetic_simplify
 1.53% :     0.000003s :     15: predicate.cast_eliminate
 0.58% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.04% :     0.000002s :     15: predicate.compare_switch_simplify
 1.48% :     0.000003s :     15: predicate.depend_value_elim
 1.03% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.86% :     0.000004s :     15: predicate.dict_set_item_eliminator
 0.85% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.28% :     0.000001s :      5: predicate.elim_not_effective
 0.58% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.04% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.49% :     0.000003s :     15: predicate.environ_get_eliminate
 1.02% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.30% :     0.000003s :     10: predicate.get_grad_eliminate
 0.37% :     0.000001s :      5: predicate.graph_param_transform
 4.72% :     0.000009s :     31: predicate.inline
 0.94% :     0.000002s :     10: predicate.inline_without_move
 0.38% :     0.000001s :     10: predicate.j_node_and_user_rematch
 2.01% :     0.000004s :     10: predicate.less_batch_normalization
 1.27% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.66% :     0.000003s :     22: predicate.load_eliminater
 1.26% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.09% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.04% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000002s :     15: predicate.merge_addn
 1.03% :     0.000002s :     15: predicate.minmaximum_grad
 1.49% :     0.000003s :      5: predicate.mutable_eliminate
 0.59% :     0.000001s :      5: predicate.opt_reshape
 2.16% :     0.000004s :     22: predicate.partial_eliminate
 1.12% :     0.000002s :     15: predicate.print_const_string_wrapper
 2.24% :     0.000004s :     15: predicate.reduce_eliminate
 1.39% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.57% :     0.000003s :     27: predicate.replace_applicator
 0.52% :     0.000001s :     10: predicate.replace_old_param
 0.43% :     0.000001s :      5: predicate.reset_defer_inline
 1.29% :     0.000002s :     15: predicate.reshape_eliminate
 1.20% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.91% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.13% :     0.000002s :     15: predicate.same_eliminate
 0.63% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.18% :     0.000002s :     10: predicate.special_op_eliminate
 1.09% :     0.000002s :     10: predicate.specialize_transform
 1.42% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.97% :     0.000004s :     15: predicate.stack_unstack_eliminate
 0.62% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.98% :     0.000004s :     21: predicate.switch_defer_inline
 1.73% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.02% :     0.000012s :     68: predicate.switch_simplify
 1.22% :     0.000002s :     15: predicate.tile_eliminate
 1.74% :     0.000003s :     15: predicate.transpose_eliminate
 1.45% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 2.15% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.62% :     0.000007s :     32: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001214    13
43.20% :     0.000524s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.69% :     0.000069s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.11% :     0.000620s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.059947    76
 0.13% :     0.000078s :      1: add_recomputation
 0.31% :     0.000184s :      1: auto_monad
 0.05% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.79% :     0.000473s :      1: bootstrap
 0.06% :     0.000035s :      1: cconv
 0.03% :     0.000017s :      1: convert_after_rewriter
 0.06% :     0.000034s :      1: cse_after_recomputation
 0.04% :     0.000021s :      1: environ_conv
 0.04% :     0.000027s :      1: event_method
 0.03% :     0.000016s :      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
14.31% :     0.008577s :      1: jit_opt_a
 0.36% :     0.000213s :      1: jit_opt_after_cconv
 0.13% :     0.000079s :      1: jit_opt_b
 0.81% :     0.000486s :      1: loop_unroll
 1.66% :     0.000994s :      1: mutable_eliminate
 2.30% :     0.001376s :     26: opt.transform.jit_opt_a
 0.15% :     0.000089s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000046s :      4: opt.transform.jit_opt_b
 0.03% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000063s :      4: opt.transform.symbol_engine_opt
 0.90% :     0.000539s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.30% :     0.000182s :      1: py_interpret_to_execute
 0.03% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000023s :      1: remove_dup_value
 1.76% :     0.001056s :      1: renormalize.infer
 1.01% :     0.000603s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.59% :     0.000351s :      1: rewriter_after_opt_a
 0.18% :     0.000107s :      1: rewriter_before_opt_a
 0.22% :     0.000131s :      1: symbol_engine_optimizer
21.53% :     0.012905s :      1: task_emit
51.63% :     0.030948s :      1: type_inference
 0.19% :     0.000115s :      1: validate


TotalTime = 0.0486335, [33]
[bootstrap]: 0.00030786
[type_inference]: 0.0272427
[event_method]: 2.059e-05
[auto_monad]: 0.00014603
[graph_reusing]: 4.15e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 0.00014742
[rewriter_before_opt_a]: 9.125e-05
[expand_dump_flag]: 2.36e-06
[jit_opt_a]: 0.00697378, [2]
    [Cycle 1]: 0.00266224, [27]
        [switch_simplify]: 5.791e-05
        [loop_unroll]: 3.804e-05
        [a_1]: 0.00083119
        [with_stream_mark]: 2.24e-05
        [recompute_prepare]: 1.176e-05
        [updatestate_depend_eliminate]: 6.63e-06
        [updatestate_assign_eliminate]: 4.81002e-06
        [updatestate_loads_eliminate]: 4.57e-06
        [parameter_eliminate]: 1.71e-06
        [specialize_transform]: 9.66e-06
        [updatestate_useless_node_eliminater]: 1.171e-05
        [accelerated_algorithm]: 9.47999e-06
        [meta_shard_fg_expand]: 2.49999e-06
        [get_grad_eliminate_]: 8.74998e-06
        [merge_forward]: 5.02e-06
        [cell_reuse_recompute_pass]: 1.59998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.754e-05
        [j_node_and_user_rematch]: 1.371e-05
        [meta_fg_expand]: 3.61001e-06
        [replace_old_param]: 1.208e-05
        [inline_without_move]: 8.32e-06
        [renormalize]: 0.00132448
        [add_forward_monad_depend]: 5.36002e-06
        [auto_monad_grad]: 1.72001e-06
        [auto_monad_eliminator]: 1.845e-05
        [cse]: 3.151e-05
        [replace_applicator]: 1.636e-05
    [Cycle 2]: 0.00047389, [27]
        [switch_simplify]: 8.92e-06
        [loop_unroll]: 8.80999e-06
        [a_1]: 0.00016933
        [with_stream_mark]: 1.242e-05
        [recompute_prepare]: 8.30999e-06
        [updatestate_depend_eliminate]: 4.94998e-06
        [updatestate_assign_eliminate]: 4.53001e-06
        [updatestate_loads_eliminate]: 3.96001e-06
        [parameter_eliminate]: 1.14003e-06
        [specialize_transform]: 8.17e-06
        [updatestate_useless_node_eliminater]: 1.101e-05
        [accelerated_algorithm]: 7.87e-06
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 7.4e-06
        [merge_forward]: 4.17e-06
        [cell_reuse_recompute_pass]: 1.98997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.622e-05
        [j_node_and_user_rematch]: 1.287e-05
        [meta_fg_expand]: 2.66999e-06
        [replace_old_param]: 9.77999e-06
        [inline_without_move]: 7.78001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 1.038e-05
        [cse]: 2.18e-05
        [replace_applicator]: 7.85998e-06
[py_interpret_to_execute_after_opt_a]: 1.285e-05
[rewriter_after_opt_a]: 0.00033083
[convert_after_rewriter]: 1.369e-05
[order_py_execute_after_rewriter]: 7.63999e-06
[mutable_eliminate]: 0.00071817
[jit_opt_b]: 7.211e-05, [1]
    [Cycle 1]: 6.532e-05, [2]
        [frontend_op_eliminate]: 2.685e-05
        [inline_after_opt_a]: 2.596e-05
[cconv]: 2.302e-05
[loop_unroll]: 0.00051758
[jit_opt_after_cconv]: 0.00020112, [1]
    [Cycle 1]: 0.00019434, [11]
        [c_1]: 4.628e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 8.07e-06
        [updatestate_assign_eliminate]: 4.79e-06
        [updatestate_loads_eliminate]: 4.37e-06
        [cse]: 3.108e-05
        [call_graph_tuple_transform]: 2.335e-05
        [tuple_list_get_item_eliminator]: 8.35999e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 3.59985e-07
        [switch_simplify]: 8.47e-06
[remove_dup_value]: 1.696e-05
[partial_unused_args_eliminate]: 2.08002e-06
[environ_conv]: 1.108e-05
[add_recomputation]: 5.589e-05
[cse_after_recomputation]: 3.007e-05, [1]
    [Cycle 1]: 2.357e-05, [1]
        [cse]: 1.768e-05
[auto_monad_reorder]: 2.424e-05
[get_jit_bprop_graph]: 1.71002e-06
[rewriter_after_jit_bprop_graph]: 4.89e-06
[opt_after_jit_grad]: 0.0005092
[symbol_engine_optimizer]: 0.00011472, [1]
    [Cycle 1]: 0.00010776, [6]
        [build]: 1.415e-05
        [elim_shapecalc]: 1.403e-05
        [elim_not_effective]: 2.212e-05
        [opt_reshape]: 1.018e-05
        [fold_const_symbol]: 1.391e-05
        [renormalize]: 4.69998e-07
[validate]: 5.365e-05
[backend_pass]: 7.30011e-07
[task_emit]: 0.0107357
[execute]: 8.45001e-06
Sums
    bootstrap                            :     0.000308s :  0.70%
    type_inference                       :     0.027243s : 61.72%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000146s :  0.33%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000147s :  0.33%
    rewriter_before_opt_a                :     0.000091s :  0.21%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000067s :  0.15%
    jit_opt_a.loop_unroll                :     0.000047s :  0.11%
    jit_opt_a.a_1                        :     0.001001s :  2.27%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    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 :     0.000034s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000022s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001325s :  3.00%
    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      :     0.000029s :  0.07%
    jit_opt_a.cse                        :     0.000053s :  0.12%
    jit_opt_a.replace_applicator         :     0.000024s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000331s :  0.75%
    convert_after_rewriter               :     0.000014s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000718s :  1.63%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000023s :  0.05%
    loop_unroll                          :     0.000518s :  1.17%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.10%
    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              :     0.000031s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000017s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000056s :  0.13%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000024s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000509s :  1.15%
    symbol_engine_optimizer.build        :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000054s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010736s : 24.32%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000305    47
 3.20% :     0.000010s :      2: substitution.depend_value_elim
 0.99% :     0.000003s :      4: substitution.elim_not_effective
 0.63% :     0.000002s :      4: substitution.fold_const_symbol
 2.03% :     0.000006s :      5: substitution.graph_param_transform
78.19% :     0.000238s :      4: substitution.inline
 1.57% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.27% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.19% :     0.000004s :      2: substitution.replace_old_param
 3.85% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.06% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.03% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.027165     2
92.31% :     0.025074s :      1: type_inference.infer
 7.69% :     0.002090s :      1: type_inference.specialize
------[replace.]   0.000107     6
65.34% :     0.000070s :      4: replace.inline
34.66% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     6
95.70% :     0.000234s :      4: match.inline
 4.30% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000238  1021
 1.12% :     0.000003s :     15: predicate.accumulaten_eliminater
 0.88% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 0.84% :     0.000002s :     15: predicate.addn_check_dump
 1.29% :     0.000003s :     15: predicate.addn_zero_filter
 1.71% :     0.000004s :     15: predicate.arithmetic_simplify
25.00% :     0.000060s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 0.86% :     0.000002s :     15: predicate.compare_switch_simplify
 1.13% :     0.000003s :     15: predicate.depend_value_elim
 1.01% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.00% :     0.000002s :     15: predicate.dict_get_item_eliminator
 0.94% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.62% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.28% :     0.000001s :      5: predicate.elim_not_effective
 0.51% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 0.90% :     0.000002s :     15: predicate.environ_add_const_eliminate
 0.90% :     0.000002s :     15: predicate.environ_get_add_eliminate
 0.84% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.20% :     0.000003s :     15: predicate.environ_get_eliminate
 0.85% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      5: predicate.fold_const_symbol
 0.83% :     0.000002s :     10: predicate.get_grad_eliminate
 0.24% :     0.000001s :      5: predicate.graph_param_transform
 3.59% :     0.000009s :     31: predicate.inline
 0.84% :     0.000002s :     10: predicate.inline_without_move
 0.30% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.20% :     0.000003s :     10: predicate.less_batch_normalization
 1.05% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.43% :     0.000003s :     22: predicate.load_eliminater
 1.07% :     0.000003s :      5: predicate.loop_unroll_after_grad
 2.56% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.82% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 0.85% :     0.000002s :     15: predicate.merge_addn
 1.06% :     0.000003s :     15: predicate.minmaximum_grad
 1.16% :     0.000003s :      5: predicate.mutable_eliminate
 0.45% :     0.000001s :      5: predicate.opt_reshape
 1.98% :     0.000005s :     22: predicate.partial_eliminate
 0.94% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.55% :     0.000004s :     15: predicate.reduce_eliminate
 1.10% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.34% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.28% :     0.000003s :     27: predicate.replace_applicator
 0.42% :     0.000001s :     10: predicate.replace_old_param
 0.24% :     0.000001s :      5: predicate.reset_defer_inline
 1.07% :     0.000003s :     15: predicate.reshape_eliminate
 0.99% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.64% :     0.000002s :      5: predicate.row_tensor_eliminate
 0.99% :     0.000002s :     15: predicate.same_eliminate
 0.43% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.87% :     0.000002s :     10: predicate.special_op_eliminate
 0.82% :     0.000002s :     10: predicate.specialize_transform
 1.03% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.05% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.53% :     0.000004s :     21: predicate.switch_defer_inline
 1.53% :     0.000004s :     21: predicate.switch_layer_defer_inline
 4.88% :     0.000012s :     68: predicate.switch_simplify
 0.95% :     0.000002s :     15: predicate.tile_eliminate
 0.98% :     0.000002s :     15: predicate.transpose_eliminate
 1.13% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 0.94% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 2.43% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.18% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.54% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 2.31% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.24% :     0.000003s :     15: predicate.value_based_eliminate
 0.31% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.41% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001055    13
42.50% :     0.000448s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.98% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.52% :     0.000554s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.051452    76
 0.11% :     0.000059s :      1: add_recomputation
 0.29% :     0.000151s :      1: auto_monad
 0.05% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.63% :     0.000323s :      1: bootstrap
 0.05% :     0.000026s :      1: cconv
 0.03% :     0.000017s :      1: convert_after_rewriter
 0.06% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.05% :     0.000025s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000007s :      1: graph_reusing
13.56% :     0.006978s :      1: jit_opt_a
 0.40% :     0.000204s :      1: jit_opt_after_cconv
 0.15% :     0.000075s :      1: jit_opt_b
 1.02% :     0.000526s :      1: loop_unroll
 1.41% :     0.000727s :      1: mutable_eliminate
 2.53% :     0.001304s :     26: opt.transform.jit_opt_a
 0.16% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000045s :      4: opt.transform.jit_opt_b
 0.03% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000056s :      4: opt.transform.symbol_engine_opt
 1.01% :     0.000517s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000151s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000020s :      1: remove_dup_value
 1.60% :     0.000825s :      1: renormalize.infer
 0.96% :     0.000491s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.65% :     0.000336s :      1: rewriter_after_opt_a
 0.18% :     0.000095s :      1: rewriter_before_opt_a
 0.23% :     0.000117s :      1: symbol_engine_optimizer
20.90% :     0.010755s :      1: task_emit
52.97% :     0.027255s :      1: type_inference
 0.15% :     0.000076s :      1: validate


TotalTime = 0.0497715, [33]
[bootstrap]: 0.00040125
[type_inference]: 0.0267155
[event_method]: 2.124e-05
[auto_monad]: 0.00017537
[graph_reusing]: 7.45003e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 0.00017949
[rewriter_before_opt_a]: 0.000101
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00790828, [2]
    [Cycle 1]: 0.00280355, [27]
        [switch_simplify]: 6.708e-05
        [loop_unroll]: 3.686e-05
        [a_1]: 0.00083417
        [with_stream_mark]: 2.345e-05
        [recompute_prepare]: 1.175e-05
        [updatestate_depend_eliminate]: 6.56999e-06
        [updatestate_assign_eliminate]: 6.32001e-06
        [updatestate_loads_eliminate]: 5.12e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 1.013e-05
        [updatestate_useless_node_eliminater]: 1.232e-05
        [accelerated_algorithm]: 9.08002e-06
        [meta_shard_fg_expand]: 3.08998e-06
        [get_grad_eliminate_]: 9.14e-06
        [merge_forward]: 5.51e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.06e-05
        [j_node_and_user_rematch]: 1.39e-05
        [meta_fg_expand]: 3.90998e-06
        [replace_old_param]: 1.33e-05
        [inline_without_move]: 8.62998e-06
        [renormalize]: 0.00143251
        [add_forward_monad_depend]: 5.76003e-06
        [auto_monad_grad]: 3.03998e-06
        [auto_monad_eliminator]: 2.296e-05
        [cse]: 3.025e-05
        [replace_applicator]: 1.741e-05
    [Cycle 2]: 0.00052862, [27]
        [switch_simplify]: 9.41998e-06
        [loop_unroll]: 8.29998e-06
        [a_1]: 0.00017127
        [with_stream_mark]: 1.516e-05
        [recompute_prepare]: 9.19e-06
        [updatestate_depend_eliminate]: 5.39998e-06
        [updatestate_assign_eliminate]: 4.78001e-06
        [updatestate_loads_eliminate]: 4.24997e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 8.17e-06
        [updatestate_useless_node_eliminater]: 1.211e-05
        [accelerated_algorithm]: 7.91001e-06
        [meta_shard_fg_expand]: 3.48e-06
        [get_grad_eliminate_]: 8.42998e-06
        [merge_forward]: 5.72999e-06
        [cell_reuse_recompute_pass]: 3.46001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.051e-05
        [j_node_and_user_rematch]: 1.475e-05
        [meta_fg_expand]: 5.82001e-06
        [replace_old_param]: 1.201e-05
        [inline_without_move]: 7.88001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 2.47001e-06
        [auto_monad_grad]: 1.91e-06
        [auto_monad_eliminator]: 1.428e-05
        [cse]: 2.644e-05
        [replace_applicator]: 1.133e-05
[py_interpret_to_execute_after_opt_a]: 1.859e-05
[rewriter_after_opt_a]: 0.00036687
[convert_after_rewriter]: 1.329e-05
[order_py_execute_after_rewriter]: 8.77999e-06
[mutable_eliminate]: 0.00091558
[jit_opt_b]: 7.366e-05, [1]
    [Cycle 1]: 6.54e-05, [2]
        [frontend_op_eliminate]: 2.635e-05
        [inline_after_opt_a]: 2.585e-05
[cconv]: 2.935e-05
[loop_unroll]: 0.00046787
[jit_opt_after_cconv]: 0.00020546, [1]
    [Cycle 1]: 0.00019827, [11]
        [c_1]: 4.68e-05
        [parameter_eliminate]: 3.06001e-06
        [updatestate_depend_eliminate]: 8.55001e-06
        [updatestate_assign_eliminate]: 4.58001e-06
        [updatestate_loads_eliminate]: 4.02e-06
        [cse]: 3.244e-05
        [call_graph_tuple_transform]: 2.495e-05
        [tuple_list_get_item_eliminator]: 8.47e-06
        [none_parameter_eliminate]: 1.96e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 9.69999e-06
[remove_dup_value]: 2.121e-05
[partial_unused_args_eliminate]: 3.04001e-06
[environ_conv]: 1.141e-05
[add_recomputation]: 7.547e-05
[cse_after_recomputation]: 3.509e-05, [1]
    [Cycle 1]: 2.843e-05, [1]
        [cse]: 2.21e-05
[auto_monad_reorder]: 2.924e-05
[get_jit_bprop_graph]: 2.12001e-06
[rewriter_after_jit_bprop_graph]: 5.91e-06
[opt_after_jit_grad]: 0.00049888
[symbol_engine_optimizer]: 0.00010991, [1]
    [Cycle 1]: 0.00010259, [6]
        [build]: 1.264e-05
        [elim_shapecalc]: 1.374e-05
        [elim_not_effective]: 1.976e-05
        [opt_reshape]: 1.08e-05
        [fold_const_symbol]: 1.482e-05
        [renormalize]: 5.19998e-07
[validate]: 6.162e-05
[backend_pass]: 1.18001e-06
[task_emit]: 0.0110404
[execute]: 8.64e-06
Sums
    bootstrap                            :     0.000401s :  0.90%
    type_inference                       :     0.026716s : 60.05%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000175s :  0.39%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000179s :  0.40%
    rewriter_before_opt_a                :     0.000101s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.10%
    jit_opt_a.a_1                        :     0.001005s :  2.26%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000018s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000010s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.06%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001433s :  3.22%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000037s :  0.08%
    jit_opt_a.cse                        :     0.000057s :  0.13%
    jit_opt_a.replace_applicator         :     0.000029s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.04%
    rewriter_after_opt_a                 :     0.000367s :  0.82%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000009s :  0.02%
    mutable_eliminate                    :     0.000916s :  2.06%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000029s :  0.07%
    loop_unroll                          :     0.000468s :  1.05%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000032s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000075s :  0.17%
    cse_after_recomputation.cse          :     0.000022s :  0.05%
    auto_monad_reorder                   :     0.000029s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000499s :  1.12%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000011s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000062s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011040s : 24.82%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000379    47
 2.48% :     0.000009s :      2: substitution.depend_value_elim
 0.84% :     0.000003s :      4: substitution.elim_not_effective
 0.67% :     0.000003s :      4: substitution.fold_const_symbol
 1.82% :     0.000007s :      5: substitution.graph_param_transform
80.37% :     0.000305s :      4: substitution.inline
 1.21% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.19% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.37% :     0.000005s :      2: substitution.replace_old_param
 3.57% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 2.62% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 2.85% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026626     2
92.65% :     0.024668s :      1: type_inference.infer
 7.35% :     0.001958s :      1: type_inference.specialize
------[replace.]   0.000107     6
65.57% :     0.000070s :      4: replace.inline
34.43% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000315     6
96.04% :     0.000302s :      4: match.inline
 3.96% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000180  1021
 1.18% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.21% :     0.000002s :     15: predicate.addn_check_dump
 1.25% :     0.000002s :     15: predicate.addn_zero_filter
 2.03% :     0.000004s :     15: predicate.arithmetic_simplify
 1.41% :     0.000003s :     15: predicate.cast_eliminate
 0.62% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     15: predicate.compare_switch_simplify
 1.38% :     0.000002s :     15: predicate.depend_value_elim
 1.10% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.31% :     0.000001s :      5: predicate.elim_not_effective
 0.89% :     0.000002s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     15: predicate.environ_get_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      5: predicate.fold_const_symbol
 1.20% :     0.000002s :     10: predicate.get_grad_eliminate
 0.41% :     0.000001s :      5: predicate.graph_param_transform
 4.69% :     0.000008s :     31: predicate.inline
 1.01% :     0.000002s :     10: predicate.inline_without_move
 0.40% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :     10: predicate.less_batch_normalization
 1.35% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.04% :     0.000004s :     22: predicate.load_eliminater
 1.31% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.34% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.23% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     15: predicate.merge_addn
 1.10% :     0.000002s :     15: predicate.minmaximum_grad
 1.72% :     0.000003s :      5: predicate.mutable_eliminate
 0.61% :     0.000001s :      5: predicate.opt_reshape
 2.29% :     0.000004s :     22: predicate.partial_eliminate
 1.28% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.71% :     0.000003s :     15: predicate.reduce_eliminate
 1.49% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.83% :     0.000003s :     27: predicate.replace_applicator
 0.55% :     0.000001s :     10: predicate.replace_old_param
 0.47% :     0.000001s :      5: predicate.reset_defer_inline
 1.32% :     0.000002s :     15: predicate.reshape_eliminate
 1.57% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.87% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.08% :     0.000002s :     10: predicate.special_op_eliminate
 1.12% :     0.000002s :     10: predicate.specialize_transform
 1.42% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.56% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.99% :     0.000004s :     21: predicate.switch_defer_inline
 1.96% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.71% :     0.000012s :     68: predicate.switch_simplify
 1.28% :     0.000002s :     15: predicate.tile_eliminate
 1.29% :     0.000002s :     15: predicate.transpose_eliminate
 1.58% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.88% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.43% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 2.06% :     0.000004s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001067    13
44.45% :     0.000474s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.38% :     0.000068s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.17% :     0.000525s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.052722    76
 0.15% :     0.000079s :      1: add_recomputation
 0.34% :     0.000181s :      1: auto_monad
 0.06% :     0.000032s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.81% :     0.000426s :      1: bootstrap
 0.06% :     0.000032s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.07% :     0.000038s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     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% :     0.000010s :      1: graph_reusing
15.01% :     0.007913s :      1: jit_opt_a
 0.40% :     0.000208s :      1: jit_opt_after_cconv
 0.14% :     0.000076s :      1: jit_opt_b
 0.90% :     0.000476s :      1: loop_unroll
 1.75% :     0.000925s :      1: mutable_eliminate
 2.53% :     0.001332s :     26: opt.transform.jit_opt_a
 0.16% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000044s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000055s :      4: opt.transform.symbol_engine_opt
 0.96% :     0.000507s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.35% :     0.000184s :      1: py_interpret_to_execute
 0.04% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000024s :      1: remove_dup_value
 1.71% :     0.000899s :      1: renormalize.infer
 0.99% :     0.000524s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.70% :     0.000372s :      1: rewriter_after_opt_a
 0.20% :     0.000105s :      1: rewriter_before_opt_a
 0.21% :     0.000113s :      1: symbol_engine_optimizer
20.98% :     0.011060s :      1: task_emit
50.70% :     0.026729s :      1: type_inference
 0.17% :     0.000091s :      1: validate


TotalTime = 0.0630884, [33]
[bootstrap]: 0.00036561
[type_inference]: 0.0344555
[event_method]: 2.265e-05
[auto_monad]: 0.00219363
[graph_reusing]: 8.92999e-06
[pre_auto_parallel]: 3.83999e-06
[py_interpret_to_execute]: 0.00026383
[rewriter_before_opt_a]: 0.00012421
[expand_dump_flag]: 4.62e-06
[jit_opt_a]: 0.00922039, [2]
    [Cycle 1]: 0.00369056, [27]
        [switch_simplify]: 9.306e-05
        [loop_unroll]: 3.885e-05
        [a_1]: 0.00092322
        [with_stream_mark]: 3.844e-05
        [recompute_prepare]: 1.986e-05
        [updatestate_depend_eliminate]: 7.12997e-06
        [updatestate_assign_eliminate]: 7.44002e-06
        [updatestate_loads_eliminate]: 5.05001e-06
        [parameter_eliminate]: 2.32999e-06
        [specialize_transform]: 1.149e-05
        [updatestate_useless_node_eliminater]: 1.673e-05
        [accelerated_algorithm]: 1.065e-05
        [meta_shard_fg_expand]: 4.3e-06
        [get_grad_eliminate_]: 1.046e-05
        [merge_forward]: 5.40001e-06
        [cell_reuse_recompute_pass]: 3.97998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.229e-05
        [j_node_and_user_rematch]: 1.617e-05
        [meta_fg_expand]: 5.30001e-06
        [replace_old_param]: 1.587e-05
        [inline_without_move]: 8.45001e-06
        [renormalize]: 0.00188543
        [add_forward_monad_depend]: 1.051e-05
        [auto_monad_grad]: 3.06999e-06
        [auto_monad_eliminator]: 2.777e-05
        [cse]: 4.803e-05
        [replace_applicator]: 2.643e-05
    [Cycle 2]: 0.00060858, [27]
        [switch_simplify]: 1.056e-05
        [loop_unroll]: 9.62999e-06
        [a_1]: 0.00020014
        [with_stream_mark]: 2.218e-05
        [recompute_prepare]: 1.188e-05
        [updatestate_depend_eliminate]: 8.13001e-06
        [updatestate_assign_eliminate]: 7.65998e-06
        [updatestate_loads_eliminate]: 4.23001e-06
        [parameter_eliminate]: 2.51998e-06
        [specialize_transform]: 1.052e-05
        [updatestate_useless_node_eliminater]: 1.44e-05
        [accelerated_algorithm]: 1.081e-05
        [meta_shard_fg_expand]: 3.59002e-06
        [get_grad_eliminate_]: 8.69e-06
        [merge_forward]: 7.15e-06
        [cell_reuse_recompute_pass]: 4.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.195e-05
        [j_node_and_user_rematch]: 1.454e-05
        [meta_fg_expand]: 3.24001e-06
        [replace_old_param]: 1.292e-05
        [inline_without_move]: 8.97999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 2.70002e-06
        [auto_monad_grad]: 2.24001e-06
        [auto_monad_eliminator]: 1.573e-05
        [cse]: 3.055e-05
        [replace_applicator]: 9.96998e-06
[py_interpret_to_execute_after_opt_a]: 2.333e-05
[rewriter_after_opt_a]: 0.00034209
[convert_after_rewriter]: 1.492e-05
[order_py_execute_after_rewriter]: 7.25e-06
[mutable_eliminate]: 0.00099536
[jit_opt_b]: 8.168e-05, [1]
    [Cycle 1]: 7.09e-05, [2]
        [frontend_op_eliminate]: 2.889e-05
        [inline_after_opt_a]: 2.782e-05
[cconv]: 3.914e-05
[loop_unroll]: 0.00058913
[jit_opt_after_cconv]: 0.00025263, [1]
    [Cycle 1]: 0.00024438, [11]
        [c_1]: 6.262e-05
        [parameter_eliminate]: 5.30999e-06
        [updatestate_depend_eliminate]: 1.238e-05
        [updatestate_assign_eliminate]: 5.51e-06
        [updatestate_loads_eliminate]: 4.2e-06
        [cse]: 5.076e-05
        [call_graph_tuple_transform]: 2.345e-05
        [tuple_list_get_item_eliminator]: 9.29e-06
        [none_parameter_eliminate]: 1.77999e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 9.97001e-06
[remove_dup_value]: 2.468e-05
[partial_unused_args_eliminate]: 3.65e-06
[environ_conv]: 1.658e-05
[add_recomputation]: 8.808e-05
[cse_after_recomputation]: 3.584e-05, [1]
    [Cycle 1]: 2.825e-05, [1]
        [cse]: 2.03e-05
[auto_monad_reorder]: 2.863e-05
[get_jit_bprop_graph]: 2.54999e-06
[rewriter_after_jit_bprop_graph]: 1.022e-05
[opt_after_jit_grad]: 0.00062422
[symbol_engine_optimizer]: 0.00012692, [1]
    [Cycle 1]: 0.00011847, [6]
        [build]: 1.988e-05
        [elim_shapecalc]: 1.468e-05
        [elim_not_effective]: 2.535e-05
        [opt_reshape]: 1.031e-05
        [fold_const_symbol]: 1.55e-05
        [renormalize]: 6.50005e-07
[validate]: 7.984e-05
[backend_pass]: 1.23002e-06
[task_emit]: 0.0127039
[execute]: 8.60999e-06
Sums
    bootstrap                            :     0.000366s :  0.64%
    type_inference                       :     0.034456s : 60.32%
    event_method                         :     0.000023s :  0.04%
    auto_monad                           :     0.002194s :  3.84%
    graph_reusing                        :     0.000009s :  0.02%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000264s :  0.46%
    rewriter_before_opt_a                :     0.000124s :  0.22%
    expand_dump_flag                     :     0.000005s :  0.01%
    jit_opt_a.switch_simplify            :     0.000104s :  0.18%
    jit_opt_a.loop_unroll                :     0.000048s :  0.08%
    jit_opt_a.a_1                        :     0.001123s :  1.97%
    jit_opt_a.with_stream_mark           :     0.000061s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000022s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000031s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000021s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000019s :  0.03%
    jit_opt_a.merge_forward              :     0.000013s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000008s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000044s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000031s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000009s :  0.01%
    jit_opt_a.replace_old_param          :     0.000029s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.03%
    jit_opt_a.renormalize                :     0.001886s :  3.30%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000044s :  0.08%
    jit_opt_a.cse                        :     0.000079s :  0.14%
    jit_opt_a.replace_applicator         :     0.000036s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000023s :  0.04%
    rewriter_after_opt_a                 :     0.000342s :  0.60%
    convert_after_rewriter               :     0.000015s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000995s :  1.74%
    jit_opt_b.frontend_op_eliminate      :     0.000029s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000028s :  0.05%
    cconv                                :     0.000039s :  0.07%
    loop_unroll                          :     0.000589s :  1.03%
    jit_opt_after_cconv.c_1              :     0.000063s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000051s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000025s :  0.04%
    partial_unused_args_eliminate        :     0.000004s :  0.01%
    environ_conv                         :     0.000017s :  0.03%
    add_recomputation                    :     0.000088s :  0.15%
    cse_after_recomputation.cse          :     0.000020s :  0.04%
    auto_monad_reorder                   :     0.000029s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.02%
    opt_after_jit_grad                   :     0.000624s :  1.09%
    symbol_engine_optimizer.build        :     0.000020s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000015s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000025s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000080s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012704s : 22.24%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000414    47
 3.04% :     0.000013s :      2: substitution.depend_value_elim
 0.97% :     0.000004s :      4: substitution.elim_not_effective
 0.69% :     0.000003s :      4: substitution.fold_const_symbol
 1.93% :     0.000008s :      5: substitution.graph_param_transform
76.66% :     0.000318s :      4: substitution.inline
 1.70% :     0.000007s :      8: substitution.j_node_and_user_rematch
 1.91% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.87% :     0.000008s :      2: substitution.replace_old_param
 3.91% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
 4.20% :     0.000017s :      3: substitution.updatestate_pure_node_eliminater
 3.13% :     0.000013s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.034351     2
94.60% :     0.032496s :      1: type_inference.infer
 5.40% :     0.001855s :      1: type_inference.specialize
------[replace.]   0.000133     6
68.92% :     0.000092s :      4: replace.inline
31.08% :     0.000041s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000328     6
95.66% :     0.000314s :      4: match.inline
 4.34% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000216  1021
 0.94% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.18% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 0.91% :     0.000002s :     15: predicate.addn_check_dump
 1.49% :     0.000003s :     15: predicate.addn_zero_filter
 1.93% :     0.000004s :     15: predicate.arithmetic_simplify
 1.42% :     0.000003s :     15: predicate.cast_eliminate
 0.54% :     0.000001s :      5: predicate.check_bprop_eliminate
 0.98% :     0.000002s :     15: predicate.compare_switch_simplify
 1.68% :     0.000004s :     15: predicate.depend_value_elim
 1.18% :     0.000003s :     15: predicate.dict_get_item_const_eliminator
 0.99% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.31% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.78% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.62% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000003s :     15: predicate.environ_add_const_eliminate
 0.90% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.01% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.50% :     0.000003s :     15: predicate.environ_get_eliminate
 1.02% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      5: predicate.fold_const_symbol
 1.25% :     0.000003s :     10: predicate.get_grad_eliminate
 0.23% :     0.000000s :      5: predicate.graph_param_transform
 4.63% :     0.000010s :     31: predicate.inline
 0.80% :     0.000002s :     10: predicate.inline_without_move
 0.34% :     0.000001s :     10: predicate.j_node_and_user_rematch
 2.21% :     0.000005s :     10: predicate.less_batch_normalization
 1.62% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000004s :     22: predicate.load_eliminater
 2.04% :     0.000004s :      5: predicate.loop_unroll_after_grad
 3.46% :     0.000007s :     42: predicate.loop_unroll_before_grad
 2.19% :     0.000005s :     20: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000002s :     15: predicate.merge_addn
 1.28% :     0.000003s :     15: predicate.minmaximum_grad
 1.89% :     0.000004s :      5: predicate.mutable_eliminate
 0.64% :     0.000001s :      5: predicate.opt_reshape
 1.92% :     0.000004s :     22: predicate.partial_eliminate
 1.53% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.89% :     0.000004s :     15: predicate.reduce_eliminate
 1.52% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.65% :     0.000004s :     27: predicate.replace_applicator
 0.56% :     0.000001s :     10: predicate.replace_old_param
 0.47% :     0.000001s :      5: predicate.reset_defer_inline
 1.66% :     0.000004s :     15: predicate.reshape_eliminate
 1.45% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.85% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.58% :     0.000003s :     15: predicate.same_eliminate
 0.89% :     0.000002s :     10: predicate.set_cell_output_no_recompute
 1.12% :     0.000002s :     10: predicate.special_op_eliminate
 1.17% :     0.000003s :     10: predicate.specialize_transform
 1.43% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.79% :     0.000004s :     21: predicate.switch_defer_inline
 1.68% :     0.000004s :     21: predicate.switch_layer_defer_inline
 7.26% :     0.000016s :     68: predicate.switch_simplify
 1.44% :     0.000003s :     15: predicate.tile_eliminate
 1.43% :     0.000003s :     15: predicate.transpose_eliminate
 1.52% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.49% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000003s :     15: predicate.value_based_eliminate
 0.38% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001126    13
44.12% :     0.000497s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.23% :     0.000081s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.66% :     0.000548s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.066696    76
 0.14% :     0.000092s :      1: add_recomputation
 3.32% :     0.002214s :      1: auto_monad
 0.05% :     0.000032s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.57% :     0.000383s :      1: bootstrap
 0.06% :     0.000042s :      1: cconv
 0.03% :     0.000018s :      1: convert_after_rewriter
 0.06% :     0.000038s :      1: cse_after_recomputation
 0.03% :     0.000020s :      1: environ_conv
 0.04% :     0.000028s :      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
 0.02% :     0.000014s :      1: graph_reusing
13.83% :     0.009226s :      1: jit_opt_a
 0.38% :     0.000256s :      1: jit_opt_after_cconv
 0.13% :     0.000085s :      1: jit_opt_b
 0.90% :     0.000601s :      1: loop_unroll
 1.51% :     0.001010s :      1: mutable_eliminate
 2.27% :     0.001513s :     26: opt.transform.jit_opt_a
 0.15% :     0.000100s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000049s :      4: opt.transform.jit_opt_b
 0.04% :     0.000024s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000028s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000039s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000061s :      4: opt.transform.symbol_engine_opt
 0.95% :     0.000635s :      1: opt_after_jit_grad
 0.01% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000006s :      1: partial_unused_args_eliminate
 0.01% :     0.000009s :      1: pre_auto_parallel
 0.40% :     0.000269s :      1: py_interpret_to_execute
 0.04% :     0.000026s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000027s :      1: remove_dup_value
 1.89% :     0.001263s :      1: renormalize.infer
 0.91% :     0.000608s :      1: renormalize.specialize
 0.02% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.52% :     0.000350s :      1: rewriter_after_opt_a
 0.19% :     0.000130s :      1: rewriter_before_opt_a
 0.20% :     0.000131s :      1: symbol_engine_optimizer
19.08% :     0.012727s :      1: task_emit
51.69% :     0.034477s :      1: type_inference
 0.17% :     0.000112s :      1: validate


TotalTime = 0.095735, [33]
[bootstrap]: 0.00042256
[type_inference]: 0.0362052
[event_method]: 2.36e-05
[auto_monad]: 0.00019421
[graph_reusing]: 6.13998e-06
[pre_auto_parallel]: 3.6e-06
[py_interpret_to_execute]: 0.00017382
[rewriter_before_opt_a]: 0.00010644
[expand_dump_flag]: 3.43999e-06
[jit_opt_a]: 0.0245863, [2]
    [Cycle 1]: 0.00311128, [27]
        [switch_simplify]: 7.507e-05
        [loop_unroll]: 3.791e-05
        [a_1]: 0.00085274
        [with_stream_mark]: 2.695e-05
        [recompute_prepare]: 1.216e-05
        [updatestate_depend_eliminate]: 6.24999e-06
        [updatestate_assign_eliminate]: 5.69e-06
        [updatestate_loads_eliminate]: 5.34e-06
        [parameter_eliminate]: 2.80997e-06
        [specialize_transform]: 1.021e-05
        [updatestate_useless_node_eliminater]: 9.653e-05
        [accelerated_algorithm]: 1.159e-05
        [meta_shard_fg_expand]: 3.53999e-06
        [get_grad_eliminate_]: 9.14e-06
        [merge_forward]: 6.49001e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.347e-05
        [j_node_and_user_rematch]: 1.468e-05
        [meta_fg_expand]: 3.91001e-06
        [replace_old_param]: 1.302e-05
        [inline_without_move]: 9.72999e-06
        [renormalize]: 0.00155241
        [add_forward_monad_depend]: 9.29e-06
        [auto_monad_grad]: 3.01999e-06
        [auto_monad_eliminator]: 2.591e-05
        [cse]: 4.715e-05
        [replace_applicator]: 2.634e-05
    [Cycle 2]: 0.00052972, [27]
        [switch_simplify]: 1.061e-05
        [loop_unroll]: 9.17999e-06
        [a_1]: 0.00019237
        [with_stream_mark]: 1.806e-05
        [recompute_prepare]: 9.78998e-06
        [updatestate_depend_eliminate]: 5.53002e-06
        [updatestate_assign_eliminate]: 5.49e-06
        [updatestate_loads_eliminate]: 4.52e-06
        [parameter_eliminate]: 1.47999e-06
        [specialize_transform]: 8.80001e-06
        [updatestate_useless_node_eliminater]: 1.092e-05
        [accelerated_algorithm]: 8.47e-06
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 7.84002e-06
        [merge_forward]: 6.09001e-06
        [cell_reuse_recompute_pass]: 3.43e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.024e-05
        [j_node_and_user_rematch]: 1.28e-05
        [meta_fg_expand]: 3.46999e-06
        [replace_old_param]: 1.098e-05
        [inline_without_move]: 7.8e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 1.59998e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 1.113e-05
        [cse]: 2.409e-05
        [replace_applicator]: 9.24e-06
[py_interpret_to_execute_after_opt_a]: 2.063e-05
[rewriter_after_opt_a]: 0.00032367
[convert_after_rewriter]: 1.338e-05
[order_py_execute_after_rewriter]: 7.93001e-06
[mutable_eliminate]: 0.00098336
[jit_opt_b]: 8.048e-05, [1]
    [Cycle 1]: 7.093e-05, [2]
        [frontend_op_eliminate]: 2.843e-05
        [inline_after_opt_a]: 2.915e-05
[cconv]: 0.00022462
[loop_unroll]: 0.00060499
[jit_opt_after_cconv]: 0.00023783, [1]
    [Cycle 1]: 0.00023059, [11]
        [c_1]: 5.615e-05
        [parameter_eliminate]: 5.51e-06
        [updatestate_depend_eliminate]: 1.076e-05
        [updatestate_assign_eliminate]: 6.50002e-06
        [updatestate_loads_eliminate]: 4.64002e-06
        [cse]: 3.735e-05
        [call_graph_tuple_transform]: 2.633e-05
        [tuple_list_get_item_eliminator]: 9.41e-06
        [none_parameter_eliminate]: 1.58002e-06
        [renormalize]: 2.69996e-07
        [switch_simplify]: 1.164e-05
[remove_dup_value]: 2.221e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 1.58e-05
[add_recomputation]: 7.595e-05
[cse_after_recomputation]: 3.381e-05, [1]
    [Cycle 1]: 2.572e-05, [1]
        [cse]: 1.858e-05
[auto_monad_reorder]: 2.721e-05
[get_jit_bprop_graph]: 2.25002e-06
[rewriter_after_jit_bprop_graph]: 9.58002e-06
[opt_after_jit_grad]: 0.00059332
[symbol_engine_optimizer]: 0.00011768, [1]
    [Cycle 1]: 0.00010973, [6]
        [build]: 1.762e-05
        [elim_shapecalc]: 1.342e-05
        [elim_not_effective]: 2.229e-05
        [opt_reshape]: 9.51003e-06
        [fold_const_symbol]: 1.539e-05
        [renormalize]: 5.00004e-07
[validate]: 7.58e-05
[backend_pass]: 1.04998e-06
[task_emit]: 0.0302322
[execute]: 1.192e-05
Sums
    bootstrap                            :     0.000423s :  0.57%
    type_inference                       :     0.036205s : 48.92%
    event_method                         :     0.000024s :  0.03%
    auto_monad                           :     0.000194s :  0.26%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000004s :  0.00%
    py_interpret_to_execute              :     0.000174s :  0.23%
    rewriter_before_opt_a                :     0.000106s :  0.14%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000086s :  0.12%
    jit_opt_a.loop_unroll                :     0.000047s :  0.06%
    jit_opt_a.a_1                        :     0.001045s :  1.41%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000107s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000020s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.02%
    jit_opt_a.merge_forward              :     0.000013s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000044s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.03%
    jit_opt_a.inline_without_move        :     0.000018s :  0.02%
    jit_opt_a.renormalize                :     0.001553s :  2.10%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000037s :  0.05%
    jit_opt_a.cse                        :     0.000071s :  0.10%
    jit_opt_a.replace_applicator         :     0.000036s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.03%
    rewriter_after_opt_a                 :     0.000324s :  0.44%
    convert_after_rewriter               :     0.000013s :  0.02%
    order_py_execute_after_rewriter      :     0.000008s :  0.01%
    mutable_eliminate                    :     0.000983s :  1.33%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000029s :  0.04%
    cconv                                :     0.000225s :  0.30%
    loop_unroll                          :     0.000605s :  0.82%
    jit_opt_after_cconv.c_1              :     0.000056s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000037s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000012s :  0.02%
    remove_dup_value                     :     0.000022s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000016s :  0.02%
    add_recomputation                    :     0.000076s :  0.10%
    cse_after_recomputation.cse          :     0.000019s :  0.03%
    auto_monad_reorder                   :     0.000027s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.01%
    opt_after_jit_grad                   :     0.000593s :  0.80%
    symbol_engine_optimizer.build        :     0.000018s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000076s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.030232s : 40.85%
    execute                              :     0.000012s :  0.02%


Time group info:
------[substitution.]   0.000376    47
 3.15% :     0.000012s :      2: substitution.depend_value_elim
 0.94% :     0.000004s :      4: substitution.elim_not_effective
 0.70% :     0.000003s :      4: substitution.fold_const_symbol
 2.14% :     0.000008s :      5: substitution.graph_param_transform
77.50% :     0.000291s :      4: substitution.inline
 1.41% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.28% :     0.000009s :      8: substitution.remove_not_recompute_node
 1.51% :     0.000006s :      2: substitution.replace_old_param
 3.87% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.36% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.14% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.036097     2
95.07% :     0.034316s :      1: type_inference.infer
 4.93% :     0.001781s :      1: type_inference.specialize
------[replace.]   0.000107     6
67.35% :     0.000072s :      4: replace.inline
32.65% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000301     6
95.54% :     0.000288s :      4: match.inline
 4.46% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000197  1021
 1.60% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.08% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.36% :     0.000003s :     15: predicate.addn_check_dump
 1.27% :     0.000003s :     15: predicate.addn_zero_filter
 2.40% :     0.000005s :     15: predicate.arithmetic_simplify
 1.28% :     0.000003s :     15: predicate.cast_eliminate
 0.48% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.04% :     0.000002s :     15: predicate.compare_switch_simplify
 1.39% :     0.000003s :     15: predicate.depend_value_elim
 1.13% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.06% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.46% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.28% :     0.000001s :      5: predicate.elim_not_effective
 0.64% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000003s :     15: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.32% :     0.000003s :     15: predicate.environ_get_depend_swap
 1.48% :     0.000003s :     15: predicate.environ_get_eliminate
 1.00% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.01% :     0.000002s :     10: predicate.get_grad_eliminate
 0.33% :     0.000001s :      5: predicate.graph_param_transform
 5.22% :     0.000010s :     31: predicate.inline
 1.15% :     0.000002s :     10: predicate.inline_without_move
 0.36% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.56% :     0.000003s :     10: predicate.less_batch_normalization
 1.58% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000004s :     22: predicate.load_eliminater
 1.25% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.16% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.20% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.88% :     0.000004s :      5: predicate.mutable_eliminate
 0.55% :     0.000001s :      5: predicate.opt_reshape
 2.08% :     0.000004s :     22: predicate.partial_eliminate
 1.15% :     0.000002s :     15: predicate.print_const_string_wrapper
 2.17% :     0.000004s :     15: predicate.reduce_eliminate
 1.48% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.71% :     0.000003s :     27: predicate.replace_applicator
 0.59% :     0.000001s :     10: predicate.replace_old_param
 0.42% :     0.000001s :      5: predicate.reset_defer_inline
 1.33% :     0.000003s :     15: predicate.reshape_eliminate
 1.20% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.90% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000003s :     15: predicate.same_eliminate
 0.61% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.18% :     0.000002s :     10: predicate.special_op_eliminate
 1.11% :     0.000002s :     10: predicate.specialize_transform
 1.59% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.26% :     0.000004s :     21: predicate.switch_defer_inline
 1.92% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.17% :     0.000012s :     68: predicate.switch_simplify
 1.14% :     0.000002s :     15: predicate.tile_eliminate
 1.38% :     0.000003s :     15: predicate.transpose_eliminate
 1.32% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.62% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.32% :     0.000007s :     32: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.82% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001357    13
56.02% :     0.000760s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.14% :     0.000070s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.84% :     0.000527s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.098969    76
 0.08% :     0.000079s :      1: add_recomputation
 0.20% :     0.000201s :      1: auto_monad
 0.03% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.45% :     0.000447s :      1: bootstrap
 0.23% :     0.000232s :      1: cconv
 0.02% :     0.000017s :      1: convert_after_rewriter
 0.04% :     0.000037s :      1: cse_after_recomputation
 0.02% :     0.000018s :      1: environ_conv
 0.03% :     0.000029s :      1: event_method
 0.02% :     0.000018s :      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
24.85% :     0.024592s :      1: jit_opt_a
 0.24% :     0.000241s :      1: jit_opt_after_cconv
 0.09% :     0.000084s :      1: jit_opt_b
 0.62% :     0.000616s :      1: loop_unroll
 1.01% :     0.000995s :      1: mutable_eliminate
 1.49% :     0.001475s :     26: opt.transform.jit_opt_a
 0.10% :     0.000099s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000049s :      4: opt.transform.jit_opt_b
 0.02% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000057s :      4: opt.transform.symbol_engine_opt
 0.61% :     0.000602s :      1: opt_after_jit_grad
 0.01% :     0.000010s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.18% :     0.000178s :      1: py_interpret_to_execute
 0.02% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000025s :      1: remove_dup_value
 0.96% :     0.000954s :      1: renormalize.infer
 0.59% :     0.000585s :      1: renormalize.specialize
 0.01% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.33% :     0.000330s :      1: rewriter_after_opt_a
 0.11% :     0.000111s :      1: rewriter_before_opt_a
 0.12% :     0.000121s :      1: symbol_engine_optimizer
30.57% :     0.030255s :      1: task_emit
36.61% :     0.036228s :      1: type_inference
 0.11% :     0.000106s :      1: validate


TotalTime = 0.0865445, [33]
[bootstrap]: 0.0003739
[type_inference]: 0.0268792
[event_method]: 2.334e-05
[auto_monad]: 0.00018622
[graph_reusing]: 7.12002e-06
[pre_auto_parallel]: 3.08998e-06
[py_interpret_to_execute]: 0.00016133
[rewriter_before_opt_a]: 0.00010382
[expand_dump_flag]: 4.02002e-06
[jit_opt_a]: 0.0325133, [2]
    [Cycle 1]: 0.0271641, [27]
        [switch_simplify]: 7.533e-05
        [loop_unroll]: 3.679e-05
        [a_1]: 0.0008449
        [with_stream_mark]: 2.741e-05
        [recompute_prepare]: 1.34e-05
        [updatestate_depend_eliminate]: 7.18998e-06
        [updatestate_assign_eliminate]: 5.91998e-06
        [updatestate_loads_eliminate]: 5.17e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 1.026e-05
        [updatestate_useless_node_eliminater]: 1.28e-05
        [accelerated_algorithm]: 9.47999e-06
        [meta_shard_fg_expand]: 3.96001e-06
        [get_grad_eliminate_]: 9.57001e-06
        [merge_forward]: 6.88e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.151e-05
        [j_node_and_user_rematch]: 1.563e-05
        [meta_fg_expand]: 3.72998e-06
        [replace_old_param]: 1.271e-05
        [inline_without_move]: 8.64e-06
        [renormalize]: 0.0256401
        [add_forward_monad_depend]: 1.351e-05
        [auto_monad_grad]: 3.45e-06
        [auto_monad_eliminator]: 3.548e-05
        [cse]: 4.918e-05
        [replace_applicator]: 3.744e-05
    [Cycle 2]: 0.0006238, [27]
        [switch_simplify]: 1.22e-05
        [loop_unroll]: 1.135e-05
        [a_1]: 0.00024014
        [with_stream_mark]: 2.201e-05
        [recompute_prepare]: 1.046e-05
        [updatestate_depend_eliminate]: 5.74999e-06
        [updatestate_assign_eliminate]: 8.75999e-06
        [updatestate_loads_eliminate]: 5.27001e-06
        [parameter_eliminate]: 2.10002e-06
        [specialize_transform]: 9.42999e-06
        [updatestate_useless_node_eliminater]: 1.273e-05
        [accelerated_algorithm]: 1.208e-05
        [meta_shard_fg_expand]: 4.97999e-06
        [get_grad_eliminate_]: 9.74999e-06
        [merge_forward]: 6.24001e-06
        [cell_reuse_recompute_pass]: 6.06998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.207e-05
        [j_node_and_user_rematch]: 1.396e-05
        [meta_fg_expand]: 3.53e-06
        [replace_old_param]: 1.226e-05
        [inline_without_move]: 8.59998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.46998e-06
        [auto_monad_grad]: 1.42e-06
        [auto_monad_eliminator]: 1.097e-05
        [cse]: 2.847e-05
        [replace_applicator]: 8.18999e-06
[py_interpret_to_execute_after_opt_a]: 2.814e-05
[rewriter_after_opt_a]: 0.00037559
[convert_after_rewriter]: 1.159e-05
[order_py_execute_after_rewriter]: 9.96e-06
[mutable_eliminate]: 0.00097813
[jit_opt_b]: 8.587e-05, [1]
    [Cycle 1]: 7.686e-05, [2]
        [frontend_op_eliminate]: 3.203e-05
        [inline_after_opt_a]: 3.068e-05
[cconv]: 3.696e-05
[loop_unroll]: 0.00052465
[jit_opt_after_cconv]: 0.00023733, [1]
    [Cycle 1]: 0.00022931, [11]
        [c_1]: 5.455e-05
        [parameter_eliminate]: 3.91001e-06
        [updatestate_depend_eliminate]: 8.82999e-06
        [updatestate_assign_eliminate]: 5.05001e-06
        [updatestate_loads_eliminate]: 4.32e-06
        [cse]: 4.166e-05
        [call_graph_tuple_transform]: 2.87e-05
        [tuple_list_get_item_eliminator]: 9.12001e-06
        [none_parameter_eliminate]: 1.88002e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 1.178e-05
[remove_dup_value]: 2.266e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 1.464e-05
[add_recomputation]: 7.925e-05
[cse_after_recomputation]: 3.209e-05, [1]
    [Cycle 1]: 2.549e-05, [1]
        [cse]: 1.939e-05
[auto_monad_reorder]: 2.659e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 1.112e-05
[opt_after_jit_grad]: 0.00058587
[symbol_engine_optimizer]: 0.00011612, [1]
    [Cycle 1]: 0.00010855, [6]
        [build]: 1.768e-05
        [elim_shapecalc]: 1.291e-05
        [elim_not_effective]: 2.23e-05
        [opt_reshape]: 9.76e-06
        [fold_const_symbol]: 1.538e-05
        [renormalize]: 6.30011e-07
[validate]: 7.255e-05
[backend_pass]: 1.05001e-06
[task_emit]: 0.022747
[execute]: 8.37e-06
Sums
    bootstrap                            :     0.000374s :  0.46%
    type_inference                       :     0.026879s : 33.18%
    event_method                         :     0.000023s :  0.03%
    auto_monad                           :     0.000186s :  0.23%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000161s :  0.20%
    rewriter_before_opt_a                :     0.000104s :  0.13%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000088s :  0.11%
    jit_opt_a.loop_unroll                :     0.000048s :  0.06%
    jit_opt_a.a_1                        :     0.001085s :  1.34%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000020s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000022s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000019s :  0.02%
    jit_opt_a.merge_forward              :     0.000013s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000008s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000044s :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000025s :  0.03%
    jit_opt_a.inline_without_move        :     0.000017s :  0.02%
    jit_opt_a.renormalize                :     0.025640s : 31.65%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000046s :  0.06%
    jit_opt_a.cse                        :     0.000078s :  0.10%
    jit_opt_a.replace_applicator         :     0.000046s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000028s :  0.03%
    rewriter_after_opt_a                 :     0.000376s :  0.46%
    convert_after_rewriter               :     0.000012s :  0.01%
    order_py_execute_after_rewriter      :     0.000010s :  0.01%
    mutable_eliminate                    :     0.000978s :  1.21%
    jit_opt_b.frontend_op_eliminate      :     0.000032s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000031s :  0.04%
    cconv                                :     0.000037s :  0.05%
    loop_unroll                          :     0.000525s :  0.65%
    jit_opt_after_cconv.c_1              :     0.000055s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    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              :     0.000042s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000012s :  0.01%
    remove_dup_value                     :     0.000023s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000015s :  0.02%
    add_recomputation                    :     0.000079s :  0.10%
    cse_after_recomputation.cse          :     0.000019s :  0.02%
    auto_monad_reorder                   :     0.000027s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.01%
    opt_after_jit_grad                   :     0.000586s :  0.72%
    symbol_engine_optimizer.build        :     0.000018s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000073s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.022747s : 28.08%
    execute                              :     0.000008s :  0.01%


Time group info:
------[substitution.]   0.000367    47
 3.59% :     0.000013s :      2: substitution.depend_value_elim
 1.03% :     0.000004s :      4: substitution.elim_not_effective
 0.72% :     0.000003s :      4: substitution.fold_const_symbol
 2.17% :     0.000008s :      5: substitution.graph_param_transform
76.09% :     0.000279s :      4: substitution.inline
 1.55% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.35% :     0.000009s :      8: substitution.remove_not_recompute_node
 1.50% :     0.000006s :      2: substitution.replace_old_param
 4.27% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
 3.69% :     0.000014s :      3: substitution.updatestate_pure_node_eliminater
 3.04% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026777     2
93.12% :     0.024935s :      1: type_inference.infer
 6.88% :     0.001841s :      1: type_inference.specialize
------[replace.]   0.000117     6
65.68% :     0.000077s :      4: replace.inline
34.32% :     0.000040s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000290     6
95.00% :     0.000276s :      4: match.inline
 5.00% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000221  1021
 1.47% :     0.000003s :     15: predicate.accumulaten_eliminater
 0.93% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.02% :     0.000002s :     15: predicate.addn_check_dump
 1.40% :     0.000003s :     15: predicate.addn_zero_filter
 2.54% :     0.000006s :     15: predicate.arithmetic_simplify
 1.51% :     0.000003s :     15: predicate.cast_eliminate
 0.43% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     15: predicate.compare_switch_simplify
 1.56% :     0.000003s :     15: predicate.depend_value_elim
 1.15% :     0.000003s :     15: predicate.dict_get_item_const_eliminator
 1.56% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.41% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.69% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      5: predicate.elim_not_effective
 0.67% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.59% :     0.000004s :     15: predicate.environ_add_const_eliminate
 0.90% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.22% :     0.000003s :     15: predicate.environ_get_depend_swap
 1.42% :     0.000003s :     15: predicate.environ_get_eliminate
 1.06% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000001s :      5: predicate.graph_param_transform
 4.53% :     0.000010s :     31: predicate.inline
 0.83% :     0.000002s :     10: predicate.inline_without_move
 0.33% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.48% :     0.000003s :     10: predicate.less_batch_normalization
 1.33% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.71% :     0.000004s :     22: predicate.load_eliminater
 1.23% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.33% :     0.000007s :     42: predicate.loop_unroll_before_grad
 2.78% :     0.000006s :     20: predicate.make_slice_get_slice_eliminator
 0.90% :     0.000002s :     15: predicate.merge_addn
 1.03% :     0.000002s :     15: predicate.minmaximum_grad
 1.60% :     0.000004s :      5: predicate.mutable_eliminate
 0.46% :     0.000001s :      5: predicate.opt_reshape
 1.92% :     0.000004s :     22: predicate.partial_eliminate
 1.61% :     0.000004s :     15: predicate.print_const_string_wrapper
 1.90% :     0.000004s :     15: predicate.reduce_eliminate
 1.44% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.65% :     0.000004s :     27: predicate.replace_applicator
 0.51% :     0.000001s :     10: predicate.replace_old_param
 0.29% :     0.000001s :      5: predicate.reset_defer_inline
 1.48% :     0.000003s :     15: predicate.reshape_eliminate
 1.43% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.77% :     0.000002s :      5: predicate.row_tensor_eliminate
 2.17% :     0.000005s :     15: predicate.same_eliminate
 0.45% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.90% :     0.000002s :     10: predicate.special_op_eliminate
 0.85% :     0.000002s :     10: predicate.specialize_transform
 1.74% :     0.000004s :     15: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.82% :     0.000004s :     21: predicate.switch_defer_inline
 1.80% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.28% :     0.000014s :     68: predicate.switch_simplify
 1.53% :     0.000003s :     15: predicate.tile_eliminate
 1.17% :     0.000003s :     15: predicate.transpose_eliminate
 1.62% :     0.000004s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.92% :     0.000004s :     15: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 2.03% :     0.000004s :     17: predicate.tuple_to_list_eliminator_
 1.46% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.34% :     0.000007s :     32: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000004s :     15: predicate.value_based_eliminate
 0.35% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.74% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001070    13
46.29% :     0.000495s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.59% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.12% :     0.000515s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.113835    76
 0.07% :     0.000082s :      1: add_recomputation
 0.17% :     0.000192s :      1: auto_monad
 0.03% :     0.000029s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.35% :     0.000395s :      1: bootstrap
 0.04% :     0.000040s :      1: cconv
 0.01% :     0.000015s :      1: convert_after_rewriter
 0.03% :     0.000034s :      1: cse_after_recomputation
 0.02% :     0.000017s :      1: environ_conv
 0.02% :     0.000028s :      1: event_method
 0.01% :     0.000013s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
28.57% :     0.032519s :      1: jit_opt_a
 0.21% :     0.000240s :      1: jit_opt_after_cconv
 0.08% :     0.000089s :      1: jit_opt_b
 0.47% :     0.000534s :      1: loop_unroll
 0.87% :     0.000990s :      1: mutable_eliminate
 1.27% :     0.001448s :     26: opt.transform.jit_opt_a
 0.09% :     0.000099s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000054s :      4: opt.transform.jit_opt_b
 0.02% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000057s :      4: opt.transform.symbol_engine_opt
 0.52% :     0.000595s :      1: opt_after_jit_grad
 0.01% :     0.000012s :      1: order_py_execute_after_rewriter
 0.00% :     0.000005s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000167s :      1: py_interpret_to_execute
 0.03% :     0.000031s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000025s :      1: remove_dup_value
 0.75% :     0.000852s :      1: renormalize.infer
21.76% :     0.024772s :      1: renormalize.specialize
 0.01% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.33% :     0.000380s :      1: rewriter_after_opt_a
 0.10% :     0.000109s :      1: rewriter_before_opt_a
 0.10% :     0.000119s :      1: symbol_engine_optimizer
20.00% :     0.022768s :      1: task_emit
23.63% :     0.026901s :      1: type_inference
 0.09% :     0.000101s :      1: validate


TotalTime = 0.092924, [33]
[bootstrap]: 0.00044898
[type_inference]: 0.0498102
[event_method]: 2.187e-05
[auto_monad]: 0.00016254
[graph_reusing]: 6.81001e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 0.00020306
[rewriter_before_opt_a]: 0.00010129
[expand_dump_flag]: 3.42997e-06
[jit_opt_a]: 0.00716517, [2]
    [Cycle 1]: 0.00264502, [27]
        [switch_simplify]: 6.758e-05
        [loop_unroll]: 3.579e-05
        [a_1]: 0.00077122
        [with_stream_mark]: 2.648e-05
        [recompute_prepare]: 1.198e-05
        [updatestate_depend_eliminate]: 7.13e-06
        [updatestate_assign_eliminate]: 5.64e-06
        [updatestate_loads_eliminate]: 4.94e-06
        [parameter_eliminate]: 2.18998e-06
        [specialize_transform]: 9.29e-06
        [updatestate_useless_node_eliminater]: 1.149e-05
        [accelerated_algorithm]: 8.95001e-06
        [meta_shard_fg_expand]: 3.27002e-06
        [get_grad_eliminate_]: 8.48999e-06
        [merge_forward]: 5.99e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.073e-05
        [j_node_and_user_rematch]: 1.349e-05
        [meta_fg_expand]: 4.27998e-06
        [replace_old_param]: 1.283e-05
        [inline_without_move]: 7.97e-06
        [renormalize]: 0.00130606
        [add_forward_monad_depend]: 7.21999e-06
        [auto_monad_grad]: 3.01001e-06
        [auto_monad_eliminator]: 2.429e-05
        [cse]: 4.426e-05
        [replace_applicator]: 2.052e-05
    [Cycle 2]: 0.00048077, [27]
        [switch_simplify]: 9.18002e-06
        [loop_unroll]: 8.3e-06
        [a_1]: 0.00017215
        [with_stream_mark]: 1.396e-05
        [recompute_prepare]: 8.11002e-06
        [updatestate_depend_eliminate]: 5.16002e-06
        [updatestate_assign_eliminate]: 5.16002e-06
        [updatestate_loads_eliminate]: 4.1e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 8.23999e-06
        [updatestate_useless_node_eliminater]: 1.14e-05
        [accelerated_algorithm]: 7.78999e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 8.04002e-06
        [merge_forward]: 4.50999e-06
        [cell_reuse_recompute_pass]: 2.35002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.632e-05
        [j_node_and_user_rematch]: 1.384e-05
        [meta_fg_expand]: 2.98e-06
        [replace_old_param]: 9.52001e-06
        [inline_without_move]: 7.48999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.74e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 1.101e-05
        [cse]: 2.101e-05
        [replace_applicator]: 7.63001e-06
[py_interpret_to_execute_after_opt_a]: 1.534e-05
[rewriter_after_opt_a]: 0.00030085
[convert_after_rewriter]: 1.22e-05
[order_py_execute_after_rewriter]: 8.15e-06
[mutable_eliminate]: 0.00082954
[jit_opt_b]: 7.477e-05, [1]
    [Cycle 1]: 6.597e-05, [2]
        [frontend_op_eliminate]: 2.618e-05
        [inline_after_opt_a]: 2.654e-05
[cconv]: 3.312e-05
[loop_unroll]: 0.0005028
[jit_opt_after_cconv]: 0.00021362, [1]
    [Cycle 1]: 0.00020609, [11]
        [c_1]: 5.037e-05
        [parameter_eliminate]: 3.25998e-06
        [updatestate_depend_eliminate]: 9.37001e-06
        [updatestate_assign_eliminate]: 4.89e-06
        [updatestate_loads_eliminate]: 4.67998e-06
        [cse]: 3.537e-05
        [call_graph_tuple_transform]: 2.352e-05
        [tuple_list_get_item_eliminator]: 8.67e-06
        [none_parameter_eliminate]: 1.94e-06
        [renormalize]: 1.02e-06
        [switch_simplify]: 1.028e-05
[remove_dup_value]: 1.994e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 1.387e-05
[add_recomputation]: 7.602e-05
[cse_after_recomputation]: 3.257e-05, [1]
    [Cycle 1]: 2.589e-05, [1]
        [cse]: 1.924e-05
[auto_monad_reorder]: 2.754e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 8.91997e-06
[opt_after_jit_grad]: 0.00056507
[symbol_engine_optimizer]: 0.00011514, [1]
    [Cycle 1]: 0.00010831, [6]
        [build]: 1.758e-05
        [elim_shapecalc]: 1.206e-05
        [elim_not_effective]: 2.02e-05
        [opt_reshape]: 8.67998e-06
        [fold_const_symbol]: 1.743e-05
        [renormalize]: 6.39993e-07
[validate]: 6.791e-05
[backend_pass]: 1.09998e-06
[task_emit]: 0.0317956
[execute]: 9.60001e-06
Sums
    bootstrap                            :     0.000449s :  0.51%
    type_inference                       :     0.049810s : 56.50%
    event_method                         :     0.000022s :  0.02%
    auto_monad                           :     0.000163s :  0.18%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000203s :  0.23%
    rewriter_before_opt_a                :     0.000101s :  0.11%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000077s :  0.09%
    jit_opt_a.loop_unroll                :     0.000044s :  0.05%
    jit_opt_a.a_1                        :     0.000943s :  1.07%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.00%
    jit_opt_a.specialize_transform       :     0.000018s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.03%
    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_        :     0.000017s :  0.02%
    jit_opt_a.merge_forward              :     0.000010s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000022s :  0.03%
    jit_opt_a.inline_without_move        :     0.000015s :  0.02%
    jit_opt_a.renormalize                :     0.001306s :  1.48%
    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      :     0.000035s :  0.04%
    jit_opt_a.cse                        :     0.000065s :  0.07%
    jit_opt_a.replace_applicator         :     0.000028s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.02%
    rewriter_after_opt_a                 :     0.000301s :  0.34%
    convert_after_rewriter               :     0.000012s :  0.01%
    order_py_execute_after_rewriter      :     0.000008s :  0.01%
    mutable_eliminate                    :     0.000830s :  0.94%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.03%
    cconv                                :     0.000033s :  0.04%
    loop_unroll                          :     0.000503s :  0.57%
    jit_opt_after_cconv.c_1              :     0.000050s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.01%
    remove_dup_value                     :     0.000020s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000014s :  0.02%
    add_recomputation                    :     0.000076s :  0.09%
    cse_after_recomputation.cse          :     0.000019s :  0.02%
    auto_monad_reorder                   :     0.000028s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.01%
    opt_after_jit_grad                   :     0.000565s :  0.64%
    symbol_engine_optimizer.build        :     0.000018s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000017s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000068s :  0.08%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.031796s : 36.06%
    execute                              :     0.000010s :  0.01%


Time group info:
------[substitution.]   0.000325    47
 3.03% :     0.000010s :      2: substitution.depend_value_elim
 0.86% :     0.000003s :      4: substitution.elim_not_effective
 0.76% :     0.000002s :      4: substitution.fold_const_symbol
 2.16% :     0.000007s :      5: substitution.graph_param_transform
77.26% :     0.000251s :      4: substitution.inline
 1.55% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.17% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.40% :     0.000005s :      2: substitution.replace_old_param
 4.23% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.66% :     0.000012s :      3: substitution.updatestate_pure_node_eliminater
 2.91% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.049722     2
96.91% :     0.048188s :      1: type_inference.infer
 3.09% :     0.001534s :      1: type_inference.specialize
------[replace.]   0.000099     6
64.39% :     0.000064s :      4: replace.inline
35.61% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
95.17% :     0.000248s :      4: match.inline
 4.83% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000181  1021
 1.20% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.15% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.07% :     0.000002s :     15: predicate.addn_check_dump
 1.16% :     0.000002s :     15: predicate.addn_zero_filter
 2.04% :     0.000004s :     15: predicate.arithmetic_simplify
 1.63% :     0.000003s :     15: predicate.cast_eliminate
 0.60% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     15: predicate.compare_switch_simplify
 1.31% :     0.000002s :     15: predicate.depend_value_elim
 1.18% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.57% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.62% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.47% :     0.000003s :     15: predicate.environ_get_eliminate
 1.23% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      5: predicate.fold_const_symbol
 1.15% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 4.99% :     0.000009s :     31: predicate.inline
 1.04% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.58% :     0.000003s :     10: predicate.less_batch_normalization
 1.46% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.85% :     0.000003s :     22: predicate.load_eliminater
 1.37% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.19% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.05% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     15: predicate.merge_addn
 1.18% :     0.000002s :     15: predicate.minmaximum_grad
 1.80% :     0.000003s :      5: predicate.mutable_eliminate
 0.63% :     0.000001s :      5: predicate.opt_reshape
 2.27% :     0.000004s :     22: predicate.partial_eliminate
 1.18% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.59% :     0.000003s :     15: predicate.reduce_eliminate
 1.64% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.86% :     0.000003s :     27: predicate.replace_applicator
 0.61% :     0.000001s :     10: predicate.replace_old_param
 0.31% :     0.000001s :      5: predicate.reset_defer_inline
 1.25% :     0.000002s :     15: predicate.reshape_eliminate
 1.48% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     15: predicate.same_eliminate
 0.79% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :     10: predicate.special_op_eliminate
 0.99% :     0.000002s :     10: predicate.specialize_transform
 1.49% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.15% :     0.000004s :     21: predicate.switch_defer_inline
 1.85% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.85% :     0.000012s :     68: predicate.switch_simplify
 1.43% :     0.000003s :     15: predicate.tile_eliminate
 1.32% :     0.000002s :     15: predicate.transpose_eliminate
 1.62% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 2.00% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.18% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000982    13
46.88% :     0.000460s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.35% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.76% :     0.000459s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.095674    76
 0.08% :     0.000080s :      1: add_recomputation
 0.18% :     0.000168s :      1: auto_monad
 0.03% :     0.000030s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.49% :     0.000472s :      1: bootstrap
 0.04% :     0.000036s :      1: cconv
 0.02% :     0.000015s :      1: convert_after_rewriter
 0.04% :     0.000035s :      1: cse_after_recomputation
 0.02% :     0.000016s :      1: environ_conv
 0.03% :     0.000027s :      1: event_method
 0.02% :     0.000015s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000006s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
 7.50% :     0.007171s :      1: jit_opt_a
 0.23% :     0.000217s :      1: jit_opt_after_cconv
 0.08% :     0.000078s :      1: jit_opt_b
 0.53% :     0.000512s :      1: loop_unroll
 0.88% :     0.000840s :      1: mutable_eliminate
 1.31% :     0.001255s :     26: opt.transform.jit_opt_a
 0.09% :     0.000089s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000044s :      4: opt.transform.jit_opt_b
 0.02% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000055s :      4: opt.transform.symbol_engine_opt
 0.60% :     0.000576s :      1: opt_after_jit_grad
 0.01% :     0.000011s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000210s :      1: py_interpret_to_execute
 0.02% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000022s :      1: remove_dup_value
 0.84% :     0.000804s :      1: renormalize.infer
 0.52% :     0.000493s :      1: renormalize.specialize
 0.01% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.32% :     0.000307s :      1: rewriter_after_opt_a
 0.11% :     0.000105s :      1: rewriter_before_opt_a
 0.12% :     0.000118s :      1: symbol_engine_optimizer
33.25% :     0.031815s :      1: task_emit
52.08% :     0.049827s :      1: type_inference
 0.10% :     0.000095s :      1: validate


TotalTime = 0.0501207, [33]
[bootstrap]: 0.00038499
[type_inference]: 0.0262051
[event_method]: 2.155e-05
[auto_monad]: 0.00016428
[graph_reusing]: 6.29999e-06
[pre_auto_parallel]: 3.21001e-06
[py_interpret_to_execute]: 0.00035348
[rewriter_before_opt_a]: 0.00010798
[expand_dump_flag]: 4.35999e-06
[jit_opt_a]: 0.00726869, [2]
    [Cycle 1]: 0.00268058, [27]
        [switch_simplify]: 6.912e-05
        [loop_unroll]: 3.718e-05
        [a_1]: 0.0007883
        [with_stream_mark]: 2.752e-05
        [recompute_prepare]: 1.196e-05
        [updatestate_depend_eliminate]: 7.03998e-06
        [updatestate_assign_eliminate]: 5.86e-06
        [updatestate_loads_eliminate]: 5.66e-06
        [parameter_eliminate]: 2.11998e-06
        [specialize_transform]: 9.06998e-06
        [updatestate_useless_node_eliminater]: 1.15e-05
        [accelerated_algorithm]: 8.69e-06
        [meta_shard_fg_expand]: 2.89001e-06
        [get_grad_eliminate_]: 8.18001e-06
        [merge_forward]: 5.72001e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.04e-05
        [j_node_and_user_rematch]: 1.385e-05
        [meta_fg_expand]: 3.6e-06
        [replace_old_param]: 1.141e-05
        [inline_without_move]: 7.98001e-06
        [renormalize]: 0.00133053
        [add_forward_monad_depend]: 7.41999e-06
        [auto_monad_grad]: 3.07002e-06
        [auto_monad_eliminator]: 2.327e-05
        [cse]: 4.281e-05
        [replace_applicator]: 1.839e-05
    [Cycle 2]: 0.00047761, [27]
        [switch_simplify]: 9.66e-06
        [loop_unroll]: 8.17e-06
        [a_1]: 0.00017362
        [with_stream_mark]: 1.336e-05
        [recompute_prepare]: 8.12e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 4.68999e-06
        [updatestate_loads_eliminate]: 4.33001e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 8.01001e-06
        [updatestate_useless_node_eliminater]: 1.048e-05
        [accelerated_algorithm]: 8.05e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 8.15999e-06
        [merge_forward]: 4.63001e-06
        [cell_reuse_recompute_pass]: 1.87999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.659e-05
        [j_node_and_user_rematch]: 1.268e-05
        [meta_fg_expand]: 2.69001e-06
        [replace_old_param]: 9.83002e-06
        [inline_without_move]: 7.51001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 2.34001e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 9.91998e-06
        [cse]: 2.148e-05
        [replace_applicator]: 7.82e-06
[py_interpret_to_execute_after_opt_a]: 1.666e-05
[rewriter_after_opt_a]: 0.00030518
[convert_after_rewriter]: 1.247e-05
[order_py_execute_after_rewriter]: 8.29998e-06
[mutable_eliminate]: 0.00078975
[jit_opt_b]: 7.016e-05, [1]
    [Cycle 1]: 6.172e-05, [2]
        [frontend_op_eliminate]: 2.478e-05
        [inline_after_opt_a]: 2.512e-05
[cconv]: 2.958e-05
[loop_unroll]: 0.00046751
[jit_opt_after_cconv]: 0.00020308, [1]
    [Cycle 1]: 0.00019652, [11]
        [c_1]: 4.64e-05
        [parameter_eliminate]: 3.16999e-06
        [updatestate_depend_eliminate]: 8.50001e-06
        [updatestate_assign_eliminate]: 4.38999e-06
        [updatestate_loads_eliminate]: 4.15999e-06
        [cse]: 3.368e-05
        [call_graph_tuple_transform]: 2.241e-05
        [tuple_list_get_item_eliminator]: 8.74e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 8.89995e-07
        [switch_simplify]: 8.39002e-06
[remove_dup_value]: 2.116e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 1.241e-05
[add_recomputation]: 7.239e-05
[cse_after_recomputation]: 3.069e-05, [1]
    [Cycle 1]: 2.522e-05, [1]
        [cse]: 1.879e-05
[auto_monad_reorder]: 2.651e-05
[get_jit_bprop_graph]: 2.46998e-06
[rewriter_after_jit_bprop_graph]: 5.56e-06
[opt_after_jit_grad]: 0.00049719
[symbol_engine_optimizer]: 0.00011277, [1]
    [Cycle 1]: 0.00010617, [6]
        [build]: 1.661e-05
        [elim_shapecalc]: 1.257e-05
        [elim_not_effective]: 1.939e-05
        [opt_reshape]: 9.37999e-06
        [fold_const_symbol]: 1.562e-05
        [renormalize]: 5.19998e-07
[validate]: 6.648e-05
[backend_pass]: 1.29e-06
[task_emit]: 0.0125781
[execute]: 1.014e-05
Sums
    bootstrap                            :     0.000385s :  0.85%
    type_inference                       :     0.026205s : 57.83%
    event_method                         :     0.000022s :  0.05%
    auto_monad                           :     0.000164s :  0.36%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000353s :  0.78%
    rewriter_before_opt_a                :     0.000108s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000079s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.10%
    jit_opt_a.a_1                        :     0.000962s :  2.12%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000015s :  0.03%
    jit_opt_a.renormalize                :     0.001331s :  2.94%
    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      :     0.000033s :  0.07%
    jit_opt_a.cse                        :     0.000064s :  0.14%
    jit_opt_a.replace_applicator         :     0.000026s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.04%
    rewriter_after_opt_a                 :     0.000305s :  0.67%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000790s :  1.74%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000030s :  0.07%
    loop_unroll                          :     0.000468s :  1.03%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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 :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000072s :  0.16%
    cse_after_recomputation.cse          :     0.000019s :  0.04%
    auto_monad_reorder                   :     0.000027s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000497s :  1.10%
    symbol_engine_optimizer.build        :     0.000017s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000066s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012578s : 27.76%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000339    47
 2.82% :     0.000010s :      2: substitution.depend_value_elim
 0.86% :     0.000003s :      4: substitution.elim_not_effective
 0.73% :     0.000002s :      4: substitution.fold_const_symbol
 2.06% :     0.000007s :      5: substitution.graph_param_transform
78.89% :     0.000267s :      4: substitution.inline
 1.37% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.08% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.27% :     0.000004s :      2: substitution.replace_old_param
 4.16% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 2.92% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 2.82% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026119     2
93.95% :     0.024538s :      1: type_inference.infer
 6.05% :     0.001581s :      1: type_inference.specialize
------[replace.]   0.000099     6
66.92% :     0.000066s :      4: replace.inline
33.08% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000277     6
95.33% :     0.000264s :      4: match.inline
 4.67% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000174  1021
 1.28% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.08% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.24% :     0.000002s :     15: predicate.addn_check_dump
 1.37% :     0.000002s :     15: predicate.addn_zero_filter
 2.04% :     0.000004s :     15: predicate.arithmetic_simplify
 1.40% :     0.000002s :     15: predicate.cast_eliminate
 0.54% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     15: predicate.compare_switch_simplify
 1.39% :     0.000002s :     15: predicate.depend_value_elim
 1.14% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.54% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.35% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.47% :     0.000003s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 4.85% :     0.000008s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :     10: predicate.less_batch_normalization
 1.51% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.10% :     0.000004s :     22: predicate.load_eliminater
 1.25% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.89% :     0.000007s :     42: predicate.loop_unroll_before_grad
 1.90% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.11% :     0.000002s :     15: predicate.minmaximum_grad
 1.47% :     0.000003s :      5: predicate.mutable_eliminate
 0.48% :     0.000001s :      5: predicate.opt_reshape
 2.34% :     0.000004s :     22: predicate.partial_eliminate
 1.37% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     15: predicate.reduce_eliminate
 1.41% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.62% :     0.000003s :     27: predicate.replace_applicator
 0.68% :     0.000001s :     10: predicate.replace_old_param
 0.29% :     0.000001s :      5: predicate.reset_defer_inline
 1.30% :     0.000002s :     15: predicate.reshape_eliminate
 1.45% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.88% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.49% :     0.000003s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.13% :     0.000002s :     10: predicate.special_op_eliminate
 1.03% :     0.000002s :     10: predicate.specialize_transform
 1.55% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.07% :     0.000004s :     21: predicate.switch_defer_inline
 1.88% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.66% :     0.000012s :     68: predicate.switch_simplify
 1.30% :     0.000002s :     15: predicate.tile_eliminate
 1.28% :     0.000002s :     15: predicate.transpose_eliminate
 1.55% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.94% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.36% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000968    13
45.75% :     0.000443s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.86% :     0.000066s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.39% :     0.000459s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.052901    76
 0.14% :     0.000076s :      1: add_recomputation
 0.32% :     0.000170s :      1: auto_monad
 0.05% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.77% :     0.000407s :      1: bootstrap
 0.06% :     0.000032s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.06% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.05% :     0.000027s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
13.75% :     0.007273s :      1: jit_opt_a
 0.39% :     0.000206s :      1: jit_opt_after_cconv
 0.14% :     0.000073s :      1: jit_opt_b
 0.90% :     0.000476s :      1: loop_unroll
 1.51% :     0.000799s :      1: mutable_eliminate
 2.40% :     0.001272s :     26: opt.transform.jit_opt_a
 0.15% :     0.000081s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000043s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000053s :      4: opt.transform.symbol_engine_opt
 0.96% :     0.000506s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.68% :     0.000361s :      1: py_interpret_to_execute
 0.04% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000025s :      1: remove_dup_value
 1.57% :     0.000828s :      1: renormalize.infer
 0.93% :     0.000493s :      1: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.59% :     0.000310s :      1: rewriter_after_opt_a
 0.21% :     0.000113s :      1: rewriter_before_opt_a
 0.22% :     0.000116s :      1: symbol_engine_optimizer
23.82% :     0.012599s :      1: task_emit
49.57% :     0.026222s :      1: type_inference
 0.18% :     0.000094s :      1: validate


TotalTime = 0.103989, [33]
[bootstrap]: 0.00047856
[type_inference]: 0.0665859
[event_method]: 2.151e-05
[auto_monad]: 0.00017018
[graph_reusing]: 7.12002e-06
[pre_auto_parallel]: 3.43e-06
[py_interpret_to_execute]: 0.00013813
[rewriter_before_opt_a]: 0.00010481
[expand_dump_flag]: 3.47997e-06
[jit_opt_a]: 0.00787295, [2]
    [Cycle 1]: 0.00272625, [27]
        [switch_simplify]: 7.782e-05
        [loop_unroll]: 3.668e-05
        [a_1]: 0.00080511
        [with_stream_mark]: 2.819e-05
        [recompute_prepare]: 1.082e-05
        [updatestate_depend_eliminate]: 7.45003e-06
        [updatestate_assign_eliminate]: 6.17001e-06
        [updatestate_loads_eliminate]: 4.82998e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 9.66e-06
        [updatestate_useless_node_eliminater]: 1.163e-05
        [accelerated_algorithm]: 8.67e-06
        [meta_shard_fg_expand]: 2.72001e-06
        [get_grad_eliminate_]: 8.33001e-06
        [merge_forward]: 5.76e-06
        [cell_reuse_recompute_pass]: 1.59998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.952e-05
        [j_node_and_user_rematch]: 1.444e-05
        [meta_fg_expand]: 3.78999e-06
        [replace_old_param]: 1.195e-05
        [inline_without_move]: 8.17e-06
        [renormalize]: 0.00134839
        [add_forward_monad_depend]: 7.43999e-06
        [auto_monad_grad]: 2.70002e-06
        [auto_monad_eliminator]: 2.456e-05
        [cse]: 4.468e-05
        [replace_applicator]: 1.864e-05
    [Cycle 2]: 0.00049222, [27]
        [switch_simplify]: 9.24998e-06
        [loop_unroll]: 8.13999e-06
        [a_1]: 0.00017222
        [with_stream_mark]: 1.357e-05
        [recompute_prepare]: 8.64e-06
        [updatestate_depend_eliminate]: 5.05999e-06
        [updatestate_assign_eliminate]: 4.90999e-06
        [updatestate_loads_eliminate]: 4.03999e-06
        [parameter_eliminate]: 1.29998e-06
        [specialize_transform]: 9.09e-06
        [updatestate_useless_node_eliminater]: 1.153e-05
        [accelerated_algorithm]: 9.95002e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 8.63001e-06
        [merge_forward]: 4.73001e-06
        [cell_reuse_recompute_pass]: 2.45002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.793e-05
        [j_node_and_user_rematch]: 1.388e-05
        [meta_fg_expand]: 2.81999e-06
        [replace_old_param]: 1.156e-05
        [inline_without_move]: 8.69e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.61998e-06
        [auto_monad_grad]: 9.79984e-07
        [auto_monad_eliminator]: 1.066e-05
        [cse]: 2.148e-05
        [replace_applicator]: 8.00999e-06
[py_interpret_to_execute_after_opt_a]: 1.719e-05
[rewriter_after_opt_a]: 0.00029991
[convert_after_rewriter]: 1.153e-05
[order_py_execute_after_rewriter]: 8.25e-06
[mutable_eliminate]: 0.00078393
[jit_opt_b]: 7.046e-05, [1]
    [Cycle 1]: 6.256e-05, [2]
        [frontend_op_eliminate]: 2.529e-05
        [inline_after_opt_a]: 2.471e-05
[cconv]: 2.991e-05
[loop_unroll]: 0.00047747
[jit_opt_after_cconv]: 0.00020909, [1]
    [Cycle 1]: 0.00020177, [11]
        [c_1]: 4.741e-05
        [parameter_eliminate]: 3.43999e-06
        [updatestate_depend_eliminate]: 8.90999e-06
        [updatestate_assign_eliminate]: 4.57e-06
        [updatestate_loads_eliminate]: 4.31002e-06
        [cse]: 3.19e-05
        [call_graph_tuple_transform]: 2.486e-05
        [tuple_list_get_item_eliminator]: 8.65001e-06
        [none_parameter_eliminate]: 1.77999e-06
        [renormalize]: 1.46002e-06
        [switch_simplify]: 9.37001e-06
[remove_dup_value]: 2.099e-05
[partial_unused_args_eliminate]: 2.29999e-06
[environ_conv]: 1.175e-05
[add_recomputation]: 7.226e-05
[cse_after_recomputation]: 3.197e-05, [1]
    [Cycle 1]: 2.414e-05, [1]
        [cse]: 1.819e-05
[auto_monad_reorder]: 2.704e-05
[get_jit_bprop_graph]: 3.26001e-06
[rewriter_after_jit_bprop_graph]: 6.14999e-06
[opt_after_jit_grad]: 0.00051063
[symbol_engine_optimizer]: 0.00010659, [1]
    [Cycle 1]: 9.907e-05, [6]
        [build]: 1.476e-05
        [elim_shapecalc]: 1.175e-05
        [elim_not_effective]: 1.904e-05
        [opt_reshape]: 8.84e-06
        [fold_const_symbol]: 1.496e-05
        [renormalize]: 3.89991e-07
[validate]: 6.408e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.0255697
[execute]: 8.80001e-06
Sums
    bootstrap                            :     0.000479s :  0.49%
    type_inference                       :     0.066586s : 67.51%
    event_method                         :     0.000022s :  0.02%
    auto_monad                           :     0.000170s :  0.17%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000138s :  0.14%
    rewriter_before_opt_a                :     0.000105s :  0.11%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000087s :  0.09%
    jit_opt_a.loop_unroll                :     0.000045s :  0.05%
    jit_opt_a.a_1                        :     0.000977s :  0.99%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000019s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.02%
    jit_opt_a.merge_forward              :     0.000010s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.02%
    jit_opt_a.inline_without_move        :     0.000017s :  0.02%
    jit_opt_a.renormalize                :     0.001348s :  1.37%
    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      :     0.000035s :  0.04%
    jit_opt_a.cse                        :     0.000066s :  0.07%
    jit_opt_a.replace_applicator         :     0.000027s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.02%
    rewriter_after_opt_a                 :     0.000300s :  0.30%
    convert_after_rewriter               :     0.000012s :  0.01%
    order_py_execute_after_rewriter      :     0.000008s :  0.01%
    mutable_eliminate                    :     0.000784s :  0.79%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.03%
    cconv                                :     0.000030s :  0.03%
    loop_unroll                          :     0.000477s :  0.48%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.cse              :     0.000032s :  0.03%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.01%
    remove_dup_value                     :     0.000021s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000012s :  0.01%
    add_recomputation                    :     0.000072s :  0.07%
    cse_after_recomputation.cse          :     0.000018s :  0.02%
    auto_monad_reorder                   :     0.000027s :  0.03%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000511s :  0.52%
    symbol_engine_optimizer.build        :     0.000015s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000064s :  0.06%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.025570s : 25.93%
    execute                              :     0.000009s :  0.01%


Time group info:
------[substitution.]   0.000349    47
 3.01% :     0.000011s :      2: substitution.depend_value_elim
 0.98% :     0.000003s :      4: substitution.elim_not_effective
 0.70% :     0.000002s :      4: substitution.fold_const_symbol
 2.11% :     0.000007s :      5: substitution.graph_param_transform
78.44% :     0.000274s :      4: substitution.inline
 1.59% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.12% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.40% :     0.000005s :      2: substitution.replace_old_param
 3.79% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 2.89% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 2.96% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.066492     2
97.62% :     0.064907s :      1: type_inference.infer
 2.38% :     0.001585s :      1: type_inference.specialize
------[replace.]   0.000098     6
67.20% :     0.000066s :      4: replace.inline
32.80% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000283     6
95.71% :     0.000271s :      4: match.inline
 4.29% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000183  1021
 1.20% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     15: predicate.addn_check_dump
 1.17% :     0.000002s :     15: predicate.addn_zero_filter
 2.13% :     0.000004s :     15: predicate.arithmetic_simplify
 1.60% :     0.000003s :     15: predicate.cast_eliminate
 0.54% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.34% :     0.000002s :     15: predicate.compare_switch_simplify
 1.57% :     0.000003s :     15: predicate.depend_value_elim
 1.04% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.82% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.57% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.22% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.39% :     0.000003s :     15: predicate.environ_get_depend_swap
 1.43% :     0.000003s :     15: predicate.environ_get_eliminate
 1.05% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.08% :     0.000002s :     10: predicate.get_grad_eliminate
 0.29% :     0.000001s :      5: predicate.graph_param_transform
 4.97% :     0.000009s :     31: predicate.inline
 0.96% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.62% :     0.000003s :     10: predicate.less_batch_normalization
 1.36% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     22: predicate.load_eliminater
 1.26% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.27% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     15: predicate.merge_addn
 1.06% :     0.000002s :     15: predicate.minmaximum_grad
 1.41% :     0.000003s :      5: predicate.mutable_eliminate
 0.43% :     0.000001s :      5: predicate.opt_reshape
 2.22% :     0.000004s :     22: predicate.partial_eliminate
 1.27% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.82% :     0.000003s :     15: predicate.reduce_eliminate
 2.03% :     0.000004s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.79% :     0.000003s :     27: predicate.replace_applicator
 0.55% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.29% :     0.000002s :     15: predicate.reshape_eliminate
 1.55% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.87% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.36% :     0.000003s :     15: predicate.same_eliminate
 0.51% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.22% :     0.000002s :     10: predicate.special_op_eliminate
 1.15% :     0.000002s :     10: predicate.specialize_transform
 2.06% :     0.000004s :     15: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.14% :     0.000004s :     21: predicate.switch_defer_inline
 1.89% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.12% :     0.000011s :     68: predicate.switch_simplify
 1.33% :     0.000002s :     15: predicate.tile_eliminate
 1.45% :     0.000003s :     15: predicate.transpose_eliminate
 1.45% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.68% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.38% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000003s :     15: predicate.value_based_eliminate
 0.80% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000999    13
49.35% :     0.000493s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.71% :     0.000067s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.95% :     0.000439s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.106818    76
 0.07% :     0.000075s :      1: add_recomputation
 0.16% :     0.000175s :      1: auto_monad
 0.03% :     0.000030s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.47% :     0.000499s :      1: bootstrap
 0.03% :     0.000033s :      1: cconv
 0.01% :     0.000015s :      1: convert_after_rewriter
 0.03% :     0.000034s :      1: cse_after_recomputation
 0.01% :     0.000014s :      1: environ_conv
 0.02% :     0.000026s :      1: event_method
 0.01% :     0.000014s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000006s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
 7.38% :     0.007879s :      1: jit_opt_a
 0.20% :     0.000212s :      1: jit_opt_after_cconv
 0.07% :     0.000073s :      1: jit_opt_b
 0.45% :     0.000486s :      1: loop_unroll
 0.74% :     0.000794s :      1: mutable_eliminate
 1.22% :     0.001298s :     26: opt.transform.jit_opt_a
 0.08% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000043s :      4: opt.transform.jit_opt_b
 0.02% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000051s :      4: opt.transform.symbol_engine_opt
 0.49% :     0.000520s :      1: opt_after_jit_grad
 0.01% :     0.000010s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.13% :     0.000142s :      1: py_interpret_to_execute
 0.02% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000024s :      1: remove_dup_value
 0.77% :     0.000822s :      1: renormalize.infer
 0.48% :     0.000516s :      1: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000305s :      1: rewriter_after_opt_a
 0.10% :     0.000109s :      1: rewriter_before_opt_a
 0.10% :     0.000110s :      1: symbol_engine_optimizer
23.96% :     0.025591s :      1: task_emit
62.35% :     0.066604s :      1: type_inference
 0.09% :     0.000093s :      1: validate


TotalTime = 0.110922, [33]
[bootstrap]: 0.00035852
[type_inference]: 0.0648719
[event_method]: 2.284e-05
[auto_monad]: 0.00017222
[graph_reusing]: 5.89999e-06
[pre_auto_parallel]: 2.91999e-06
[py_interpret_to_execute]: 0.00013706
[rewriter_before_opt_a]: 0.00010081
[expand_dump_flag]: 3.93001e-06
[jit_opt_a]: 0.00759908, [2]
    [Cycle 1]: 0.00281171, [27]
        [switch_simplify]: 7.585e-05
        [loop_unroll]: 3.791e-05
        [a_1]: 0.00087924
        [with_stream_mark]: 2.783e-05
        [recompute_prepare]: 1.172e-05
        [updatestate_depend_eliminate]: 7.31001e-06
        [updatestate_assign_eliminate]: 6.01e-06
        [updatestate_loads_eliminate]: 5.01002e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 1.05e-05
        [updatestate_useless_node_eliminater]: 1.156e-05
        [accelerated_algorithm]: 9.27999e-06
        [meta_shard_fg_expand]: 2.88e-06
        [get_grad_eliminate_]: 8.75001e-06
        [merge_forward]: 5.99e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.108e-05
        [j_node_and_user_rematch]: 1.418e-05
        [meta_fg_expand]: 3.6e-06
        [replace_old_param]: 1.38e-05
        [inline_without_move]: 7.83001e-06
        [renormalize]: 0.00133187
        [add_forward_monad_depend]: 7.4e-06
        [auto_monad_grad]: 3.65e-06
        [auto_monad_eliminator]: 2.621e-05
        [cse]: 4.635e-05
        [replace_applicator]: 2.053e-05
    [Cycle 2]: 0.00048444, [27]
        [switch_simplify]: 9.14e-06
        [loop_unroll]: 8.51002e-06
        [a_1]: 0.00017695
        [with_stream_mark]: 1.357e-05
        [recompute_prepare]: 8.43001e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 5.21002e-06
        [updatestate_loads_eliminate]: 4.13999e-06
        [parameter_eliminate]: 1.41998e-06
        [specialize_transform]: 8.28001e-06
        [updatestate_useless_node_eliminater]: 1.031e-05
        [accelerated_algorithm]: 8.02e-06
        [meta_shard_fg_expand]: 2.01998e-06
        [get_grad_eliminate_]: 8.04002e-06
        [merge_forward]: 4.37e-06
        [cell_reuse_recompute_pass]: 3.01001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.683e-05
        [j_node_and_user_rematch]: 1.234e-05
        [meta_fg_expand]: 2.76e-06
        [replace_old_param]: 1.031e-05
        [inline_without_move]: 7.28999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 9.99979e-07
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 1.122e-05
        [cse]: 2.204e-05
        [replace_applicator]: 7.98999e-06
[py_interpret_to_execute_after_opt_a]: 1.77e-05
[rewriter_after_opt_a]: 0.00028658
[convert_after_rewriter]: 1.201e-05
[order_py_execute_after_rewriter]: 7.07002e-06
[mutable_eliminate]: 0.00087407
[jit_opt_b]: 7.279e-05, [1]
    [Cycle 1]: 6.418e-05, [2]
        [frontend_op_eliminate]: 2.636e-05
        [inline_after_opt_a]: 2.456e-05
[cconv]: 3.124e-05
[loop_unroll]: 0.0222628
[jit_opt_after_cconv]: 0.00028714, [1]
    [Cycle 1]: 0.00027421, [11]
        [c_1]: 6.953e-05
        [parameter_eliminate]: 7.77e-06
        [updatestate_depend_eliminate]: 1.532e-05
        [updatestate_assign_eliminate]: 5.58002e-06
        [updatestate_loads_eliminate]: 4.80001e-06
        [cse]: 5.911e-05
        [call_graph_tuple_transform]: 2.71e-05
        [tuple_list_get_item_eliminator]: 9.46003e-06
        [none_parameter_eliminate]: 1.56998e-06
        [renormalize]: 1.30999e-06
        [switch_simplify]: 1.193e-05
[remove_dup_value]: 2.43e-05
[partial_unused_args_eliminate]: 2.72001e-06
[environ_conv]: 1.809e-05
[add_recomputation]: 8.143e-05
[cse_after_recomputation]: 3.808e-05, [1]
    [Cycle 1]: 2.912e-05, [1]
        [cse]: 2.177e-05
[auto_monad_reorder]: 2.82e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 1.399e-05
[opt_after_jit_grad]: 0.0008259
[symbol_engine_optimizer]: 0.00012534, [1]
    [Cycle 1]: 0.00011672, [6]
        [build]: 2.085e-05
        [elim_shapecalc]: 1.33e-05
        [elim_not_effective]: 2.454e-05
        [opt_reshape]: 9.31e-06
        [fold_const_symbol]: 1.509e-05
        [renormalize]: 1.02e-06
[validate]: 7.888e-05
[backend_pass]: 1.17e-06
[task_emit]: 0.0122507
[execute]: 9.36e-06
Sums
    bootstrap                            :     0.000359s :  0.34%
    type_inference                       :     0.064872s : 61.29%
    event_method                         :     0.000023s :  0.02%
    auto_monad                           :     0.000172s :  0.16%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000137s :  0.13%
    rewriter_before_opt_a                :     0.000101s :  0.10%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000085s :  0.08%
    jit_opt_a.loop_unroll                :     0.000046s :  0.04%
    jit_opt_a.a_1                        :     0.001056s :  1.00%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000019s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.02%
    jit_opt_a.merge_forward              :     0.000010s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.02%
    jit_opt_a.inline_without_move        :     0.000015s :  0.01%
    jit_opt_a.renormalize                :     0.001332s :  1.26%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000037s :  0.04%
    jit_opt_a.cse                        :     0.000068s :  0.06%
    jit_opt_a.replace_applicator         :     0.000029s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.02%
    rewriter_after_opt_a                 :     0.000287s :  0.27%
    convert_after_rewriter               :     0.000012s :  0.01%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000874s :  0.83%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.02%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.02%
    cconv                                :     0.000031s :  0.03%
    loop_unroll                          :     0.022263s : 21.03%
    jit_opt_after_cconv.c_1              :     0.000070s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000008s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000015s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.00%
    jit_opt_after_cconv.cse              :     0.000059s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000012s :  0.01%
    remove_dup_value                     :     0.000024s :  0.02%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000018s :  0.02%
    add_recomputation                    :     0.000081s :  0.08%
    cse_after_recomputation.cse          :     0.000022s :  0.02%
    auto_monad_reorder                   :     0.000028s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000014s :  0.01%
    opt_after_jit_grad                   :     0.000826s :  0.78%
    symbol_engine_optimizer.build        :     0.000021s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000025s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000079s :  0.07%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012251s : 11.57%
    execute                              :     0.000009s :  0.01%


Time group info:
------[substitution.]   0.000362    47
 3.14% :     0.000011s :      2: substitution.depend_value_elim
 0.96% :     0.000003s :      4: substitution.elim_not_effective
 0.81% :     0.000003s :      4: substitution.fold_const_symbol
 2.10% :     0.000008s :      5: substitution.graph_param_transform
76.47% :     0.000277s :      4: substitution.inline
 1.37% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.05% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.66% :     0.000006s :      2: substitution.replace_old_param
 3.78% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 4.30% :     0.000016s :      3: substitution.updatestate_pure_node_eliminater
 3.36% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.064776     2
97.46% :     0.063131s :      1: type_inference.infer
 2.54% :     0.001646s :      1: type_inference.specialize
------[replace.]   0.000102     6
68.41% :     0.000070s :      4: replace.inline
31.59% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000286     6
95.62% :     0.000273s :      4: match.inline
 4.38% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000197  1021
 1.13% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.78% :     0.000004s :      5: predicate.ad_related_special_op_eliminate
 1.03% :     0.000002s :     15: predicate.addn_check_dump
 1.14% :     0.000002s :     15: predicate.addn_zero_filter
 2.21% :     0.000004s :     15: predicate.arithmetic_simplify
 1.61% :     0.000003s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 0.97% :     0.000002s :     15: predicate.compare_switch_simplify
 1.29% :     0.000003s :     15: predicate.depend_value_elim
 0.99% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.28% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.97% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      5: predicate.elim_not_effective
 0.63% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.03% :     0.000002s :     15: predicate.environ_get_add_eliminate
 0.99% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.82% :     0.000004s :     15: predicate.environ_get_eliminate
 0.95% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      5: predicate.fold_const_symbol
 1.20% :     0.000002s :     10: predicate.get_grad_eliminate
 0.33% :     0.000001s :      5: predicate.graph_param_transform
 5.39% :     0.000011s :     31: predicate.inline
 0.90% :     0.000002s :     10: predicate.inline_without_move
 0.38% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :     10: predicate.less_batch_normalization
 1.32% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.57% :     0.000003s :     22: predicate.load_eliminater
 3.31% :     0.000007s :      5: predicate.loop_unroll_after_grad
 3.31% :     0.000007s :     42: predicate.loop_unroll_before_grad
 2.39% :     0.000005s :     20: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000002s :     15: predicate.merge_addn
 0.97% :     0.000002s :     15: predicate.minmaximum_grad
 1.45% :     0.000003s :      5: predicate.mutable_eliminate
 0.52% :     0.000001s :      5: predicate.opt_reshape
 2.15% :     0.000004s :     22: predicate.partial_eliminate
 1.24% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.95% :     0.000004s :     15: predicate.reduce_eliminate
 1.32% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.75% :     0.000003s :     27: predicate.replace_applicator
 0.51% :     0.000001s :     10: predicate.replace_old_param
 0.28% :     0.000001s :      5: predicate.reset_defer_inline
 1.20% :     0.000002s :     15: predicate.reshape_eliminate
 1.67% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.94% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     15: predicate.same_eliminate
 0.50% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :     10: predicate.special_op_eliminate
 1.00% :     0.000002s :     10: predicate.specialize_transform
 1.63% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.94% :     0.000004s :     21: predicate.switch_defer_inline
 1.81% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.53% :     0.000013s :     68: predicate.switch_simplify
 1.27% :     0.000002s :     15: predicate.tile_eliminate
 1.37% :     0.000003s :     15: predicate.transpose_eliminate
 1.43% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 4.05% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.55% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.96% :     0.000004s :     15: predicate.value_based_eliminate
 0.49% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000919    13
40.63% :     0.000374s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.59% :     0.000070s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.78% :     0.000476s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.113896    76
 0.07% :     0.000085s :      1: add_recomputation
 0.16% :     0.000178s :      1: auto_monad
 0.03% :     0.000031s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.33% :     0.000377s :      1: bootstrap
 0.03% :     0.000034s :      1: cconv
 0.01% :     0.000015s :      1: convert_after_rewriter
 0.04% :     0.000041s :      1: cse_after_recomputation
 0.02% :     0.000020s :      1: environ_conv
 0.02% :     0.000028s :      1: event_method
 0.01% :     0.000015s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000006s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
 6.68% :     0.007605s :      1: jit_opt_a
 0.26% :     0.000292s :      1: jit_opt_after_cconv
 0.07% :     0.000075s :      1: jit_opt_b
19.57% :     0.022291s :      1: loop_unroll
 0.78% :     0.000885s :      1: mutable_eliminate
 1.21% :     0.001381s :     26: opt.transform.jit_opt_a
 0.10% :     0.000113s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000044s :      4: opt.transform.jit_opt_b
 0.04% :     0.000050s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000044s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000058s :      4: opt.transform.symbol_engine_opt
 0.74% :     0.000840s :      1: opt_after_jit_grad
 0.01% :     0.000009s :      1: order_py_execute_after_rewriter
 0.00% :     0.000005s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000141s :      1: py_interpret_to_execute
 0.02% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000028s :      1: remove_dup_value
 0.72% :     0.000822s :      1: renormalize.infer
 0.44% :     0.000500s :      1: renormalize.specialize
 0.01% :     0.000016s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000291s :      1: rewriter_after_opt_a
 0.09% :     0.000105s :      1: rewriter_before_opt_a
 0.11% :     0.000128s :      1: symbol_engine_optimizer
10.78% :     0.012274s :      1: task_emit
56.97% :     0.064891s :      1: type_inference
 0.10% :     0.000111s :      1: validate


TotalTime = 0.0595385, [33]
[bootstrap]: 0.00060226
[type_inference]: 0.0288744
[event_method]: 2.467e-05
[auto_monad]: 0.00019081
[graph_reusing]: 7.46001e-06
[pre_auto_parallel]: 3.36001e-06
[py_interpret_to_execute]: 0.00014671
[rewriter_before_opt_a]: 0.00010489
[expand_dump_flag]: 4.05e-06
[jit_opt_a]: 0.00869487, [2]
    [Cycle 1]: 0.00304736, [27]
        [switch_simplify]: 7.733e-05
        [loop_unroll]: 3.604e-05
        [a_1]: 0.00083697
        [with_stream_mark]: 3.262e-05
        [recompute_prepare]: 1.356e-05
        [updatestate_depend_eliminate]: 6.70002e-06
        [updatestate_assign_eliminate]: 1.068e-05
        [updatestate_loads_eliminate]: 5.42999e-06
        [parameter_eliminate]: 2.96999e-06
        [specialize_transform]: 9.89999e-06
        [updatestate_useless_node_eliminater]: 1.245e-05
        [accelerated_algorithm]: 9.94999e-06
        [meta_shard_fg_expand]: 3.43e-06
        [get_grad_eliminate_]: 8.54002e-06
        [merge_forward]: 6.21e-06
        [cell_reuse_recompute_pass]: 2.77002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 4.377e-05
        [j_node_and_user_rematch]: 1.728e-05
        [meta_fg_expand]: 4.75001e-06
        [replace_old_param]: 1.431e-05
        [inline_without_move]: 8.21002e-06
        [renormalize]: 0.0015298
        [add_forward_monad_depend]: 1.098e-05
        [auto_monad_grad]: 2.93e-06
        [auto_monad_eliminator]: 3.025e-05
        [cse]: 4.638e-05
        [replace_applicator]: 2.624e-05
    [Cycle 2]: 0.00059837, [27]
        [switch_simplify]: 1.069e-05
        [loop_unroll]: 9.58997e-06
        [a_1]: 0.00019826
        [with_stream_mark]: 2.115e-05
        [recompute_prepare]: 1.125e-05
        [updatestate_depend_eliminate]: 6.81999e-06
        [updatestate_assign_eliminate]: 6.02001e-06
        [updatestate_loads_eliminate]: 4.73001e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 1.003e-05
        [updatestate_useless_node_eliminater]: 1.253e-05
        [accelerated_algorithm]: 8.45001e-06
        [meta_shard_fg_expand]: 2.79999e-06
        [get_grad_eliminate_]: 8.75999e-06
        [merge_forward]: 6.50997e-06
        [cell_reuse_recompute_pass]: 3.45e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.101e-05
        [j_node_and_user_rematch]: 1.394e-05
        [meta_fg_expand]: 4.08001e-06
        [replace_old_param]: 1.321e-05
        [inline_without_move]: 8.33001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.62001e-06
        [auto_monad_grad]: 2.65002e-06
        [auto_monad_eliminator]: 1.692e-05
        [cse]: 3.128e-05
        [replace_applicator]: 1.137e-05
[py_interpret_to_execute_after_opt_a]: 2.234e-05
[rewriter_after_opt_a]: 0.00034556
[convert_after_rewriter]: 1.985e-05
[order_py_execute_after_rewriter]: 8.42998e-06
[mutable_eliminate]: 0.00122491
[jit_opt_b]: 9.03e-05, [1]
    [Cycle 1]: 7.7e-05, [2]
        [frontend_op_eliminate]: 3.101e-05
        [inline_after_opt_a]: 3.064e-05
[cconv]: 4.544e-05
[loop_unroll]: 0.00303574
[jit_opt_after_cconv]: 0.00028621, [1]
    [Cycle 1]: 0.00027413, [11]
        [c_1]: 6.72e-05
        [parameter_eliminate]: 7.67998e-06
        [updatestate_depend_eliminate]: 1.589e-05
        [updatestate_assign_eliminate]: 5.89e-06
        [updatestate_loads_eliminate]: 5.19e-06
        [cse]: 5.34e-05
        [call_graph_tuple_transform]: 3.014e-05
        [tuple_list_get_item_eliminator]: 1.037e-05
        [none_parameter_eliminate]: 2.01998e-06
        [renormalize]: 1.07e-06
        [switch_simplify]: 1.091e-05
[remove_dup_value]: 2.487e-05
[partial_unused_args_eliminate]: 3.68e-06
[environ_conv]: 1.834e-05
[add_recomputation]: 9.366e-05
[cse_after_recomputation]: 4.516e-05, [1]
    [Cycle 1]: 3.538e-05, [1]
        [cse]: 2.578e-05
[auto_monad_reorder]: 3.272e-05
[get_jit_bprop_graph]: 2.61999e-06
[rewriter_after_jit_bprop_graph]: 1.297e-05
[opt_after_jit_grad]: 0.00080082
[symbol_engine_optimizer]: 0.00012915, [1]
    [Cycle 1]: 0.00011983, [6]
        [build]: 2.081e-05
        [elim_shapecalc]: 1.35e-05
        [elim_not_effective]: 2.507e-05
        [opt_reshape]: 1.022e-05
        [fold_const_symbol]: 1.598e-05
        [renormalize]: 8.39995e-07
[validate]: 8.913e-05
[backend_pass]: 1.27e-06
[task_emit]: 0.014202
[execute]: 1.084e-05
Sums
    bootstrap                            :     0.000602s :  1.12%
    type_inference                       :     0.028874s : 53.87%
    event_method                         :     0.000025s :  0.05%
    auto_monad                           :     0.000191s :  0.36%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000147s :  0.27%
    rewriter_before_opt_a                :     0.000105s :  0.20%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000088s :  0.16%
    jit_opt_a.loop_unroll                :     0.000046s :  0.09%
    jit_opt_a.a_1                        :     0.001035s :  1.93%
    jit_opt_a.with_stream_mark           :     0.000054s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000025s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000017s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000020s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.03%
    jit_opt_a.merge_forward              :     0.000013s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000065s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000031s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000009s :  0.02%
    jit_opt_a.replace_old_param          :     0.000028s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.03%
    jit_opt_a.renormalize                :     0.001530s :  2.85%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000047s :  0.09%
    jit_opt_a.cse                        :     0.000078s :  0.14%
    jit_opt_a.replace_applicator         :     0.000038s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000022s :  0.04%
    rewriter_after_opt_a                 :     0.000346s :  0.64%
    convert_after_rewriter               :     0.000020s :  0.04%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.001225s :  2.29%
    jit_opt_b.frontend_op_eliminate      :     0.000031s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000031s :  0.06%
    cconv                                :     0.000045s :  0.08%
    loop_unroll                          :     0.003036s :  5.66%
    jit_opt_after_cconv.c_1              :     0.000067s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000008s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000016s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000053s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000030s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000010s :  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  :     0.000011s :  0.02%
    remove_dup_value                     :     0.000025s :  0.05%
    partial_unused_args_eliminate        :     0.000004s :  0.01%
    environ_conv                         :     0.000018s :  0.03%
    add_recomputation                    :     0.000094s :  0.17%
    cse_after_recomputation.cse          :     0.000026s :  0.05%
    auto_monad_reorder                   :     0.000033s :  0.06%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.02%
    opt_after_jit_grad                   :     0.000801s :  1.49%
    symbol_engine_optimizer.build        :     0.000021s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000025s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000089s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.014202s : 26.50%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000385    47
 3.20% :     0.000012s :      2: substitution.depend_value_elim
 0.94% :     0.000004s :      4: substitution.elim_not_effective
 0.65% :     0.000003s :      4: substitution.fold_const_symbol
 2.14% :     0.000008s :      5: substitution.graph_param_transform
75.50% :     0.000290s :      4: substitution.inline
 1.38% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.04% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.97% :     0.000008s :      2: substitution.replace_old_param
 4.11% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
 4.81% :     0.000018s :      3: substitution.updatestate_pure_node_eliminater
 3.26% :     0.000013s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.028762     2
94.01% :     0.027040s :      1: type_inference.infer
 5.99% :     0.001722s :      1: type_inference.specialize
------[replace.]   0.000106     6
64.81% :     0.000069s :      4: replace.inline
35.19% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000302     6
95.14% :     0.000287s :      4: match.inline
 4.86% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000206  1021
 1.42% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.29% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 0.93% :     0.000002s :     15: predicate.addn_check_dump
 1.40% :     0.000003s :     15: predicate.addn_zero_filter
 2.19% :     0.000005s :     15: predicate.arithmetic_simplify
 1.16% :     0.000002s :     15: predicate.cast_eliminate
 0.61% :     0.000001s :      5: predicate.check_bprop_eliminate
 0.91% :     0.000002s :     15: predicate.compare_switch_simplify
 1.52% :     0.000003s :     15: predicate.depend_value_elim
 1.01% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.09% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.27% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.99% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000003s :     15: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.00% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.50% :     0.000003s :     15: predicate.environ_get_eliminate
 1.66% :     0.000003s :     15: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      5: predicate.fold_const_symbol
 1.24% :     0.000003s :     10: predicate.get_grad_eliminate
 0.37% :     0.000001s :      5: predicate.graph_param_transform
 5.27% :     0.000011s :     31: predicate.inline
 0.85% :     0.000002s :     10: predicate.inline_without_move
 0.34% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.43% :     0.000003s :     10: predicate.less_batch_normalization
 1.61% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.84% :     0.000004s :     22: predicate.load_eliminater
 2.84% :     0.000006s :      5: predicate.loop_unroll_after_grad
 3.25% :     0.000007s :     42: predicate.loop_unroll_before_grad
 2.34% :     0.000005s :     20: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000002s :     15: predicate.merge_addn
 1.02% :     0.000002s :     15: predicate.minmaximum_grad
 2.15% :     0.000004s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.00% :     0.000004s :     22: predicate.partial_eliminate
 1.18% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.61% :     0.000003s :     15: predicate.reduce_eliminate
 1.50% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.53% :     0.000003s :     27: predicate.replace_applicator
 0.64% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000003s :     15: predicate.reshape_eliminate
 1.16% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.16% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.58% :     0.000003s :     15: predicate.same_eliminate
 0.50% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.39% :     0.000003s :     10: predicate.special_op_eliminate
 1.06% :     0.000002s :     10: predicate.specialize_transform
 1.36% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.65% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.22% :     0.000005s :     21: predicate.switch_defer_inline
 1.58% :     0.000003s :     21: predicate.switch_layer_defer_inline
 5.84% :     0.000012s :     68: predicate.switch_simplify
 1.33% :     0.000003s :     15: predicate.tile_eliminate
 1.40% :     0.000003s :     15: predicate.transpose_eliminate
 1.32% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.52% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 2.17% :     0.000004s :     15: predicate.value_based_eliminate
 0.59% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001050    13
45.82% :     0.000481s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.55% :     0.000069s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.63% :     0.000500s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.062795    76
 0.16% :     0.000100s :      1: add_recomputation
 0.31% :     0.000196s :      1: auto_monad
 0.06% :     0.000037s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 1.00% :     0.000629s :      1: bootstrap
 0.08% :     0.000049s :      1: cconv
 0.04% :     0.000024s :      1: convert_after_rewriter
 0.08% :     0.000048s :      1: cse_after_recomputation
 0.03% :     0.000021s :      1: environ_conv
 0.05% :     0.000029s :      1: event_method
 0.03% :     0.000018s :      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
13.86% :     0.008701s :      1: jit_opt_a
 0.46% :     0.000290s :      1: jit_opt_after_cconv
 0.15% :     0.000094s :      1: jit_opt_b
 4.86% :     0.003055s :      1: loop_unroll
 1.99% :     0.001248s :      1: mutable_eliminate
 2.24% :     0.001409s :     26: opt.transform.jit_opt_a
 0.18% :     0.000113s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000053s :      4: opt.transform.jit_opt_b
 0.06% :     0.000035s :      1: opt.transform.loop_unroll_optimizer
 0.15% :     0.000094s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000047s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000060s :      4: opt.transform.symbol_engine_opt
 1.30% :     0.000817s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000006s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000151s :      1: py_interpret_to_execute
 0.04% :     0.000025s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000028s :      1: remove_dup_value
 1.50% :     0.000942s :      1: renormalize.infer
 0.91% :     0.000574s :      1: renormalize.specialize
 0.02% :     0.000015s :      1: rewriter_after_jit_bprop_graph
 0.57% :     0.000356s :      1: rewriter_after_opt_a
 0.18% :     0.000110s :      1: rewriter_before_opt_a
 0.21% :     0.000133s :      1: symbol_engine_optimizer
22.66% :     0.014228s :      1: task_emit
46.02% :     0.028895s :      1: type_inference
 0.20% :     0.000123s :      1: validate


TotalTime = 0.078087, [33]
[bootstrap]: 0.00049281
[type_inference]: 0.0426649
[event_method]: 2.618e-05
[auto_monad]: 0.00019857
[graph_reusing]: 7.85e-06
[pre_auto_parallel]: 3.4e-06
[py_interpret_to_execute]: 0.00016226
[rewriter_before_opt_a]: 0.0001169
[expand_dump_flag]: 3.81001e-06
[jit_opt_a]: 0.0157272, [2]
    [Cycle 1]: 0.00993065, [27]
        [switch_simplify]: 7.614e-05
        [loop_unroll]: 3.786e-05
        [a_1]: 0.006352
        [with_stream_mark]: 2.986e-05
        [recompute_prepare]: 1.522e-05
        [updatestate_depend_eliminate]: 8.32e-06
        [updatestate_assign_eliminate]: 6.76e-06
        [updatestate_loads_eliminate]: 4.62998e-06
        [parameter_eliminate]: 2.41998e-06
        [specialize_transform]: 1.099e-05
        [updatestate_useless_node_eliminater]: 1.223e-05
        [accelerated_algorithm]: 9.44e-06
        [meta_shard_fg_expand]: 3.93001e-06
        [get_grad_eliminate_]: 8.72e-06
        [merge_forward]: 5.45001e-06
        [cell_reuse_recompute_pass]: 2.61e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.065e-05
        [j_node_and_user_rematch]: 1.425e-05
        [meta_fg_expand]: 4.35e-06
        [replace_old_param]: 1.354e-05
        [inline_without_move]: 8.35001e-06
        [renormalize]: 0.0029007
        [add_forward_monad_depend]: 1.148e-05
        [auto_monad_grad]: 3.37997e-06
        [auto_monad_eliminator]: 2.865e-05
        [cse]: 4.56e-05
        [replace_applicator]: 2.583e-05
    [Cycle 2]: 0.00051344, [27]
        [switch_simplify]: 1.011e-05
        [loop_unroll]: 8.63001e-06
        [a_1]: 0.00018671
        [with_stream_mark]: 1.691e-05
        [recompute_prepare]: 8.48001e-06
        [updatestate_depend_eliminate]: 5.57999e-06
        [updatestate_assign_eliminate]: 6.04001e-06
        [updatestate_loads_eliminate]: 4.43001e-06
        [parameter_eliminate]: 1.62001e-06
        [specialize_transform]: 8.43001e-06
        [updatestate_useless_node_eliminater]: 1.142e-05
        [accelerated_algorithm]: 8.32e-06
        [meta_shard_fg_expand]: 2.90998e-06
        [get_grad_eliminate_]: 7.75e-06
        [merge_forward]: 6.09999e-06
        [cell_reuse_recompute_pass]: 3.08e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.905e-05
        [j_node_and_user_rematch]: 1.335e-05
        [meta_fg_expand]: 3.53e-06
        [replace_old_param]: 1.186e-05
        [inline_without_move]: 7.65e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 1.49998e-06
        [auto_monad_eliminator]: 1.128e-05
        [cse]: 1.991e-05
        [replace_applicator]: 8.22e-06
[py_interpret_to_execute_after_opt_a]: 1.906e-05
[rewriter_after_opt_a]: 0.00032345
[convert_after_rewriter]: 1.231e-05
[order_py_execute_after_rewriter]: 7.7e-06
[mutable_eliminate]: 0.00110322
[jit_opt_b]: 7.529e-05, [1]
    [Cycle 1]: 6.541e-05, [2]
        [frontend_op_eliminate]: 2.681e-05
        [inline_after_opt_a]: 2.628e-05
[cconv]: 3.181e-05
[loop_unroll]: 0.0005229
[jit_opt_after_cconv]: 0.00022198, [1]
    [Cycle 1]: 0.00021503, [11]
        [c_1]: 4.894e-05
        [parameter_eliminate]: 5.13002e-06
        [updatestate_depend_eliminate]: 1.063e-05
        [updatestate_assign_eliminate]: 4.58999e-06
        [updatestate_loads_eliminate]: 4.34002e-06
        [cse]: 4.074e-05
        [call_graph_tuple_transform]: 2.585e-05
        [tuple_list_get_item_eliminator]: 9.21002e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 1.02998e-06
        [switch_simplify]: 9.54e-06
[remove_dup_value]: 2.025e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 1.406e-05
[add_recomputation]: 7.65e-05
[cse_after_recomputation]: 3.079e-05, [1]
    [Cycle 1]: 2.363e-05, [1]
        [cse]: 1.762e-05
[auto_monad_reorder]: 2.75e-05
[get_jit_bprop_graph]: 2.42001e-06
[rewriter_after_jit_bprop_graph]: 8.98002e-06
[opt_after_jit_grad]: 0.00097495
[symbol_engine_optimizer]: 0.00011668, [1]
    [Cycle 1]: 0.00010975, [6]
        [build]: 1.822e-05
        [elim_shapecalc]: 1.323e-05
        [elim_not_effective]: 2.136e-05
        [opt_reshape]: 9.20999e-06
        [fold_const_symbol]: 1.523e-05
        [renormalize]: 6.50005e-07
[validate]: 6.99e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.0147167
[execute]: 8.72998e-06
Sums
    bootstrap                            :     0.000493s :  0.68%
    type_inference                       :     0.042665s : 59.27%
    event_method                         :     0.000026s :  0.04%
    auto_monad                           :     0.000199s :  0.28%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000162s :  0.23%
    rewriter_before_opt_a                :     0.000117s :  0.16%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000086s :  0.12%
    jit_opt_a.loop_unroll                :     0.000046s :  0.06%
    jit_opt_a.a_1                        :     0.006539s :  9.08%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.02%
    jit_opt_a.merge_forward              :     0.000012s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000008s :  0.01%
    jit_opt_a.replace_old_param          :     0.000025s :  0.04%
    jit_opt_a.inline_without_move        :     0.000016s :  0.02%
    jit_opt_a.renormalize                :     0.002901s :  4.03%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000040s :  0.06%
    jit_opt_a.cse                        :     0.000066s :  0.09%
    jit_opt_a.replace_applicator         :     0.000034s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.03%
    rewriter_after_opt_a                 :     0.000323s :  0.45%
    convert_after_rewriter               :     0.000012s :  0.02%
    order_py_execute_after_rewriter      :     0.000008s :  0.01%
    mutable_eliminate                    :     0.001103s :  1.53%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.04%
    cconv                                :     0.000032s :  0.04%
    loop_unroll                          :     0.000523s :  0.73%
    jit_opt_after_cconv.c_1              :     0.000049s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.01%
    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              :     0.000041s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.01%
    remove_dup_value                     :     0.000020s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000014s :  0.02%
    add_recomputation                    :     0.000077s :  0.11%
    cse_after_recomputation.cse          :     0.000018s :  0.02%
    auto_monad_reorder                   :     0.000028s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.01%
    opt_after_jit_grad                   :     0.000975s :  1.35%
    symbol_engine_optimizer.build        :     0.000018s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000070s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.014717s : 20.44%
    execute                              :     0.000009s :  0.01%


Time group info:
------[substitution.]   0.000410    47
 3.31% :     0.000014s :      2: substitution.depend_value_elim
 0.81% :     0.000003s :      4: substitution.elim_not_effective
 0.56% :     0.000002s :      4: substitution.fold_const_symbol
 1.88% :     0.000008s :      5: substitution.graph_param_transform
78.30% :     0.000321s :      4: substitution.inline
 1.28% :     0.000005s :      8: substitution.j_node_and_user_rematch
 1.88% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.58% :     0.000006s :      2: substitution.replace_old_param
 5.05% :     0.000021s :      2: substitution.tuple_list_get_item_eliminator
 2.87% :     0.000012s :      3: substitution.updatestate_pure_node_eliminater
 2.49% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.042553     2
95.27% :     0.040539s :      1: type_inference.infer
 4.73% :     0.002014s :      1: type_inference.specialize
------[replace.]   0.000184     6
73.22% :     0.000135s :      4: replace.inline
26.78% :     0.000049s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000337     6
94.21% :     0.000317s :      4: match.inline
 5.79% :     0.000019s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000221  1021
 1.42% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.37% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.00% :     0.000002s :     15: predicate.addn_check_dump
 1.59% :     0.000004s :     15: predicate.addn_zero_filter
 2.44% :     0.000005s :     15: predicate.arithmetic_simplify
 2.00% :     0.000004s :     15: predicate.cast_eliminate
 0.47% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.32% :     0.000003s :     15: predicate.compare_switch_simplify
 1.47% :     0.000003s :     15: predicate.depend_value_elim
 1.51% :     0.000003s :     15: predicate.dict_get_item_const_eliminator
 1.34% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.29% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.86% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.25% :     0.000001s :      5: predicate.elim_not_effective
 0.76% :     0.000002s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.56% :     0.000003s :     15: predicate.environ_add_const_eliminate
 0.97% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.14% :     0.000003s :     15: predicate.environ_get_depend_swap
 1.52% :     0.000003s :     15: predicate.environ_get_eliminate
 0.96% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000001s :      5: predicate.graph_param_transform
 5.19% :     0.000011s :     31: predicate.inline
 0.76% :     0.000002s :     10: predicate.inline_without_move
 0.33% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.45% :     0.000003s :     10: predicate.less_batch_normalization
 1.79% :     0.000004s :     17: predicate.list_to_tuple_eliminator_
 1.71% :     0.000004s :     22: predicate.load_eliminater
 1.10% :     0.000002s :      5: predicate.loop_unroll_after_grad
 2.86% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.78% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000002s :     15: predicate.merge_addn
 1.05% :     0.000002s :     15: predicate.minmaximum_grad
 1.61% :     0.000004s :      5: predicate.mutable_eliminate
 0.39% :     0.000001s :      5: predicate.opt_reshape
 1.97% :     0.000004s :     22: predicate.partial_eliminate
 1.35% :     0.000003s :     15: predicate.print_const_string_wrapper
 2.11% :     0.000005s :     15: predicate.reduce_eliminate
 1.77% :     0.000004s :     17: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.52% :     0.000003s :     27: predicate.replace_applicator
 0.67% :     0.000001s :     10: predicate.replace_old_param
 0.30% :     0.000001s :      5: predicate.reset_defer_inline
 1.62% :     0.000004s :     15: predicate.reshape_eliminate
 1.48% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.24% :     0.000003s :     15: predicate.same_eliminate
 0.47% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.01% :     0.000002s :     10: predicate.special_op_eliminate
 1.05% :     0.000002s :     10: predicate.specialize_transform
 1.87% :     0.000004s :     15: predicate.split_environ_get_set_with_tuple_value
 1.04% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.47% :     0.000005s :     21: predicate.switch_defer_inline
 2.12% :     0.000005s :     21: predicate.switch_layer_defer_inline
 5.79% :     0.000013s :     68: predicate.switch_simplify
 1.26% :     0.000003s :     15: predicate.tile_eliminate
 1.25% :     0.000003s :     15: predicate.transpose_eliminate
 1.28% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 4.09% :     0.000009s :     27: predicate.tuple_list_get_item_eliminator
 2.14% :     0.000005s :     15: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000004s :     17: predicate.tuple_to_list_eliminator_
 1.63% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000004s :     15: predicate.value_based_eliminate
 0.33% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.84% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001588    13
61.00% :     0.000969s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.65% :     0.000074s :      1: func_graph_cloner_run.FuncGraphClonerNode
34.36% :     0.000546s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.088054    76
 0.09% :     0.000080s :      1: add_recomputation
 0.23% :     0.000206s :      1: auto_monad
 0.03% :     0.000030s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.59% :     0.000516s :      1: bootstrap
 0.04% :     0.000035s :      1: cconv
 0.02% :     0.000015s :      1: convert_after_rewriter
 0.04% :     0.000033s :      1: cse_after_recomputation
 0.02% :     0.000016s :      1: environ_conv
 0.04% :     0.000031s :      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% :     0.000010s :      1: graph_reusing
17.87% :     0.015731s :      1: jit_opt_a
 0.26% :     0.000225s :      1: jit_opt_after_cconv
 0.09% :     0.000079s :      1: jit_opt_b
 0.60% :     0.000532s :      1: loop_unroll
 1.27% :     0.001116s :      1: mutable_eliminate
 7.81% :     0.006875s :     26: opt.transform.jit_opt_a
 0.10% :     0.000089s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000045s :      4: opt.transform.jit_opt_b
 0.02% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000038s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000055s :      4: opt.transform.symbol_engine_opt
 1.12% :     0.000986s :      1: opt_after_jit_grad
 0.01% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.19% :     0.000169s :      1: py_interpret_to_execute
 0.02% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000023s :      1: remove_dup_value
 2.61% :     0.002298s :      1: renormalize.infer
 0.67% :     0.000590s :      1: renormalize.specialize
 0.01% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.37% :     0.000328s :      1: rewriter_after_opt_a
 0.14% :     0.000123s :      1: rewriter_before_opt_a
 0.14% :     0.000119s :      1: symbol_engine_optimizer
16.74% :     0.014744s :      1: task_emit
48.48% :     0.042690s :      1: type_inference
 0.11% :     0.000101s :      1: validate


TotalTime = 0.0813437, [33]
[bootstrap]: 0.00034909
[type_inference]: 0.0356299
[event_method]: 2.095e-05
[auto_monad]: 0.00016389
[graph_reusing]: 6.95002e-06
[pre_auto_parallel]: 3.70998e-06
[py_interpret_to_execute]: 0.00014787
[rewriter_before_opt_a]: 9.793e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.00864049, [2]
    [Cycle 1]: 0.00264058, [27]
        [switch_simplify]: 7.553e-05
        [loop_unroll]: 3.693e-05
        [a_1]: 0.00078322
        [with_stream_mark]: 2.562e-05
        [recompute_prepare]: 1.251e-05
        [updatestate_depend_eliminate]: 6.96001e-06
        [updatestate_assign_eliminate]: 6.81001e-06
        [updatestate_loads_eliminate]: 5.01002e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 1.029e-05
        [updatestate_useless_node_eliminater]: 1.153e-05
        [accelerated_algorithm]: 8.91997e-06
        [meta_shard_fg_expand]: 2.69999e-06
        [get_grad_eliminate_]: 9.69999e-06
        [merge_forward]: 5.48002e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.125e-05
        [j_node_and_user_rematch]: 1.442e-05
        [meta_fg_expand]: 4.4e-06
        [replace_old_param]: 1.242e-05
        [inline_without_move]: 8.02e-06
        [renormalize]: 0.0012677
        [add_forward_monad_depend]: 7.76001e-06
        [auto_monad_grad]: 3.26001e-06
        [auto_monad_eliminator]: 2.226e-05
        [cse]: 4.559e-05
        [replace_applicator]: 2.043e-05
    [Cycle 2]: 0.00050967, [27]
        [switch_simplify]: 8.93002e-06
        [loop_unroll]: 7.91001e-06
        [a_1]: 0.00017251
        [with_stream_mark]: 1.545e-05
        [recompute_prepare]: 8.87e-06
        [updatestate_depend_eliminate]: 5.39998e-06
        [updatestate_assign_eliminate]: 5.32001e-06
        [updatestate_loads_eliminate]: 4.03999e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 8.3e-06
        [updatestate_useless_node_eliminater]: 1.122e-05
        [accelerated_algorithm]: 8.03999e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 7.92e-06
        [merge_forward]: 5.16002e-06
        [cell_reuse_recompute_pass]: 3.06001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.858e-05
        [j_node_and_user_rematch]: 1.286e-05
        [meta_fg_expand]: 2.83998e-06
        [replace_old_param]: 2.686e-05
        [inline_without_move]: 8.17998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 2.14e-06
        [auto_monad_grad]: 1.01002e-06
        [auto_monad_eliminator]: 1.121e-05
        [cse]: 2.224e-05
        [replace_applicator]: 8.22e-06
[py_interpret_to_execute_after_opt_a]: 1.676e-05
[rewriter_after_opt_a]: 0.00029995
[convert_after_rewriter]: 1.17e-05
[order_py_execute_after_rewriter]: 8.2e-06
[mutable_eliminate]: 0.00082448
[jit_opt_b]: 7.381e-05, [1]
    [Cycle 1]: 6.485e-05, [2]
        [frontend_op_eliminate]: 2.636e-05
        [inline_after_opt_a]: 2.548e-05
[cconv]: 3.104e-05
[loop_unroll]: 0.00050024
[jit_opt_after_cconv]: 0.00021338, [1]
    [Cycle 1]: 0.00020633, [11]
        [c_1]: 4.832e-05
        [parameter_eliminate]: 3.2e-06
        [updatestate_depend_eliminate]: 8.85999e-06
        [updatestate_assign_eliminate]: 4.77998e-06
        [updatestate_loads_eliminate]: 3.9e-06
        [cse]: 3.779e-05
        [call_graph_tuple_transform]: 2.384e-05
        [tuple_list_get_item_eliminator]: 8.65999e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 1.129e-05
[remove_dup_value]: 2.108e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 1.111e-05
[add_recomputation]: 7.411e-05
[cse_after_recomputation]: 3.191e-05, [1]
    [Cycle 1]: 2.461e-05, [1]
        [cse]: 1.812e-05
[auto_monad_reorder]: 2.95e-05
[get_jit_bprop_graph]: 2.03002e-06
[rewriter_after_jit_bprop_graph]: 8.53001e-06
[opt_after_jit_grad]: 0.00054481
[symbol_engine_optimizer]: 0.00134169, [1]
    [Cycle 1]: 0.00133163, [6]
        [build]: 0.00056471
        [elim_shapecalc]: 2.155e-05
        [elim_not_effective]: 3.246e-05
        [opt_reshape]: 0.00018694
        [fold_const_symbol]: 0.00024884
        [renormalize]: 8.50006e-07
[validate]: 0.00026256
[backend_pass]: 3.93001e-06
[task_emit]: 0.0313289
[execute]: 8.69998e-06
Sums
    bootstrap                            :     0.000349s :  0.47%
    type_inference                       :     0.035630s : 47.82%
    event_method                         :     0.000021s :  0.03%
    auto_monad                           :     0.000164s :  0.22%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000004s :  0.00%
    py_interpret_to_execute              :     0.000148s :  0.20%
    rewriter_before_opt_a                :     0.000098s :  0.13%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000084s :  0.11%
    jit_opt_a.loop_unroll                :     0.000045s :  0.06%
    jit_opt_a.a_1                        :     0.000956s :  1.28%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.03%
    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 :     0.000040s :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000039s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.02%
    jit_opt_a.renormalize                :     0.001268s :  1.70%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000033s :  0.04%
    jit_opt_a.cse                        :     0.000068s :  0.09%
    jit_opt_a.replace_applicator         :     0.000029s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.02%
    rewriter_after_opt_a                 :     0.000300s :  0.40%
    convert_after_rewriter               :     0.000012s :  0.02%
    order_py_execute_after_rewriter      :     0.000008s :  0.01%
    mutable_eliminate                    :     0.000824s :  1.11%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.03%
    cconv                                :     0.000031s :  0.04%
    loop_unroll                          :     0.000500s :  0.67%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    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              :     0.000038s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000011s :  0.02%
    remove_dup_value                     :     0.000021s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000011s :  0.01%
    add_recomputation                    :     0.000074s :  0.10%
    cse_after_recomputation.cse          :     0.000018s :  0.02%
    auto_monad_reorder                   :     0.000030s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.01%
    opt_after_jit_grad                   :     0.000545s :  0.73%
    symbol_engine_optimizer.build        :     0.000565s :  0.76%
    symbol_engine_optimizer.elim_shapecalc :     0.000022s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000032s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000187s :  0.25%
    symbol_engine_optimizer.fold_const_symbol :     0.000249s :  0.33%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000263s :  0.35%
    backend_pass                         :     0.000004s :  0.01%
    task_emit                            :     0.031329s : 42.04%
    execute                              :     0.000009s :  0.01%


Time group info:
------[substitution.]   0.000337    47
 3.45% :     0.000012s :      2: substitution.depend_value_elim
 0.97% :     0.000003s :      4: substitution.elim_not_effective
 1.07% :     0.000004s :      4: substitution.fold_const_symbol
 2.07% :     0.000007s :      5: substitution.graph_param_transform
76.56% :     0.000258s :      4: substitution.inline
 1.29% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.51% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.60% :     0.000005s :      2: substitution.replace_old_param
 4.24% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.19% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.04% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.035535     2
93.14% :     0.033097s :      1: type_inference.infer
 6.86% :     0.002438s :      1: type_inference.specialize
------[replace.]   0.000099     6
65.97% :     0.000065s :      4: replace.inline
34.03% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000268     6
95.10% :     0.000255s :      4: match.inline
 4.90% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000187  1021
 1.23% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     15: predicate.addn_check_dump
 1.43% :     0.000003s :     15: predicate.addn_zero_filter
 2.07% :     0.000004s :     15: predicate.arithmetic_simplify
 1.18% :     0.000002s :     15: predicate.cast_eliminate
 0.61% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     15: predicate.compare_switch_simplify
 1.36% :     0.000003s :     15: predicate.depend_value_elim
 1.09% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.54% :     0.000001s :      5: predicate.elim_not_effective
 1.30% :     0.000002s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.03% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     15: predicate.environ_get_eliminate
 1.07% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.75% :     0.000001s :      5: predicate.fold_const_symbol
 1.27% :     0.000002s :     10: predicate.get_grad_eliminate
 0.26% :     0.000000s :      5: predicate.graph_param_transform
 4.90% :     0.000009s :     31: predicate.inline
 1.04% :     0.000002s :     10: predicate.inline_without_move
 0.40% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.42% :     0.000003s :     10: predicate.less_batch_normalization
 1.61% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.99% :     0.000004s :     22: predicate.load_eliminater
 1.16% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.37% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.05% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     15: predicate.merge_addn
 1.05% :     0.000002s :     15: predicate.minmaximum_grad
 1.61% :     0.000003s :      5: predicate.mutable_eliminate
 1.54% :     0.000003s :      5: predicate.opt_reshape
 2.27% :     0.000004s :     22: predicate.partial_eliminate
 1.28% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.71% :     0.000003s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.69% :     0.000003s :     27: predicate.replace_applicator
 0.71% :     0.000001s :     10: predicate.replace_old_param
 0.44% :     0.000001s :      5: predicate.reset_defer_inline
 1.21% :     0.000002s :     15: predicate.reshape_eliminate
 1.45% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.87% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.36% :     0.000003s :     15: predicate.same_eliminate
 0.60% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.94% :     0.000002s :     10: predicate.special_op_eliminate
 0.98% :     0.000002s :     10: predicate.specialize_transform
 1.55% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.04% :     0.000004s :     21: predicate.switch_defer_inline
 1.84% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.49% :     0.000012s :     68: predicate.switch_simplify
 1.20% :     0.000002s :     15: predicate.tile_eliminate
 1.35% :     0.000003s :     15: predicate.transpose_eliminate
 1.51% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.33% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.67% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.84% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000938    13
43.28% :     0.000406s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.19% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.53% :     0.000474s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.084343    76
 0.09% :     0.000077s :      1: add_recomputation
 0.20% :     0.000170s :      1: auto_monad
 0.04% :     0.000033s :      1: auto_monad_reorder
 0.02% :     0.000014s :      1: backend_pass
 0.44% :     0.000369s :      1: bootstrap
 0.04% :     0.000034s :      1: cconv
 0.02% :     0.000015s :      1: convert_after_rewriter
 0.04% :     0.000034s :      1: cse_after_recomputation
 0.02% :     0.000013s :      1: environ_conv
 0.03% :     0.000026s :      1: event_method
 0.02% :     0.000013s :      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
10.25% :     0.008645s :      1: jit_opt_a
 0.26% :     0.000216s :      1: jit_opt_after_cconv
 0.09% :     0.000077s :      1: jit_opt_b
 0.60% :     0.000509s :      1: loop_unroll
 0.99% :     0.000836s :      1: mutable_eliminate
 1.54% :     0.001301s :     26: opt.transform.jit_opt_a
 0.10% :     0.000088s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000044s :      4: opt.transform.jit_opt_b
 0.02% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.57% :     0.000480s :      4: opt.transform.symbol_engine_opt
 0.66% :     0.000554s :      1: opt_after_jit_grad
 0.01% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.18% :     0.000153s :      1: py_interpret_to_execute
 0.02% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000024s :      1: remove_dup_value
 0.90% :     0.000762s :      1: renormalize.infer
 0.59% :     0.000495s :      1: renormalize.specialize
 0.01% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.36% :     0.000306s :      1: rewriter_after_opt_a
 0.12% :     0.000102s :      1: rewriter_before_opt_a
 1.60% :     0.001348s :      1: symbol_engine_optimizer
37.16% :     0.031343s :      1: task_emit
42.27% :     0.035649s :      1: type_inference
 0.56% :     0.000473s :      1: validate


TotalTime = 0.0527203, [33]
[bootstrap]: 0.00040448
[type_inference]: 0.0285479
[event_method]: 2.283e-05
[auto_monad]: 0.00018393
[graph_reusing]: 7.01999e-06
[pre_auto_parallel]: 3.06001e-06
[py_interpret_to_execute]: 0.00014962
[rewriter_before_opt_a]: 0.00010352
[expand_dump_flag]: 3.83999e-06
[jit_opt_a]: 0.0084755, [2]
    [Cycle 1]: 0.00295549, [27]
        [switch_simplify]: 7.633e-05
        [loop_unroll]: 3.81e-05
        [a_1]: 0.00083955
        [with_stream_mark]: 2.909e-05
        [recompute_prepare]: 1.242e-05
        [updatestate_depend_eliminate]: 6.45997e-06
        [updatestate_assign_eliminate]: 6.47001e-06
        [updatestate_loads_eliminate]: 5.24e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 1.046e-05
        [updatestate_useless_node_eliminater]: 1.292e-05
        [accelerated_algorithm]: 8.75999e-06
        [meta_shard_fg_expand]: 3.33e-06
        [get_grad_eliminate_]: 8.12998e-06
        [merge_forward]: 6.34999e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.084e-05
        [j_node_and_user_rematch]: 1.429e-05
        [meta_fg_expand]: 3.83999e-06
        [replace_old_param]: 1.352e-05
        [inline_without_move]: 8.33001e-06
        [renormalize]: 0.00149296
        [add_forward_monad_depend]: 9.37001e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 2.507e-05
        [cse]: 4.563e-05
        [replace_applicator]: 2.41e-05
    [Cycle 2]: 0.00050547, [27]
        [switch_simplify]: 9.58002e-06
        [loop_unroll]: 9.29e-06
        [a_1]: 0.00018215
        [with_stream_mark]: 1.69e-05
        [recompute_prepare]: 7.93999e-06
        [updatestate_depend_eliminate]: 5.15001e-06
        [updatestate_assign_eliminate]: 5.55001e-06
        [updatestate_loads_eliminate]: 4.92e-06
        [parameter_eliminate]: 2.29999e-06
        [specialize_transform]: 7.97998e-06
        [updatestate_useless_node_eliminater]: 1.225e-05
        [accelerated_algorithm]: 7.66001e-06
        [meta_shard_fg_expand]: 2.10002e-06
        [get_grad_eliminate_]: 7.48e-06
        [merge_forward]: 4.77998e-06
        [cell_reuse_recompute_pass]: 2.78e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.725e-05
        [j_node_and_user_rematch]: 1.371e-05
        [meta_fg_expand]: 3.26999e-06
        [replace_old_param]: 1.168e-05
        [inline_without_move]: 7.53999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 1.02998e-06
        [auto_monad_eliminator]: 1.165e-05
        [cse]: 2.239e-05
        [replace_applicator]: 8.27e-06
[py_interpret_to_execute_after_opt_a]: 1.901e-05
[rewriter_after_opt_a]: 0.0003156
[convert_after_rewriter]: 1.407e-05
[order_py_execute_after_rewriter]: 8.37e-06
[mutable_eliminate]: 0.00091711
[jit_opt_b]: 7.574e-05, [1]
    [Cycle 1]: 6.567e-05, [2]
        [frontend_op_eliminate]: 2.602e-05
        [inline_after_opt_a]: 2.689e-05
[cconv]: 4.184e-05
[loop_unroll]: 0.0005659
[jit_opt_after_cconv]: 0.00024643, [1]
    [Cycle 1]: 0.00023802, [11]
        [c_1]: 5.768e-05
        [parameter_eliminate]: 6.36e-06
        [updatestate_depend_eliminate]: 1.376e-05
        [updatestate_assign_eliminate]: 5.70001e-06
        [updatestate_loads_eliminate]: 4.74e-06
        [cse]: 4.527e-05
        [call_graph_tuple_transform]: 2.495e-05
        [tuple_list_get_item_eliminator]: 8.54998e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 1.106e-05
[remove_dup_value]: 2.144e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 1.351e-05
[add_recomputation]: 7.958e-05
[cse_after_recomputation]: 3.537e-05, [1]
    [Cycle 1]: 2.878e-05, [1]
        [cse]: 2.03e-05
[auto_monad_reorder]: 2.836e-05
[get_jit_bprop_graph]: 2.34001e-06
[rewriter_after_jit_bprop_graph]: 1.183e-05
[opt_after_jit_grad]: 0.00070094
[symbol_engine_optimizer]: 0.00011272, [1]
    [Cycle 1]: 0.00010497, [6]
        [build]: 1.826e-05
        [elim_shapecalc]: 1.221e-05
        [elim_not_effective]: 2.066e-05
        [opt_reshape]: 9.15999e-06
        [fold_const_symbol]: 1.44e-05
        [renormalize]: 6.29982e-07
[validate]: 7.084e-05
[backend_pass]: 1.29e-06
[task_emit]: 0.0112449
[execute]: 1.04e-05
Sums
    bootstrap                            :     0.000404s :  0.86%
    type_inference                       :     0.028548s : 60.82%
    event_method                         :     0.000023s :  0.05%
    auto_monad                           :     0.000184s :  0.39%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000150s :  0.32%
    rewriter_before_opt_a                :     0.000104s :  0.22%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000086s :  0.18%
    jit_opt_a.loop_unroll                :     0.000047s :  0.10%
    jit_opt_a.a_1                        :     0.001022s :  2.18%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.03%
    jit_opt_a.merge_forward              :     0.000011s :  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 :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.03%
    jit_opt_a.renormalize                :     0.001493s :  3.18%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000037s :  0.08%
    jit_opt_a.cse                        :     0.000068s :  0.14%
    jit_opt_a.replace_applicator         :     0.000032s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.04%
    rewriter_after_opt_a                 :     0.000316s :  0.67%
    convert_after_rewriter               :     0.000014s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000917s :  1.95%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.06%
    cconv                                :     0.000042s :  0.09%
    loop_unroll                          :     0.000566s :  1.21%
    jit_opt_after_cconv.c_1              :     0.000058s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000014s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000045s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000011s :  0.02%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000014s :  0.03%
    add_recomputation                    :     0.000080s :  0.17%
    cse_after_recomputation.cse          :     0.000020s :  0.04%
    auto_monad_reorder                   :     0.000028s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.03%
    opt_after_jit_grad                   :     0.000701s :  1.49%
    symbol_engine_optimizer.build        :     0.000018s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000071s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011245s : 23.96%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000375    47
 3.02% :     0.000011s :      2: substitution.depend_value_elim
 0.86% :     0.000003s :      4: substitution.elim_not_effective
 0.54% :     0.000002s :      4: substitution.fold_const_symbol
 2.01% :     0.000008s :      5: substitution.graph_param_transform
77.85% :     0.000292s :      4: substitution.inline
 1.37% :     0.000005s :      8: substitution.j_node_and_user_rematch
 1.88% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.67% :     0.000006s :      2: substitution.replace_old_param
 4.06% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.68% :     0.000014s :      3: substitution.updatestate_pure_node_eliminater
 3.07% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.028439     2
94.14% :     0.026773s :      1: type_inference.infer
 5.86% :     0.001667s :      1: type_inference.specialize
------[replace.]   0.000106     6
66.46% :     0.000070s :      4: replace.inline
33.54% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000302     6
95.45% :     0.000288s :      4: match.inline
 4.55% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000197  1021
 1.82% :     0.000004s :     15: predicate.accumulaten_eliminater
 1.27% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.06% :     0.000002s :     15: predicate.addn_check_dump
 1.17% :     0.000002s :     15: predicate.addn_zero_filter
 2.26% :     0.000004s :     15: predicate.arithmetic_simplify
 1.30% :     0.000003s :     15: predicate.cast_eliminate
 0.60% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.39% :     0.000003s :     15: predicate.compare_switch_simplify
 1.21% :     0.000002s :     15: predicate.depend_value_elim
 1.00% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.93% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.30% :     0.000001s :      5: predicate.elim_not_effective
 0.70% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.27% :     0.000003s :     15: predicate.environ_get_add_eliminate
 0.97% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     15: predicate.environ_get_eliminate
 1.01% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 0.91% :     0.000002s :     10: predicate.get_grad_eliminate
 0.34% :     0.000001s :      5: predicate.graph_param_transform
 4.81% :     0.000009s :     31: predicate.inline
 0.91% :     0.000002s :     10: predicate.inline_without_move
 0.37% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :     10: predicate.less_batch_normalization
 1.54% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.55% :     0.000005s :     22: predicate.load_eliminater
 1.77% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.18% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.89% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000002s :     15: predicate.merge_addn
 1.05% :     0.000002s :     15: predicate.minmaximum_grad
 2.34% :     0.000005s :      5: predicate.mutable_eliminate
 0.66% :     0.000001s :      5: predicate.opt_reshape
 2.06% :     0.000004s :     22: predicate.partial_eliminate
 1.38% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.80% :     0.000004s :     15: predicate.reduce_eliminate
 1.86% :     0.000004s :     17: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.67% :     0.000003s :     27: predicate.replace_applicator
 0.72% :     0.000001s :     10: predicate.replace_old_param
 0.41% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000002s :     15: predicate.reshape_eliminate
 1.47% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 1.02% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.96% :     0.000004s :     15: predicate.same_eliminate
 0.51% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.07% :     0.000002s :     10: predicate.special_op_eliminate
 0.97% :     0.000002s :     10: predicate.specialize_transform
 1.77% :     0.000004s :     15: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.93% :     0.000004s :     21: predicate.switch_defer_inline
 1.68% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.24% :     0.000012s :     68: predicate.switch_simplify
 1.35% :     0.000003s :     15: predicate.tile_eliminate
 1.35% :     0.000003s :     15: predicate.transpose_eliminate
 1.26% :     0.000002s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 4.26% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 2.01% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.19% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.62% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.93% :     0.000004s :     15: predicate.value_based_eliminate
 0.58% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001026    13
44.70% :     0.000459s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.88% :     0.000071s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.42% :     0.000497s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.055781    76
 0.15% :     0.000083s :      1: add_recomputation
 0.34% :     0.000190s :      1: auto_monad
 0.06% :     0.000032s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.76% :     0.000423s :      1: bootstrap
 0.08% :     0.000044s :      1: cconv
 0.03% :     0.000018s :      1: convert_after_rewriter
 0.07% :     0.000038s :      1: cse_after_recomputation
 0.03% :     0.000016s :      1: environ_conv
 0.05% :     0.000028s :      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
15.20% :     0.008480s :      1: jit_opt_a
 0.45% :     0.000249s :      1: jit_opt_after_cconv
 0.14% :     0.000079s :      1: jit_opt_b
 1.04% :     0.000579s :      1: loop_unroll
 1.67% :     0.000931s :      1: mutable_eliminate
 2.43% :     0.001355s :     26: opt.transform.jit_opt_a
 0.18% :     0.000098s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000045s :      4: opt.transform.jit_opt_b
 0.04% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000032s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000036s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.28% :     0.000713s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000154s :      1: py_interpret_to_execute
 0.04% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000025s :      1: remove_dup_value
 1.54% :     0.000862s :      1: renormalize.infer
 1.11% :     0.000621s :      1: renormalize.specialize
 0.03% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.58% :     0.000322s :      1: rewriter_after_opt_a
 0.19% :     0.000108s :      1: rewriter_before_opt_a
 0.21% :     0.000116s :      1: symbol_engine_optimizer
20.19% :     0.011264s :      1: task_emit
51.22% :     0.028572s :      1: type_inference
 0.18% :     0.000098s :      1: validate


TotalTime = 0.0521306, [33]
[bootstrap]: 0.00044347
[type_inference]: 0.0278519
[event_method]: 2.308e-05
[auto_monad]: 0.00018122
[graph_reusing]: 6.98e-06
[pre_auto_parallel]: 2.79999e-06
[py_interpret_to_execute]: 0.00014968
[rewriter_before_opt_a]: 9.923e-05
[expand_dump_flag]: 3.54002e-06
[jit_opt_a]: 0.00755504, [2]
    [Cycle 1]: 0.00272392, [27]
        [switch_simplify]: 6.997e-05
        [loop_unroll]: 3.681e-05
        [a_1]: 0.00077909
        [with_stream_mark]: 2.871e-05
        [recompute_prepare]: 1.201e-05
        [updatestate_depend_eliminate]: 7e-06
        [updatestate_assign_eliminate]: 6.91001e-06
        [updatestate_loads_eliminate]: 4.86002e-06
        [parameter_eliminate]: 1.90001e-06
        [specialize_transform]: 9.15001e-06
        [updatestate_useless_node_eliminater]: 1.208e-05
        [accelerated_algorithm]: 9.04e-06
        [meta_shard_fg_expand]: 3.35998e-06
        [get_grad_eliminate_]: 1.134e-05
        [merge_forward]: 6.75002e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.272e-05
        [j_node_and_user_rematch]: 1.546e-05
        [meta_fg_expand]: 3.9e-06
        [replace_old_param]: 1.439e-05
        [inline_without_move]: 9.13002e-06
        [renormalize]: 0.00127799
        [add_forward_monad_depend]: 6.89001e-06
        [auto_monad_grad]: 3.03e-06
        [auto_monad_eliminator]: 2.391e-05
        [cse]: 4.225e-05
        [replace_applicator]: 2.04e-05
    [Cycle 2]: 0.00048701, [27]
        [switch_simplify]: 9.86998e-06
        [loop_unroll]: 7.79002e-06
        [a_1]: 0.0001719
        [with_stream_mark]: 1.436e-05
        [recompute_prepare]: 8.44002e-06
        [updatestate_depend_eliminate]: 5.84999e-06
        [updatestate_assign_eliminate]: 5.49e-06
        [updatestate_loads_eliminate]: 4e-06
        [parameter_eliminate]: 1.65001e-06
        [specialize_transform]: 7.84002e-06
        [updatestate_useless_node_eliminater]: 1.101e-05
        [accelerated_algorithm]: 7.78001e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 7.66999e-06
        [merge_forward]: 4.28001e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.827e-05
        [j_node_and_user_rematch]: 1.335e-05
        [meta_fg_expand]: 3.23e-06
        [replace_old_param]: 1.087e-05
        [inline_without_move]: 7.68001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.60999e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 1.201e-05
        [cse]: 2.099e-05
        [replace_applicator]: 8.23999e-06
[py_interpret_to_execute_after_opt_a]: 1.82e-05
[rewriter_after_opt_a]: 0.00029108
[convert_after_rewriter]: 1.211e-05
[order_py_execute_after_rewriter]: 6.98e-06
[mutable_eliminate]: 0.0007885
[jit_opt_b]: 7.002e-05, [1]
    [Cycle 1]: 6.225e-05, [2]
        [frontend_op_eliminate]: 2.578e-05
        [inline_after_opt_a]: 2.417e-05
[cconv]: 3.26e-05
[loop_unroll]: 0.00063123
[jit_opt_after_cconv]: 0.00021476, [1]
    [Cycle 1]: 0.00020757, [11]
        [c_1]: 5.268e-05
        [parameter_eliminate]: 3.48999e-06
        [updatestate_depend_eliminate]: 9.47001e-06
        [updatestate_assign_eliminate]: 6.19999e-06
        [updatestate_loads_eliminate]: 4.70001e-06
        [cse]: 3.476e-05
        [call_graph_tuple_transform]: 2.308e-05
        [tuple_list_get_item_eliminator]: 9.30001e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 8.47e-06
[remove_dup_value]: 3.981e-05
[partial_unused_args_eliminate]: 2.35002e-06
[environ_conv]: 1.225e-05
[add_recomputation]: 8.698e-05
[cse_after_recomputation]: 3.159e-05, [1]
    [Cycle 1]: 2.513e-05, [1]
        [cse]: 1.89e-05
[auto_monad_reorder]: 2.919e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 7.56999e-06
[opt_after_jit_grad]: 0.00059505
[symbol_engine_optimizer]: 0.00010641, [1]
    [Cycle 1]: 9.984e-05, [6]
        [build]: 1.566e-05
        [elim_shapecalc]: 1.235e-05
        [elim_not_effective]: 1.919e-05
        [opt_reshape]: 9.20999e-06
        [fold_const_symbol]: 1.356e-05
        [renormalize]: 6.30011e-07
[validate]: 6.339e-05
[backend_pass]: 1.20999e-06
[task_emit]: 0.0124917
[execute]: 1.116e-05
Sums
    bootstrap                            :     0.000443s :  0.94%
    type_inference                       :     0.027852s : 59.27%
    event_method                         :     0.000023s :  0.05%
    auto_monad                           :     0.000181s :  0.39%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000150s :  0.32%
    rewriter_before_opt_a                :     0.000099s :  0.21%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.09%
    jit_opt_a.a_1                        :     0.000951s :  2.02%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000019s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001278s :  2.72%
    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      :     0.000036s :  0.08%
    jit_opt_a.cse                        :     0.000063s :  0.13%
    jit_opt_a.replace_applicator         :     0.000029s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.04%
    rewriter_after_opt_a                 :     0.000291s :  0.62%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000788s :  1.68%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.05%
    cconv                                :     0.000033s :  0.07%
    loop_unroll                          :     0.000631s :  1.34%
    jit_opt_after_cconv.c_1              :     0.000053s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000040s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000087s :  0.19%
    cse_after_recomputation.cse          :     0.000019s :  0.04%
    auto_monad_reorder                   :     0.000029s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000595s :  1.27%
    symbol_engine_optimizer.build        :     0.000016s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000063s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012492s : 26.58%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000333    47
 3.28% :     0.000011s :      2: substitution.depend_value_elim
 0.80% :     0.000003s :      4: substitution.elim_not_effective
 0.58% :     0.000002s :      4: substitution.fold_const_symbol
 2.16% :     0.000007s :      5: substitution.graph_param_transform
77.06% :     0.000256s :      4: substitution.inline
 1.60% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.41% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.80% :     0.000006s :      2: substitution.replace_old_param
 4.10% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.21% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.01% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.027749     2
94.32% :     0.026174s :      1: type_inference.infer
 5.68% :     0.001575s :      1: type_inference.specialize
------[replace.]   0.000102     6
69.01% :     0.000070s :      4: replace.inline
30.99% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     6
95.28% :     0.000253s :      4: match.inline
 4.72% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000180  1021
 1.44% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.09% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.17% :     0.000002s :     15: predicate.addn_zero_filter
 2.00% :     0.000004s :     15: predicate.arithmetic_simplify
 1.28% :     0.000002s :     15: predicate.cast_eliminate
 0.60% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     15: predicate.compare_switch_simplify
 1.37% :     0.000002s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.94% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.56% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.49% :     0.000003s :     15: predicate.environ_get_eliminate
 1.07% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.30% :     0.000002s :     10: predicate.get_grad_eliminate
 0.41% :     0.000001s :      5: predicate.graph_param_transform
 5.41% :     0.000010s :     31: predicate.inline
 1.05% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.61% :     0.000003s :     10: predicate.less_batch_normalization
 1.44% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.88% :     0.000003s :     22: predicate.load_eliminater
 1.34% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.27% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.90% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     15: predicate.merge_addn
 1.14% :     0.000002s :     15: predicate.minmaximum_grad
 1.99% :     0.000004s :      5: predicate.mutable_eliminate
 0.66% :     0.000001s :      5: predicate.opt_reshape
 2.28% :     0.000004s :     22: predicate.partial_eliminate
 1.18% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.78% :     0.000003s :     15: predicate.reduce_eliminate
 1.50% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.89% :     0.000003s :     27: predicate.replace_applicator
 0.60% :     0.000001s :     10: predicate.replace_old_param
 0.31% :     0.000001s :      5: predicate.reset_defer_inline
 1.30% :     0.000002s :     15: predicate.reshape_eliminate
 1.35% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.25% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     15: predicate.same_eliminate
 0.66% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.04% :     0.000002s :     10: predicate.special_op_eliminate
 1.01% :     0.000002s :     10: predicate.specialize_transform
 1.62% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.18% :     0.000004s :     21: predicate.switch_defer_inline
 1.79% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.93% :     0.000012s :     68: predicate.switch_simplify
 1.35% :     0.000002s :     15: predicate.tile_eliminate
 1.28% :     0.000002s :     15: predicate.transpose_eliminate
 1.57% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.64% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.57% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000893    13
44.72% :     0.000399s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.38% :     0.000066s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.90% :     0.000428s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.054872    76
 0.16% :     0.000090s :      1: add_recomputation
 0.34% :     0.000187s :      1: auto_monad
 0.06% :     0.000032s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.84% :     0.000464s :      1: bootstrap
 0.07% :     0.000036s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.06% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.05% :     0.000028s :      1: event_method
 0.03% :     0.000018s :      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
13.78% :     0.007559s :      1: jit_opt_a
 0.40% :     0.000219s :      1: jit_opt_after_cconv
 0.13% :     0.000073s :      1: jit_opt_b
 1.17% :     0.000643s :      1: loop_unroll
 1.45% :     0.000798s :      1: mutable_eliminate
 2.33% :     0.001278s :     26: opt.transform.jit_opt_a
 0.16% :     0.000089s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000043s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.10% :     0.000604s :      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% :     0.000006s :      1: pre_auto_parallel
 0.28% :     0.000154s :      1: py_interpret_to_execute
 0.04% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000043s :      1: remove_dup_value
 1.47% :     0.000809s :      1: renormalize.infer
 0.84% :     0.000459s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.54% :     0.000296s :      1: rewriter_after_opt_a
 0.19% :     0.000103s :      1: rewriter_before_opt_a
 0.20% :     0.000109s :      1: symbol_engine_optimizer
22.80% :     0.012513s :      1: task_emit
50.80% :     0.027876s :      1: type_inference
 0.17% :     0.000091s :      1: validate


TotalTime = 0.0731675, [33]
[bootstrap]: 0.00043832
[type_inference]: 0.0440406
[event_method]: 2.478e-05
[auto_monad]: 0.00019346
[graph_reusing]: 6.49999e-06
[pre_auto_parallel]: 3.16999e-06
[py_interpret_to_execute]: 0.00015868
[rewriter_before_opt_a]: 0.00011643
[expand_dump_flag]: 3.91001e-06
[jit_opt_a]: 0.00935861, [2]
    [Cycle 1]: 0.00353949, [27]
        [switch_simplify]: 7.937e-05
        [loop_unroll]: 3.687e-05
        [a_1]: 0.00087253
        [with_stream_mark]: 3.161e-05
        [recompute_prepare]: 1.465e-05
        [updatestate_depend_eliminate]: 8.37e-06
        [updatestate_assign_eliminate]: 6.86001e-06
        [updatestate_loads_eliminate]: 4.74e-06
        [parameter_eliminate]: 3.11999e-06
        [specialize_transform]: 1.141e-05
        [updatestate_useless_node_eliminater]: 1.316e-05
        [accelerated_algorithm]: 9.49999e-06
        [meta_shard_fg_expand]: 4.47e-06
        [get_grad_eliminate_]: 8.27e-06
        [merge_forward]: 5.83002e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.199e-05
        [j_node_and_user_rematch]: 1.48e-05
        [meta_fg_expand]: 4.13999e-06
        [replace_old_param]: 1.464e-05
        [inline_without_move]: 9.24998e-06
        [renormalize]: 0.00201359
        [add_forward_monad_depend]: 9.44e-06
        [auto_monad_grad]: 2.74001e-06
        [auto_monad_eliminator]: 2.596e-05
        [cse]: 4.682e-05
        [replace_applicator]: 2.552e-05
    [Cycle 2]: 0.0005357, [27]
        [switch_simplify]: 1.052e-05
        [loop_unroll]: 8.55001e-06
        [a_1]: 0.00019963
        [with_stream_mark]: 1.809e-05
        [recompute_prepare]: 8.53001e-06
        [updatestate_depend_eliminate]: 5.45001e-06
        [updatestate_assign_eliminate]: 5.58997e-06
        [updatestate_loads_eliminate]: 5.01002e-06
        [parameter_eliminate]: 1.67001e-06
        [specialize_transform]: 7.97998e-06
        [updatestate_useless_node_eliminater]: 1.196e-05
        [accelerated_algorithm]: 8.37998e-06
        [meta_shard_fg_expand]: 2.91e-06
        [get_grad_eliminate_]: 7.93999e-06
        [merge_forward]: 5.57001e-06
        [cell_reuse_recompute_pass]: 4.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.939e-05
        [j_node_and_user_rematch]: 1.286e-05
        [meta_fg_expand]: 4.15999e-06
        [replace_old_param]: 1.132e-05
        [inline_without_move]: 7.49002e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.83002e-06
        [auto_monad_grad]: 1.54998e-06
        [auto_monad_eliminator]: 1.237e-05
        [cse]: 2.227e-05
        [replace_applicator]: 8.43999e-06
[py_interpret_to_execute_after_opt_a]: 2.055e-05
[rewriter_after_opt_a]: 0.00032805
[convert_after_rewriter]: 1.492e-05
[order_py_execute_after_rewriter]: 8.63001e-06
[mutable_eliminate]: 0.0009365
[jit_opt_b]: 8.422e-05, [1]
    [Cycle 1]: 7.407e-05, [2]
        [frontend_op_eliminate]: 2.943e-05
        [inline_after_opt_a]: 3.029e-05
[cconv]: 4.284e-05
[loop_unroll]: 0.00064068
[jit_opt_after_cconv]: 0.00025878, [1]
    [Cycle 1]: 0.00024984, [11]
        [c_1]: 6.144e-05
        [parameter_eliminate]: 7.08e-06
        [updatestate_depend_eliminate]: 1.3e-05
        [updatestate_assign_eliminate]: 5.67001e-06
        [updatestate_loads_eliminate]: 4.74e-06
        [cse]: 4.866e-05
        [call_graph_tuple_transform]: 2.781e-05
        [tuple_list_get_item_eliminator]: 9.48002e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 1.099e-05
[remove_dup_value]: 2.055e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 1.603e-05
[add_recomputation]: 8.214e-05
[cse_after_recomputation]: 3.463e-05, [1]
    [Cycle 1]: 2.642e-05, [1]
        [cse]: 1.846e-05
[auto_monad_reorder]: 2.911e-05
[get_jit_bprop_graph]: 2.24001e-06
[rewriter_after_jit_bprop_graph]: 1.1e-05
[opt_after_jit_grad]: 0.00072694
[symbol_engine_optimizer]: 0.00012127, [1]
    [Cycle 1]: 0.00011261, [6]
        [build]: 2.081e-05
        [elim_shapecalc]: 1.357e-05
        [elim_not_effective]: 2.257e-05
        [opt_reshape]: 9.12001e-06
        [fold_const_symbol]: 1.469e-05
        [renormalize]: 6.29982e-07
[validate]: 7.936e-05
[backend_pass]: 1.20999e-06
[task_emit]: 0.0150515
[execute]: 1.053e-05
Sums
    bootstrap                            :     0.000438s :  0.65%
    type_inference                       :     0.044041s : 65.66%
    event_method                         :     0.000025s :  0.04%
    auto_monad                           :     0.000193s :  0.29%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000159s :  0.24%
    rewriter_before_opt_a                :     0.000116s :  0.17%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000090s :  0.13%
    jit_opt_a.loop_unroll                :     0.000045s :  0.07%
    jit_opt_a.a_1                        :     0.001072s :  1.60%
    jit_opt_a.with_stream_mark           :     0.000050s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.02%
    jit_opt_a.merge_forward              :     0.000011s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000008s :  0.01%
    jit_opt_a.replace_old_param          :     0.000026s :  0.04%
    jit_opt_a.inline_without_move        :     0.000017s :  0.02%
    jit_opt_a.renormalize                :     0.002014s :  3.00%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000038s :  0.06%
    jit_opt_a.cse                        :     0.000069s :  0.10%
    jit_opt_a.replace_applicator         :     0.000034s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.03%
    rewriter_after_opt_a                 :     0.000328s :  0.49%
    convert_after_rewriter               :     0.000015s :  0.02%
    order_py_execute_after_rewriter      :     0.000009s :  0.01%
    mutable_eliminate                    :     0.000937s :  1.40%
    jit_opt_b.frontend_op_eliminate      :     0.000029s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000030s :  0.05%
    cconv                                :     0.000043s :  0.06%
    loop_unroll                          :     0.000641s :  0.96%
    jit_opt_after_cconv.c_1              :     0.000061s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000049s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000011s :  0.02%
    remove_dup_value                     :     0.000021s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000016s :  0.02%
    add_recomputation                    :     0.000082s :  0.12%
    cse_after_recomputation.cse          :     0.000018s :  0.03%
    auto_monad_reorder                   :     0.000029s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.02%
    opt_after_jit_grad                   :     0.000727s :  1.08%
    symbol_engine_optimizer.build        :     0.000021s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000079s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.015052s : 22.44%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000382    47
 3.30% :     0.000013s :      2: substitution.depend_value_elim
 0.83% :     0.000003s :      4: substitution.elim_not_effective
 0.55% :     0.000002s :      4: substitution.fold_const_symbol
 2.29% :     0.000009s :      5: substitution.graph_param_transform
77.11% :     0.000295s :      4: substitution.inline
 1.40% :     0.000005s :      8: substitution.j_node_and_user_rematch
 1.91% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.86% :     0.000007s :      2: substitution.replace_old_param
 3.57% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 4.10% :     0.000016s :      3: substitution.updatestate_pure_node_eliminater
 3.08% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.043929     2
96.22% :     0.042268s :      1: type_inference.infer
 3.78% :     0.001661s :      1: type_inference.specialize
------[replace.]   0.000106     6
66.83% :     0.000071s :      4: replace.inline
33.17% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000304     6
95.92% :     0.000291s :      4: match.inline
 4.08% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000207  1021
 1.33% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.30% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 0.99% :     0.000002s :     15: predicate.addn_check_dump
 1.89% :     0.000004s :     15: predicate.addn_zero_filter
 2.43% :     0.000005s :     15: predicate.arithmetic_simplify
 1.47% :     0.000003s :     15: predicate.cast_eliminate
 0.45% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     15: predicate.compare_switch_simplify
 1.63% :     0.000003s :     15: predicate.depend_value_elim
 1.09% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.31% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.58% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.89% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      5: predicate.elim_not_effective
 0.52% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000003s :     15: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.13% :     0.000002s :     15: predicate.environ_get_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.28% :     0.000003s :     10: predicate.get_grad_eliminate
 0.23% :     0.000000s :      5: predicate.graph_param_transform
 5.85% :     0.000012s :     31: predicate.inline
 0.91% :     0.000002s :     10: predicate.inline_without_move
 0.35% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.49% :     0.000003s :     10: predicate.less_batch_normalization
 1.30% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.97% :     0.000004s :     22: predicate.load_eliminater
 1.76% :     0.000004s :      5: predicate.loop_unroll_after_grad
 3.24% :     0.000007s :     42: predicate.loop_unroll_before_grad
 2.08% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     15: predicate.merge_addn
 1.17% :     0.000002s :     15: predicate.minmaximum_grad
 1.92% :     0.000004s :      5: predicate.mutable_eliminate
 0.57% :     0.000001s :      5: predicate.opt_reshape
 2.10% :     0.000004s :     22: predicate.partial_eliminate
 1.32% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.75% :     0.000004s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.76% :     0.000004s :     27: predicate.replace_applicator
 0.54% :     0.000001s :     10: predicate.replace_old_param
 0.58% :     0.000001s :      5: predicate.reset_defer_inline
 1.29% :     0.000003s :     15: predicate.reshape_eliminate
 1.54% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 1.26% :     0.000003s :      5: predicate.row_tensor_eliminate
 1.25% :     0.000003s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.04% :     0.000002s :     10: predicate.special_op_eliminate
 0.98% :     0.000002s :     10: predicate.specialize_transform
 1.71% :     0.000004s :     15: predicate.split_environ_get_set_with_tuple_value
 1.05% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.89% :     0.000004s :     21: predicate.switch_defer_inline
 1.79% :     0.000004s :     21: predicate.switch_layer_defer_inline
 5.89% :     0.000012s :     68: predicate.switch_simplify
 1.21% :     0.000002s :     15: predicate.tile_eliminate
 1.09% :     0.000002s :     15: predicate.transpose_eliminate
 1.55% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.61% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.28% :     0.000007s :     32: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000003s :     15: predicate.value_based_eliminate
 0.37% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.74% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001029    13
45.08% :     0.000464s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.11% :     0.000073s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.82% :     0.000492s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.076831    76
 0.11% :     0.000086s :      1: add_recomputation
 0.26% :     0.000200s :      1: auto_monad
 0.04% :     0.000032s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.60% :     0.000464s :      1: bootstrap
 0.06% :     0.000045s :      1: cconv
 0.02% :     0.000018s :      1: convert_after_rewriter
 0.05% :     0.000037s :      1: cse_after_recomputation
 0.02% :     0.000018s :      1: environ_conv
 0.04% :     0.000029s :      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
12.19% :     0.009362s :      1: jit_opt_a
 0.34% :     0.000262s :      1: jit_opt_after_cconv
 0.11% :     0.000088s :      1: jit_opt_b
 0.85% :     0.000655s :      1: loop_unroll
 1.24% :     0.000955s :      1: mutable_eliminate
 1.84% :     0.001416s :     26: opt.transform.jit_opt_a
 0.14% :     0.000105s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000050s :      4: opt.transform.jit_opt_b
 0.03% :     0.000024s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000031s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000040s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000055s :      4: opt.transform.symbol_engine_opt
 0.96% :     0.000741s :      1: opt_after_jit_grad
 0.01% :     0.000011s :      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% :     0.000164s :      1: py_interpret_to_execute
 0.03% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000023s :      1: remove_dup_value
 1.90% :     0.001462s :      1: renormalize.infer
 0.70% :     0.000539s :      1: renormalize.specialize
 0.02% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.44% :     0.000335s :      1: rewriter_after_opt_a
 0.16% :     0.000122s :      1: rewriter_before_opt_a
 0.16% :     0.000125s :      1: symbol_engine_optimizer
19.62% :     0.015071s :      1: task_emit
57.35% :     0.044065s :      1: type_inference
 0.14% :     0.000111s :      1: validate


TotalTime = 0.0578211, [33]
[bootstrap]: 0.0004535
[type_inference]: 0.0291396
[event_method]: 2.335e-05
[auto_monad]: 0.00023533
[graph_reusing]: 7.53999e-06
[pre_auto_parallel]: 3.66001e-06
[py_interpret_to_execute]: 0.0002219
[rewriter_before_opt_a]: 0.00010981
[expand_dump_flag]: 4.28999e-06
[jit_opt_a]: 0.0095026, [2]
    [Cycle 1]: 0.00322236, [27]
        [switch_simplify]: 7.659e-05
        [loop_unroll]: 3.681e-05
        [a_1]: 0.00097631
        [with_stream_mark]: 3.286e-05
        [recompute_prepare]: 1.684e-05
        [updatestate_depend_eliminate]: 8.13999e-06
        [updatestate_assign_eliminate]: 6.58e-06
        [updatestate_loads_eliminate]: 5.12999e-06
        [parameter_eliminate]: 3.19001e-06
        [specialize_transform]: 1.094e-05
        [updatestate_useless_node_eliminater]: 1.462e-05
        [accelerated_algorithm]: 1.007e-05
        [meta_shard_fg_expand]: 3.18e-06
        [get_grad_eliminate_]: 8.53001e-06
        [merge_forward]: 6.01998e-06
        [cell_reuse_recompute_pass]: 2.58003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.186e-05
        [j_node_and_user_rematch]: 1.425e-05
        [meta_fg_expand]: 3.4e-06
        [replace_old_param]: 1.415e-05
        [inline_without_move]: 9.27001e-06
        [renormalize]: 0.00155534
        [add_forward_monad_depend]: 1.101e-05
        [auto_monad_grad]: 2.95998e-06
        [auto_monad_eliminator]: 3.091e-05
        [cse]: 4.697e-05
        [replace_applicator]: 2.313e-05
    [Cycle 2]: 0.00051816, [27]
        [switch_simplify]: 9.46998e-06
        [loop_unroll]: 8.45999e-06
        [a_1]: 0.00018749
        [with_stream_mark]: 1.833e-05
        [recompute_prepare]: 8.37e-06
        [updatestate_depend_eliminate]: 5.82999e-06
        [updatestate_assign_eliminate]: 6.72002e-06
        [updatestate_loads_eliminate]: 4.24997e-06
        [parameter_eliminate]: 1.99999e-06
        [specialize_transform]: 8.58001e-06
        [updatestate_useless_node_eliminater]: 1.117e-05
        [accelerated_algorithm]: 8.42e-06
        [meta_shard_fg_expand]: 3.02002e-06
        [get_grad_eliminate_]: 7.35e-06
        [merge_forward]: 5.24e-06
        [cell_reuse_recompute_pass]: 3.04001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.809e-05
        [j_node_and_user_rematch]: 1.266e-05
        [meta_fg_expand]: 2.93998e-06
        [replace_old_param]: 1.105e-05
        [inline_without_move]: 7.71001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.26e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 1.054e-05
        [cse]: 2.166e-05
        [replace_applicator]: 8.57e-06
[py_interpret_to_execute_after_opt_a]: 2.14e-05
[rewriter_after_opt_a]: 0.0003528
[convert_after_rewriter]: 1.335e-05
[order_py_execute_after_rewriter]: 7.73001e-06
[mutable_eliminate]: 0.0008888
[jit_opt_b]: 7.396e-05, [1]
    [Cycle 1]: 6.558e-05, [2]
        [frontend_op_eliminate]: 2.558e-05
        [inline_after_opt_a]: 2.707e-05
[cconv]: 3.336e-05
[loop_unroll]: 0.00048215
[jit_opt_after_cconv]: 0.00021072, [1]
    [Cycle 1]: 0.00020403, [11]
        [c_1]: 4.679e-05
        [parameter_eliminate]: 4.04002e-06
        [updatestate_depend_eliminate]: 1.04e-05
        [updatestate_assign_eliminate]: 4.84e-06
        [updatestate_loads_eliminate]: 4.12e-06
        [cse]: 3.48e-05
        [call_graph_tuple_transform]: 2.43e-05
        [tuple_list_get_item_eliminator]: 8.79e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 9.15999e-06
[remove_dup_value]: 6.775e-05
[partial_unused_args_eliminate]: 2.46998e-06
[environ_conv]: 4.767e-05
[add_recomputation]: 0.00014155
[cse_after_recomputation]: 3.657e-05, [1]
    [Cycle 1]: 2.943e-05, [1]
        [cse]: 2.114e-05
[auto_monad_reorder]: 2.832e-05
[get_jit_bprop_graph]: 2.61e-06
[rewriter_after_jit_bprop_graph]: 8.48999e-06
[opt_after_jit_grad]: 0.00057198
[symbol_engine_optimizer]: 0.0001159, [1]
    [Cycle 1]: 0.00010822, [6]
        [build]: 1.879e-05
        [elim_shapecalc]: 1.382e-05
        [elim_not_effective]: 2.029e-05
        [opt_reshape]: 8.38001e-06
        [fold_const_symbol]: 1.575e-05
        [renormalize]: 6.39993e-07
[validate]: 9.439e-05
[backend_pass]: 1.35999e-06
[task_emit]: 0.0146128
[execute]: 1.072e-05
Sums
    bootstrap                            :     0.000454s :  0.89%
    type_inference                       :     0.029140s : 56.87%
    event_method                         :     0.000023s :  0.05%
    auto_monad                           :     0.000235s :  0.46%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000222s :  0.43%
    rewriter_before_opt_a                :     0.000110s :  0.21%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000086s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.09%
    jit_opt_a.a_1                        :     0.001164s :  2.27%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000025s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000020s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.03%
    jit_opt_a.merge_forward              :     0.000011s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000025s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.03%
    jit_opt_a.renormalize                :     0.001555s :  3.04%
    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      :     0.000041s :  0.08%
    jit_opt_a.cse                        :     0.000069s :  0.13%
    jit_opt_a.replace_applicator         :     0.000032s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.04%
    rewriter_after_opt_a                 :     0.000353s :  0.69%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000889s :  1.73%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.05%
    cconv                                :     0.000033s :  0.07%
    loop_unroll                          :     0.000482s :  0.94%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.09%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000068s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000048s :  0.09%
    add_recomputation                    :     0.000142s :  0.28%
    cse_after_recomputation.cse          :     0.000021s :  0.04%
    auto_monad_reorder                   :     0.000028s :  0.06%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000572s :  1.12%
    symbol_engine_optimizer.build        :     0.000019s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000094s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.014613s : 28.52%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000419    47
 3.04% :     0.000013s :      2: substitution.depend_value_elim
 0.75% :     0.000003s :      4: substitution.elim_not_effective
 0.51% :     0.000002s :      4: substitution.fold_const_symbol
 1.75% :     0.000007s :      5: substitution.graph_param_transform
80.31% :     0.000336s :      4: substitution.inline
 1.26% :     0.000005s :      8: substitution.j_node_and_user_rematch
 1.87% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.49% :     0.000006s :      2: substitution.replace_old_param
 3.68% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 2.69% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 2.65% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.029048     2
94.50% :     0.027452s :      1: type_inference.infer
 5.50% :     0.001596s :      1: type_inference.specialize
------[replace.]   0.000111     6
65.13% :     0.000073s :      4: replace.inline
34.87% :     0.000039s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000347     6
96.01% :     0.000333s :      4: match.inline
 3.99% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000194  1021
 1.40% :     0.000003s :     15: predicate.accumulaten_eliminater
 0.98% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     15: predicate.addn_check_dump
 1.54% :     0.000003s :     15: predicate.addn_zero_filter
 2.80% :     0.000005s :     15: predicate.arithmetic_simplify
 1.30% :     0.000003s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 0.99% :     0.000002s :     15: predicate.compare_switch_simplify
 1.52% :     0.000003s :     15: predicate.depend_value_elim
 1.04% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.30% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.38% :     0.000003s :     15: predicate.dict_set_item_eliminator
 1.08% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.30% :     0.000001s :      5: predicate.elim_not_effective
 0.61% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000002s :     15: predicate.environ_add_const_eliminate
 0.99% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.05% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.76% :     0.000003s :     15: predicate.environ_get_eliminate
 1.01% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.20% :     0.000002s :     10: predicate.get_grad_eliminate
 0.24% :     0.000000s :      5: predicate.graph_param_transform
 5.52% :     0.000011s :     31: predicate.inline
 1.09% :     0.000002s :     10: predicate.inline_without_move
 0.38% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.73% :     0.000003s :     10: predicate.less_batch_normalization
 1.75% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.91% :     0.000004s :     22: predicate.load_eliminater
 1.19% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.16% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.02% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000002s :     15: predicate.merge_addn
 1.05% :     0.000002s :     15: predicate.minmaximum_grad
 1.60% :     0.000003s :      5: predicate.mutable_eliminate
 0.48% :     0.000001s :      5: predicate.opt_reshape
 2.16% :     0.000004s :     22: predicate.partial_eliminate
 1.61% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.79% :     0.000003s :     15: predicate.reduce_eliminate
 1.55% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.75% :     0.000003s :     27: predicate.replace_applicator
 0.57% :     0.000001s :     10: predicate.replace_old_param
 0.45% :     0.000001s :      5: predicate.reset_defer_inline
 1.50% :     0.000003s :     15: predicate.reshape_eliminate
 1.50% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.79% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.65% :     0.000003s :     15: predicate.same_eliminate
 0.68% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.98% :     0.000002s :     10: predicate.special_op_eliminate
 1.10% :     0.000002s :     10: predicate.specialize_transform
 1.64% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.05% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.35% :     0.000005s :     21: predicate.switch_defer_inline
 1.78% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.11% :     0.000012s :     68: predicate.switch_simplify
 1.23% :     0.000002s :     15: predicate.tile_eliminate
 1.21% :     0.000002s :     15: predicate.transpose_eliminate
 1.44% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.67% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 2.18% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.72% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001018    13
45.64% :     0.000465s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.29% :     0.000074s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.07% :     0.000479s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.061071    76
 0.24% :     0.000145s :      1: add_recomputation
 0.40% :     0.000242s :      1: auto_monad
 0.05% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.82% :     0.000499s :      1: bootstrap
 0.06% :     0.000036s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.06% :     0.000039s :      1: cse_after_recomputation
 0.08% :     0.000051s :      1: environ_conv
 0.05% :     0.000028s :      1: event_method
 0.03% :     0.000017s :      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
15.57% :     0.009507s :      1: jit_opt_a
 0.35% :     0.000214s :      1: jit_opt_after_cconv
 0.13% :     0.000077s :      1: jit_opt_b
 0.80% :     0.000491s :      1: loop_unroll
 1.47% :     0.000899s :      1: mutable_eliminate
 2.46% :     0.001502s :     26: opt.transform.jit_opt_a
 0.14% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000044s :      4: opt.transform.jit_opt_b
 0.03% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000055s :      4: opt.transform.symbol_engine_opt
 0.95% :     0.000581s :      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
 0.38% :     0.000230s :      1: py_interpret_to_execute
 0.04% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000071s :      1: remove_dup_value
 1.59% :     0.000973s :      1: renormalize.infer
 0.93% :     0.000570s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.59% :     0.000359s :      1: rewriter_after_opt_a
 0.19% :     0.000116s :      1: rewriter_before_opt_a
 0.19% :     0.000119s :      1: symbol_engine_optimizer
23.96% :     0.014633s :      1: task_emit
47.75% :     0.029159s :      1: type_inference
 0.20% :     0.000124s :      1: validate


TotalTime = 0.129006, [33]
[bootstrap]: 0.00046623
[type_inference]: 0.0287452
[event_method]: 2.508e-05
[auto_monad]: 0.00019319
[graph_reusing]: 8.05999e-06
[pre_auto_parallel]: 3.32002e-06
[py_interpret_to_execute]: 0.00015656
[rewriter_before_opt_a]: 0.00010732
[expand_dump_flag]: 3.75e-06
[jit_opt_a]: 0.0359642, [2]
    [Cycle 1]: 0.00313608, [27]
        [switch_simplify]: 0.00011517
        [loop_unroll]: 4.036e-05
        [a_1]: 0.00083292
        [with_stream_mark]: 2.719e-05
        [recompute_prepare]: 1.122e-05
        [updatestate_depend_eliminate]: 6.91001e-06
        [updatestate_assign_eliminate]: 6.66e-06
        [updatestate_loads_eliminate]: 4.81002e-06
        [parameter_eliminate]: 2.34001e-06
        [specialize_transform]: 1.017e-05
        [updatestate_useless_node_eliminater]: 1.254e-05
        [accelerated_algorithm]: 1.226e-05
        [meta_shard_fg_expand]: 3.08e-06
        [get_grad_eliminate_]: 8.99e-06
        [merge_forward]: 5.64e-06
        [cell_reuse_recompute_pass]: 1.46998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.057e-05
        [j_node_and_user_rematch]: 1.426e-05
        [meta_fg_expand]: 3.51999e-06
        [replace_old_param]: 1.325e-05
        [inline_without_move]: 8.85999e-06
        [renormalize]: 0.00162064
        [add_forward_monad_depend]: 7.90998e-06
        [auto_monad_grad]: 2.92002e-06
        [auto_monad_eliminator]: 2.679e-05
        [cse]: 4.394e-05
        [replace_applicator]: 2.239e-05
    [Cycle 2]: 0.00051127, [27]
        [switch_simplify]: 9.24e-06
        [loop_unroll]: 8.64e-06
        [a_1]: 0.00018101
        [with_stream_mark]: 1.702e-05
        [recompute_prepare]: 8.43999e-06
        [updatestate_depend_eliminate]: 6.18002e-06
        [updatestate_assign_eliminate]: 5.74999e-06
        [updatestate_loads_eliminate]: 5.05001e-06
        [parameter_eliminate]: 1.62001e-06
        [specialize_transform]: 8.65999e-06
        [updatestate_useless_node_eliminater]: 1.184e-05
        [accelerated_algorithm]: 9.70002e-06
        [meta_shard_fg_expand]: 2.81999e-06
        [get_grad_eliminate_]: 7.8e-06
        [merge_forward]: 4.99998e-06
        [cell_reuse_recompute_pass]: 2.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.884e-05
        [j_node_and_user_rematch]: 1.367e-05
        [meta_fg_expand]: 3.33e-06
        [replace_old_param]: 1.075e-05
        [inline_without_move]: 7.82e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.63002e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 1.155e-05
        [cse]: 2.272e-05
        [replace_applicator]: 7.66999e-06
[py_interpret_to_execute_after_opt_a]: 1.82e-05
[rewriter_after_opt_a]: 0.0467962
[convert_after_rewriter]: 2.621e-05
[order_py_execute_after_rewriter]: 1.07e-05
[mutable_eliminate]: 0.00096181
[jit_opt_b]: 8.976e-05, [1]
    [Cycle 1]: 7.774e-05, [2]
        [frontend_op_eliminate]: 3.093e-05
        [inline_after_opt_a]: 3.016e-05
[cconv]: 3.496e-05
[loop_unroll]: 0.00057361
[jit_opt_after_cconv]: 0.00024795, [1]
    [Cycle 1]: 0.00024037, [11]
        [c_1]: 5.63e-05
        [parameter_eliminate]: 5.10001e-06
        [updatestate_depend_eliminate]: 1.03e-05
        [updatestate_assign_eliminate]: 5.38002e-06
        [updatestate_loads_eliminate]: 4.84e-06
        [cse]: 5.38e-05
        [call_graph_tuple_transform]: 2.438e-05
        [tuple_list_get_item_eliminator]: 8.42e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 1.267e-05
[remove_dup_value]: 2.276e-05
[partial_unused_args_eliminate]: 2.79999e-06
[environ_conv]: 1.418e-05
[add_recomputation]: 7.845e-05
[cse_after_recomputation]: 3.758e-05, [1]
    [Cycle 1]: 2.959e-05, [1]
        [cse]: 2.219e-05
[auto_monad_reorder]: 2.672e-05
[get_jit_bprop_graph]: 2.39001e-06
[rewriter_after_jit_bprop_graph]: 8.92e-06
[opt_after_jit_grad]: 0.00062878
[symbol_engine_optimizer]: 0.00011723, [1]
    [Cycle 1]: 0.00010937, [6]
        [build]: 1.899e-05
        [elim_shapecalc]: 1.285e-05
        [elim_not_effective]: 1.997e-05
        [opt_reshape]: 1.047e-05
        [fold_const_symbol]: 1.488e-05
        [renormalize]: 7.2e-07
[validate]: 0.00010282
[backend_pass]: 1.29e-06
[task_emit]: 0.0130364
[execute]: 1.099e-05
Sums
    bootstrap                            :     0.000466s :  0.49%
    type_inference                       :     0.028745s : 30.04%
    event_method                         :     0.000025s :  0.03%
    auto_monad                           :     0.000193s :  0.20%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000157s :  0.16%
    rewriter_before_opt_a                :     0.000107s :  0.11%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000124s :  0.13%
    jit_opt_a.loop_unroll                :     0.000049s :  0.05%
    jit_opt_a.a_1                        :     0.001014s :  1.06%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  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 :     0.000024s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000022s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.02%
    jit_opt_a.merge_forward              :     0.000011s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.03%
    jit_opt_a.inline_without_move        :     0.000017s :  0.02%
    jit_opt_a.renormalize                :     0.001621s :  1.69%
    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      :     0.000038s :  0.04%
    jit_opt_a.cse                        :     0.000067s :  0.07%
    jit_opt_a.replace_applicator         :     0.000030s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.02%
    rewriter_after_opt_a                 :     0.046796s : 48.90%
    convert_after_rewriter               :     0.000026s :  0.03%
    order_py_execute_after_rewriter      :     0.000011s :  0.01%
    mutable_eliminate                    :     0.000962s :  1.01%
    jit_opt_b.frontend_op_eliminate      :     0.000031s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000030s :  0.03%
    cconv                                :     0.000035s :  0.04%
    loop_unroll                          :     0.000574s :  0.60%
    jit_opt_after_cconv.c_1              :     0.000056s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000054s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  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  :     0.000013s :  0.01%
    remove_dup_value                     :     0.000023s :  0.02%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000014s :  0.01%
    add_recomputation                    :     0.000078s :  0.08%
    cse_after_recomputation.cse          :     0.000022s :  0.02%
    auto_monad_reorder                   :     0.000027s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.01%
    opt_after_jit_grad                   :     0.000629s :  0.66%
    symbol_engine_optimizer.build        :     0.000019s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000103s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.013036s : 13.62%
    execute                              :     0.000011s :  0.01%


Time group info:
------[substitution.]   0.000371    47
 3.48% :     0.000013s :      2: substitution.depend_value_elim
 0.75% :     0.000003s :      4: substitution.elim_not_effective
 0.58% :     0.000002s :      4: substitution.fold_const_symbol
 1.99% :     0.000007s :      5: substitution.graph_param_transform
77.59% :     0.000288s :      4: substitution.inline
 1.49% :     0.000006s :      8: substitution.j_node_and_user_rematch
 1.99% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.40% :     0.000005s :      2: substitution.replace_old_param
 4.02% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.40% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.30% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.028633     2
93.58% :     0.026796s :      1: type_inference.infer
 6.42% :     0.001838s :      1: type_inference.specialize
------[replace.]   0.000108     6
65.49% :     0.000071s :      4: replace.inline
34.51% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000299     6
95.40% :     0.000285s :      4: match.inline
 4.60% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000194  1021
 1.40% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.40% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.28% :     0.000002s :     15: predicate.addn_check_dump
 1.33% :     0.000003s :     15: predicate.addn_zero_filter
 2.95% :     0.000006s :     15: predicate.arithmetic_simplify
 1.37% :     0.000003s :     15: predicate.cast_eliminate
 0.57% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     15: predicate.compare_switch_simplify
 1.33% :     0.000003s :     15: predicate.depend_value_elim
 0.98% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.40% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.35% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.79% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.69% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.42% :     0.000003s :     15: predicate.environ_get_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000002s :     10: predicate.get_grad_eliminate
 0.37% :     0.000001s :      5: predicate.graph_param_transform
 5.44% :     0.000011s :     31: predicate.inline
 0.87% :     0.000002s :     10: predicate.inline_without_move
 0.39% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.41% :     0.000003s :     10: predicate.less_batch_normalization
 1.24% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 2.22% :     0.000004s :     22: predicate.load_eliminater
 1.48% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.38% :     0.000007s :     42: predicate.loop_unroll_before_grad
 1.97% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.11% :     0.000002s :     15: predicate.minmaximum_grad
 1.62% :     0.000003s :      5: predicate.mutable_eliminate
 0.58% :     0.000001s :      5: predicate.opt_reshape
 2.13% :     0.000004s :     22: predicate.partial_eliminate
 1.29% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     15: predicate.reduce_eliminate
 1.50% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.70% :     0.000003s :     27: predicate.replace_applicator
 0.72% :     0.000001s :     10: predicate.replace_old_param
 0.40% :     0.000001s :      5: predicate.reset_defer_inline
 1.46% :     0.000003s :     15: predicate.reshape_eliminate
 1.31% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.93% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.40% :     0.000003s :     15: predicate.same_eliminate
 0.53% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.04% :     0.000002s :     10: predicate.special_op_eliminate
 1.07% :     0.000002s :     10: predicate.specialize_transform
 1.48% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.56% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.95% :     0.000004s :     21: predicate.switch_defer_inline
 1.76% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.42% :     0.000012s :     68: predicate.switch_simplify
 1.37% :     0.000003s :     15: predicate.tile_eliminate
 1.17% :     0.000002s :     15: predicate.transpose_eliminate
 1.72% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.72% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     15: predicate.value_based_eliminate
 0.36% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001043    13
46.96% :     0.000490s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.32% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.73% :     0.000508s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.132124    76
 0.06% :     0.000082s :      1: add_recomputation
 0.15% :     0.000201s :      1: auto_monad
 0.02% :     0.000030s :      1: auto_monad_reorder
 0.00% :     0.000005s :      1: backend_pass
 0.38% :     0.000496s :      1: bootstrap
 0.03% :     0.000039s :      1: cconv
 0.02% :     0.000031s :      1: convert_after_rewriter
 0.03% :     0.000040s :      1: cse_after_recomputation
 0.01% :     0.000017s :      1: environ_conv
 0.02% :     0.000030s :      1: event_method
 0.01% :     0.000017s :      1: execute
 0.04% :     0.000047s :      1: expand_dump_flag
 0.00% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000011s :      1: graph_reusing
27.23% :     0.035974s :      1: jit_opt_a
 0.19% :     0.000251s :      1: jit_opt_after_cconv
 0.07% :     0.000093s :      1: jit_opt_b
 0.44% :     0.000584s :      1: loop_unroll
 0.74% :     0.000975s :      1: mutable_eliminate
 1.05% :     0.001390s :     26: opt.transform.jit_opt_a
 0.07% :     0.000097s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000051s :      4: opt.transform.jit_opt_b
 0.02% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000033s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000054s :      4: opt.transform.symbol_engine_opt
 0.48% :     0.000640s :      1: opt_after_jit_grad
 0.01% :     0.000013s :      1: order_py_execute_after_rewriter
 0.00% :     0.000005s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000162s :      1: py_interpret_to_execute
 0.02% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000027s :      1: remove_dup_value
 0.75% :     0.000988s :      1: renormalize.infer
 0.47% :     0.000623s :      1: renormalize.specialize
 0.01% :     0.000011s :      1: rewriter_after_jit_bprop_graph
35.44% :     0.046822s :      1: rewriter_after_opt_a
 0.09% :     0.000114s :      1: rewriter_before_opt_a
 0.09% :     0.000121s :      1: symbol_engine_optimizer
 9.88% :     0.013058s :      1: task_emit
21.78% :     0.028770s :      1: type_inference
 0.10% :     0.000134s :      1: validate


TotalTime = 0.0735544, [33]
[bootstrap]: 0.00041064
[type_inference]: 0.0428109
[event_method]: 2.451e-05
[auto_monad]: 0.00027454
[graph_reusing]: 7.58001e-06
[pre_auto_parallel]: 3.38e-06
[py_interpret_to_execute]: 0.000292
[rewriter_before_opt_a]: 0.00011247
[expand_dump_flag]: 3.51001e-06
[jit_opt_a]: 0.0133497, [2]
    [Cycle 1]: 0.00358687, [27]
        [switch_simplify]: 7.796e-05
        [loop_unroll]: 3.594e-05
        [a_1]: 0.00119386
        [with_stream_mark]: 2.055e-05
        [recompute_prepare]: 1.259e-05
        [updatestate_depend_eliminate]: 5.57001e-06
        [updatestate_assign_eliminate]: 4.85999e-06
        [updatestate_loads_eliminate]: 4.1e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 9.99001e-06
        [updatestate_useless_node_eliminater]: 1.252e-05
        [accelerated_algorithm]: 9.97001e-06
        [meta_shard_fg_expand]: 4.69002e-06
        [get_grad_eliminate_]: 9.64999e-06
        [merge_forward]: 4.68999e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.047e-05
        [j_node_and_user_rematch]: 1.581e-05
        [meta_fg_expand]: 4.85001e-06
        [replace_old_param]: 1.26e-05
        [inline_without_move]: 9.02e-06
        [renormalize]: 0.00179901
        [add_forward_monad_depend]: 5.86998e-06
        [auto_monad_grad]: 1.79e-06
        [auto_monad_eliminator]: 2.072e-05
        [cse]: 3.589e-05
        [replace_applicator]: 1.862e-05
    [Cycle 2]: 0.00050765, [27]
        [switch_simplify]: 1.059e-05
        [loop_unroll]: 9.15999e-06
        [a_1]: 0.00018565
        [with_stream_mark]: 1.561e-05
        [recompute_prepare]: 9.04998e-06
        [updatestate_depend_eliminate]: 5.72999e-06
        [updatestate_assign_eliminate]: 4.92999e-06
        [updatestate_loads_eliminate]: 4.3e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 8.85999e-06
        [updatestate_useless_node_eliminater]: 1.147e-05
        [accelerated_algorithm]: 8.50999e-06
        [meta_shard_fg_expand]: 3.23e-06
        [get_grad_eliminate_]: 7.82e-06
        [merge_forward]: 4.67998e-06
        [cell_reuse_recompute_pass]: 2.56e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.661e-05
        [j_node_and_user_rematch]: 1.337e-05
        [meta_fg_expand]: 3.43999e-06
        [replace_old_param]: 1.091e-05
        [inline_without_move]: 7.78001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 1.03e-05
        [cse]: 2.211e-05
        [replace_applicator]: 7.68001e-06
[py_interpret_to_execute_after_opt_a]: 1.518e-05
[rewriter_after_opt_a]: 0.00030816
[convert_after_rewriter]: 1.116e-05
[order_py_execute_after_rewriter]: 7.28e-06
[mutable_eliminate]: 0.00077882
[jit_opt_b]: 7.712e-05, [1]
    [Cycle 1]: 6.922e-05, [2]
        [frontend_op_eliminate]: 2.969e-05
        [inline_after_opt_a]: 2.654e-05
[cconv]: 2.504e-05
[loop_unroll]: 0.00047933
[jit_opt_after_cconv]: 0.00020654, [1]
    [Cycle 1]: 0.00019924, [11]
        [c_1]: 5.068e-05
        [parameter_eliminate]: 2.94001e-06
        [updatestate_depend_eliminate]: 8.59e-06
        [updatestate_assign_eliminate]: 4.42e-06
        [updatestate_loads_eliminate]: 4e-06
        [cse]: 2.858e-05
        [call_graph_tuple_transform]: 2.323e-05
        [tuple_list_get_item_eliminator]: 8.77e-06
        [none_parameter_eliminate]: 1.25999e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 9.20999e-06
[remove_dup_value]: 1.587e-05
[partial_unused_args_eliminate]: 1.89e-06
[environ_conv]: 1.162e-05
[add_recomputation]: 0.00011005
[cse_after_recomputation]: 3.805e-05, [1]
    [Cycle 1]: 3.121e-05, [1]
        [cse]: 2.301e-05
[auto_monad_reorder]: 2.502e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 7.44002e-06
[opt_after_jit_grad]: 0.00054604
[symbol_engine_optimizer]: 0.00011818, [1]
    [Cycle 1]: 0.00011084, [6]
        [build]: 1.601e-05
        [elim_shapecalc]: 1.325e-05
        [elim_not_effective]: 2.204e-05
        [opt_reshape]: 9.73002e-06
        [fold_const_symbol]: 1.739e-05
        [renormalize]: 7.80012e-07
[validate]: 6.282e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.0131435
[execute]: 8.23001e-06
Sums
    bootstrap                            :     0.000411s :  0.65%
    type_inference                       :     0.042811s : 67.37%
    event_method                         :     0.000025s :  0.04%
    auto_monad                           :     0.000275s :  0.43%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000292s :  0.46%
    rewriter_before_opt_a                :     0.000112s :  0.18%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000089s :  0.14%
    jit_opt_a.loop_unroll                :     0.000045s :  0.07%
    jit_opt_a.a_1                        :     0.001380s :  2.17%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.00%
    jit_opt_a.specialize_transform       :     0.000019s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.03%
    jit_opt_a.merge_forward              :     0.000009s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000008s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.04%
    jit_opt_a.inline_without_move        :     0.000017s :  0.03%
    jit_opt_a.renormalize                :     0.001799s :  2.83%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000031s :  0.05%
    jit_opt_a.cse                        :     0.000058s :  0.09%
    jit_opt_a.replace_applicator         :     0.000026s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.02%
    rewriter_after_opt_a                 :     0.000308s :  0.48%
    convert_after_rewriter               :     0.000011s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000779s :  1.23%
    jit_opt_b.frontend_op_eliminate      :     0.000030s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.04%
    cconv                                :     0.000025s :  0.04%
    loop_unroll                          :     0.000479s :  0.75%
    jit_opt_after_cconv.c_1              :     0.000051s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    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              :     0.000029s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.01%
    remove_dup_value                     :     0.000016s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000012s :  0.02%
    add_recomputation                    :     0.000110s :  0.17%
    cse_after_recomputation.cse          :     0.000023s :  0.04%
    auto_monad_reorder                   :     0.000025s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000546s :  0.86%
    symbol_engine_optimizer.build        :     0.000016s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000017s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000063s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.013143s : 20.68%
    execute                              :     0.000008s :  0.01%


Time group info:
------[substitution.]   0.000452    47
 2.45% :     0.000011s :      2: substitution.depend_value_elim
 0.75% :     0.000003s :      4: substitution.elim_not_effective
 0.74% :     0.000003s :      4: substitution.fold_const_symbol
 1.24% :     0.000006s :      5: substitution.graph_param_transform
81.88% :     0.000370s :      4: substitution.inline
 1.27% :     0.000006s :      8: substitution.j_node_and_user_rematch
 1.63% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.14% :     0.000005s :      2: substitution.replace_old_param
 3.59% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
 2.93% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 2.39% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.042698     2
65.57% :     0.027996s :      1: type_inference.infer
34.43% :     0.014702s :      1: type_inference.specialize
------[replace.]   0.000255     6
80.44% :     0.000205s :      4: replace.inline
19.56% :     0.000050s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000381     6
96.15% :     0.000366s :      4: match.inline
 3.85% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000219  1021
 1.57% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.08% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.02% :     0.000002s :     15: predicate.addn_check_dump
 1.65% :     0.000004s :     15: predicate.addn_zero_filter
 2.70% :     0.000006s :     15: predicate.arithmetic_simplify
 1.82% :     0.000004s :     15: predicate.cast_eliminate
 0.46% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     15: predicate.compare_switch_simplify
 1.68% :     0.000004s :     15: predicate.depend_value_elim
 1.13% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.24% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.22% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.76% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.27% :     0.000001s :      5: predicate.elim_not_effective
 0.64% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000003s :     15: predicate.environ_add_const_eliminate
 1.20% :     0.000003s :     15: predicate.environ_get_add_eliminate
 1.01% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.65% :     0.000004s :     15: predicate.environ_get_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      5: predicate.fold_const_symbol
 1.11% :     0.000002s :     10: predicate.get_grad_eliminate
 0.22% :     0.000000s :      5: predicate.graph_param_transform
 5.02% :     0.000011s :     31: predicate.inline
 0.82% :     0.000002s :     10: predicate.inline_without_move
 0.33% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.27% :     0.000003s :     10: predicate.less_batch_normalization
 1.92% :     0.000004s :     17: predicate.list_to_tuple_eliminator_
 1.94% :     0.000004s :     22: predicate.load_eliminater
 1.07% :     0.000002s :      5: predicate.loop_unroll_after_grad
 2.73% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.02% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000002s :     15: predicate.merge_addn
 1.08% :     0.000002s :     15: predicate.minmaximum_grad
 1.17% :     0.000003s :      5: predicate.mutable_eliminate
 0.43% :     0.000001s :      5: predicate.opt_reshape
 1.93% :     0.000004s :     22: predicate.partial_eliminate
 1.20% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.93% :     0.000004s :     15: predicate.reduce_eliminate
 1.52% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.38% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.40% :     0.000003s :     27: predicate.replace_applicator
 0.59% :     0.000001s :     10: predicate.replace_old_param
 0.27% :     0.000001s :      5: predicate.reset_defer_inline
 1.58% :     0.000003s :     15: predicate.reshape_eliminate
 1.53% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.75% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.34% :     0.000003s :     15: predicate.same_eliminate
 0.46% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.80% :     0.000002s :     10: predicate.special_op_eliminate
 0.85% :     0.000002s :     10: predicate.specialize_transform
 1.55% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 4.37% :     0.000010s :     21: predicate.switch_defer_inline
 2.10% :     0.000005s :     21: predicate.switch_layer_defer_inline
 5.89% :     0.000013s :     68: predicate.switch_simplify
 1.40% :     0.000003s :     15: predicate.tile_eliminate
 1.28% :     0.000003s :     15: predicate.transpose_eliminate
 1.68% :     0.000004s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000004s :     15: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 2.33% :     0.000005s :     15: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.65% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 2.06% :     0.000005s :     15: predicate.value_based_eliminate
 0.33% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.78% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000990    13
43.49% :     0.000430s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.26% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.24% :     0.000497s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.077259    76
 0.15% :     0.000114s :      1: add_recomputation
 0.36% :     0.000280s :      1: auto_monad
 0.04% :     0.000028s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.56% :     0.000432s :      1: bootstrap
 0.04% :     0.000028s :      1: cconv
 0.02% :     0.000014s :      1: convert_after_rewriter
 0.05% :     0.000041s :      1: cse_after_recomputation
 0.02% :     0.000014s :      1: environ_conv
 0.04% :     0.000030s :      1: event_method
 0.02% :     0.000013s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
17.29% :     0.013354s :      1: jit_opt_a
 0.27% :     0.000210s :      1: jit_opt_after_cconv
 0.10% :     0.000080s :      1: jit_opt_b
 0.63% :     0.000488s :      1: loop_unroll
 1.02% :     0.000790s :      1: mutable_eliminate
 2.21% :     0.001710s :     26: opt.transform.jit_opt_a
 0.11% :     0.000088s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000048s :      4: opt.transform.jit_opt_b
 0.02% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000058s :      4: opt.transform.symbol_engine_opt
 0.72% :     0.000557s :      1: opt_after_jit_grad
 0.01% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.38% :     0.000297s :      1: py_interpret_to_execute
 0.02% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000019s :      1: remove_dup_value
 1.48% :     0.001140s :      1: renormalize.infer
 0.84% :     0.000648s :      1: renormalize.specialize
 0.01% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.40% :     0.000313s :      1: rewriter_after_opt_a
 0.15% :     0.000117s :      1: rewriter_before_opt_a
 0.16% :     0.000121s :      1: symbol_engine_optimizer
17.04% :     0.013163s :      1: task_emit
55.44% :     0.042833s :      1: type_inference
 0.11% :     0.000089s :      1: validate


TotalTime = 0.060612, [33]
[bootstrap]: 0.00043227
[type_inference]: 0.0332238
[event_method]: 2.544e-05
[auto_monad]: 0.00019687
[graph_reusing]: 7.08998e-06
[pre_auto_parallel]: 3.36001e-06
[py_interpret_to_execute]: 0.00016853
[rewriter_before_opt_a]: 0.00010851
[expand_dump_flag]: 3.86001e-06
[jit_opt_a]: 0.00878431, [2]
    [Cycle 1]: 0.00297549, [27]
        [switch_simplify]: 8.49e-05
        [loop_unroll]: 3.972e-05
        [a_1]: 0.00087802
        [with_stream_mark]: 3.23e-05
        [recompute_prepare]: 1.413e-05
        [updatestate_depend_eliminate]: 8.50001e-06
        [updatestate_assign_eliminate]: 5.81e-06
        [updatestate_loads_eliminate]: 4.67e-06
        [parameter_eliminate]: 2.70997e-06
        [specialize_transform]: 1.031e-05
        [updatestate_useless_node_eliminater]: 1.414e-05
        [accelerated_algorithm]: 9.39998e-06
        [meta_shard_fg_expand]: 3.14999e-06
        [get_grad_eliminate_]: 8.58001e-06
        [merge_forward]: 5.59998e-06
        [cell_reuse_recompute_pass]: 1.66998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.278e-05
        [j_node_and_user_rematch]: 1.468e-05
        [meta_fg_expand]: 3.70998e-06
        [replace_old_param]: 1.407e-05
        [inline_without_move]: 9.30001e-06
        [renormalize]: 0.00143681
        [add_forward_monad_depend]: 8.75999e-06
        [auto_monad_grad]: 2.68e-06
        [auto_monad_eliminator]: 2.691e-05
        [cse]: 4.409e-05
        [replace_applicator]: 2.463e-05
    [Cycle 2]: 0.00110091, [27]
        [switch_simplify]: 9.41998e-06
        [loop_unroll]: 9.02e-06
        [a_1]: 0.00018217
        [with_stream_mark]: 1.979e-05
        [recompute_prepare]: 9.66998e-06
        [updatestate_depend_eliminate]: 5.91e-06
        [updatestate_assign_eliminate]: 5.22999e-06
        [updatestate_loads_eliminate]: 4.28999e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 9.61e-06
        [updatestate_useless_node_eliminater]: 1.171e-05
        [accelerated_algorithm]: 2.543e-05
        [meta_shard_fg_expand]: 3.87998e-06
        [get_grad_eliminate_]: 9.37001e-06
        [merge_forward]: 9.96e-06
        [cell_reuse_recompute_pass]: 3.75e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.459e-05
        [j_node_and_user_rematch]: 1.472e-05
        [meta_fg_expand]: 3.51001e-06
        [replace_old_param]: 1.192e-05
        [inline_without_move]: 8.06001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 4.82998e-06
        [auto_monad_grad]: 1.67999e-06
        [auto_monad_eliminator]: 1.957e-05
        [cse]: 3.583e-05
        [replace_applicator]: 9.56e-06
[py_interpret_to_execute_after_opt_a]: 2.1e-05
[rewriter_after_opt_a]: 0.00032389
[convert_after_rewriter]: 1.618e-05
[order_py_execute_after_rewriter]: 8.62998e-06
[mutable_eliminate]: 0.00092634
[jit_opt_b]: 7.821e-05, [1]
    [Cycle 1]: 6.925e-05, [2]
        [frontend_op_eliminate]: 2.566e-05
        [inline_after_opt_a]: 2.908e-05
[cconv]: 4.113e-05
[loop_unroll]: 0.00059284
[jit_opt_after_cconv]: 0.00024555, [1]
    [Cycle 1]: 0.00023674, [11]
        [c_1]: 5.563e-05
        [parameter_eliminate]: 6.41e-06
        [updatestate_depend_eliminate]: 1.196e-05
        [updatestate_assign_eliminate]: 5.46002e-06
        [updatestate_loads_eliminate]: 4.78001e-06
        [cse]: 4.922e-05
        [call_graph_tuple_transform]: 2.526e-05
        [tuple_list_get_item_eliminator]: 8.22998e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 9.99979e-07
        [switch_simplify]: 1.025e-05
[remove_dup_value]: 2.295e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 1.497e-05
[add_recomputation]: 7.616e-05
[cse_after_recomputation]: 3.693e-05, [1]
    [Cycle 1]: 2.94e-05, [1]
        [cse]: 2.212e-05
[auto_monad_reorder]: 2.99e-05
[get_jit_bprop_graph]: 2.69001e-06
[rewriter_after_jit_bprop_graph]: 1.135e-05
[opt_after_jit_grad]: 0.00064816
[symbol_engine_optimizer]: 0.00012744, [1]
    [Cycle 1]: 0.00011757, [6]
        [build]: 2.136e-05
        [elim_shapecalc]: 1.427e-05
        [elim_not_effective]: 2.225e-05
        [opt_reshape]: 9.62999e-06
        [fold_const_symbol]: 1.523e-05
        [renormalize]: 7.60017e-07
[validate]: 7.389e-05
[backend_pass]: 1.01002e-06
[task_emit]: 0.0140049
[execute]: 1.165e-05
Sums
    bootstrap                            :     0.000432s :  0.79%
    type_inference                       :     0.033224s : 60.93%
    event_method                         :     0.000025s :  0.05%
    auto_monad                           :     0.000197s :  0.36%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000169s :  0.31%
    rewriter_before_opt_a                :     0.000109s :  0.20%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000094s :  0.17%
    jit_opt_a.loop_unroll                :     0.000049s :  0.09%
    jit_opt_a.a_1                        :     0.001060s :  1.94%
    jit_opt_a.with_stream_mark           :     0.000052s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000020s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000035s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000018s :  0.03%
    jit_opt_a.merge_forward              :     0.000016s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000047s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000026s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.03%
    jit_opt_a.renormalize                :     0.001437s :  2.64%
    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      :     0.000046s :  0.09%
    jit_opt_a.cse                        :     0.000080s :  0.15%
    jit_opt_a.replace_applicator         :     0.000034s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.04%
    rewriter_after_opt_a                 :     0.000324s :  0.59%
    convert_after_rewriter               :     0.000016s :  0.03%
    order_py_execute_after_rewriter      :     0.000009s :  0.02%
    mutable_eliminate                    :     0.000926s :  1.70%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000029s :  0.05%
    cconv                                :     0.000041s :  0.08%
    loop_unroll                          :     0.000593s :  1.09%
    jit_opt_after_cconv.c_1              :     0.000056s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000049s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000023s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000015s :  0.03%
    add_recomputation                    :     0.000076s :  0.14%
    cse_after_recomputation.cse          :     0.000022s :  0.04%
    auto_monad_reorder                   :     0.000030s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.02%
    opt_after_jit_grad                   :     0.000648s :  1.19%
    symbol_engine_optimizer.build        :     0.000021s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000074s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.014005s : 25.68%
    execute                              :     0.000012s :  0.02%


Time group info:
------[substitution.]   0.000392    47
 2.85% :     0.000011s :      2: substitution.depend_value_elim
 0.80% :     0.000003s :      4: substitution.elim_not_effective
 0.56% :     0.000002s :      4: substitution.fold_const_symbol
 1.82% :     0.000007s :      5: substitution.graph_param_transform
77.63% :     0.000305s :      4: substitution.inline
 1.80% :     0.000007s :      8: substitution.j_node_and_user_rematch
 2.26% :     0.000009s :      8: substitution.remove_not_recompute_node
 1.76% :     0.000007s :      2: substitution.replace_old_param
 3.83% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.57% :     0.000014s :      3: substitution.updatestate_pure_node_eliminater
 3.12% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.033114     2
93.80% :     0.031061s :      1: type_inference.infer
 6.20% :     0.002052s :      1: type_inference.specialize
------[replace.]   0.000117     6
64.34% :     0.000075s :      4: replace.inline
35.66% :     0.000042s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000315     6
95.59% :     0.000301s :      4: match.inline
 4.41% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000202  1021
 1.24% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.32% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.08% :     0.000002s :     15: predicate.addn_check_dump
 1.54% :     0.000003s :     15: predicate.addn_zero_filter
 2.40% :     0.000005s :     15: predicate.arithmetic_simplify
 1.79% :     0.000004s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.04% :     0.000002s :     15: predicate.compare_switch_simplify
 1.26% :     0.000003s :     15: predicate.depend_value_elim
 1.02% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.63% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.15% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.27% :     0.000001s :      5: predicate.elim_not_effective
 0.74% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.07% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.33% :     0.000003s :     15: predicate.environ_get_eliminate
 1.01% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      5: predicate.fold_const_symbol
 1.19% :     0.000002s :     10: predicate.get_grad_eliminate
 0.36% :     0.000001s :      5: predicate.graph_param_transform
 5.10% :     0.000010s :     31: predicate.inline
 1.18% :     0.000002s :     10: predicate.inline_without_move
 0.36% :     0.000001s :     10: predicate.j_node_and_user_rematch
 2.39% :     0.000005s :     10: predicate.less_batch_normalization
 1.24% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.91% :     0.000004s :     22: predicate.load_eliminater
 1.20% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.35% :     0.000007s :     42: predicate.loop_unroll_before_grad
 1.83% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000002s :     15: predicate.merge_addn
 1.14% :     0.000002s :     15: predicate.minmaximum_grad
 2.14% :     0.000004s :      5: predicate.mutable_eliminate
 0.51% :     0.000001s :      5: predicate.opt_reshape
 2.04% :     0.000004s :     22: predicate.partial_eliminate
 1.34% :     0.000003s :     15: predicate.print_const_string_wrapper
 2.03% :     0.000004s :     15: predicate.reduce_eliminate
 1.58% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.51% :     0.000003s :     27: predicate.replace_applicator
 0.61% :     0.000001s :     10: predicate.replace_old_param
 0.26% :     0.000001s :      5: predicate.reset_defer_inline
 1.26% :     0.000003s :     15: predicate.reshape_eliminate
 1.20% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.87% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.52% :     0.000003s :     15: predicate.same_eliminate
 0.48% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.12% :     0.000002s :     10: predicate.special_op_eliminate
 1.19% :     0.000002s :     10: predicate.specialize_transform
 1.47% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.96% :     0.000004s :     21: predicate.switch_defer_inline
 2.01% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.29% :     0.000013s :     68: predicate.switch_simplify
 1.17% :     0.000002s :     15: predicate.tile_eliminate
 1.22% :     0.000002s :     15: predicate.transpose_eliminate
 1.60% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 2.01% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.58% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.36% :     0.000007s :     32: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000004s :     15: predicate.value_based_eliminate
 0.36% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001164    13
42.02% :     0.000489s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.14% :     0.000071s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.84% :     0.000603s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.063692    76
 0.13% :     0.000080s :      1: add_recomputation
 0.32% :     0.000205s :      1: auto_monad
 0.05% :     0.000034s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.72% :     0.000458s :      1: bootstrap
 0.07% :     0.000045s :      1: cconv
 0.03% :     0.000020s :      1: convert_after_rewriter
 0.06% :     0.000040s :      1: cse_after_recomputation
 0.03% :     0.000018s :      1: environ_conv
 0.05% :     0.000031s :      1: event_method
 0.03% :     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
13.80% :     0.008789s :      1: jit_opt_a
 0.39% :     0.000249s :      1: jit_opt_after_cconv
 0.13% :     0.000082s :      1: jit_opt_b
 0.95% :     0.000606s :      1: loop_unroll
 1.48% :     0.000941s :      1: mutable_eliminate
 2.25% :     0.001435s :     26: opt.transform.jit_opt_a
 0.15% :     0.000095s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000045s :      4: opt.transform.jit_opt_b
 0.03% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000030s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000038s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000057s :      4: opt.transform.symbol_engine_opt
 1.07% :     0.000679s :      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
 0.28% :     0.000176s :      1: py_interpret_to_execute
 0.04% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000026s :      1: remove_dup_value
 1.35% :     0.000863s :      1: renormalize.infer
 0.88% :     0.000560s :      1: renormalize.specialize
 0.02% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.52% :     0.000334s :      1: rewriter_after_opt_a
 0.18% :     0.000113s :      1: rewriter_before_opt_a
 0.21% :     0.000131s :      1: symbol_engine_optimizer
22.03% :     0.014030s :      1: task_emit
52.21% :     0.033252s :      1: type_inference
 0.17% :     0.000105s :      1: validate


TotalTime = 0.0910374, [33]
[bootstrap]: 0.00051527
[type_inference]: 0.0506736
[event_method]: 2.511e-05
[auto_monad]: 0.00018708
[graph_reusing]: 7.2e-06
[pre_auto_parallel]: 3.35003e-06
[py_interpret_to_execute]: 0.00014027
[rewriter_before_opt_a]: 0.00010489
[expand_dump_flag]: 3.86999e-06
[jit_opt_a]: 0.0188847, [2]
    [Cycle 1]: 0.00464625, [27]
        [switch_simplify]: 0.00016675
        [loop_unroll]: 3.861e-05
        [a_1]: 0.00098225
        [with_stream_mark]: 2.521e-05
        [recompute_prepare]: 1.337e-05
        [updatestate_depend_eliminate]: 7.08998e-06
        [updatestate_assign_eliminate]: 5.99e-06
        [updatestate_loads_eliminate]: 4.77e-06
        [parameter_eliminate]: 2.01998e-06
        [specialize_transform]: 9.74e-06
        [updatestate_useless_node_eliminater]: 1.209e-05
        [accelerated_algorithm]: 1.013e-05
        [meta_shard_fg_expand]: 3.58e-06
        [get_grad_eliminate_]: 8.85001e-06
        [merge_forward]: 6.31998e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.165e-05
        [j_node_and_user_rematch]: 1.613e-05
        [meta_fg_expand]: 3.61999e-06
        [replace_old_param]: 1.376e-05
        [inline_without_move]: 9.31e-06
        [renormalize]: 0.00289815
        [add_forward_monad_depend]: 1.435e-05
        [auto_monad_grad]: 2.86999e-06
        [auto_monad_eliminator]: 3.424e-05
        [cse]: 5.023e-05
        [replace_applicator]: 3.526e-05
    [Cycle 2]: 0.0005919, [27]
        [switch_simplify]: 1.282e-05
        [loop_unroll]: 1.032e-05
        [a_1]: 0.00022219
        [with_stream_mark]: 2.262e-05
        [recompute_prepare]: 1.006e-05
        [updatestate_depend_eliminate]: 6.59001e-06
        [updatestate_assign_eliminate]: 6.46e-06
        [updatestate_loads_eliminate]: 4.35e-06
        [parameter_eliminate]: 2.37001e-06
        [specialize_transform]: 9.56998e-06
        [updatestate_useless_node_eliminater]: 1.141e-05
        [accelerated_algorithm]: 1.062e-05
        [meta_shard_fg_expand]: 3.32002e-06
        [get_grad_eliminate_]: 8.55001e-06
        [merge_forward]: 5.81e-06
        [cell_reuse_recompute_pass]: 4.00998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.233e-05
        [j_node_and_user_rematch]: 1.448e-05
        [meta_fg_expand]: 4.52e-06
        [replace_old_param]: 1.267e-05
        [inline_without_move]: 8.54998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 2.37999e-06
        [auto_monad_grad]: 1.19998e-06
        [auto_monad_eliminator]: 1.17e-05
        [cse]: 2.537e-05
        [replace_applicator]: 8.20999e-06
[py_interpret_to_execute_after_opt_a]: 2.09e-05
[rewriter_after_opt_a]: 0.00033068
[convert_after_rewriter]: 1.279e-05
[order_py_execute_after_rewriter]: 8.10999e-06
[mutable_eliminate]: 0.00092632
[jit_opt_b]: 8.103e-05, [1]
    [Cycle 1]: 7.181e-05, [2]
        [frontend_op_eliminate]: 2.771e-05
        [inline_after_opt_a]: 3.051e-05
[cconv]: 3.681e-05
[loop_unroll]: 0.00049881
[jit_opt_after_cconv]: 0.00022927, [1]
    [Cycle 1]: 0.00022166, [11]
        [c_1]: 5.224e-05
        [parameter_eliminate]: 3.40998e-06
        [updatestate_depend_eliminate]: 1.154e-05
        [updatestate_assign_eliminate]: 4.55001e-06
        [updatestate_loads_eliminate]: 4.36002e-06
        [cse]: 3.719e-05
        [call_graph_tuple_transform]: 2.775e-05
        [tuple_list_get_item_eliminator]: 9.35001e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 8.99978e-07
        [switch_simplify]: 1.032e-05
[remove_dup_value]: 2.105e-05
[partial_unused_args_eliminate]: 2.47001e-06
[environ_conv]: 1.39e-05
[add_recomputation]: 7.34e-05
[cse_after_recomputation]: 3.484e-05, [1]
    [Cycle 1]: 2.727e-05, [1]
        [cse]: 1.969e-05
[auto_monad_reorder]: 2.698e-05
[get_jit_bprop_graph]: 1.92999e-06
[rewriter_after_jit_bprop_graph]: 7.43999e-06
[opt_after_jit_grad]: 0.00056907
[symbol_engine_optimizer]: 0.00011752, [1]
    [Cycle 1]: 0.00011046, [6]
        [build]: 1.412e-05
        [elim_shapecalc]: 1.215e-05
        [elim_not_effective]: 2.362e-05
        [opt_reshape]: 9.79e-06
        [fold_const_symbol]: 1.694e-05
        [renormalize]: 4.80009e-07
[validate]: 6.605e-05
[backend_pass]: 1.05999e-06
[task_emit]: 0.0171135
[execute]: 1.13e-05
Sums
    bootstrap                            :     0.000515s :  0.67%
    type_inference                       :     0.050674s : 66.17%
    event_method                         :     0.000025s :  0.03%
    auto_monad                           :     0.000187s :  0.24%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000140s :  0.18%
    rewriter_before_opt_a                :     0.000105s :  0.14%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000180s :  0.23%
    jit_opt_a.loop_unroll                :     0.000049s :  0.06%
    jit_opt_a.a_1                        :     0.001204s :  1.57%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000021s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.02%
    jit_opt_a.merge_forward              :     0.000012s :  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 :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000031s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000008s :  0.01%
    jit_opt_a.replace_old_param          :     0.000026s :  0.03%
    jit_opt_a.inline_without_move        :     0.000018s :  0.02%
    jit_opt_a.renormalize                :     0.002898s :  3.78%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000046s :  0.06%
    jit_opt_a.cse                        :     0.000076s :  0.10%
    jit_opt_a.replace_applicator         :     0.000043s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.03%
    rewriter_after_opt_a                 :     0.000331s :  0.43%
    convert_after_rewriter               :     0.000013s :  0.02%
    order_py_execute_after_rewriter      :     0.000008s :  0.01%
    mutable_eliminate                    :     0.000926s :  1.21%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000031s :  0.04%
    cconv                                :     0.000037s :  0.05%
    loop_unroll                          :     0.000499s :  0.65%
    jit_opt_after_cconv.c_1              :     0.000052s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  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              :     0.000037s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.01%
    remove_dup_value                     :     0.000021s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000014s :  0.02%
    add_recomputation                    :     0.000073s :  0.10%
    cse_after_recomputation.cse          :     0.000020s :  0.03%
    auto_monad_reorder                   :     0.000027s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000569s :  0.74%
    symbol_engine_optimizer.build        :     0.000014s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000024s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000017s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000066s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.017114s : 22.35%
    execute                              :     0.000011s :  0.01%


Time group info:
------[substitution.]   0.000496    47
 2.80% :     0.000014s :      2: substitution.depend_value_elim
 0.65% :     0.000003s :      4: substitution.elim_not_effective
 0.58% :     0.000003s :      4: substitution.fold_const_symbol
 1.47% :     0.000007s :      5: substitution.graph_param_transform
82.60% :     0.000410s :      4: substitution.inline
 1.27% :     0.000006s :      8: substitution.j_node_and_user_rematch
 1.74% :     0.000009s :      8: substitution.remove_not_recompute_node
 1.39% :     0.000007s :      2: substitution.replace_old_param
 2.91% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 2.20% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 2.40% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.050570     2
96.38% :     0.048741s :      1: type_inference.infer
 3.62% :     0.001829s :      1: type_inference.specialize
------[replace.]   0.000123     6
65.49% :     0.000081s :      4: replace.inline
34.51% :     0.000043s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000420     6
96.84% :     0.000407s :      4: match.inline
 3.16% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000203  1021
 1.30% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.31% :     0.000003s :     15: predicate.addn_check_dump
 1.36% :     0.000003s :     15: predicate.addn_zero_filter
 2.29% :     0.000005s :     15: predicate.arithmetic_simplify
 1.62% :     0.000003s :     15: predicate.cast_eliminate
 0.54% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     15: predicate.compare_switch_simplify
 1.29% :     0.000003s :     15: predicate.depend_value_elim
 1.07% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.26% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.27% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.37% :     0.000003s :     15: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     15: predicate.environ_get_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.22% :     0.000000s :      5: predicate.graph_param_transform
 5.46% :     0.000011s :     31: predicate.inline
 1.02% :     0.000002s :     10: predicate.inline_without_move
 0.35% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.60% :     0.000003s :     10: predicate.less_batch_normalization
 1.47% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.10% :     0.000004s :     22: predicate.load_eliminater
 1.33% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.02% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.25% :     0.000005s :     20: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     15: predicate.merge_addn
 1.13% :     0.000002s :     15: predicate.minmaximum_grad
 1.90% :     0.000004s :      5: predicate.mutable_eliminate
 0.44% :     0.000001s :      5: predicate.opt_reshape
 2.10% :     0.000004s :     22: predicate.partial_eliminate
 1.25% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.82% :     0.000004s :     15: predicate.reduce_eliminate
 1.46% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.75% :     0.000004s :     27: predicate.replace_applicator
 0.54% :     0.000001s :     10: predicate.replace_old_param
 0.46% :     0.000001s :      5: predicate.reset_defer_inline
 1.36% :     0.000003s :     15: predicate.reshape_eliminate
 1.33% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.90% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.36% :     0.000003s :     15: predicate.same_eliminate
 0.48% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.09% :     0.000002s :     10: predicate.special_op_eliminate
 0.97% :     0.000002s :     10: predicate.specialize_transform
 1.45% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.19% :     0.000004s :     21: predicate.switch_defer_inline
 1.89% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.70% :     0.000014s :     68: predicate.switch_simplify
 1.17% :     0.000002s :     15: predicate.tile_eliminate
 1.37% :     0.000003s :     15: predicate.transpose_eliminate
 1.45% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.62% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 2.40% :     0.000005s :     15: predicate.value_based_eliminate
 0.34% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002531    13
64.36% :     0.001629s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 2.53% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
33.11% :     0.000838s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.095770    76
 0.08% :     0.000077s :      1: add_recomputation
 0.20% :     0.000193s :      1: auto_monad
 0.03% :     0.000030s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.56% :     0.000540s :      1: bootstrap
 0.04% :     0.000040s :      1: cconv
 0.02% :     0.000016s :      1: convert_after_rewriter
 0.04% :     0.000037s :      1: cse_after_recomputation
 0.02% :     0.000016s :      1: environ_conv
 0.03% :     0.000031s :      1: event_method
 0.02% :     0.000017s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000006s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
19.72% :     0.018890s :      1: jit_opt_a
 0.24% :     0.000232s :      1: jit_opt_after_cconv
 0.09% :     0.000084s :      1: jit_opt_b
 0.53% :     0.000508s :      1: loop_unroll
 0.98% :     0.000940s :      1: mutable_eliminate
 1.73% :     0.001655s :     26: opt.transform.jit_opt_a
 0.10% :     0.000095s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000049s :      4: opt.transform.jit_opt_b
 0.02% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000058s :      4: opt.transform.symbol_engine_opt
 0.60% :     0.000579s :      1: opt_after_jit_grad
 0.01% :     0.000011s :      1: order_py_execute_after_rewriter
 0.00% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000145s :      1: py_interpret_to_execute
 0.02% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000024s :      1: remove_dup_value
 1.53% :     0.001463s :      1: renormalize.infer
 1.45% :     0.001393s :      1: renormalize.specialize
 0.01% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.35% :     0.000337s :      1: rewriter_after_opt_a
 0.11% :     0.000110s :      1: rewriter_before_opt_a
 0.13% :     0.000121s :      1: symbol_engine_optimizer
17.90% :     0.017140s :      1: task_emit
52.94% :     0.050696s :      1: type_inference
 0.10% :     0.000094s :      1: validate


TotalTime = 0.0912784, [33]
[bootstrap]: 0.00047593
[type_inference]: 0.0605418
[event_method]: 2.722e-05
[auto_monad]: 0.00020536
[graph_reusing]: 8.25e-06
[pre_auto_parallel]: 2.99999e-06
[py_interpret_to_execute]: 0.00016732
[rewriter_before_opt_a]: 0.00011419
[expand_dump_flag]: 3.71999e-06
[jit_opt_a]: 0.00980303, [2]
    [Cycle 1]: 0.00415922, [27]
        [switch_simplify]: 8.576e-05
        [loop_unroll]: 3.848e-05
        [a_1]: 0.00093614
        [with_stream_mark]: 3.45e-05
        [recompute_prepare]: 1.795e-05
        [updatestate_depend_eliminate]: 8.15999e-06
        [updatestate_assign_eliminate]: 6.73e-06
        [updatestate_loads_eliminate]: 5.17999e-06
        [parameter_eliminate]: 2.96001e-06
        [specialize_transform]: 1.28e-05
        [updatestate_useless_node_eliminater]: 1.606e-05
        [accelerated_algorithm]: 8.99e-06
        [meta_shard_fg_expand]: 3.80998e-06
        [get_grad_eliminate_]: 9.52999e-06
        [merge_forward]: 6.43e-06
        [cell_reuse_recompute_pass]: 2.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.372e-05
        [j_node_and_user_rematch]: 1.529e-05
        [meta_fg_expand]: 4.56002e-06
        [replace_old_param]: 1.427e-05
        [inline_without_move]: 8.38999e-06
        [renormalize]: 0.00247407
        [add_forward_monad_depend]: 1.385e-05
        [auto_monad_grad]: 3.55e-06
        [auto_monad_eliminator]: 3.265e-05
        [cse]: 5.031e-05
        [replace_applicator]: 3.364e-05
    [Cycle 2]: 0.00060719, [27]
        [switch_simplify]: 1.083e-05
        [loop_unroll]: 9.05001e-06
        [a_1]: 0.00020273
        [with_stream_mark]: 2.509e-05
        [recompute_prepare]: 1.018e-05
        [updatestate_depend_eliminate]: 7.01999e-06
        [updatestate_assign_eliminate]: 6.73e-06
        [updatestate_loads_eliminate]: 5.08002e-06
        [parameter_eliminate]: 3.02002e-06
        [specialize_transform]: 9.05001e-06
        [updatestate_useless_node_eliminater]: 1.266e-05
        [accelerated_algorithm]: 8.79e-06
        [meta_shard_fg_expand]: 2.71e-06
        [get_grad_eliminate_]: 8.50999e-06
        [merge_forward]: 7.32002e-06
        [cell_reuse_recompute_pass]: 4.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.034e-05
        [j_node_and_user_rematch]: 1.493e-05
        [meta_fg_expand]: 3.61999e-06
        [replace_old_param]: 1.297e-05
        [inline_without_move]: 7.70998e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 3.38e-06
        [auto_monad_grad]: 2.76999e-06
        [auto_monad_eliminator]: 1.719e-05
        [cse]: 3.052e-05
        [replace_applicator]: 9.93002e-06
[py_interpret_to_execute_after_opt_a]: 2.699e-05
[rewriter_after_opt_a]: 0.00042683
[convert_after_rewriter]: 1.945e-05
[order_py_execute_after_rewriter]: 9.52999e-06
[mutable_eliminate]: 0.00115819
[jit_opt_b]: 8.626e-05, [1]
    [Cycle 1]: 7.588e-05, [2]
        [frontend_op_eliminate]: 2.958e-05
        [inline_after_opt_a]: 3.096e-05
[cconv]: 4.252e-05
[loop_unroll]: 0.00079551
[jit_opt_after_cconv]: 0.00028463, [1]
    [Cycle 1]: 0.00027532, [11]
        [c_1]: 6.901e-05
        [parameter_eliminate]: 7.55e-06
        [updatestate_depend_eliminate]: 1.641e-05
        [updatestate_assign_eliminate]: 6.10002e-06
        [updatestate_loads_eliminate]: 5.20999e-06
        [cse]: 5.409e-05
        [call_graph_tuple_transform]: 2.858e-05
        [tuple_list_get_item_eliminator]: 9.69e-06
        [none_parameter_eliminate]: 2.47001e-06
        [renormalize]: 1.25001e-06
        [switch_simplify]: 1.17e-05
[remove_dup_value]: 2.748e-05
[partial_unused_args_eliminate]: 2.53e-06
[environ_conv]: 1.9e-05
[add_recomputation]: 8.55e-05
[cse_after_recomputation]: 3.812e-05, [1]
    [Cycle 1]: 2.832e-05, [1]
        [cse]: 2.057e-05
[auto_monad_reorder]: 3.215e-05
[get_jit_bprop_graph]: 2.64001e-06
[rewriter_after_jit_bprop_graph]: 1.549e-05
[opt_after_jit_grad]: 0.00097741
[symbol_engine_optimizer]: 0.00013279, [1]
    [Cycle 1]: 0.00012214, [6]
        [build]: 2.309e-05
        [elim_shapecalc]: 1.306e-05
        [elim_not_effective]: 2.842e-05
        [opt_reshape]: 9.49e-06
        [fold_const_symbol]: 1.46e-05
        [renormalize]: 1.05001e-06
[validate]: 8.324e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.0152903
[execute]: 9.17001e-06
Sums
    bootstrap                            :     0.000476s :  0.56%
    type_inference                       :     0.060542s : 70.99%
    event_method                         :     0.000027s :  0.03%
    auto_monad                           :     0.000205s :  0.24%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000167s :  0.20%
    rewriter_before_opt_a                :     0.000114s :  0.13%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000097s :  0.11%
    jit_opt_a.loop_unroll                :     0.000048s :  0.06%
    jit_opt_a.a_1                        :     0.001139s :  1.34%
    jit_opt_a.with_stream_mark           :     0.000060s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000028s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.01%
    jit_opt_a.specialize_transform       :     0.000022s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000029s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000018s :  0.02%
    jit_opt_a.merge_forward              :     0.000014s :  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 :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000008s :  0.01%
    jit_opt_a.replace_old_param          :     0.000027s :  0.03%
    jit_opt_a.inline_without_move        :     0.000016s :  0.02%
    jit_opt_a.renormalize                :     0.002474s :  2.90%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000050s :  0.06%
    jit_opt_a.cse                        :     0.000081s :  0.09%
    jit_opt_a.replace_applicator         :     0.000044s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000027s :  0.03%
    rewriter_after_opt_a                 :     0.000427s :  0.50%
    convert_after_rewriter               :     0.000019s :  0.02%
    order_py_execute_after_rewriter      :     0.000010s :  0.01%
    mutable_eliminate                    :     0.001158s :  1.36%
    jit_opt_b.frontend_op_eliminate      :     0.000030s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000031s :  0.04%
    cconv                                :     0.000043s :  0.05%
    loop_unroll                          :     0.000796s :  0.93%
    jit_opt_after_cconv.c_1              :     0.000069s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000008s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000016s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000054s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000010s :  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  :     0.000012s :  0.01%
    remove_dup_value                     :     0.000027s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000019s :  0.02%
    add_recomputation                    :     0.000086s :  0.10%
    cse_after_recomputation.cse          :     0.000021s :  0.02%
    auto_monad_reorder                   :     0.000032s :  0.04%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000015s :  0.02%
    opt_after_jit_grad                   :     0.000977s :  1.15%
    symbol_engine_optimizer.build        :     0.000023s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000028s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000083s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.015290s : 17.93%
    execute                              :     0.000009s :  0.01%


Time group info:
------[substitution.]   0.000436    47
 3.14% :     0.000014s :      2: substitution.depend_value_elim
 0.73% :     0.000003s :      4: substitution.elim_not_effective
 0.66% :     0.000003s :      4: substitution.fold_const_symbol
 1.93% :     0.000008s :      5: substitution.graph_param_transform
76.75% :     0.000335s :      4: substitution.inline
 1.44% :     0.000006s :      8: substitution.j_node_and_user_rematch
 1.98% :     0.000009s :      8: substitution.remove_not_recompute_node
 1.60% :     0.000007s :      2: substitution.replace_old_param
 3.75% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
 4.40% :     0.000019s :      3: substitution.updatestate_pure_node_eliminater
 3.61% :     0.000016s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.060413     2
96.88% :     0.058530s :      1: type_inference.infer
 3.12% :     0.001884s :      1: type_inference.specialize
------[replace.]   0.000129     6
64.55% :     0.000084s :      4: replace.inline
35.45% :     0.000046s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000346     6
95.61% :     0.000331s :      4: match.inline
 4.39% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000217  1021
 1.89% :     0.000004s :     15: predicate.accumulaten_eliminater
 1.72% :     0.000004s :      5: predicate.ad_related_special_op_eliminate
 0.90% :     0.000002s :     15: predicate.addn_check_dump
 1.14% :     0.000002s :     15: predicate.addn_zero_filter
 2.08% :     0.000005s :     15: predicate.arithmetic_simplify
 1.31% :     0.000003s :     15: predicate.cast_eliminate
 0.42% :     0.000001s :      5: predicate.check_bprop_eliminate
 0.94% :     0.000002s :     15: predicate.compare_switch_simplify
 1.30% :     0.000003s :     15: predicate.depend_value_elim
 0.95% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.04% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.40% :     0.000003s :      5: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      5: predicate.elim_not_effective
 0.60% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000002s :     15: predicate.environ_add_const_eliminate
 0.90% :     0.000002s :     15: predicate.environ_get_add_eliminate
 0.90% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.48% :     0.000003s :     15: predicate.environ_get_eliminate
 0.89% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      5: predicate.fold_const_symbol
 1.01% :     0.000002s :     10: predicate.get_grad_eliminate
 0.23% :     0.000001s :      5: predicate.graph_param_transform
 5.38% :     0.000012s :     31: predicate.inline
 0.89% :     0.000002s :     10: predicate.inline_without_move
 0.35% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.39% :     0.000003s :     10: predicate.less_batch_normalization
 1.45% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.11% :     0.000005s :     22: predicate.load_eliminater
 2.51% :     0.000005s :      5: predicate.loop_unroll_after_grad
 3.07% :     0.000007s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000003s :     15: predicate.merge_addn
 1.04% :     0.000002s :     15: predicate.minmaximum_grad
 2.64% :     0.000006s :      5: predicate.mutable_eliminate
 0.51% :     0.000001s :      5: predicate.opt_reshape
 2.05% :     0.000004s :     22: predicate.partial_eliminate
 1.23% :     0.000003s :     15: predicate.print_const_string_wrapper
 2.20% :     0.000005s :     15: predicate.reduce_eliminate
 1.45% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.83% :     0.000004s :     27: predicate.replace_applicator
 0.56% :     0.000001s :     10: predicate.replace_old_param
 0.53% :     0.000001s :      5: predicate.reset_defer_inline
 1.60% :     0.000003s :     15: predicate.reshape_eliminate
 1.13% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.18% :     0.000003s :      5: predicate.row_tensor_eliminate
 1.28% :     0.000003s :     15: predicate.same_eliminate
 0.78% :     0.000002s :     10: predicate.set_cell_output_no_recompute
 1.16% :     0.000003s :     10: predicate.special_op_eliminate
 1.03% :     0.000002s :     10: predicate.specialize_transform
 1.49% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.63% :     0.000004s :     15: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.05% :     0.000004s :     21: predicate.switch_defer_inline
 1.68% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.06% :     0.000013s :     68: predicate.switch_simplify
 1.16% :     0.000003s :     15: predicate.tile_eliminate
 1.32% :     0.000003s :     15: predicate.transpose_eliminate
 1.25% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000009s :     27: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.47% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.19% :     0.000007s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000004s :     15: predicate.value_based_eliminate
 0.45% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001288    13
48.06% :     0.000619s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.09% :     0.000078s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.86% :     0.000591s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.095533    76
 0.09% :     0.000090s :      1: add_recomputation
 0.22% :     0.000212s :      1: auto_monad
 0.04% :     0.000035s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.53% :     0.000505s :      1: bootstrap
 0.05% :     0.000046s :      1: cconv
 0.02% :     0.000024s :      1: convert_after_rewriter
 0.04% :     0.000041s :      1: cse_after_recomputation
 0.02% :     0.000022s :      1: environ_conv
 0.03% :     0.000033s :      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% :     0.000011s :      1: graph_reusing
10.27% :     0.009809s :      1: jit_opt_a
 0.30% :     0.000289s :      1: jit_opt_after_cconv
 0.09% :     0.000090s :      1: jit_opt_b
 0.85% :     0.000814s :      1: loop_unroll
 1.24% :     0.001183s :      1: mutable_eliminate
 1.59% :     0.001515s :     26: opt.transform.jit_opt_a
 0.12% :     0.000114s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000050s :      4: opt.transform.jit_opt_b
 0.04% :     0.000034s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000041s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000056s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000061s :      4: opt.transform.symbol_engine_opt
 1.05% :     0.001000s :      1: opt_after_jit_grad
 0.01% :     0.000012s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000173s :      1: py_interpret_to_execute
 0.03% :     0.000031s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000030s :      1: remove_dup_value
 1.82% :     0.001735s :      1: renormalize.infer
 0.75% :     0.000721s :      1: renormalize.specialize
 0.02% :     0.000018s :      1: rewriter_after_jit_bprop_graph
 0.46% :     0.000439s :      1: rewriter_after_opt_a
 0.13% :     0.000121s :      1: rewriter_before_opt_a
 0.14% :     0.000137s :      1: symbol_engine_optimizer
16.03% :     0.015314s :      1: task_emit
63.40% :     0.060568s :      1: type_inference
 0.12% :     0.000119s :      1: validate


TotalTime = 0.0675436, [33]
[bootstrap]: 0.00051226
[type_inference]: 0.030858
[event_method]: 2.26e-05
[auto_monad]: 0.0001761
[graph_reusing]: 7.2e-06
[pre_auto_parallel]: 3.90998e-06
[py_interpret_to_execute]: 0.00019928
[rewriter_before_opt_a]: 0.00014046
[expand_dump_flag]: 3.56001e-06
[jit_opt_a]: 0.0160158, [2]
    [Cycle 1]: 0.0111402, [27]
        [switch_simplify]: 7.759e-05
        [loop_unroll]: 3.68e-05
        [a_1]: 0.00076895
        [with_stream_mark]: 2.511e-05
        [recompute_prepare]: 1.24e-05
        [updatestate_depend_eliminate]: 6.55002e-06
        [updatestate_assign_eliminate]: 6.39001e-06
        [updatestate_loads_eliminate]: 4.86002e-06
        [parameter_eliminate]: 2.03997e-06
        [specialize_transform]: 1.028e-05
        [updatestate_useless_node_eliminater]: 1.173e-05
        [accelerated_algorithm]: 9.98002e-06
        [meta_shard_fg_expand]: 3.56001e-06
        [get_grad_eliminate_]: 8.98002e-06
        [merge_forward]: 5.67001e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.138e-05
        [j_node_and_user_rematch]: 1.457e-05
        [meta_fg_expand]: 3.67002e-06
        [replace_old_param]: 1.194e-05
        [inline_without_move]: 8.29998e-06
        [renormalize]: 0.00959515
        [add_forward_monad_depend]: 1.378e-05
        [auto_monad_grad]: 2.79001e-06
        [auto_monad_eliminator]: 4.697e-05
        [cse]: 4.997e-05
        [replace_applicator]: 0.00013825
    [Cycle 2]: 0.00057954, [27]
        [switch_simplify]: 1.193e-05
        [loop_unroll]: 1.007e-05
        [a_1]: 0.00020485
        [with_stream_mark]: 2.378e-05
        [recompute_prepare]: 1.05e-05
        [updatestate_depend_eliminate]: 6.76e-06
        [updatestate_assign_eliminate]: 7.05e-06
        [updatestate_loads_eliminate]: 5.17999e-06
        [parameter_eliminate]: 2.27001e-06
        [specialize_transform]: 8.77e-06
        [updatestate_useless_node_eliminater]: 1.09e-05
        [accelerated_algorithm]: 1.03e-05
        [meta_shard_fg_expand]: 3.43e-06
        [get_grad_eliminate_]: 8.36002e-06
        [merge_forward]: 6.54999e-06
        [cell_reuse_recompute_pass]: 4.35e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.56e-05
        [j_node_and_user_rematch]: 1.392e-05
        [meta_fg_expand]: 3.33e-06
        [replace_old_param]: 1.321e-05
        [inline_without_move]: 9.81e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 2.00002e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 1.252e-05
        [cse]: 2.678e-05
        [replace_applicator]: 9.08002e-06
[py_interpret_to_execute_after_opt_a]: 3.774e-05
[rewriter_after_opt_a]: 0.00034591
[convert_after_rewriter]: 1.321e-05
[order_py_execute_after_rewriter]: 8.51002e-06
[mutable_eliminate]: 0.00246379
[jit_opt_b]: 8.513e-05, [1]
    [Cycle 1]: 7.326e-05, [2]
        [frontend_op_eliminate]: 2.859e-05
        [inline_after_opt_a]: 3.004e-05
[cconv]: 3.766e-05
[loop_unroll]: 0.0006496
[jit_opt_after_cconv]: 0.00024242, [1]
    [Cycle 1]: 0.00023433, [11]
        [c_1]: 5.44e-05
        [parameter_eliminate]: 4.62998e-06
        [updatestate_depend_eliminate]: 1.189e-05
        [updatestate_assign_eliminate]: 5.76998e-06
        [updatestate_loads_eliminate]: 4.55999e-06
        [cse]: 4.749e-05
        [call_graph_tuple_transform]: 2.676e-05
        [tuple_list_get_item_eliminator]: 9.39998e-06
        [none_parameter_eliminate]: 1.53002e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 9.97001e-06
[remove_dup_value]: 2.334e-05
[partial_unused_args_eliminate]: 2.35002e-06
[environ_conv]: 1.67e-05
[add_recomputation]: 7.983e-05
[cse_after_recomputation]: 3.243e-05, [1]
    [Cycle 1]: 2.547e-05, [1]
        [cse]: 1.813e-05
[auto_monad_reorder]: 2.856e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 1.061e-05
[opt_after_jit_grad]: 0.00059133
[symbol_engine_optimizer]: 0.00011784, [1]
    [Cycle 1]: 0.00011105, [6]
        [build]: 1.91e-05
        [elim_shapecalc]: 1.306e-05
        [elim_not_effective]: 2.142e-05
        [opt_reshape]: 9.14998e-06
        [fold_const_symbol]: 1.573e-05
        [renormalize]: 6.69999e-07
[validate]: 0.00080095
[backend_pass]: 1.59e-06
[task_emit]: 0.0136847
[execute]: 8.95001e-06
Sums
    bootstrap                            :     0.000512s :  0.82%
    type_inference                       :     0.030858s : 49.44%
    event_method                         :     0.000023s :  0.04%
    auto_monad                           :     0.000176s :  0.28%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000199s :  0.32%
    rewriter_before_opt_a                :     0.000140s :  0.23%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000090s :  0.14%
    jit_opt_a.loop_unroll                :     0.000047s :  0.08%
    jit_opt_a.a_1                        :     0.000974s :  1.56%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000020s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.03%
    jit_opt_a.merge_forward              :     0.000012s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000047s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000025s :  0.04%
    jit_opt_a.inline_without_move        :     0.000018s :  0.03%
    jit_opt_a.renormalize                :     0.009595s : 15.37%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000059s :  0.10%
    jit_opt_a.cse                        :     0.000077s :  0.12%
    jit_opt_a.replace_applicator         :     0.000147s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000038s :  0.06%
    rewriter_after_opt_a                 :     0.000346s :  0.55%
    convert_after_rewriter               :     0.000013s :  0.02%
    order_py_execute_after_rewriter      :     0.000009s :  0.01%
    mutable_eliminate                    :     0.002464s :  3.95%
    jit_opt_b.frontend_op_eliminate      :     0.000029s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000030s :  0.05%
    cconv                                :     0.000038s :  0.06%
    loop_unroll                          :     0.000650s :  1.04%
    jit_opt_after_cconv.c_1              :     0.000054s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000047s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000023s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000017s :  0.03%
    add_recomputation                    :     0.000080s :  0.13%
    cse_after_recomputation.cse          :     0.000018s :  0.03%
    auto_monad_reorder                   :     0.000029s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.02%
    opt_after_jit_grad                   :     0.000591s :  0.95%
    symbol_engine_optimizer.build        :     0.000019s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000801s :  1.28%
    backend_pass                         :     0.000002s :  0.00%
    task_emit                            :     0.013685s : 21.93%
    execute                              :     0.000009s :  0.01%


Time group info:
------[substitution.]   0.000325    47
 4.00% :     0.000013s :      2: substitution.depend_value_elim
 0.93% :     0.000003s :      4: substitution.elim_not_effective
 0.75% :     0.000002s :      4: substitution.fold_const_symbol
 2.15% :     0.000007s :      5: substitution.graph_param_transform
73.48% :     0.000238s :      4: substitution.inline
 1.78% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.63% :     0.000009s :      8: substitution.remove_not_recompute_node
 1.86% :     0.000006s :      2: substitution.replace_old_param
 4.46% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 4.25% :     0.000014s :      3: substitution.updatestate_pure_node_eliminater
 3.72% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.030715     2
94.61% :     0.029061s :      1: type_inference.infer
 5.39% :     0.001654s :      1: type_inference.specialize
------[replace.]   0.000102     6
66.50% :     0.000068s :      4: replace.inline
33.50% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000248     6
94.64% :     0.000235s :      4: match.inline
 5.36% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000208  1021
 1.85% :     0.000004s :     15: predicate.accumulaten_eliminater
 1.21% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 0.99% :     0.000002s :     15: predicate.addn_check_dump
 1.60% :     0.000003s :     15: predicate.addn_zero_filter
 2.31% :     0.000005s :     15: predicate.arithmetic_simplify
 1.52% :     0.000003s :     15: predicate.cast_eliminate
 0.53% :     0.000001s :      5: predicate.check_bprop_eliminate
 0.99% :     0.000002s :     15: predicate.compare_switch_simplify
 1.13% :     0.000002s :     15: predicate.depend_value_elim
 1.01% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.27% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.37% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.80% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.25% :     0.000001s :      5: predicate.elim_not_effective
 0.56% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000003s :     15: predicate.environ_add_const_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.02% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.40% :     0.000003s :     15: predicate.environ_get_eliminate
 1.03% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.33% :     0.000001s :      5: predicate.fold_const_symbol
 1.60% :     0.000003s :     10: predicate.get_grad_eliminate
 0.25% :     0.000001s :      5: predicate.graph_param_transform
 4.88% :     0.000010s :     31: predicate.inline
 1.01% :     0.000002s :     10: predicate.inline_without_move
 0.46% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.67% :     0.000003s :     10: predicate.less_batch_normalization
 1.46% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.01% :     0.000004s :     22: predicate.load_eliminater
 1.42% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.02% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     15: predicate.merge_addn
 1.03% :     0.000002s :     15: predicate.minmaximum_grad
 2.51% :     0.000005s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.12% :     0.000004s :     22: predicate.partial_eliminate
 1.63% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.93% :     0.000004s :     15: predicate.reduce_eliminate
 1.79% :     0.000004s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.65% :     0.000003s :     27: predicate.replace_applicator
 0.68% :     0.000001s :     10: predicate.replace_old_param
 0.47% :     0.000001s :      5: predicate.reset_defer_inline
 1.38% :     0.000003s :     15: predicate.reshape_eliminate
 1.44% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.81% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.36% :     0.000003s :     15: predicate.same_eliminate
 0.46% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.01% :     0.000002s :     10: predicate.special_op_eliminate
 0.96% :     0.000002s :     10: predicate.specialize_transform
 1.53% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.91% :     0.000004s :     21: predicate.switch_defer_inline
 1.72% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.03% :     0.000013s :     68: predicate.switch_simplify
 1.35% :     0.000003s :     15: predicate.tile_eliminate
 1.42% :     0.000003s :     15: predicate.transpose_eliminate
 1.57% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.58% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 2.12% :     0.000004s :     15: predicate.value_based_eliminate
 0.33% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.74% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001032    13
41.86% :     0.000432s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.61% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
53.53% :     0.000552s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.078684    76
 0.11% :     0.000083s :      1: add_recomputation
 0.23% :     0.000182s :      1: auto_monad
 0.04% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000006s :      1: backend_pass
 0.68% :     0.000536s :      1: bootstrap
 0.05% :     0.000041s :      1: cconv
 0.02% :     0.000016s :      1: convert_after_rewriter
 0.04% :     0.000035s :      1: cse_after_recomputation
 0.02% :     0.000019s :      1: environ_conv
 0.04% :     0.000028s :      1: event_method
 0.02% :     0.000014s :      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
20.36% :     0.016021s :      1: jit_opt_a
 0.31% :     0.000245s :      1: jit_opt_after_cconv
 0.11% :     0.000089s :      1: jit_opt_b
 0.84% :     0.000661s :      1: loop_unroll
 3.16% :     0.002487s :      1: mutable_eliminate
 1.70% :     0.001335s :     26: opt.transform.jit_opt_a
 0.12% :     0.000096s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000050s :      4: opt.transform.jit_opt_b
 0.03% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000036s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000055s :      4: opt.transform.symbol_engine_opt
 0.76% :     0.000601s :      1: opt_after_jit_grad
 0.01% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.26% :     0.000205s :      1: py_interpret_to_execute
 0.05% :     0.000041s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000026s :      1: remove_dup_value
11.18% :     0.008796s :      1: renormalize.infer
 0.99% :     0.000782s :      1: renormalize.specialize
 0.02% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.45% :     0.000351s :      1: rewriter_after_opt_a
 0.20% :     0.000157s :      1: rewriter_before_opt_a
 0.15% :     0.000121s :      1: symbol_engine_optimizer
17.42% :     0.013705s :      1: task_emit
39.24% :     0.030879s :      1: type_inference
 1.07% :     0.000844s :      1: validate


TotalTime = 0.0740244, [33]
[bootstrap]: 0.00040815
[type_inference]: 0.0378737
[event_method]: 2.295e-05
[auto_monad]: 0.00018879
[graph_reusing]: 6.96001e-06
[pre_auto_parallel]: 3.25e-06
[py_interpret_to_execute]: 0.00015713
[rewriter_before_opt_a]: 0.00010669
[expand_dump_flag]: 3.41001e-06
[jit_opt_a]: 0.0144685, [2]
    [Cycle 1]: 0.00691164, [27]
        [switch_simplify]: 7.288e-05
        [loop_unroll]: 3.712e-05
        [a_1]: 0.00077894
        [with_stream_mark]: 2.307e-05
        [recompute_prepare]: 1.073e-05
        [updatestate_depend_eliminate]: 6.01e-06
        [updatestate_assign_eliminate]: 5.22e-06
        [updatestate_loads_eliminate]: 5.29e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 1.02e-05
        [updatestate_useless_node_eliminater]: 1.154e-05
        [accelerated_algorithm]: 8.95001e-06
        [meta_shard_fg_expand]: 3.13e-06
        [get_grad_eliminate_]: 8.74e-06
        [merge_forward]: 5.16998e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.928e-05
        [j_node_and_user_rematch]: 1.422e-05
        [meta_fg_expand]: 3.76999e-06
        [replace_old_param]: 1.235e-05
        [inline_without_move]: 8.1e-06
        [renormalize]: 0.00555126
        [add_forward_monad_depend]: 1.004e-05
        [auto_monad_grad]: 2.75002e-06
        [auto_monad_eliminator]: 2.746e-05
        [cse]: 4.224e-05
        [replace_applicator]: 3.152e-05
    [Cycle 2]: 0.00059264, [27]
        [switch_simplify]: 1.29e-05
        [loop_unroll]: 1.043e-05
        [a_1]: 0.0002211
        [with_stream_mark]: 2.105e-05
        [recompute_prepare]: 1.062e-05
        [updatestate_depend_eliminate]: 6.23998e-06
        [updatestate_assign_eliminate]: 7.24001e-06
        [updatestate_loads_eliminate]: 4.66002e-06
        [parameter_eliminate]: 2.91e-06
        [specialize_transform]: 8.92999e-06
        [updatestate_useless_node_eliminater]: 1.292e-05
        [accelerated_algorithm]: 9.04e-06
        [meta_shard_fg_expand]: 3.68e-06
        [get_grad_eliminate_]: 8.17e-06
        [merge_forward]: 5.67999e-06
        [cell_reuse_recompute_pass]: 4.63001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.138e-05
        [j_node_and_user_rematch]: 1.324e-05
        [meta_fg_expand]: 4e-06
        [replace_old_param]: 1.241e-05
        [inline_without_move]: 7.65998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.72999e-06
        [auto_monad_grad]: 1.62001e-06
        [auto_monad_eliminator]: 1.291e-05
        [cse]: 2.712e-05
        [replace_applicator]: 9.18002e-06
[py_interpret_to_execute_after_opt_a]: 2.036e-05
[rewriter_after_opt_a]: 0.00034769
[convert_after_rewriter]: 1.426e-05
[order_py_execute_after_rewriter]: 8.12e-06
[mutable_eliminate]: 0.00092781
[jit_opt_b]: 8.304e-05, [1]
    [Cycle 1]: 7.244e-05, [2]
        [frontend_op_eliminate]: 2.924e-05
        [inline_after_opt_a]: 2.923e-05
[cconv]: 3.893e-05
[loop_unroll]: 0.00053435
[jit_opt_after_cconv]: 0.00023758, [1]
    [Cycle 1]: 0.00023012, [11]
        [c_1]: 5.423e-05
        [parameter_eliminate]: 4.52e-06
        [updatestate_depend_eliminate]: 1.19e-05
        [updatestate_assign_eliminate]: 4.47998e-06
        [updatestate_loads_eliminate]: 4.77998e-06
        [cse]: 4.524e-05
        [call_graph_tuple_transform]: 2.612e-05
        [tuple_list_get_item_eliminator]: 8.75999e-06
        [none_parameter_eliminate]: 1.64998e-06
        [renormalize]: 4.90021e-07
        [switch_simplify]: 1.171e-05
[remove_dup_value]: 2.374e-05
[partial_unused_args_eliminate]: 2.61e-06
[environ_conv]: 1.551e-05
[add_recomputation]: 7.705e-05
[cse_after_recomputation]: 3.34e-05, [1]
    [Cycle 1]: 2.572e-05, [1]
        [cse]: 1.936e-05
[auto_monad_reorder]: 3.033e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 7.60998e-06
[opt_after_jit_grad]: 0.0005724
[symbol_engine_optimizer]: 0.00012282, [1]
    [Cycle 1]: 0.00011482, [6]
        [build]: 1.816e-05
        [elim_shapecalc]: 1.307e-05
        [elim_not_effective]: 2.33e-05
        [opt_reshape]: 1.024e-05
        [fold_const_symbol]: 1.575e-05
        [renormalize]: 6.09987e-07
[validate]: 7.316e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.0172663
[execute]: 1.045e-05
Sums
    bootstrap                            :     0.000408s :  0.62%
    type_inference                       :     0.037874s : 57.17%
    event_method                         :     0.000023s :  0.03%
    auto_monad                           :     0.000189s :  0.28%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000157s :  0.24%
    rewriter_before_opt_a                :     0.000107s :  0.16%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000086s :  0.13%
    jit_opt_a.loop_unroll                :     0.000048s :  0.07%
    jit_opt_a.a_1                        :     0.001000s :  1.51%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.03%
    jit_opt_a.merge_forward              :     0.000011s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000008s :  0.01%
    jit_opt_a.replace_old_param          :     0.000025s :  0.04%
    jit_opt_a.inline_without_move        :     0.000016s :  0.02%
    jit_opt_a.renormalize                :     0.005551s :  8.38%
    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      :     0.000040s :  0.06%
    jit_opt_a.cse                        :     0.000069s :  0.10%
    jit_opt_a.replace_applicator         :     0.000041s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.03%
    rewriter_after_opt_a                 :     0.000348s :  0.52%
    convert_after_rewriter               :     0.000014s :  0.02%
    order_py_execute_after_rewriter      :     0.000008s :  0.01%
    mutable_eliminate                    :     0.000928s :  1.40%
    jit_opt_b.frontend_op_eliminate      :     0.000029s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000029s :  0.04%
    cconv                                :     0.000039s :  0.06%
    loop_unroll                          :     0.000534s :  0.81%
    jit_opt_after_cconv.c_1              :     0.000054s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000045s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000012s :  0.02%
    remove_dup_value                     :     0.000024s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000016s :  0.02%
    add_recomputation                    :     0.000077s :  0.12%
    cse_after_recomputation.cse          :     0.000019s :  0.03%
    auto_monad_reorder                   :     0.000030s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.01%
    opt_after_jit_grad                   :     0.000572s :  0.86%
    symbol_engine_optimizer.build        :     0.000018s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000073s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.017266s : 26.06%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000334    47
 3.98% :     0.000013s :      2: substitution.depend_value_elim
 1.05% :     0.000003s :      4: substitution.elim_not_effective
 0.89% :     0.000003s :      4: substitution.fold_const_symbol
 2.18% :     0.000007s :      5: substitution.graph_param_transform
75.88% :     0.000253s :      4: substitution.inline
 1.61% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.25% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.81% :     0.000006s :      2: substitution.replace_old_param
 3.66% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.44% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.25% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.037775     2
93.78% :     0.035426s :      1: type_inference.infer
 6.22% :     0.002350s :      1: type_inference.specialize
------[replace.]   0.000111     6
68.09% :     0.000075s :      4: replace.inline
31.91% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
95.73% :     0.000250s :      4: match.inline
 4.27% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000204  1021
 1.32% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.65% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     15: predicate.addn_check_dump
 1.39% :     0.000003s :     15: predicate.addn_zero_filter
 2.23% :     0.000005s :     15: predicate.arithmetic_simplify
 1.43% :     0.000003s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.00% :     0.000002s :     15: predicate.compare_switch_simplify
 1.20% :     0.000002s :     15: predicate.depend_value_elim
 1.24% :     0.000003s :     15: predicate.dict_get_item_const_eliminator
 1.06% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.81% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.52% :     0.000001s :      5: predicate.elim_not_effective
 0.51% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.00% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.45% :     0.000003s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      5: predicate.fold_const_symbol
 1.17% :     0.000002s :     10: predicate.get_grad_eliminate
 0.44% :     0.000001s :      5: predicate.graph_param_transform
 4.86% :     0.000010s :     31: predicate.inline
 0.86% :     0.000002s :     10: predicate.inline_without_move
 0.35% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.40% :     0.000003s :     10: predicate.less_batch_normalization
 1.65% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.59% :     0.000003s :     22: predicate.load_eliminater
 1.20% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.17% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.98% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000002s :     15: predicate.merge_addn
 1.02% :     0.000002s :     15: predicate.minmaximum_grad
 2.01% :     0.000004s :      5: predicate.mutable_eliminate
 0.49% :     0.000001s :      5: predicate.opt_reshape
 2.06% :     0.000004s :     22: predicate.partial_eliminate
 1.33% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.80% :     0.000004s :     15: predicate.reduce_eliminate
 1.49% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.95% :     0.000004s :     27: predicate.replace_applicator
 0.62% :     0.000001s :     10: predicate.replace_old_param
 0.28% :     0.000001s :      5: predicate.reset_defer_inline
 1.35% :     0.000003s :     15: predicate.reshape_eliminate
 1.23% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 1.19% :     0.000002s :      5: predicate.row_tensor_eliminate
 4.73% :     0.000010s :     15: predicate.same_eliminate
 0.88% :     0.000002s :     10: predicate.set_cell_output_no_recompute
 1.03% :     0.000002s :     10: predicate.special_op_eliminate
 0.96% :     0.000002s :     10: predicate.specialize_transform
 1.43% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.03% :     0.000004s :     21: predicate.switch_defer_inline
 1.71% :     0.000003s :     21: predicate.switch_layer_defer_inline
 5.79% :     0.000012s :     68: predicate.switch_simplify
 1.18% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000003s :     15: predicate.transpose_eliminate
 1.42% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.40% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.78% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.14% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000003s :     15: predicate.value_based_eliminate
 0.33% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001056    13
37.81% :     0.000399s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.44% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
57.75% :     0.000610s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.081133    76
 0.10% :     0.000081s :      1: add_recomputation
 0.24% :     0.000195s :      1: auto_monad
 0.04% :     0.000034s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.54% :     0.000434s :      1: bootstrap
 0.05% :     0.000042s :      1: cconv
 0.02% :     0.000018s :      1: convert_after_rewriter
 0.04% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000018s :      1: environ_conv
 0.03% :     0.000027s :      1: event_method
 0.02% :     0.000016s :      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
17.84% :     0.014474s :      1: jit_opt_a
 0.30% :     0.000241s :      1: jit_opt_after_cconv
 0.11% :     0.000086s :      1: jit_opt_b
 0.67% :     0.000547s :      1: loop_unroll
 1.16% :     0.000943s :      1: mutable_eliminate
 1.66% :     0.001344s :     26: opt.transform.jit_opt_a
 0.12% :     0.000095s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000049s :      4: opt.transform.jit_opt_b
 0.02% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000031s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000057s :      4: opt.transform.symbol_engine_opt
 0.72% :     0.000585s :      1: opt_after_jit_grad
 0.01% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000163s :      1: py_interpret_to_execute
 0.08% :     0.000062s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000027s :      1: remove_dup_value
 5.87% :     0.004763s :      1: renormalize.infer
 0.95% :     0.000774s :      1: renormalize.specialize
 0.01% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.44% :     0.000354s :      1: rewriter_after_opt_a
 0.14% :     0.000111s :      1: rewriter_before_opt_a
 0.16% :     0.000126s :      1: symbol_engine_optimizer
21.31% :     0.017289s :      1: task_emit
46.71% :     0.037897s :      1: type_inference
 0.13% :     0.000104s :      1: validate


TotalTime = 0.0924916, [33]
[bootstrap]: 0.00044739
[type_inference]: 0.0616285
[event_method]: 2.619e-05
[auto_monad]: 0.00019757
[graph_reusing]: 9.85e-05
[pre_auto_parallel]: 4.1e-06
[py_interpret_to_execute]: 0.00016242
[rewriter_before_opt_a]: 0.00011033
[expand_dump_flag]: 4.03999e-06
[jit_opt_a]: 0.00944189, [2]
    [Cycle 1]: 0.00397188, [27]
        [switch_simplify]: 8.521e-05
        [loop_unroll]: 3.875e-05
        [a_1]: 0.00090766
        [with_stream_mark]: 3.884e-05
        [recompute_prepare]: 1.95e-05
        [updatestate_depend_eliminate]: 7.96001e-06
        [updatestate_assign_eliminate]: 6.74999e-06
        [updatestate_loads_eliminate]: 5.27001e-06
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 1.225e-05
        [updatestate_useless_node_eliminater]: 1.524e-05
        [accelerated_algorithm]: 9.05999e-06
        [meta_shard_fg_expand]: 3.91001e-06
        [get_grad_eliminate_]: 9.76998e-06
        [merge_forward]: 6.51999e-06
        [cell_reuse_recompute_pass]: 2.83e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.41e-05
        [j_node_and_user_rematch]: 1.587e-05
        [meta_fg_expand]: 4.47998e-06
        [replace_old_param]: 1.488e-05
        [inline_without_move]: 8.69e-06
        [renormalize]: 0.00234576
        [add_forward_monad_depend]: 1.278e-05
        [auto_monad_grad]: 3.11001e-06
        [auto_monad_eliminator]: 3.09e-05
        [cse]: 4.854e-05
        [replace_applicator]: 2.832e-05
    [Cycle 2]: 0.00054436, [27]
        [switch_simplify]: 1.03e-05
        [loop_unroll]: 8.87999e-06
        [a_1]: 0.00019436
        [with_stream_mark]: 2.155e-05
        [recompute_prepare]: 8.33001e-06
        [updatestate_depend_eliminate]: 6.07001e-06
        [updatestate_assign_eliminate]: 6.74001e-06
        [updatestate_loads_eliminate]: 4.55001e-06
        [parameter_eliminate]: 2.41998e-06
        [specialize_transform]: 8.33999e-06
        [updatestate_useless_node_eliminater]: 1.091e-05
        [accelerated_algorithm]: 9.29e-06
        [meta_shard_fg_expand]: 3.51001e-06
        [get_grad_eliminate_]: 7.89997e-06
        [merge_forward]: 5.74e-06
        [cell_reuse_recompute_pass]: 4.06001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.157e-05
        [j_node_and_user_rematch]: 1.378e-05
        [meta_fg_expand]: 4.50001e-06
        [replace_old_param]: 1.173e-05
        [inline_without_move]: 7.53e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.64998e-06
        [auto_monad_grad]: 1.94999e-06
        [auto_monad_eliminator]: 1.227e-05
        [cse]: 2.496e-05
        [replace_applicator]: 8.40999e-06
[py_interpret_to_execute_after_opt_a]: 2.153e-05
[rewriter_after_opt_a]: 0.00031823
[convert_after_rewriter]: 1.478e-05
[order_py_execute_after_rewriter]: 8.82e-06
[mutable_eliminate]: 0.00093148
[jit_opt_b]: 7.788e-05, [1]
    [Cycle 1]: 6.846e-05, [2]
        [frontend_op_eliminate]: 2.603e-05
        [inline_after_opt_a]: 2.755e-05
[cconv]: 4.026e-05
[loop_unroll]: 0.00055907
[jit_opt_after_cconv]: 0.00030133, [1]
    [Cycle 1]: 0.00029333, [11]
        [c_1]: 5.381e-05
        [parameter_eliminate]: 5.09e-06
        [updatestate_depend_eliminate]: 1.335e-05
        [updatestate_assign_eliminate]: 5.81e-06
        [updatestate_loads_eliminate]: 4.67e-06
        [cse]: 0.00010449
        [call_graph_tuple_transform]: 2.56e-05
        [tuple_list_get_item_eliminator]: 8.57998e-06
        [none_parameter_eliminate]: 2.19999e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 1.085e-05
[remove_dup_value]: 2.054e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 1.654e-05
[add_recomputation]: 7.725e-05
[cse_after_recomputation]: 3.454e-05, [1]
    [Cycle 1]: 2.787e-05, [1]
        [cse]: 2.114e-05
[auto_monad_reorder]: 2.893e-05
[get_jit_bprop_graph]: 2.48e-06
[rewriter_after_jit_bprop_graph]: 8.79e-06
[opt_after_jit_grad]: 0.00067594
[symbol_engine_optimizer]: 0.00012337, [1]
    [Cycle 1]: 0.00011507, [6]
        [build]: 1.854e-05
        [elim_shapecalc]: 1.371e-05
        [elim_not_effective]: 2.295e-05
        [opt_reshape]: 1.038e-05
        [fold_const_symbol]: 1.638e-05
        [renormalize]: 9.70002e-07
[validate]: 7.725e-05
[backend_pass]: 1.74e-06
[task_emit]: 0.0167156
[execute]: 1.009e-05
Sums
    bootstrap                            :     0.000447s :  0.52%
    type_inference                       :     0.061628s : 71.05%
    event_method                         :     0.000026s :  0.03%
    auto_monad                           :     0.000198s :  0.23%
    graph_reusing                        :     0.000099s :  0.11%
    pre_auto_parallel                    :     0.000004s :  0.00%
    py_interpret_to_execute              :     0.000162s :  0.19%
    rewriter_before_opt_a                :     0.000110s :  0.13%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000096s :  0.11%
    jit_opt_a.loop_unroll                :     0.000048s :  0.05%
    jit_opt_a.a_1                        :     0.001102s :  1.27%
    jit_opt_a.with_stream_mark           :     0.000060s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000028s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000021s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000018s :  0.02%
    jit_opt_a.merge_forward              :     0.000012s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000046s :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000009s :  0.01%
    jit_opt_a.replace_old_param          :     0.000027s :  0.03%
    jit_opt_a.inline_without_move        :     0.000016s :  0.02%
    jit_opt_a.renormalize                :     0.002346s :  2.70%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000043s :  0.05%
    jit_opt_a.cse                        :     0.000074s :  0.08%
    jit_opt_a.replace_applicator         :     0.000037s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000022s :  0.02%
    rewriter_after_opt_a                 :     0.000318s :  0.37%
    convert_after_rewriter               :     0.000015s :  0.02%
    order_py_execute_after_rewriter      :     0.000009s :  0.01%
    mutable_eliminate                    :     0.000931s :  1.07%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000028s :  0.03%
    cconv                                :     0.000040s :  0.05%
    loop_unroll                          :     0.000559s :  0.64%
    jit_opt_after_cconv.c_1              :     0.000054s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000104s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000011s :  0.01%
    remove_dup_value                     :     0.000021s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000017s :  0.02%
    add_recomputation                    :     0.000077s :  0.09%
    cse_after_recomputation.cse          :     0.000021s :  0.02%
    auto_monad_reorder                   :     0.000029s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.01%
    opt_after_jit_grad                   :     0.000676s :  0.78%
    symbol_engine_optimizer.build        :     0.000019s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000077s :  0.09%
    backend_pass                         :     0.000002s :  0.00%
    task_emit                            :     0.016716s : 19.27%
    execute                              :     0.000010s :  0.01%


Time group info:
------[substitution.]   0.000417    47
 3.51% :     0.000015s :      2: substitution.depend_value_elim
 0.87% :     0.000004s :      4: substitution.elim_not_effective
 0.64% :     0.000003s :      4: substitution.fold_const_symbol
 1.86% :     0.000008s :      5: substitution.graph_param_transform
77.09% :     0.000322s :      4: substitution.inline
 1.52% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.17% :     0.000009s :      8: substitution.remove_not_recompute_node
 1.76% :     0.000007s :      2: substitution.replace_old_param
 4.27% :     0.000018s :      2: substitution.tuple_list_get_item_eliminator
 3.39% :     0.000014s :      3: substitution.updatestate_pure_node_eliminater
 2.93% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.061519     2
97.08% :     0.059723s :      1: type_inference.infer
 2.92% :     0.001795s :      1: type_inference.specialize
------[replace.]   0.000125     6
62.21% :     0.000078s :      4: replace.inline
37.79% :     0.000047s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000334     6
95.18% :     0.000318s :      4: match.inline
 4.82% :     0.000016s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000211  1021
 1.41% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.34% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 0.90% :     0.000002s :     15: predicate.addn_check_dump
 1.24% :     0.000003s :     15: predicate.addn_zero_filter
 2.56% :     0.000005s :     15: predicate.arithmetic_simplify
 1.42% :     0.000003s :     15: predicate.cast_eliminate
 0.55% :     0.000001s :      5: predicate.check_bprop_eliminate
 0.90% :     0.000002s :     15: predicate.compare_switch_simplify
 1.83% :     0.000004s :     15: predicate.depend_value_elim
 0.98% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.20% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.85% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.28% :     0.000001s :      5: predicate.elim_not_effective
 0.68% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000003s :     15: predicate.environ_add_const_eliminate
 0.93% :     0.000002s :     15: predicate.environ_get_add_eliminate
 0.98% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.39% :     0.000003s :     15: predicate.environ_get_eliminate
 0.95% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      5: predicate.fold_const_symbol
 1.32% :     0.000003s :     10: predicate.get_grad_eliminate
 0.30% :     0.000001s :      5: predicate.graph_param_transform
 5.24% :     0.000011s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.37% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.57% :     0.000003s :     10: predicate.less_batch_normalization
 1.75% :     0.000004s :     17: predicate.list_to_tuple_eliminator_
 1.61% :     0.000003s :     22: predicate.load_eliminater
 1.60% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.25% :     0.000007s :     42: predicate.loop_unroll_before_grad
 1.67% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000002s :     15: predicate.merge_addn
 0.95% :     0.000002s :     15: predicate.minmaximum_grad
 1.65% :     0.000003s :      5: predicate.mutable_eliminate
 0.72% :     0.000002s :      5: predicate.opt_reshape
 2.06% :     0.000004s :     22: predicate.partial_eliminate
 1.60% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.93% :     0.000004s :     15: predicate.reduce_eliminate
 1.68% :     0.000004s :     17: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.84% :     0.000004s :     27: predicate.replace_applicator
 0.55% :     0.000001s :     10: predicate.replace_old_param
 0.49% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000003s :     15: predicate.reshape_eliminate
 1.54% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.91% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.70% :     0.000004s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.85% :     0.000002s :     10: predicate.special_op_eliminate
 1.18% :     0.000002s :     10: predicate.specialize_transform
 1.39% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.28% :     0.000005s :     21: predicate.switch_defer_inline
 1.65% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.80% :     0.000014s :     68: predicate.switch_simplify
 1.14% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000003s :     15: predicate.transpose_eliminate
 1.44% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 4.75% :     0.000010s :     27: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.71% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000004s :     15: predicate.value_based_eliminate
 0.35% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.46% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001135    13
43.64% :     0.000495s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.68% :     0.000076s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.68% :     0.000564s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.096516    76
 0.08% :     0.000081s :      1: add_recomputation
 0.21% :     0.000205s :      1: auto_monad
 0.03% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.49% :     0.000471s :      1: bootstrap
 0.05% :     0.000044s :      1: cconv
 0.02% :     0.000018s :      1: convert_after_rewriter
 0.04% :     0.000037s :      1: cse_after_recomputation
 0.02% :     0.000019s :      1: environ_conv
 0.03% :     0.000032s :      1: event_method
 0.02% :     0.000016s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.11% :     0.000104s :      1: graph_reusing
 9.79% :     0.009447s :      1: jit_opt_a
 0.32% :     0.000305s :      1: jit_opt_after_cconv
 0.08% :     0.000082s :      1: jit_opt_b
 0.59% :     0.000570s :      1: loop_unroll
 0.98% :     0.000947s :      1: mutable_eliminate
 1.52% :     0.001469s :     26: opt.transform.jit_opt_a
 0.10% :     0.000095s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000046s :      4: opt.transform.jit_opt_b
 0.02% :     0.000022s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000041s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000060s :      4: opt.transform.symbol_engine_opt
 0.71% :     0.000690s :      1: opt_after_jit_grad
 0.01% :     0.000012s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.17% :     0.000167s :      1: py_interpret_to_execute
 0.03% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000023s :      1: remove_dup_value
 1.73% :     0.001670s :      1: renormalize.infer
 0.68% :     0.000657s :      1: renormalize.specialize
 0.01% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.34% :     0.000325s :      1: rewriter_after_opt_a
 0.12% :     0.000115s :      1: rewriter_before_opt_a
 0.13% :     0.000127s :      1: symbol_engine_optimizer
17.34% :     0.016739s :      1: task_emit
63.88% :     0.061653s :      1: type_inference
 0.11% :     0.000108s :      1: validate


TotalTime = 0.0990773, [33]
[bootstrap]: 0.00034623
[type_inference]: 0.0468401
[event_method]: 2.076e-05
[auto_monad]: 0.00017079
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 0.00016896
[rewriter_before_opt_a]: 0.00011314
[expand_dump_flag]: 4.25e-06
[jit_opt_a]: 0.00757544, [2]
    [Cycle 1]: 0.00271289, [27]
        [switch_simplify]: 7.274e-05
        [loop_unroll]: 3.592e-05
        [a_1]: 0.00073857
        [with_stream_mark]: 1.763e-05
        [recompute_prepare]: 1.07e-05
        [updatestate_depend_eliminate]: 5.81e-06
        [updatestate_assign_eliminate]: 4.97e-06
        [updatestate_loads_eliminate]: 4.18001e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 9.36998e-06
        [updatestate_useless_node_eliminater]: 1.164e-05
        [accelerated_algorithm]: 9.30001e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 8.82e-06
        [merge_forward]: 4.44998e-06
        [cell_reuse_recompute_pass]: 1.61002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.794e-05
        [j_node_and_user_rematch]: 1.446e-05
        [meta_fg_expand]: 3.12002e-06
        [replace_old_param]: 1.226e-05
        [inline_without_move]: 8.79998e-06
        [renormalize]: 0.00145443
        [add_forward_monad_depend]: 5.74e-06
        [auto_monad_grad]: 2.19001e-06
        [auto_monad_eliminator]: 1.938e-05
        [cse]: 3.201e-05
        [replace_applicator]: 1.745e-05
    [Cycle 2]: 0.00049698, [27]
        [switch_simplify]: 9.96e-06
        [loop_unroll]: 9.36e-06
        [a_1]: 0.00018115
        [with_stream_mark]: 1.493e-05
        [recompute_prepare]: 8.27e-06
        [updatestate_depend_eliminate]: 4.93001e-06
        [updatestate_assign_eliminate]: 5.02e-06
        [updatestate_loads_eliminate]: 3.97e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 8.26002e-06
        [updatestate_useless_node_eliminater]: 1.165e-05
        [accelerated_algorithm]: 8.25e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 7.63999e-06
        [merge_forward]: 4.31002e-06
        [cell_reuse_recompute_pass]: 2.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.64e-05
        [j_node_and_user_rematch]: 1.28e-05
        [meta_fg_expand]: 3.21001e-06
        [replace_old_param]: 1.1e-05
        [inline_without_move]: 7.53e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.62001e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 1.081e-05
        [cse]: 2.22e-05
        [replace_applicator]: 7.66999e-06
[py_interpret_to_execute_after_opt_a]: 1.414e-05
[rewriter_after_opt_a]: 0.0002944
[convert_after_rewriter]: 1.092e-05
[order_py_execute_after_rewriter]: 5.81e-06
[mutable_eliminate]: 0.00070706
[jit_opt_b]: 7.596e-05, [1]
    [Cycle 1]: 6.676e-05, [2]
        [frontend_op_eliminate]: 2.693e-05
        [inline_after_opt_a]: 2.64e-05
[cconv]: 2.187e-05
[loop_unroll]: 0.00048095
[jit_opt_after_cconv]: 0.0002024, [1]
    [Cycle 1]: 0.00019454, [11]
        [c_1]: 4.849e-05
        [parameter_eliminate]: 2.76e-06
        [updatestate_depend_eliminate]: 9.49e-06
        [updatestate_assign_eliminate]: 4.45e-06
        [updatestate_loads_eliminate]: 4.43999e-06
        [cse]: 3.094e-05
        [call_graph_tuple_transform]: 2.164e-05
        [tuple_list_get_item_eliminator]: 8.52e-06
        [none_parameter_eliminate]: 9.5999e-07
        [renormalize]: 3.69997e-07
        [switch_simplify]: 8.65001e-06
[remove_dup_value]: 1.64e-05
[partial_unused_args_eliminate]: 1.59998e-06
[environ_conv]: 1.148e-05
[add_recomputation]: 5.736e-05
[cse_after_recomputation]: 3.006e-05, [1]
    [Cycle 1]: 2.447e-05, [1]
        [cse]: 1.788e-05
[auto_monad_reorder]: 2.417e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 6.51e-06
[opt_after_jit_grad]: 0.00092953
[symbol_engine_optimizer]: 0.00011296, [1]
    [Cycle 1]: 0.00010515, [6]
        [build]: 1.525e-05
        [elim_shapecalc]: 1.299e-05
        [elim_not_effective]: 2.024e-05
        [opt_reshape]: 9.96998e-06
        [fold_const_symbol]: 1.513e-05
        [renormalize]: 1.24e-06
[validate]: 5.702e-05
[backend_pass]: 8.60018e-07
[task_emit]: 0.0404901
[execute]: 9.84001e-06
Sums
    bootstrap                            :     0.000346s :  0.37%
    type_inference                       :     0.046840s : 49.83%
    event_method                         :     0.000021s :  0.02%
    auto_monad                           :     0.000171s :  0.18%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000169s :  0.18%
    rewriter_before_opt_a                :     0.000113s :  0.12%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000083s :  0.09%
    jit_opt_a.loop_unroll                :     0.000045s :  0.05%
    jit_opt_a.a_1                        :     0.000920s :  0.98%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.03%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.00%
    jit_opt_a.specialize_transform       :     0.000018s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.02%
    jit_opt_a.merge_forward              :     0.000009s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000023s :  0.02%
    jit_opt_a.inline_without_move        :     0.000016s :  0.02%
    jit_opt_a.renormalize                :     0.001455s :  1.55%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000030s :  0.03%
    jit_opt_a.cse                        :     0.000054s :  0.06%
    jit_opt_a.replace_applicator         :     0.000025s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.02%
    rewriter_after_opt_a                 :     0.000294s :  0.31%
    convert_after_rewriter               :     0.000011s :  0.01%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000707s :  0.75%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.03%
    cconv                                :     0.000022s :  0.02%
    loop_unroll                          :     0.000481s :  0.51%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.cse              :     0.000031s :  0.03%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.02%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.01%
    remove_dup_value                     :     0.000016s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000011s :  0.01%
    add_recomputation                    :     0.000057s :  0.06%
    cse_after_recomputation.cse          :     0.000018s :  0.02%
    auto_monad_reorder                   :     0.000024s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000930s :  0.99%
    symbol_engine_optimizer.build        :     0.000015s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000057s :  0.06%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.040490s : 43.07%
    execute                              :     0.000010s :  0.01%


Time group info:
------[substitution.]   0.000279    47
 3.86% :     0.000011s :      2: substitution.depend_value_elim
 1.16% :     0.000003s :      4: substitution.elim_not_effective
 0.89% :     0.000002s :      4: substitution.fold_const_symbol
 2.18% :     0.000006s :      5: substitution.graph_param_transform
74.58% :     0.000208s :      4: substitution.inline
 1.77% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.60% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.89% :     0.000005s :      2: substitution.replace_old_param
 3.62% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
 3.91% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.55% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.046755     2
95.87% :     0.044822s :      1: type_inference.infer
 4.13% :     0.001932s :      1: type_inference.specialize
------[replace.]   0.000112     6
69.45% :     0.000078s :      4: replace.inline
30.55% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
95.83% :     0.000205s :      4: match.inline
 4.17% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000179  1021
 1.21% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.49% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     15: predicate.addn_check_dump
 1.46% :     0.000003s :     15: predicate.addn_zero_filter
 2.07% :     0.000004s :     15: predicate.arithmetic_simplify
 1.53% :     0.000003s :     15: predicate.cast_eliminate
 0.73% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     15: predicate.compare_switch_simplify
 1.45% :     0.000003s :     15: predicate.depend_value_elim
 1.16% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.02% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      5: predicate.elim_not_effective
 0.60% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     15: predicate.environ_get_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.16% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 4.62% :     0.000008s :     31: predicate.inline
 1.01% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.42% :     0.000003s :     10: predicate.less_batch_normalization
 1.48% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.16% :     0.000004s :     22: predicate.load_eliminater
 1.42% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.65% :     0.000007s :     42: predicate.loop_unroll_before_grad
 2.05% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     15: predicate.merge_addn
 1.15% :     0.000002s :     15: predicate.minmaximum_grad
 1.49% :     0.000003s :      5: predicate.mutable_eliminate
 0.65% :     0.000001s :      5: predicate.opt_reshape
 2.17% :     0.000004s :     22: predicate.partial_eliminate
 1.29% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.79% :     0.000003s :     15: predicate.reduce_eliminate
 1.37% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.72% :     0.000003s :     27: predicate.replace_applicator
 0.55% :     0.000001s :     10: predicate.replace_old_param
 0.36% :     0.000001s :      5: predicate.reset_defer_inline
 1.46% :     0.000003s :     15: predicate.reshape_eliminate
 1.26% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     15: predicate.same_eliminate
 0.53% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.09% :     0.000002s :     10: predicate.special_op_eliminate
 1.08% :     0.000002s :     10: predicate.specialize_transform
 1.58% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.12% :     0.000004s :     21: predicate.switch_defer_inline
 1.77% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.51% :     0.000012s :     68: predicate.switch_simplify
 1.53% :     0.000003s :     15: predicate.tile_eliminate
 1.31% :     0.000002s :     15: predicate.transpose_eliminate
 1.81% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 2.05% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.19% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000994    13
43.75% :     0.000435s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.60% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.66% :     0.000514s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.101949    76
 0.06% :     0.000061s :      1: add_recomputation
 0.17% :     0.000176s :      1: auto_monad
 0.03% :     0.000028s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.36% :     0.000367s :      1: bootstrap
 0.02% :     0.000025s :      1: cconv
 0.01% :     0.000014s :      1: convert_after_rewriter
 0.03% :     0.000032s :      1: cse_after_recomputation
 0.01% :     0.000014s :      1: environ_conv
 0.02% :     0.000025s :      1: event_method
 0.02% :     0.000015s :      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
 7.43% :     0.007580s :      1: jit_opt_a
 0.20% :     0.000205s :      1: jit_opt_after_cconv
 0.08% :     0.000079s :      1: jit_opt_b
 0.48% :     0.000490s :      1: loop_unroll
 0.70% :     0.000717s :      1: mutable_eliminate
 1.21% :     0.001234s :     26: opt.transform.jit_opt_a
 0.08% :     0.000083s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000046s :      4: opt.transform.jit_opt_b
 0.02% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000054s :      4: opt.transform.symbol_engine_opt
 0.92% :     0.000941s :      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
 0.17% :     0.000175s :      1: py_interpret_to_execute
 0.02% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000019s :      1: remove_dup_value
 0.90% :     0.000922s :      1: renormalize.infer
 0.51% :     0.000522s :      1: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000299s :      1: rewriter_after_opt_a
 0.12% :     0.000118s :      1: rewriter_before_opt_a
 0.11% :     0.000116s :      1: symbol_engine_optimizer
39.74% :     0.040513s :      1: task_emit
45.96% :     0.046858s :      1: type_inference
 0.08% :     0.000082s :      1: validate


TotalTime = 0.114344, [33]
[bootstrap]: 0.00032833
[type_inference]: 0.0638532
[event_method]: 2.4e-05
[auto_monad]: 0.00017157
[graph_reusing]: 6.81001e-06
[pre_auto_parallel]: 3.09999e-06
[py_interpret_to_execute]: 0.0001591
[rewriter_before_opt_a]: 0.00010472
[expand_dump_flag]: 3.68e-06
[jit_opt_a]: 0.00807653, [2]
    [Cycle 1]: 0.0033571, [27]
        [switch_simplify]: 7.099e-05
        [loop_unroll]: 3.697e-05
        [a_1]: 0.00080633
        [with_stream_mark]: 2.839e-05
        [recompute_prepare]: 1.234e-05
        [updatestate_depend_eliminate]: 7.43e-06
        [updatestate_assign_eliminate]: 5.47999e-06
        [updatestate_loads_eliminate]: 5.20999e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 1.001e-05
        [updatestate_useless_node_eliminater]: 1.202e-05
        [accelerated_algorithm]: 8.99998e-06
        [meta_shard_fg_expand]: 3.01001e-06
        [get_grad_eliminate_]: 8.76002e-06
        [merge_forward]: 5.75001e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.123e-05
        [j_node_and_user_rematch]: 1.454e-05
        [meta_fg_expand]: 3.68999e-06
        [replace_old_param]: 1.248e-05
        [inline_without_move]: 8.02e-06
        [renormalize]: 0.00195804
        [add_forward_monad_depend]: 8.47998e-06
        [auto_monad_grad]: 2.93e-06
        [auto_monad_eliminator]: 2.564e-05
        [cse]: 4.704e-05
        [replace_applicator]: 2.279e-05
    [Cycle 2]: 0.00052632, [27]
        [switch_simplify]: 9.77999e-06
        [loop_unroll]: 8.97e-06
        [a_1]: 0.00018537
        [with_stream_mark]: 1.786e-05
        [recompute_prepare]: 8.84e-06
        [updatestate_depend_eliminate]: 6.16e-06
        [updatestate_assign_eliminate]: 5.66e-06
        [updatestate_loads_eliminate]: 4.05e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 8.56002e-06
        [updatestate_useless_node_eliminater]: 1.106e-05
        [accelerated_algorithm]: 8.08001e-06
        [meta_shard_fg_expand]: 2.62001e-06
        [get_grad_eliminate_]: 8.01001e-06
        [merge_forward]: 5.09e-06
        [cell_reuse_recompute_pass]: 3.65e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.122e-05
        [j_node_and_user_rematch]: 1.409e-05
        [meta_fg_expand]: 3.54002e-06
        [replace_old_param]: 1.301e-05
        [inline_without_move]: 8.48001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 1.41998e-06
        [auto_monad_eliminator]: 1.22e-05
        [cse]: 2.396e-05
        [replace_applicator]: 8.60999e-06
[py_interpret_to_execute_after_opt_a]: 1.837e-05
[rewriter_after_opt_a]: 0.00031511
[convert_after_rewriter]: 1.236e-05
[order_py_execute_after_rewriter]: 7.28e-06
[mutable_eliminate]: 0.00080651
[jit_opt_b]: 7.558e-05, [1]
    [Cycle 1]: 6.721e-05, [2]
        [frontend_op_eliminate]: 2.661e-05
        [inline_after_opt_a]: 2.708e-05
[cconv]: 3.376e-05
[loop_unroll]: 0.00050033
[jit_opt_after_cconv]: 0.00023843, [1]
    [Cycle 1]: 0.00023013, [11]
        [c_1]: 5.478e-05
        [parameter_eliminate]: 3.48e-06
        [updatestate_depend_eliminate]: 1.094e-05
        [updatestate_assign_eliminate]: 5.12e-06
        [updatestate_loads_eliminate]: 4.70999e-06
        [cse]: 4.208e-05
        [call_graph_tuple_transform]: 2.83e-05
        [tuple_list_get_item_eliminator]: 9.20001e-06
        [none_parameter_eliminate]: 1.86998e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 1.011e-05
[remove_dup_value]: 2.086e-05
[partial_unused_args_eliminate]: 2.63e-06
[environ_conv]: 1.557e-05
[add_recomputation]: 7.515e-05
[cse_after_recomputation]: 3.341e-05, [1]
    [Cycle 1]: 2.638e-05, [1]
        [cse]: 1.93e-05
[auto_monad_reorder]: 2.832e-05
[get_jit_bprop_graph]: 2.53e-06
[rewriter_after_jit_bprop_graph]: 6.99001e-06
[opt_after_jit_grad]: 0.00056426
[symbol_engine_optimizer]: 0.00011987, [1]
    [Cycle 1]: 0.00011243, [6]
        [build]: 2.005e-05
        [elim_shapecalc]: 1.293e-05
        [elim_not_effective]: 2.272e-05
        [opt_reshape]: 9.96998e-06
        [fold_const_symbol]: 1.538e-05
        [renormalize]: 5.09986e-07
[validate]: 7.204e-05
[backend_pass]: 1.17999e-06
[task_emit]: 0.0383824
[execute]: 1.053e-05
Sums
    bootstrap                            :     0.000328s :  0.30%
    type_inference                       :     0.063853s : 58.36%
    event_method                         :     0.000024s :  0.02%
    auto_monad                           :     0.000172s :  0.16%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000159s :  0.15%
    rewriter_before_opt_a                :     0.000105s :  0.10%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000081s :  0.07%
    jit_opt_a.loop_unroll                :     0.000046s :  0.04%
    jit_opt_a.a_1                        :     0.000992s :  0.91%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000019s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.02%
    jit_opt_a.merge_forward              :     0.000011s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000042s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000025s :  0.02%
    jit_opt_a.inline_without_move        :     0.000017s :  0.02%
    jit_opt_a.renormalize                :     0.001958s :  1.79%
    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      :     0.000038s :  0.03%
    jit_opt_a.cse                        :     0.000071s :  0.06%
    jit_opt_a.replace_applicator         :     0.000031s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.02%
    rewriter_after_opt_a                 :     0.000315s :  0.29%
    convert_after_rewriter               :     0.000012s :  0.01%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000807s :  0.74%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.02%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.02%
    cconv                                :     0.000034s :  0.03%
    loop_unroll                          :     0.000500s :  0.46%
    jit_opt_after_cconv.c_1              :     0.000055s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.00%
    jit_opt_after_cconv.cse              :     0.000042s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.01%
    remove_dup_value                     :     0.000021s :  0.02%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000016s :  0.01%
    add_recomputation                    :     0.000075s :  0.07%
    cse_after_recomputation.cse          :     0.000019s :  0.02%
    auto_monad_reorder                   :     0.000028s :  0.03%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000564s :  0.52%
    symbol_engine_optimizer.build        :     0.000020s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000072s :  0.07%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.038382s : 35.08%
    execute                              :     0.000011s :  0.01%


Time group info:
------[substitution.]   0.000353    47
 3.01% :     0.000011s :      2: substitution.depend_value_elim
 0.93% :     0.000003s :      4: substitution.elim_not_effective
 0.70% :     0.000002s :      4: substitution.fold_const_symbol
 2.22% :     0.000008s :      5: substitution.graph_param_transform
77.55% :     0.000274s :      4: substitution.inline
 1.56% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.38% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.81% :     0.000006s :      2: substitution.replace_old_param
 3.76% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.16% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 2.92% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.063770     2
96.90% :     0.061794s :      1: type_inference.infer
 3.10% :     0.001976s :      1: type_inference.specialize
------[replace.]   0.000107     6
66.39% :     0.000071s :      4: replace.inline
33.61% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000283     6
95.70% :     0.000270s :      4: match.inline
 4.30% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000183  1021
 1.26% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.39% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.08% :     0.000002s :     15: predicate.addn_check_dump
 1.45% :     0.000003s :     15: predicate.addn_zero_filter
 1.86% :     0.000003s :     15: predicate.arithmetic_simplify
 1.56% :     0.000003s :     15: predicate.cast_eliminate
 0.57% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     15: predicate.compare_switch_simplify
 1.49% :     0.000003s :     15: predicate.depend_value_elim
 1.32% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.02% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.71% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     15: predicate.environ_get_eliminate
 1.44% :     0.000003s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.01% :     0.000002s :     10: predicate.get_grad_eliminate
 0.35% :     0.000001s :      5: predicate.graph_param_transform
 5.75% :     0.000011s :     31: predicate.inline
 1.00% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.42% :     0.000003s :     10: predicate.less_batch_normalization
 1.54% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.42% :     0.000004s :     22: predicate.load_eliminater
 1.37% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.24% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.03% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000002s :     15: predicate.merge_addn
 1.09% :     0.000002s :     15: predicate.minmaximum_grad
 1.56% :     0.000003s :      5: predicate.mutable_eliminate
 0.46% :     0.000001s :      5: predicate.opt_reshape
 2.26% :     0.000004s :     22: predicate.partial_eliminate
 1.12% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.76% :     0.000003s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.66% :     0.000003s :     27: predicate.replace_applicator
 0.67% :     0.000001s :     10: predicate.replace_old_param
 0.70% :     0.000001s :      5: predicate.reset_defer_inline
 1.31% :     0.000002s :     15: predicate.reshape_eliminate
 1.17% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.38% :     0.000003s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.01% :     0.000002s :     10: predicate.special_op_eliminate
 1.02% :     0.000002s :     10: predicate.specialize_transform
 1.43% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.20% :     0.000004s :     21: predicate.switch_defer_inline
 1.78% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.22% :     0.000011s :     68: predicate.switch_simplify
 1.16% :     0.000002s :     15: predicate.tile_eliminate
 1.30% :     0.000002s :     15: predicate.transpose_eliminate
 1.44% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 4.26% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.30% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.80% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001143    13
39.98% :     0.000457s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.21% :     0.000071s :      1: func_graph_cloner_run.FuncGraphClonerNode
53.81% :     0.000615s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.117829    76
 0.07% :     0.000078s :      1: add_recomputation
 0.15% :     0.000177s :      1: auto_monad
 0.03% :     0.000031s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.30% :     0.000349s :      1: bootstrap
 0.03% :     0.000036s :      1: cconv
 0.01% :     0.000015s :      1: convert_after_rewriter
 0.03% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000018s :      1: environ_conv
 0.02% :     0.000029s :      1: event_method
 0.01% :     0.000017s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
 6.86% :     0.008082s :      1: jit_opt_a
 0.21% :     0.000242s :      1: jit_opt_after_cconv
 0.07% :     0.000078s :      1: jit_opt_b
 0.43% :     0.000510s :      1: loop_unroll
 0.69% :     0.000818s :      1: mutable_eliminate
 1.12% :     0.001325s :     26: opt.transform.jit_opt_a
 0.08% :     0.000098s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000045s :      4: opt.transform.jit_opt_b
 0.02% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000056s :      4: opt.transform.symbol_engine_opt
 0.49% :     0.000575s :      1: opt_after_jit_grad
 0.01% :     0.000010s :      1: order_py_execute_after_rewriter
 0.00% :     0.000005s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000163s :      1: py_interpret_to_execute
 0.02% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000024s :      1: remove_dup_value
 1.08% :     0.001275s :      1: renormalize.infer
 0.57% :     0.000671s :      1: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000320s :      1: rewriter_after_opt_a
 0.09% :     0.000109s :      1: rewriter_before_opt_a
 0.10% :     0.000123s :      1: symbol_engine_optimizer
32.59% :     0.038405s :      1: task_emit
54.20% :     0.063869s :      1: type_inference
 0.09% :     0.000101s :      1: validate


TotalTime = 0.075361, [33]
[bootstrap]: 0.00032416
[type_inference]: 0.0373395
[event_method]: 2.717e-05
[auto_monad]: 0.00018518
[graph_reusing]: 7.63001e-06
[pre_auto_parallel]: 3.11001e-06
[py_interpret_to_execute]: 0.00016883
[rewriter_before_opt_a]: 0.00011241
[expand_dump_flag]: 3.61001e-06
[jit_opt_a]: 0.00750212, [2]
    [Cycle 1]: 0.00288235, [27]
        [switch_simplify]: 7.428e-05
        [loop_unroll]: 3.813e-05
        [a_1]: 0.00078699
        [with_stream_mark]: 2.374e-05
        [recompute_prepare]: 1.107e-05
        [updatestate_depend_eliminate]: 6.12999e-06
        [updatestate_assign_eliminate]: 6.17001e-06
        [updatestate_loads_eliminate]: 5.03002e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 9.82999e-06
        [updatestate_useless_node_eliminater]: 1.205e-05
        [accelerated_algorithm]: 8.62998e-06
        [meta_shard_fg_expand]: 3.08e-06
        [get_grad_eliminate_]: 8.42e-06
        [merge_forward]: 5.25999e-06
        [cell_reuse_recompute_pass]: 1.79998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.108e-05
        [j_node_and_user_rematch]: 1.326e-05
        [meta_fg_expand]: 3.73001e-06
        [replace_old_param]: 1.248e-05
        [inline_without_move]: 8.60999e-06
        [renormalize]: 0.00153022
        [add_forward_monad_depend]: 6.32001e-06
        [auto_monad_grad]: 2.78e-06
        [auto_monad_eliminator]: 2.325e-05
        [cse]: 4.35e-05
        [replace_applicator]: 1.963e-05
    [Cycle 2]: 0.00048673, [27]
        [switch_simplify]: 9.42001e-06
        [loop_unroll]: 7.81001e-06
        [a_1]: 0.0001772
        [with_stream_mark]: 1.31e-05
        [recompute_prepare]: 8.41002e-06
        [updatestate_depend_eliminate]: 5.34e-06
        [updatestate_assign_eliminate]: 4.97999e-06
        [updatestate_loads_eliminate]: 4.20999e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 8.17e-06
        [updatestate_useless_node_eliminater]: 1.164e-05
        [accelerated_algorithm]: 8.14002e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 7.38999e-06
        [merge_forward]: 5.05999e-06
        [cell_reuse_recompute_pass]: 2.48e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.797e-05
        [j_node_and_user_rematch]: 1.292e-05
        [meta_fg_expand]: 2.94001e-06
        [replace_old_param]: 1.041e-05
        [inline_without_move]: 7.72998e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 9.59e-06
        [cse]: 2.11e-05
        [replace_applicator]: 7.78001e-06
[py_interpret_to_execute_after_opt_a]: 1.447e-05
[rewriter_after_opt_a]: 0.00030268
[convert_after_rewriter]: 1.211e-05
[order_py_execute_after_rewriter]: 7.44002e-06
[mutable_eliminate]: 0.00078284
[jit_opt_b]: 7.119e-05, [1]
    [Cycle 1]: 6.31e-05, [2]
        [frontend_op_eliminate]: 2.539e-05
        [inline_after_opt_a]: 2.554e-05
[cconv]: 2.782e-05
[loop_unroll]: 0.00047847
[jit_opt_after_cconv]: 0.00020397, [1]
    [Cycle 1]: 0.00019718, [11]
        [c_1]: 4.758e-05
        [parameter_eliminate]: 2.78e-06
        [updatestate_depend_eliminate]: 8.32e-06
        [updatestate_assign_eliminate]: 4.87e-06
        [updatestate_loads_eliminate]: 3.98999e-06
        [cse]: 3.081e-05
        [call_graph_tuple_transform]: 2.487e-05
        [tuple_list_get_item_eliminator]: 8.70001e-06
        [none_parameter_eliminate]: 1.74998e-06
        [renormalize]: 4.49974e-07
        [switch_simplify]: 9.25999e-06
[remove_dup_value]: 2.272e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 1.224e-05
[add_recomputation]: 6.773e-05
[cse_after_recomputation]: 0.0001905, [1]
    [Cycle 1]: 0.0001455, [1]
        [cse]: 0.00010374
[auto_monad_reorder]: 4.52e-05
[get_jit_bprop_graph]: 3.86999e-06
[rewriter_after_jit_bprop_graph]: 1.414e-05
[opt_after_jit_grad]: 0.0009872
[symbol_engine_optimizer]: 0.00014861, [1]
    [Cycle 1]: 0.00013939, [6]
        [build]: 2.488e-05
        [elim_shapecalc]: 1.528e-05
        [elim_not_effective]: 3.356e-05
        [opt_reshape]: 1.359e-05
        [fold_const_symbol]: 1.701e-05
        [renormalize]: 9.39996e-07
[validate]: 8.063e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.0137996
[execute]: 1.028e-05
Sums
    bootstrap                            :     0.000324s :  0.56%
    type_inference                       :     0.037339s : 64.04%
    event_method                         :     0.000027s :  0.05%
    auto_monad                           :     0.000185s :  0.32%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000169s :  0.29%
    rewriter_before_opt_a                :     0.000112s :  0.19%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000084s :  0.14%
    jit_opt_a.loop_unroll                :     0.000046s :  0.08%
    jit_opt_a.a_1                        :     0.000964s :  1.65%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.03%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000023s :  0.04%
    jit_opt_a.inline_without_move        :     0.000016s :  0.03%
    jit_opt_a.renormalize                :     0.001530s :  2.62%
    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      :     0.000033s :  0.06%
    jit_opt_a.cse                        :     0.000065s :  0.11%
    jit_opt_a.replace_applicator         :     0.000027s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.02%
    rewriter_after_opt_a                 :     0.000303s :  0.52%
    convert_after_rewriter               :     0.000012s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000783s :  1.34%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.04%
    cconv                                :     0.000028s :  0.05%
    loop_unroll                          :     0.000478s :  0.82%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.08%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000023s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000012s :  0.02%
    add_recomputation                    :     0.000068s :  0.12%
    cse_after_recomputation.cse          :     0.000104s :  0.18%
    auto_monad_reorder                   :     0.000045s :  0.08%
    get_jit_bprop_graph                  :     0.000004s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000014s :  0.02%
    opt_after_jit_grad                   :     0.000987s :  1.69%
    symbol_engine_optimizer.build        :     0.000025s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000015s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000034s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000014s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000017s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000081s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.013800s : 23.67%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000327    47
 3.42% :     0.000011s :      2: substitution.depend_value_elim
 1.30% :     0.000004s :      4: substitution.elim_not_effective
 0.87% :     0.000003s :      4: substitution.fold_const_symbol
 2.10% :     0.000007s :      5: substitution.graph_param_transform
76.62% :     0.000250s :      4: substitution.inline
 1.56% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.32% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.61% :     0.000005s :      2: substitution.replace_old_param
 4.25% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 2.92% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 3.02% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.037235     2
60.91% :     0.022681s :      1: type_inference.infer
39.09% :     0.014555s :      1: type_inference.specialize
------[replace.]   0.000104     6
65.13% :     0.000068s :      4: replace.inline
34.87% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000260     6
95.09% :     0.000247s :      4: match.inline
 4.91% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000185  1021
 1.41% :     0.000003s :     15: predicate.accumulaten_eliminater
 2.01% :     0.000004s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.52% :     0.000003s :     15: predicate.addn_zero_filter
 2.14% :     0.000004s :     15: predicate.arithmetic_simplify
 1.22% :     0.000002s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.04% :     0.000002s :     15: predicate.compare_switch_simplify
 1.76% :     0.000003s :     15: predicate.depend_value_elim
 1.12% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.18% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.54% :     0.000001s :      5: predicate.elim_not_effective
 0.69% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.06% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     15: predicate.environ_get_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      5: predicate.fold_const_symbol
 1.02% :     0.000002s :     10: predicate.get_grad_eliminate
 0.26% :     0.000000s :      5: predicate.graph_param_transform
 5.00% :     0.000009s :     31: predicate.inline
 0.95% :     0.000002s :     10: predicate.inline_without_move
 0.39% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.41% :     0.000003s :     10: predicate.less_batch_normalization
 1.55% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.84% :     0.000003s :     22: predicate.load_eliminater
 1.36% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.39% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.10% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     15: predicate.merge_addn
 1.11% :     0.000002s :     15: predicate.minmaximum_grad
 1.75% :     0.000003s :      5: predicate.mutable_eliminate
 0.61% :     0.000001s :      5: predicate.opt_reshape
 2.17% :     0.000004s :     22: predicate.partial_eliminate
 1.26% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     15: predicate.reduce_eliminate
 1.68% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.60% :     0.000003s :     27: predicate.replace_applicator
 0.83% :     0.000002s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.17% :     0.000002s :     15: predicate.reshape_eliminate
 1.23% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     15: predicate.same_eliminate
 0.55% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.11% :     0.000002s :     10: predicate.special_op_eliminate
 1.09% :     0.000002s :     10: predicate.specialize_transform
 1.56% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.08% :     0.000004s :     21: predicate.switch_defer_inline
 1.84% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.91% :     0.000013s :     68: predicate.switch_simplify
 1.19% :     0.000002s :     15: predicate.tile_eliminate
 1.18% :     0.000002s :     15: predicate.transpose_eliminate
 1.60% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.67% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000003s :     15: predicate.value_based_eliminate
 0.43% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001083    13
41.05% :     0.000445s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.43% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
54.51% :     0.000591s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.078374    76
15.51% :     0.012156s :      1: add_recomputation
 0.24% :     0.000192s :      1: auto_monad
 0.06% :     0.000049s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.44% :     0.000345s :      1: bootstrap
 0.04% :     0.000031s :      1: cconv
 0.02% :     0.000015s :      1: convert_after_rewriter
 0.25% :     0.000198s :      1: cse_after_recomputation
 0.02% :     0.000015s :      1: environ_conv
 0.04% :     0.000033s :      1: event_method
 0.02% :     0.000016s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000009s :      1: get_jit_bprop_graph
 0.01% :     0.000011s :      1: graph_reusing
 9.58% :     0.007506s :      1: jit_opt_a
 0.26% :     0.000207s :      1: jit_opt_after_cconv
 0.09% :     0.000074s :      1: jit_opt_b
 0.62% :     0.000488s :      1: loop_unroll
 1.01% :     0.000794s :      1: mutable_eliminate
 1.64% :     0.001286s :     26: opt.transform.jit_opt_a
 0.11% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000043s :      4: opt.transform.jit_opt_b
 0.02% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000048s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000072s :      4: opt.transform.symbol_engine_opt
 1.27% :     0.000999s :      1: opt_after_jit_grad
 0.01% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000173s :      1: py_interpret_to_execute
 0.02% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000025s :      1: remove_dup_value
 1.24% :     0.000969s :      1: renormalize.infer
 0.70% :     0.000551s :      1: renormalize.specialize
 0.02% :     0.000017s :      1: rewriter_after_jit_bprop_graph
 0.39% :     0.000307s :      1: rewriter_after_opt_a
 0.15% :     0.000117s :      1: rewriter_before_opt_a
 0.19% :     0.000152s :      1: symbol_engine_optimizer
17.63% :     0.013820s :      1: task_emit
47.68% :     0.037368s :      1: type_inference
 0.15% :     0.000118s :      1: validate


TotalTime = 0.104092, [33]
[bootstrap]: 0.00032354
[type_inference]: 0.0483365
[event_method]: 2.178e-05
[auto_monad]: 0.00017032
[graph_reusing]: 5.97999e-06
[pre_auto_parallel]: 2.86e-06
[py_interpret_to_execute]: 0.00014644
[rewriter_before_opt_a]: 0.00010342
[expand_dump_flag]: 4.1e-06
[jit_opt_a]: 0.0235272, [2]
    [Cycle 1]: 0.00275999, [27]
        [switch_simplify]: 7.635e-05
        [loop_unroll]: 3.665e-05
        [a_1]: 0.00079095
        [with_stream_mark]: 2.128e-05
        [recompute_prepare]: 1.07e-05
        [updatestate_depend_eliminate]: 6.14999e-06
        [updatestate_assign_eliminate]: 8.81002e-06
        [updatestate_loads_eliminate]: 4.94e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 9.87999e-06
        [updatestate_useless_node_eliminater]: 1.171e-05
        [accelerated_algorithm]: 8.2e-06
        [meta_shard_fg_expand]: 2.67001e-06
        [get_grad_eliminate_]: 8.23001e-06
        [merge_forward]: 5.39e-06
        [cell_reuse_recompute_pass]: 1.16997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.055e-05
        [j_node_and_user_rematch]: 1.388e-05
        [meta_fg_expand]: 3.43999e-06
        [replace_old_param]: 1.291e-05
        [inline_without_move]: 8.02e-06
        [renormalize]: 0.00141617
        [add_forward_monad_depend]: 6.80998e-06
        [auto_monad_grad]: 2.84999e-06
        [auto_monad_eliminator]: 2.212e-05
        [cse]: 4.285e-05
        [replace_applicator]: 1.788e-05
    [Cycle 2]: 0.00047685, [27]
        [switch_simplify]: 9.81e-06
        [loop_unroll]: 7.98001e-06
        [a_1]: 0.00017218
        [with_stream_mark]: 1.281e-05
        [recompute_prepare]: 8.11002e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 4.70001e-06
        [updatestate_loads_eliminate]: 3.94002e-06
        [parameter_eliminate]: 1.43002e-06
        [specialize_transform]: 8.23999e-06
        [updatestate_useless_node_eliminater]: 1.126e-05
        [accelerated_algorithm]: 8.40001e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 7.36001e-06
        [merge_forward]: 4.80001e-06
        [cell_reuse_recompute_pass]: 2.30002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.633e-05
        [j_node_and_user_rematch]: 1.279e-05
        [meta_fg_expand]: 2.81e-06
        [replace_old_param]: 1.001e-05
        [inline_without_move]: 7.68999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 1.009e-05
        [cse]: 2.12e-05
        [replace_applicator]: 8.03001e-06
[py_interpret_to_execute_after_opt_a]: 1.422e-05
[rewriter_after_opt_a]: 0.00028853
[convert_after_rewriter]: 1.207e-05
[order_py_execute_after_rewriter]: 7.9e-06
[mutable_eliminate]: 0.00074021
[jit_opt_b]: 7.154e-05, [1]
    [Cycle 1]: 6.416e-05, [2]
        [frontend_op_eliminate]: 2.576e-05
        [inline_after_opt_a]: 2.543e-05
[cconv]: 2.781e-05
[loop_unroll]: 0.00047624
[jit_opt_after_cconv]: 0.00020388, [1]
    [Cycle 1]: 0.00019645, [11]
        [c_1]: 4.626e-05
        [parameter_eliminate]: 2.76e-06
        [updatestate_depend_eliminate]: 8.45001e-06
        [updatestate_assign_eliminate]: 4.74998e-06
        [updatestate_loads_eliminate]: 4.55001e-06
        [cse]: 3.142e-05
        [call_graph_tuple_transform]: 2.438e-05
        [tuple_list_get_item_eliminator]: 8.64003e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 9.20001e-07
        [switch_simplify]: 9.37999e-06
[remove_dup_value]: 2.071e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 1.203e-05
[add_recomputation]: 7.359e-05
[cse_after_recomputation]: 3.041e-05, [1]
    [Cycle 1]: 2.421e-05, [1]
        [cse]: 1.804e-05
[auto_monad_reorder]: 2.528e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 5.81e-06
[opt_after_jit_grad]: 0.000528
[symbol_engine_optimizer]: 0.00012001, [1]
    [Cycle 1]: 0.00011259, [6]
        [build]: 1.566e-05
        [elim_shapecalc]: 1.289e-05
        [elim_not_effective]: 2.257e-05
        [opt_reshape]: 1.1e-05
        [fold_const_symbol]: 1.711e-05
        [renormalize]: 5.39992e-07
[validate]: 7.035e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.0284455
[execute]: 9.52001e-06
Sums
    bootstrap                            :     0.000324s :  0.39%
    type_inference                       :     0.048336s : 58.16%
    event_method                         :     0.000022s :  0.03%
    auto_monad                           :     0.000170s :  0.20%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000146s :  0.18%
    rewriter_before_opt_a                :     0.000103s :  0.12%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000086s :  0.10%
    jit_opt_a.loop_unroll                :     0.000045s :  0.05%
    jit_opt_a.a_1                        :     0.000963s :  1.16%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000018s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.03%
    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_        :     0.000016s :  0.02%
    jit_opt_a.merge_forward              :     0.000010s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000023s :  0.03%
    jit_opt_a.inline_without_move        :     0.000016s :  0.02%
    jit_opt_a.renormalize                :     0.001416s :  1.70%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000032s :  0.04%
    jit_opt_a.cse                        :     0.000064s :  0.08%
    jit_opt_a.replace_applicator         :     0.000026s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.02%
    rewriter_after_opt_a                 :     0.000289s :  0.35%
    convert_after_rewriter               :     0.000012s :  0.01%
    order_py_execute_after_rewriter      :     0.000008s :  0.01%
    mutable_eliminate                    :     0.000740s :  0.89%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.03%
    cconv                                :     0.000028s :  0.03%
    loop_unroll                          :     0.000476s :  0.57%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.06%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.01%
    remove_dup_value                     :     0.000021s :  0.02%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000012s :  0.01%
    add_recomputation                    :     0.000074s :  0.09%
    cse_after_recomputation.cse          :     0.000018s :  0.02%
    auto_monad_reorder                   :     0.000025s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000528s :  0.64%
    symbol_engine_optimizer.build        :     0.000016s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000011s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000017s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000070s :  0.08%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.028445s : 34.23%
    execute                              :     0.000010s :  0.01%


Time group info:
------[substitution.]   0.000333    47
 2.86% :     0.000010s :      2: substitution.depend_value_elim
 0.95% :     0.000003s :      4: substitution.elim_not_effective
 0.67% :     0.000002s :      4: substitution.fold_const_symbol
 2.16% :     0.000007s :      5: substitution.graph_param_transform
78.49% :     0.000261s :      4: substitution.inline
 1.53% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.08% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.45% :     0.000005s :      2: substitution.replace_old_param
 4.19% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.02% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 2.61% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.048245     2
96.26% :     0.046441s :      1: type_inference.infer
 3.74% :     0.001804s :      1: type_inference.specialize
------[replace.]   0.000100     6
66.47% :     0.000067s :      4: replace.inline
33.53% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000271     6
95.27% :     0.000259s :      4: match.inline
 4.73% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000181  1021
 1.24% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.15% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     15: predicate.addn_check_dump
 1.31% :     0.000002s :     15: predicate.addn_zero_filter
 1.97% :     0.000004s :     15: predicate.arithmetic_simplify
 1.20% :     0.000002s :     15: predicate.cast_eliminate
 0.48% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     15: predicate.compare_switch_simplify
 1.57% :     0.000003s :     15: predicate.depend_value_elim
 1.52% :     0.000003s :     15: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.44% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.97% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.61% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.07% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     15: predicate.environ_get_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.28% :     0.000001s :      5: predicate.fold_const_symbol
 1.13% :     0.000002s :     10: predicate.get_grad_eliminate
 0.55% :     0.000001s :      5: predicate.graph_param_transform
 5.07% :     0.000009s :     31: predicate.inline
 1.00% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :     10: predicate.less_batch_normalization
 1.41% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.83% :     0.000003s :     22: predicate.load_eliminater
 1.45% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.24% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.16% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     15: predicate.merge_addn
 1.20% :     0.000002s :     15: predicate.minmaximum_grad
 1.53% :     0.000003s :      5: predicate.mutable_eliminate
 0.55% :     0.000001s :      5: predicate.opt_reshape
 2.45% :     0.000004s :     22: predicate.partial_eliminate
 1.22% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.73% :     0.000003s :     15: predicate.reduce_eliminate
 1.44% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.56% :     0.000003s :     27: predicate.replace_applicator
 0.58% :     0.000001s :     10: predicate.replace_old_param
 0.35% :     0.000001s :      5: predicate.reset_defer_inline
 1.54% :     0.000003s :     15: predicate.reshape_eliminate
 1.33% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.92% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.46% :     0.000003s :     15: predicate.same_eliminate
 0.53% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.08% :     0.000002s :     10: predicate.special_op_eliminate
 1.02% :     0.000002s :     10: predicate.specialize_transform
 1.36% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.59% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.20% :     0.000004s :     21: predicate.switch_defer_inline
 1.78% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.35% :     0.000011s :     68: predicate.switch_simplify
 1.26% :     0.000002s :     15: predicate.tile_eliminate
 1.28% :     0.000002s :     15: predicate.transpose_eliminate
 1.63% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.62% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.66% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.21% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 2.05% :     0.000004s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000983    13
46.39% :     0.000456s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.88% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.73% :     0.000479s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.106979    76
 0.07% :     0.000077s :      1: add_recomputation
 0.17% :     0.000177s :      1: auto_monad
 0.03% :     0.000028s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.32% :     0.000344s :      1: bootstrap
 0.03% :     0.000031s :      1: cconv
 0.01% :     0.000015s :      1: convert_after_rewriter
 0.03% :     0.000033s :      1: cse_after_recomputation
 0.01% :     0.000014s :      1: environ_conv
 0.02% :     0.000026s :      1: event_method
 0.02% :     0.000016s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
22.00% :     0.023531s :      1: jit_opt_a
 0.19% :     0.000207s :      1: jit_opt_after_cconv
 0.07% :     0.000075s :      1: jit_opt_b
 0.45% :     0.000485s :      1: loop_unroll
 0.70% :     0.000751s :      1: mutable_eliminate
 1.20% :     0.001282s :     26: opt.transform.jit_opt_a
 0.08% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000044s :      4: opt.transform.jit_opt_b
 0.02% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000060s :      4: opt.transform.symbol_engine_opt
 0.50% :     0.000537s :      1: opt_after_jit_grad
 0.01% :     0.000010s :      1: order_py_execute_after_rewriter
 0.00% :     0.000005s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000151s :      1: py_interpret_to_execute
 0.02% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000023s :      1: remove_dup_value
 0.77% :     0.000821s :      1: renormalize.infer
 0.55% :     0.000586s :      1: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000294s :      1: rewriter_after_opt_a
 0.10% :     0.000108s :      1: rewriter_before_opt_a
 0.11% :     0.000123s :      1: symbol_engine_optimizer
26.61% :     0.028468s :      1: task_emit
45.20% :     0.048355s :      1: type_inference
 0.09% :     0.000096s :      1: validate


TotalTime = 0.129551, [33]
[bootstrap]: 0.00033795
[type_inference]: 0.0704707
[event_method]: 2.504e-05
[auto_monad]: 0.00018267
[graph_reusing]: 7.51999e-06
[pre_auto_parallel]: 3.25e-06
[py_interpret_to_execute]: 0.0001724
[rewriter_before_opt_a]: 0.00010592
[expand_dump_flag]: 3.73999e-06
[jit_opt_a]: 0.0249686, [2]
    [Cycle 1]: 0.0196632, [27]
        [switch_simplify]: 7.962e-05
        [loop_unroll]: 3.697e-05
        [a_1]: 0.0171892
        [with_stream_mark]: 3.883e-05
        [recompute_prepare]: 1.758e-05
        [updatestate_depend_eliminate]: 6.93e-06
        [updatestate_assign_eliminate]: 6.14001e-06
        [updatestate_loads_eliminate]: 5.33002e-06
        [parameter_eliminate]: 3.26999e-06
        [specialize_transform]: 1.12e-05
        [updatestate_useless_node_eliminater]: 1.419e-05
        [accelerated_algorithm]: 1.156e-05
        [meta_shard_fg_expand]: 4.45999e-06
        [get_grad_eliminate_]: 9.94001e-06
        [merge_forward]: 7.15e-06
        [cell_reuse_recompute_pass]: 5.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.285e-05
        [j_node_and_user_rematch]: 1.622e-05
        [meta_fg_expand]: 4.22e-06
        [replace_old_param]: 1.619e-05
        [inline_without_move]: 8.62e-06
        [renormalize]: 0.00175787
        [add_forward_monad_depend]: 1.001e-05
        [auto_monad_grad]: 3.40998e-06
        [auto_monad_eliminator]: 2.762e-05
        [cse]: 4.823e-05
        [replace_applicator]: 2.853e-05
    [Cycle 2]: 0.00057388, [27]
        [switch_simplify]: 1.136e-05
        [loop_unroll]: 9.27001e-06
        [a_1]: 0.00020424
        [with_stream_mark]: 2.085e-05
        [recompute_prepare]: 9.02e-06
        [updatestate_depend_eliminate]: 5.96998e-06
        [updatestate_assign_eliminate]: 6.63e-06
        [updatestate_loads_eliminate]: 4.27e-06
        [parameter_eliminate]: 2.36998e-06
        [specialize_transform]: 1.022e-05
        [updatestate_useless_node_eliminater]: 1.156e-05
        [accelerated_algorithm]: 9.62999e-06
        [meta_shard_fg_expand]: 3.55e-06
        [get_grad_eliminate_]: 8.23001e-06
        [merge_forward]: 5.38002e-06
        [cell_reuse_recompute_pass]: 4.45e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.28e-05
        [j_node_and_user_rematch]: 1.438e-05
        [meta_fg_expand]: 3.5e-06
        [replace_old_param]: 1.28e-05
        [inline_without_move]: 8.05999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.72999e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 1.27e-05
        [cse]: 2.933e-05
        [replace_applicator]: 9.59e-06
[py_interpret_to_execute_after_opt_a]: 2.161e-05
[rewriter_after_opt_a]: 0.00037653
[convert_after_rewriter]: 1.86e-05
[order_py_execute_after_rewriter]: 3.013e-05
[mutable_eliminate]: 0.00107168
[jit_opt_b]: 9.544e-05, [1]
    [Cycle 1]: 8.435e-05, [2]
        [frontend_op_eliminate]: 3.297e-05
        [inline_after_opt_a]: 3.504e-05
[cconv]: 3.857e-05
[loop_unroll]: 0.00070385
[jit_opt_after_cconv]: 0.00027681, [1]
    [Cycle 1]: 0.00026759, [11]
        [c_1]: 6.754e-05
        [parameter_eliminate]: 6.25002e-06
        [updatestate_depend_eliminate]: 1.233e-05
        [updatestate_assign_eliminate]: 5.91998e-06
        [updatestate_loads_eliminate]: 4.42e-06
        [cse]: 5.217e-05
        [call_graph_tuple_transform]: 2.847e-05
        [tuple_list_get_item_eliminator]: 1.101e-05
        [none_parameter_eliminate]: 2.07001e-06
        [renormalize]: 1.03001e-06
        [switch_simplify]: 1.295e-05
[remove_dup_value]: 2.5e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 1.587e-05
[add_recomputation]: 7.985e-05
[cse_after_recomputation]: 4.236e-05, [1]
    [Cycle 1]: 3.498e-05, [1]
        [cse]: 2.67e-05
[auto_monad_reorder]: 3.365e-05
[get_jit_bprop_graph]: 2.66999e-06
[rewriter_after_jit_bprop_graph]: 1.2e-05
[opt_after_jit_grad]: 0.00077369
[symbol_engine_optimizer]: 0.00014176, [1]
    [Cycle 1]: 0.00013141, [6]
        [build]: 2.3e-05
        [elim_shapecalc]: 1.504e-05
        [elim_not_effective]: 2.581e-05
        [opt_reshape]: 1.046e-05
        [fold_const_symbol]: 1.883e-05
        [renormalize]: 1.12999e-06
[validate]: 7.816e-05
[backend_pass]: 1.35999e-06
[task_emit]: 0.0290914
[execute]: 9.83002e-06
Sums
    bootstrap                            :     0.000338s :  0.27%
    type_inference                       :     0.070471s : 56.87%
    event_method                         :     0.000025s :  0.02%
    auto_monad                           :     0.000183s :  0.15%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000172s :  0.14%
    rewriter_before_opt_a                :     0.000106s :  0.09%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000091s :  0.07%
    jit_opt_a.loop_unroll                :     0.000046s :  0.04%
    jit_opt_a.a_1                        :     0.017393s : 14.04%
    jit_opt_a.with_stream_mark           :     0.000060s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000027s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.00%
    jit_opt_a.specialize_transform       :     0.000021s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000021s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000018s :  0.01%
    jit_opt_a.merge_forward              :     0.000013s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000010s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000046s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000031s :  0.02%
    jit_opt_a.meta_fg_expand             :     0.000008s :  0.01%
    jit_opt_a.replace_old_param          :     0.000029s :  0.02%
    jit_opt_a.inline_without_move        :     0.000017s :  0.01%
    jit_opt_a.renormalize                :     0.001758s :  1.42%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000040s :  0.03%
    jit_opt_a.cse                        :     0.000078s :  0.06%
    jit_opt_a.replace_applicator         :     0.000038s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000022s :  0.02%
    rewriter_after_opt_a                 :     0.000377s :  0.30%
    convert_after_rewriter               :     0.000019s :  0.02%
    order_py_execute_after_rewriter      :     0.000030s :  0.02%
    mutable_eliminate                    :     0.001072s :  0.86%
    jit_opt_b.frontend_op_eliminate      :     0.000033s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000035s :  0.03%
    cconv                                :     0.000039s :  0.03%
    loop_unroll                          :     0.000704s :  0.57%
    jit_opt_after_cconv.c_1              :     0.000068s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.cse              :     0.000052s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.02%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000011s :  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  :     0.000013s :  0.01%
    remove_dup_value                     :     0.000025s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000016s :  0.01%
    add_recomputation                    :     0.000080s :  0.06%
    cse_after_recomputation.cse          :     0.000027s :  0.02%
    auto_monad_reorder                   :     0.000034s :  0.03%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.01%
    opt_after_jit_grad                   :     0.000774s :  0.62%
    symbol_engine_optimizer.build        :     0.000023s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000015s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000026s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000019s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000078s :  0.06%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.029091s : 23.48%
    execute                              :     0.000010s :  0.01%


Time group info:
------[substitution.]   0.016556    47
 0.08% :     0.000013s :      2: substitution.depend_value_elim
 0.02% :     0.000004s :      4: substitution.elim_not_effective
 0.02% :     0.000004s :      4: substitution.fold_const_symbol
 0.05% :     0.000008s :      5: substitution.graph_param_transform
99.36% :     0.016451s :      4: substitution.inline
 0.04% :     0.000007s :      8: substitution.j_node_and_user_rematch
 0.05% :     0.000009s :      8: substitution.remove_not_recompute_node
 0.06% :     0.000009s :      2: substitution.replace_old_param
 0.12% :     0.000020s :      2: substitution.tuple_list_get_item_eliminator
 0.10% :     0.000017s :      3: substitution.updatestate_pure_node_eliminater
 0.09% :     0.000014s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.070372     2
97.10% :     0.068334s :      1: type_inference.infer
 2.90% :     0.002037s :      1: type_inference.specialize
------[replace.]   0.000185     6
65.30% :     0.000121s :      4: replace.inline
34.70% :     0.000064s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.016464     6
99.89% :     0.016446s :      4: match.inline
 0.11% :     0.000019s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000229  1021
 1.06% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.40% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.16% :     0.000003s :     15: predicate.addn_check_dump
 1.61% :     0.000004s :     15: predicate.addn_zero_filter
 2.20% :     0.000005s :     15: predicate.arithmetic_simplify
 1.64% :     0.000004s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     15: predicate.compare_switch_simplify
 1.52% :     0.000003s :     15: predicate.depend_value_elim
 1.12% :     0.000003s :     15: predicate.dict_get_item_const_eliminator
 1.11% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.23% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.89% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      5: predicate.elim_not_effective
 0.55% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.07% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.17% :     0.000003s :     15: predicate.environ_get_depend_swap
 1.32% :     0.000003s :     15: predicate.environ_get_eliminate
 1.01% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      5: predicate.fold_const_symbol
 1.10% :     0.000003s :     10: predicate.get_grad_eliminate
 0.32% :     0.000001s :      5: predicate.graph_param_transform
 5.97% :     0.000014s :     31: predicate.inline
 0.78% :     0.000002s :     10: predicate.inline_without_move
 0.31% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.26% :     0.000003s :     10: predicate.less_batch_normalization
 1.54% :     0.000004s :     17: predicate.list_to_tuple_eliminator_
 2.29% :     0.000005s :     22: predicate.load_eliminater
 1.98% :     0.000005s :      5: predicate.loop_unroll_after_grad
 2.68% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.05% :     0.000005s :     20: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000002s :     15: predicate.merge_addn
 1.01% :     0.000002s :     15: predicate.minmaximum_grad
 2.47% :     0.000006s :      5: predicate.mutable_eliminate
 0.49% :     0.000001s :      5: predicate.opt_reshape
 2.14% :     0.000005s :     22: predicate.partial_eliminate
 1.18% :     0.000003s :     15: predicate.print_const_string_wrapper
 2.34% :     0.000005s :     15: predicate.reduce_eliminate
 1.64% :     0.000004s :     17: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.32% :     0.000003s :     27: predicate.replace_applicator
 0.54% :     0.000001s :     10: predicate.replace_old_param
 0.35% :     0.000001s :      5: predicate.reset_defer_inline
 1.49% :     0.000003s :     15: predicate.reshape_eliminate
 1.24% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 1.04% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.29% :     0.000003s :     15: predicate.same_eliminate
 0.43% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.35% :     0.000003s :     10: predicate.special_op_eliminate
 0.82% :     0.000002s :     10: predicate.specialize_transform
 1.27% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.49% :     0.000006s :     21: predicate.switch_defer_inline
 1.79% :     0.000004s :     21: predicate.switch_layer_defer_inline
 5.49% :     0.000013s :     68: predicate.switch_simplify
 1.31% :     0.000003s :     15: predicate.tile_eliminate
 1.32% :     0.000003s :     15: predicate.transpose_eliminate
 1.57% :     0.000004s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 4.71% :     0.000011s :     27: predicate.tuple_list_get_item_eliminator
 1.98% :     0.000005s :     15: predicate.tuple_list_set_item_eliminator
 1.17% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.95% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000007s :     32: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000004s :     15: predicate.value_based_eliminate
 0.37% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.017270    13
 2.62% :     0.000453s :      6: func_graph_cloner_run.FuncGraphClonerGraph
93.86% :     0.016210s :      1: func_graph_cloner_run.FuncGraphClonerNode
 3.52% :     0.000608s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.149333    76
 0.06% :     0.000084s :      1: add_recomputation
 0.13% :     0.000189s :      1: auto_monad
 0.02% :     0.000037s :      1: auto_monad_reorder
 0.00% :     0.000005s :      1: backend_pass
 0.24% :     0.000360s :      1: bootstrap
 0.03% :     0.000042s :      1: cconv
 0.02% :     0.000023s :      1: convert_after_rewriter
 0.03% :     0.000045s :      1: cse_after_recomputation
 0.01% :     0.000018s :      1: environ_conv
 0.02% :     0.000030s :      1: event_method
 0.01% :     0.000016s :      1: execute
 0.00% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
16.72% :     0.024974s :      1: jit_opt_a
 0.19% :     0.000281s :      1: jit_opt_after_cconv
 0.07% :     0.000100s :      1: jit_opt_b
 0.48% :     0.000716s :      1: loop_unroll
 0.73% :     0.001091s :      1: mutable_eliminate
11.89% :     0.017757s :     26: opt.transform.jit_opt_a
 0.08% :     0.000115s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000057s :      4: opt.transform.jit_opt_b
 0.02% :     0.000029s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000039s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000047s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000066s :      4: opt.transform.symbol_engine_opt
 0.53% :     0.000791s :      1: opt_after_jit_grad
 0.02% :     0.000034s :      1: order_py_execute_after_rewriter
 0.00% :     0.000005s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000177s :      1: py_interpret_to_execute
 0.02% :     0.000026s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000028s :      1: remove_dup_value
 0.72% :     0.001069s :      1: renormalize.infer
 0.45% :     0.000675s :      1: renormalize.specialize
 0.01% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000387s :      1: rewriter_after_opt_a
 0.07% :     0.000111s :      1: rewriter_before_opt_a
 0.10% :     0.000146s :      1: symbol_engine_optimizer
19.50% :     0.029113s :      1: task_emit
47.21% :     0.070496s :      1: type_inference
 0.08% :     0.000115s :      1: validate


TotalTime = 0.0478086, [33]
[bootstrap]: 0.00033566
[type_inference]: 0.0260227
[event_method]: 4.984e-05
[auto_monad]: 0.00017977
[graph_reusing]: 6.34001e-06
[pre_auto_parallel]: 3.12002e-06
[py_interpret_to_execute]: 0.00015102
[rewriter_before_opt_a]: 9.771e-05
[expand_dump_flag]: 4.12e-06
[jit_opt_a]: 0.00716427, [2]
    [Cycle 1]: 0.00273254, [27]
        [switch_simplify]: 6.91e-05
        [loop_unroll]: 3.62e-05
        [a_1]: 0.00077453
        [with_stream_mark]: 2.653e-05
        [recompute_prepare]: 1.15e-05
        [updatestate_depend_eliminate]: 6.88998e-06
        [updatestate_assign_eliminate]: 6.14001e-06
        [updatestate_loads_eliminate]: 4.92999e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 9.97999e-06
        [updatestate_useless_node_eliminater]: 1.178e-05
        [accelerated_algorithm]: 9.24e-06
        [meta_shard_fg_expand]: 2.96001e-06
        [get_grad_eliminate_]: 8.70001e-06
        [merge_forward]: 5.62001e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.167e-05
        [j_node_and_user_rematch]: 1.506e-05
        [meta_fg_expand]: 3.63e-06
        [replace_old_param]: 1.296e-05
        [inline_without_move]: 8.28999e-06
        [renormalize]: 0.00138954
        [add_forward_monad_depend]: 6.64001e-06
        [auto_monad_grad]: 3.35e-06
        [auto_monad_eliminator]: 2.314e-05
        [cse]: 4.461e-05
        [replace_applicator]: 1.888e-05
    [Cycle 2]: 0.00048159, [27]
        [switch_simplify]: 9.19e-06
        [loop_unroll]: 8.62e-06
        [a_1]: 0.00017654
        [with_stream_mark]: 1.251e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 5.31002e-06
        [updatestate_assign_eliminate]: 5.33002e-06
        [updatestate_loads_eliminate]: 3.98999e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 7.97998e-06
        [updatestate_useless_node_eliminater]: 1.062e-05
        [accelerated_algorithm]: 8.09002e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 7.80998e-06
        [merge_forward]: 4.68001e-06
        [cell_reuse_recompute_pass]: 2.34001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.673e-05
        [j_node_and_user_rematch]: 1.226e-05
        [meta_fg_expand]: 2.88e-06
        [replace_old_param]: 1.033e-05
        [inline_without_move]: 7.36999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.66002e-06
        [auto_monad_grad]: 1.22999e-06
        [auto_monad_eliminator]: 1.064e-05
        [cse]: 2.109e-05
        [replace_applicator]: 8.39002e-06
[py_interpret_to_execute_after_opt_a]: 1.607e-05
[rewriter_after_opt_a]: 0.00029921
[convert_after_rewriter]: 1.314e-05
[order_py_execute_after_rewriter]: 8e-06
[mutable_eliminate]: 0.00077521
[jit_opt_b]: 7.417e-05, [1]
    [Cycle 1]: 6.538e-05, [2]
        [frontend_op_eliminate]: 2.636e-05
        [inline_after_opt_a]: 2.624e-05
[cconv]: 2.976e-05
[loop_unroll]: 0.00047765
[jit_opt_after_cconv]: 0.00021009, [1]
    [Cycle 1]: 0.00020271, [11]
        [c_1]: 4.935e-05
        [parameter_eliminate]: 3.25998e-06
        [updatestate_depend_eliminate]: 8.23999e-06
        [updatestate_assign_eliminate]: 4.62e-06
        [updatestate_loads_eliminate]: 4.39998e-06
        [cse]: 3.419e-05
        [call_graph_tuple_transform]: 2.447e-05
        [tuple_list_get_item_eliminator]: 8.83001e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 1.18001e-06
        [switch_simplify]: 9.08002e-06
[remove_dup_value]: 2.081e-05
[partial_unused_args_eliminate]: 2.74001e-06
[environ_conv]: 1.201e-05
[add_recomputation]: 7.273e-05
[cse_after_recomputation]: 3.394e-05, [1]
    [Cycle 1]: 2.759e-05, [1]
        [cse]: 2.081e-05
[auto_monad_reorder]: 2.698e-05
[get_jit_bprop_graph]: 2.26e-06
[rewriter_after_jit_bprop_graph]: 8.10999e-06
[opt_after_jit_grad]: 0.00051488
[symbol_engine_optimizer]: 0.00011031, [1]
    [Cycle 1]: 0.00010333, [6]
        [build]: 1.519e-05
        [elim_shapecalc]: 1.253e-05
        [elim_not_effective]: 2.056e-05
        [opt_reshape]: 8.56002e-06
        [fold_const_symbol]: 1.409e-05
        [renormalize]: 6.50005e-07
[validate]: 6.654e-05
[backend_pass]: 1.30999e-06
[task_emit]: 0.01071
[execute]: 7.66001e-06
Sums
    bootstrap                            :     0.000336s :  0.78%
    type_inference                       :     0.026023s : 60.36%
    event_method                         :     0.000050s :  0.12%
    auto_monad                           :     0.000180s :  0.42%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000151s :  0.35%
    rewriter_before_opt_a                :     0.000098s :  0.23%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000078s :  0.18%
    jit_opt_a.loop_unroll                :     0.000045s :  0.10%
    jit_opt_a.a_1                        :     0.000951s :  2.21%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  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    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001390s :  3.22%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000034s :  0.08%
    jit_opt_a.cse                        :     0.000066s :  0.15%
    jit_opt_a.replace_applicator         :     0.000027s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.04%
    rewriter_after_opt_a                 :     0.000299s :  0.69%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000775s :  1.80%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000030s :  0.07%
    loop_unroll                          :     0.000478s :  1.11%
    jit_opt_after_cconv.c_1              :     0.000049s :  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              :     0.000034s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000073s :  0.17%
    cse_after_recomputation.cse          :     0.000021s :  0.05%
    auto_monad_reorder                   :     0.000027s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000515s :  1.19%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000067s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010710s : 24.84%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000330    47
 3.27% :     0.000011s :      2: substitution.depend_value_elim
 0.91% :     0.000003s :      4: substitution.elim_not_effective
 0.79% :     0.000003s :      4: substitution.fold_const_symbol
 2.03% :     0.000007s :      5: substitution.graph_param_transform
77.15% :     0.000255s :      4: substitution.inline
 1.35% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.48% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.61% :     0.000005s :      2: substitution.replace_old_param
 3.94% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.26% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.21% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.025936     2
93.51% :     0.024253s :      1: type_inference.infer
 6.49% :     0.001683s :      1: type_inference.specialize
------[replace.]   0.000099     6
67.47% :     0.000067s :      4: replace.inline
32.53% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000263     6
95.48% :     0.000251s :      4: match.inline
 4.52% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1021
 1.51% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.07% :     0.000002s :     15: predicate.addn_check_dump
 1.46% :     0.000003s :     15: predicate.addn_zero_filter
 1.91% :     0.000003s :     15: predicate.arithmetic_simplify
 1.44% :     0.000003s :     15: predicate.cast_eliminate
 0.71% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.44% :     0.000003s :     15: predicate.compare_switch_simplify
 1.24% :     0.000002s :     15: predicate.depend_value_elim
 1.22% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.42% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.98% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      5: predicate.elim_not_effective
 0.66% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.30% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     15: predicate.environ_get_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.03% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 5.25% :     0.000009s :     31: predicate.inline
 0.97% :     0.000002s :     10: predicate.inline_without_move
 0.40% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.50% :     0.000003s :     10: predicate.less_batch_normalization
 1.62% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.08% :     0.000004s :     22: predicate.load_eliminater
 1.40% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.31% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.15% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     15: predicate.merge_addn
 1.31% :     0.000002s :     15: predicate.minmaximum_grad
 1.60% :     0.000003s :      5: predicate.mutable_eliminate
 0.45% :     0.000001s :      5: predicate.opt_reshape
 2.27% :     0.000004s :     22: predicate.partial_eliminate
 1.19% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.73% :     0.000003s :     15: predicate.reduce_eliminate
 1.83% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.62% :     0.000003s :     27: predicate.replace_applicator
 0.54% :     0.000001s :     10: predicate.replace_old_param
 0.46% :     0.000001s :      5: predicate.reset_defer_inline
 1.24% :     0.000002s :     15: predicate.reshape_eliminate
 1.28% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     15: predicate.same_eliminate
 0.55% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.96% :     0.000002s :     10: predicate.special_op_eliminate
 1.07% :     0.000002s :     10: predicate.specialize_transform
 1.42% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.05% :     0.000004s :     21: predicate.switch_defer_inline
 1.76% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.44% :     0.000011s :     68: predicate.switch_simplify
 1.24% :     0.000002s :     15: predicate.tile_eliminate
 1.16% :     0.000002s :     15: predicate.transpose_eliminate
 1.55% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.79% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.45% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001020    13
45.93% :     0.000469s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.91% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.16% :     0.000491s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050654    76
 0.15% :     0.000076s :      1: add_recomputation
 0.37% :     0.000186s :      1: auto_monad
 0.06% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.70% :     0.000355s :      1: bootstrap
 0.06% :     0.000033s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.07% :     0.000037s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.11% :     0.000054s :      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
14.15% :     0.007169s :      1: jit_opt_a
 0.42% :     0.000213s :      1: jit_opt_after_cconv
 0.15% :     0.000077s :      1: jit_opt_b
 0.96% :     0.000486s :      1: loop_unroll
 1.55% :     0.000785s :      1: mutable_eliminate
 2.50% :     0.001269s :     26: opt.transform.jit_opt_a
 0.17% :     0.000087s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000045s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.04% :     0.000525s :      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
 0.31% :     0.000156s :      1: py_interpret_to_execute
 0.04% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000023s :      1: remove_dup_value
 1.68% :     0.000852s :      1: renormalize.infer
 1.04% :     0.000526s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.68% :     0.000345s :      1: rewriter_after_opt_a
 0.20% :     0.000101s :      1: rewriter_before_opt_a
 0.22% :     0.000114s :      1: symbol_engine_optimizer
21.18% :     0.010730s :      1: task_emit
51.41% :     0.026042s :      1: type_inference
 0.19% :     0.000096s :      1: validate


TotalTime = 0.0500091, [33]
[bootstrap]: 0.00032678
[type_inference]: 0.0263683
[event_method]: 2.679e-05
[auto_monad]: 0.00017947
[graph_reusing]: 7.08e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 0.00015247
[rewriter_before_opt_a]: 0.00010229
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.00753797, [2]
    [Cycle 1]: 0.00273708, [27]
        [switch_simplify]: 6.785e-05
        [loop_unroll]: 3.708e-05
        [a_1]: 0.00077693
        [with_stream_mark]: 2.5e-05
        [recompute_prepare]: 1.384e-05
        [updatestate_depend_eliminate]: 7.13e-06
        [updatestate_assign_eliminate]: 6.11998e-06
        [updatestate_loads_eliminate]: 5.12e-06
        [parameter_eliminate]: 1.99999e-06
        [specialize_transform]: 9.87999e-06
        [updatestate_useless_node_eliminater]: 1.277e-05
        [accelerated_algorithm]: 8.75001e-06
        [meta_shard_fg_expand]: 3.23998e-06
        [get_grad_eliminate_]: 8.89998e-06
        [merge_forward]: 5.17e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.065e-05
        [j_node_and_user_rematch]: 1.441e-05
        [meta_fg_expand]: 3.48e-06
        [replace_old_param]: 1.311e-05
        [inline_without_move]: 8.43001e-06
        [renormalize]: 0.00136645
        [add_forward_monad_depend]: 8.88002e-06
        [auto_monad_grad]: 2.78e-06
        [auto_monad_eliminator]: 2.479e-05
        [cse]: 4.449e-05
        [replace_applicator]: 2.329e-05
    [Cycle 2]: 0.00051751, [27]
        [switch_simplify]: 9.50001e-06
        [loop_unroll]: 8.61002e-06
        [a_1]: 0.00018277
        [with_stream_mark]: 1.757e-05
        [recompute_prepare]: 8.39998e-06
        [updatestate_depend_eliminate]: 6.64001e-06
        [updatestate_assign_eliminate]: 5.77001e-06
        [updatestate_loads_eliminate]: 4.02998e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 8.38999e-06
        [updatestate_useless_node_eliminater]: 1.073e-05
        [accelerated_algorithm]: 8.33999e-06
        [meta_shard_fg_expand]: 2.34001e-06
        [get_grad_eliminate_]: 8.05e-06
        [merge_forward]: 5.24e-06
        [cell_reuse_recompute_pass]: 3.30003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.957e-05
        [j_node_and_user_rematch]: 1.341e-05
        [meta_fg_expand]: 3.25002e-06
        [replace_old_param]: 1.103e-05
        [inline_without_move]: 7.65998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 1.12e-05
        [cse]: 2.389e-05
        [replace_applicator]: 8.08001e-06
[py_interpret_to_execute_after_opt_a]: 1.658e-05
[rewriter_after_opt_a]: 0.00030651
[convert_after_rewriter]: 1.172e-05
[order_py_execute_after_rewriter]: 7.8e-06
[mutable_eliminate]: 0.00082113
[jit_opt_b]: 7.607e-05, [1]
    [Cycle 1]: 6.778e-05, [2]
        [frontend_op_eliminate]: 2.558e-05
        [inline_after_opt_a]: 2.878e-05
[cconv]: 2.949e-05
[loop_unroll]: 0.00050697
[jit_opt_after_cconv]: 0.00022569, [1]
    [Cycle 1]: 0.00021804, [11]
        [c_1]: 5.25e-05
        [parameter_eliminate]: 3.68999e-06
        [updatestate_depend_eliminate]: 9.64e-06
        [updatestate_assign_eliminate]: 5.40999e-06
        [updatestate_loads_eliminate]: 4.70001e-06
        [cse]: 4.425e-05
        [call_graph_tuple_transform]: 2.441e-05
        [tuple_list_get_item_eliminator]: 8.55001e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 9.80013e-07
        [switch_simplify]: 9.04998e-06
[remove_dup_value]: 2.005e-05
[partial_unused_args_eliminate]: 2.64001e-06
[environ_conv]: 1.232e-05
[add_recomputation]: 7.745e-05
[cse_after_recomputation]: 3.402e-05, [1]
    [Cycle 1]: 2.641e-05, [1]
        [cse]: 1.963e-05
[auto_monad_reorder]: 2.647e-05
[get_jit_bprop_graph]: 2.31998e-06
[rewriter_after_jit_bprop_graph]: 7.87e-06
[opt_after_jit_grad]: 0.00054043
[symbol_engine_optimizer]: 0.00017151, [1]
    [Cycle 1]: 0.00016304, [6]
        [build]: 1.669e-05
        [elim_shapecalc]: 1.229e-05
        [elim_not_effective]: 5.941e-05
        [opt_reshape]: 1.008e-05
        [fold_const_symbol]: 2.891e-05
        [renormalize]: 6.10016e-07
[validate]: 6.546e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.0120607
[execute]: 9.25999e-06
Sums
    bootstrap                            :     0.000327s :  0.73%
    type_inference                       :     0.026368s : 58.63%
    event_method                         :     0.000027s :  0.06%
    auto_monad                           :     0.000179s :  0.40%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000152s :  0.34%
    rewriter_before_opt_a                :     0.000102s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.17%
    jit_opt_a.loop_unroll                :     0.000046s :  0.10%
    jit_opt_a.a_1                        :     0.000960s :  2.13%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.05%
    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 :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001367s :  3.04%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000036s :  0.08%
    jit_opt_a.cse                        :     0.000068s :  0.15%
    jit_opt_a.replace_applicator         :     0.000031s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.04%
    rewriter_after_opt_a                 :     0.000307s :  0.68%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000821s :  1.83%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000029s :  0.06%
    cconv                                :     0.000029s :  0.07%
    loop_unroll                          :     0.000507s :  1.13%
    jit_opt_after_cconv.c_1              :     0.000052s :  0.12%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000044s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000077s :  0.17%
    cse_after_recomputation.cse          :     0.000020s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000540s :  1.20%
    symbol_engine_optimizer.build        :     0.000017s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000059s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000029s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000065s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012061s : 26.82%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000326    47
 3.27% :     0.000011s :      2: substitution.depend_value_elim
 0.92% :     0.000003s :      4: substitution.elim_not_effective
 0.94% :     0.000003s :      4: substitution.fold_const_symbol
 2.32% :     0.000008s :      5: substitution.graph_param_transform
75.66% :     0.000247s :      4: substitution.inline
 1.61% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.25% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.62% :     0.000005s :      2: substitution.replace_old_param
 3.96% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.77% :     0.000012s :      3: substitution.updatestate_pure_node_eliminater
 3.68% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026269     2
93.14% :     0.024468s :      1: type_inference.infer
 6.86% :     0.001801s :      1: type_inference.specialize
------[replace.]   0.000095     6
66.52% :     0.000063s :      4: replace.inline
33.48% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000256     6
95.39% :     0.000244s :      4: match.inline
 4.61% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000198  1021
 1.33% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.05% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.32% :     0.000003s :     15: predicate.addn_check_dump
 1.47% :     0.000003s :     15: predicate.addn_zero_filter
 1.95% :     0.000004s :     15: predicate.arithmetic_simplify
 1.16% :     0.000002s :     15: predicate.cast_eliminate
 0.44% :     0.000001s :      5: predicate.check_bprop_eliminate
 0.95% :     0.000002s :     15: predicate.compare_switch_simplify
 1.12% :     0.000002s :     15: predicate.depend_value_elim
 1.20% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.48% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.01% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.89% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.61% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.01% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     15: predicate.environ_get_eliminate
 1.25% :     0.000002s :     15: predicate.environ_get_set_eliminate
 6.99% :     0.000014s :      5: predicate.fold_const_symbol
 1.03% :     0.000002s :     10: predicate.get_grad_eliminate
 0.30% :     0.000001s :      5: predicate.graph_param_transform
 5.08% :     0.000010s :     31: predicate.inline
 0.88% :     0.000002s :     10: predicate.inline_without_move
 0.36% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :     10: predicate.less_batch_normalization
 1.60% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.67% :     0.000003s :     22: predicate.load_eliminater
 1.32% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.03% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.08% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.32% :     0.000003s :     15: predicate.merge_addn
 0.97% :     0.000002s :     15: predicate.minmaximum_grad
 1.77% :     0.000004s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.24% :     0.000004s :     22: predicate.partial_eliminate
 1.11% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.61% :     0.000003s :     15: predicate.reduce_eliminate
 1.24% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.56% :     0.000003s :     27: predicate.replace_applicator
 0.70% :     0.000001s :     10: predicate.replace_old_param
 0.31% :     0.000001s :      5: predicate.reset_defer_inline
 1.12% :     0.000002s :     15: predicate.reshape_eliminate
 1.12% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.89% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.34% :     0.000003s :     15: predicate.same_eliminate
 0.52% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.88% :     0.000002s :     10: predicate.special_op_eliminate
 0.94% :     0.000002s :     10: predicate.specialize_transform
 1.34% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.15% :     0.000004s :     21: predicate.switch_defer_inline
 1.65% :     0.000003s :     21: predicate.switch_layer_defer_inline
 5.94% :     0.000012s :     68: predicate.switch_simplify
 1.19% :     0.000002s :     15: predicate.tile_eliminate
 1.08% :     0.000002s :     15: predicate.transpose_eliminate
 1.33% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.64% :     0.000009s :     27: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.23% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.53% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000003s :     15: predicate.value_based_eliminate
 0.39% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000998    13
43.13% :     0.000430s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.16% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.71% :     0.000506s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.052916    76
 0.15% :     0.000081s :      1: add_recomputation
 0.35% :     0.000186s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000344s :      1: bootstrap
 0.06% :     0.000032s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000036s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.06% :     0.000032s :      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
14.25% :     0.007543s :      1: jit_opt_a
 0.43% :     0.000229s :      1: jit_opt_after_cconv
 0.15% :     0.000080s :      1: jit_opt_b
 0.97% :     0.000516s :      1: loop_unroll
 1.58% :     0.000836s :      1: mutable_eliminate
 2.43% :     0.001284s :     26: opt.transform.jit_opt_a
 0.17% :     0.000090s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000046s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000028s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000107s :      4: opt.transform.symbol_engine_opt
 1.04% :     0.000550s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000157s :      1: py_interpret_to_execute
 0.04% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000023s :      1: remove_dup_value
 1.48% :     0.000785s :      1: renormalize.infer
 1.08% :     0.000572s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.59% :     0.000312s :      1: rewriter_after_opt_a
 0.20% :     0.000107s :      1: rewriter_before_opt_a
 0.33% :     0.000175s :      1: symbol_engine_optimizer
22.83% :     0.012081s :      1: task_emit
49.88% :     0.026394s :      1: type_inference
 0.17% :     0.000092s :      1: validate


TotalTime = 0.0466082, [33]
[bootstrap]: 0.00032245
[type_inference]: 0.025369
[event_method]: 2.15e-05
[auto_monad]: 0.00016805
[graph_reusing]: 6.29001e-06
[pre_auto_parallel]: 3.27997e-06
[py_interpret_to_execute]: 0.00013288
[rewriter_before_opt_a]: 9.956e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.00714882, [2]
    [Cycle 1]: 0.00273714, [27]
        [switch_simplify]: 6.871e-05
        [loop_unroll]: 3.646e-05
        [a_1]: 0.00076834
        [with_stream_mark]: 2.677e-05
        [recompute_prepare]: 1.061e-05
        [updatestate_depend_eliminate]: 6.63e-06
        [updatestate_assign_eliminate]: 5.56998e-06
        [updatestate_loads_eliminate]: 4.85001e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 9.73002e-06
        [updatestate_useless_node_eliminater]: 1.234e-05
        [accelerated_algorithm]: 8.62998e-06
        [meta_shard_fg_expand]: 2.64001e-06
        [get_grad_eliminate_]: 8.35001e-06
        [merge_forward]: 5.70001e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.044e-05
        [j_node_and_user_rematch]: 1.392e-05
        [meta_fg_expand]: 3.55e-06
        [replace_old_param]: 1.308e-05
        [inline_without_move]: 8.23001e-06
        [renormalize]: 0.00140364
        [add_forward_monad_depend]: 8.63001e-06
        [auto_monad_grad]: 3.31999e-06
        [auto_monad_eliminator]: 2.281e-05
        [cse]: 4.327e-05
        [replace_applicator]: 2.044e-05
    [Cycle 2]: 0.00050354, [27]
        [switch_simplify]: 9.37999e-06
        [loop_unroll]: 8.62e-06
        [a_1]: 0.00018146
        [with_stream_mark]: 1.587e-05
        [recompute_prepare]: 8.87e-06
        [updatestate_depend_eliminate]: 5.77001e-06
        [updatestate_assign_eliminate]: 5.10001e-06
        [updatestate_loads_eliminate]: 4.18001e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 8.07e-06
        [updatestate_useless_node_eliminater]: 1.076e-05
        [accelerated_algorithm]: 8.12e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 7.38e-06
        [merge_forward]: 5.54998e-06
        [cell_reuse_recompute_pass]: 3.33e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.831e-05
        [j_node_and_user_rematch]: 1.299e-05
        [meta_fg_expand]: 2.91999e-06
        [replace_old_param]: 1.124e-05
        [inline_without_move]: 7.55e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.60999e-06
        [auto_monad_grad]: 1.06997e-06
        [auto_monad_eliminator]: 1.027e-05
        [cse]: 2.416e-05
        [replace_applicator]: 8.31002e-06
[py_interpret_to_execute_after_opt_a]: 1.657e-05
[rewriter_after_opt_a]: 0.00030275
[convert_after_rewriter]: 1.348e-05
[order_py_execute_after_rewriter]: 8.03001e-06
[mutable_eliminate]: 0.00070484
[jit_opt_b]: 7.385e-05, [1]
    [Cycle 1]: 6.536e-05, [2]
        [frontend_op_eliminate]: 2.55e-05
        [inline_after_opt_a]: 2.725e-05
[cconv]: 3.518e-05
[loop_unroll]: 0.0004808
[jit_opt_after_cconv]: 0.0002241, [1]
    [Cycle 1]: 0.00021707, [11]
        [c_1]: 5.362e-05
        [parameter_eliminate]: 4.42e-06
        [updatestate_depend_eliminate]: 9.82999e-06
        [updatestate_assign_eliminate]: 4.65001e-06
        [updatestate_loads_eliminate]: 4.91002e-06
        [cse]: 4.192e-05
        [call_graph_tuple_transform]: 2.39e-05
        [tuple_list_get_item_eliminator]: 8.85001e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 9.05999e-06
[remove_dup_value]: 1.986e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 1.392e-05
[add_recomputation]: 7.342e-05
[cse_after_recomputation]: 3.345e-05, [1]
    [Cycle 1]: 2.728e-05, [1]
        [cse]: 2.034e-05
[auto_monad_reorder]: 2.677e-05
[get_jit_bprop_graph]: 2.22001e-06
[rewriter_after_jit_bprop_graph]: 7.51999e-06
[opt_after_jit_grad]: 0.00055702
[symbol_engine_optimizer]: 0.00011207, [1]
    [Cycle 1]: 0.00010419, [6]
        [build]: 1.722e-05
        [elim_shapecalc]: 1.252e-05
        [elim_not_effective]: 2.013e-05
        [opt_reshape]: 9.14e-06
        [fold_const_symbol]: 1.381e-05
        [renormalize]: 4.7998e-07
[validate]: 6.663e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.0102877
[execute]: 5.62999e-06
Sums
    bootstrap                            :     0.000322s :  0.77%
    type_inference                       :     0.025369s : 60.42%
    event_method                         :     0.000022s :  0.05%
    auto_monad                           :     0.000168s :  0.40%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000133s :  0.32%
    rewriter_before_opt_a                :     0.000100s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000078s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000950s :  2.26%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    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  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000024s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001404s :  3.34%
    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      :     0.000033s :  0.08%
    jit_opt_a.cse                        :     0.000067s :  0.16%
    jit_opt_a.replace_applicator         :     0.000029s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.04%
    rewriter_after_opt_a                 :     0.000303s :  0.72%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000705s :  1.68%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.06%
    cconv                                :     0.000035s :  0.08%
    loop_unroll                          :     0.000481s :  1.15%
    jit_opt_after_cconv.c_1              :     0.000054s :  0.13%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000042s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000014s :  0.03%
    add_recomputation                    :     0.000073s :  0.17%
    cse_after_recomputation.cse          :     0.000020s :  0.05%
    auto_monad_reorder                   :     0.000027s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000557s :  1.33%
    symbol_engine_optimizer.build        :     0.000017s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000067s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010288s : 24.50%
    execute                              :     0.000006s :  0.01%


Time group info:
------[substitution.]   0.000336    47
 3.09% :     0.000010s :      2: substitution.depend_value_elim
 0.93% :     0.000003s :      4: substitution.elim_not_effective
 0.63% :     0.000002s :      4: substitution.fold_const_symbol
 2.22% :     0.000007s :      5: substitution.graph_param_transform
76.38% :     0.000257s :      4: substitution.inline
 1.48% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.23% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.94% :     0.000007s :      2: substitution.replace_old_param
 4.15% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.85% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.10% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.025286     2
93.80% :     0.023719s :      1: type_inference.infer
 6.20% :     0.001567s :      1: type_inference.specialize
------[replace.]   0.000095     6
67.09% :     0.000064s :      4: replace.inline
32.91% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     6
95.23% :     0.000254s :      4: match.inline
 4.77% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000177  1021
 1.26% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.09% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.34% :     0.000002s :     15: predicate.addn_zero_filter
 2.26% :     0.000004s :     15: predicate.arithmetic_simplify
 1.33% :     0.000002s :     15: predicate.cast_eliminate
 0.60% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.23% :     0.000002s :     15: predicate.compare_switch_simplify
 1.40% :     0.000002s :     15: predicate.depend_value_elim
 1.09% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.96% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      5: predicate.elim_not_effective
 0.54% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.26% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     15: predicate.environ_get_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.00% :     0.000002s :     10: predicate.get_grad_eliminate
 0.29% :     0.000001s :      5: predicate.graph_param_transform
 5.83% :     0.000010s :     31: predicate.inline
 1.00% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.49% :     0.000003s :     10: predicate.less_batch_normalization
 1.47% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.79% :     0.000003s :     22: predicate.load_eliminater
 1.34% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.48% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.01% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.09% :     0.000002s :     15: predicate.minmaximum_grad
 1.97% :     0.000003s :      5: predicate.mutable_eliminate
 0.58% :     0.000001s :      5: predicate.opt_reshape
 2.43% :     0.000004s :     22: predicate.partial_eliminate
 1.13% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.60% :     0.000003s :     15: predicate.reduce_eliminate
 1.56% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.76% :     0.000003s :     27: predicate.replace_applicator
 0.76% :     0.000001s :     10: predicate.replace_old_param
 0.52% :     0.000001s :      5: predicate.reset_defer_inline
 1.27% :     0.000002s :     15: predicate.reshape_eliminate
 1.21% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.86% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     15: predicate.same_eliminate
 0.60% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.04% :     0.000002s :     10: predicate.special_op_eliminate
 1.10% :     0.000002s :     10: predicate.specialize_transform
 1.41% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.60% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.09% :     0.000004s :     21: predicate.switch_defer_inline
 1.94% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.54% :     0.000012s :     68: predicate.switch_simplify
 1.13% :     0.000002s :     15: predicate.tile_eliminate
 1.21% :     0.000002s :     15: predicate.transpose_eliminate
 1.68% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.35% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000969    13
43.84% :     0.000425s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.61% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.55% :     0.000480s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049462    76
 0.15% :     0.000076s :      1: add_recomputation
 0.35% :     0.000174s :      1: auto_monad
 0.06% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000343s :      1: bootstrap
 0.08% :     0.000038s :      1: cconv
 0.03% :     0.000017s :      1: convert_after_rewriter
 0.07% :     0.000036s :      1: cse_after_recomputation
 0.03% :     0.000016s :      1: environ_conv
 0.05% :     0.000027s :      1: event_method
 0.02% :     0.000010s :      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
14.46% :     0.007153s :      1: jit_opt_a
 0.46% :     0.000227s :      1: jit_opt_after_cconv
 0.16% :     0.000077s :      1: jit_opt_b
 0.99% :     0.000490s :      1: loop_unroll
 1.45% :     0.000716s :      1: mutable_eliminate
 2.56% :     0.001266s :     26: opt.transform.jit_opt_a
 0.19% :     0.000092s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000045s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000568s :      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
 0.28% :     0.000137s :      1: py_interpret_to_execute
 0.04% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000023s :      1: remove_dup_value
 1.81% :     0.000895s :      1: renormalize.infer
 1.01% :     0.000498s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000310s :      1: rewriter_after_opt_a
 0.21% :     0.000103s :      1: rewriter_before_opt_a
 0.23% :     0.000115s :      1: symbol_engine_optimizer
20.83% :     0.010302s :      1: task_emit
51.33% :     0.025387s :      1: type_inference
 0.19% :     0.000094s :      1: validate


TotalTime = 0.0485469, [33]
[bootstrap]: 0.00033608
[type_inference]: 0.0250874
[event_method]: 2.336e-05
[auto_monad]: 0.00018175
[graph_reusing]: 6.36e-06
[pre_auto_parallel]: 2.95002e-06
[py_interpret_to_execute]: 0.00014468
[rewriter_before_opt_a]: 0.0001028
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00760098, [2]
    [Cycle 1]: 0.00273581, [27]
        [switch_simplify]: 7.559e-05
        [loop_unroll]: 3.899e-05
        [a_1]: 0.0008156
        [with_stream_mark]: 2.742e-05
        [recompute_prepare]: 1.289e-05
        [updatestate_depend_eliminate]: 7.14001e-06
        [updatestate_assign_eliminate]: 6.47001e-06
        [updatestate_loads_eliminate]: 5.17e-06
        [parameter_eliminate]: 2.76e-06
        [specialize_transform]: 1.071e-05
        [updatestate_useless_node_eliminater]: 1.311e-05
        [accelerated_algorithm]: 9.24998e-06
        [meta_shard_fg_expand]: 3.14999e-06
        [get_grad_eliminate_]: 9.24e-06
        [merge_forward]: 5.81e-06
        [cell_reuse_recompute_pass]: 1.82001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.265e-05
        [j_node_and_user_rematch]: 1.514e-05
        [meta_fg_expand]: 3.87002e-06
        [replace_old_param]: 1.388e-05
        [inline_without_move]: 8.78001e-06
        [renormalize]: 0.00129997
        [add_forward_monad_depend]: 7.92e-06
        [auto_monad_grad]: 3.14999e-06
        [auto_monad_eliminator]: 2.528e-05
        [cse]: 4.132e-05
        [replace_applicator]: 2.399e-05
    [Cycle 2]: 0.00052584, [27]
        [switch_simplify]: 1.056e-05
        [loop_unroll]: 8.35999e-06
        [a_1]: 0.00018114
        [with_stream_mark]: 1.787e-05
        [recompute_prepare]: 9.17001e-06
        [updatestate_depend_eliminate]: 6.54001e-06
        [updatestate_assign_eliminate]: 5.84999e-06
        [updatestate_loads_eliminate]: 4.47e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 8.04002e-06
        [updatestate_useless_node_eliminater]: 1.115e-05
        [accelerated_algorithm]: 8.36002e-06
        [meta_shard_fg_expand]: 3.09999e-06
        [get_grad_eliminate_]: 7.65e-06
        [merge_forward]: 5.52001e-06
        [cell_reuse_recompute_pass]: 4.03999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.979e-05
        [j_node_and_user_rematch]: 1.447e-05
        [meta_fg_expand]: 3.23e-06
        [replace_old_param]: 1.199e-05
        [inline_without_move]: 7.74997e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.79e-06
        [auto_monad_grad]: 1.91e-06
        [auto_monad_eliminator]: 1.201e-05
        [cse]: 2.494e-05
        [replace_applicator]: 9.46e-06
[py_interpret_to_execute_after_opt_a]: 1.815e-05
[rewriter_after_opt_a]: 0.00031165
[convert_after_rewriter]: 1.252e-05
[order_py_execute_after_rewriter]: 8.2e-06
[mutable_eliminate]: 0.00083165
[jit_opt_b]: 7.885e-05, [1]
    [Cycle 1]: 6.823e-05, [2]
        [frontend_op_eliminate]: 2.489e-05
        [inline_after_opt_a]: 2.949e-05
[cconv]: 3.095e-05
[loop_unroll]: 0.00051056
[jit_opt_after_cconv]: 0.00022469, [1]
    [Cycle 1]: 0.00021736, [11]
        [c_1]: 4.908e-05
        [parameter_eliminate]: 4.18001e-06
        [updatestate_depend_eliminate]: 9.46e-06
        [updatestate_assign_eliminate]: 5.37001e-06
        [updatestate_loads_eliminate]: 5.04998e-06
        [cse]: 4.353e-05
        [call_graph_tuple_transform]: 2.423e-05
        [tuple_list_get_item_eliminator]: 8.47e-06
        [none_parameter_eliminate]: 2.20002e-06
        [renormalize]: 1.02e-06
        [switch_simplify]: 9.41003e-06
[remove_dup_value]: 2.089e-05
[partial_unused_args_eliminate]: 3.01001e-06
[environ_conv]: 1.247e-05
[add_recomputation]: 7.331e-05
[cse_after_recomputation]: 3.378e-05, [1]
    [Cycle 1]: 2.734e-05, [1]
        [cse]: 1.971e-05
[auto_monad_reorder]: 2.544e-05
[get_jit_bprop_graph]: 2.03002e-06
[rewriter_after_jit_bprop_graph]: 9.04e-06
[opt_after_jit_grad]: 0.00055691
[symbol_engine_optimizer]: 0.00011138, [1]
    [Cycle 1]: 0.0001036, [6]
        [build]: 1.658e-05
        [elim_shapecalc]: 1.309e-05
        [elim_not_effective]: 2.002e-05
        [opt_reshape]: 8.83001e-06
        [fold_const_symbol]: 1.461e-05
        [renormalize]: 9.50007e-07
[validate]: 6.109e-05
[backend_pass]: 1.39998e-06
[task_emit]: 0.0118427
[execute]: 8.74e-06
Sums
    bootstrap                            :     0.000336s :  0.77%
    type_inference                       :     0.025087s : 57.74%
    event_method                         :     0.000023s :  0.05%
    auto_monad                           :     0.000182s :  0.42%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000145s :  0.33%
    rewriter_before_opt_a                :     0.000103s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000086s :  0.20%
    jit_opt_a.loop_unroll                :     0.000047s :  0.11%
    jit_opt_a.a_1                        :     0.000997s :  2.29%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.05%
    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 :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000042s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000026s :  0.06%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001300s :  2.99%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000037s :  0.09%
    jit_opt_a.cse                        :     0.000066s :  0.15%
    jit_opt_a.replace_applicator         :     0.000033s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.04%
    rewriter_after_opt_a                 :     0.000312s :  0.72%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000832s :  1.91%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000029s :  0.07%
    cconv                                :     0.000031s :  0.07%
    loop_unroll                          :     0.000511s :  1.18%
    jit_opt_after_cconv.c_1              :     0.000049s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000044s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000073s :  0.17%
    cse_after_recomputation.cse          :     0.000020s :  0.05%
    auto_monad_reorder                   :     0.000025s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000557s :  1.28%
    symbol_engine_optimizer.build        :     0.000017s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000061s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011843s : 27.26%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000360    47
 3.16% :     0.000011s :      2: substitution.depend_value_elim
 0.73% :     0.000003s :      4: substitution.elim_not_effective
 0.60% :     0.000002s :      4: substitution.fold_const_symbol
 2.07% :     0.000007s :      5: substitution.graph_param_transform
77.12% :     0.000278s :      4: substitution.inline
 1.55% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.40% :     0.000009s :      8: 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
 3.28% :     0.000012s :      3: substitution.updatestate_pure_node_eliminater
 3.32% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024994     2
93.44% :     0.023354s :      1: type_inference.infer
 6.56% :     0.001640s :      1: type_inference.specialize
------[replace.]   0.000096     6
67.46% :     0.000065s :      4: replace.inline
32.54% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000288     6
95.53% :     0.000275s :      4: match.inline
 4.47% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000187  1021
 1.63% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.11% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.06% :     0.000002s :     15: predicate.addn_check_dump
 1.41% :     0.000003s :     15: predicate.addn_zero_filter
 1.99% :     0.000004s :     15: predicate.arithmetic_simplify
 1.59% :     0.000003s :     15: predicate.cast_eliminate
 0.57% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.05% :     0.000002s :     15: predicate.compare_switch_simplify
 1.27% :     0.000002s :     15: predicate.depend_value_elim
 1.02% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.13% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.83% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      5: predicate.elim_not_effective
 0.64% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000003s :     15: predicate.environ_add_const_eliminate
 1.06% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.32% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     15: predicate.environ_get_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.13% :     0.000002s :     10: predicate.get_grad_eliminate
 0.26% :     0.000000s :      5: predicate.graph_param_transform
 5.29% :     0.000010s :     31: predicate.inline
 0.92% :     0.000002s :     10: predicate.inline_without_move
 0.38% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.41% :     0.000003s :     10: predicate.less_batch_normalization
 1.49% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.74% :     0.000003s :     22: predicate.load_eliminater
 1.51% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.30% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.93% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     15: predicate.merge_addn
 1.05% :     0.000002s :     15: predicate.minmaximum_grad
 2.28% :     0.000004s :      5: predicate.mutable_eliminate
 0.49% :     0.000001s :      5: predicate.opt_reshape
 2.27% :     0.000004s :     22: predicate.partial_eliminate
 1.14% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.80% :     0.000003s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.93% :     0.000004s :     27: predicate.replace_applicator
 0.65% :     0.000001s :     10: predicate.replace_old_param
 0.29% :     0.000001s :      5: predicate.reset_defer_inline
 1.37% :     0.000003s :     15: predicate.reshape_eliminate
 1.12% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.82% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.42% :     0.000003s :     15: predicate.same_eliminate
 0.54% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.99% :     0.000002s :     10: predicate.special_op_eliminate
 1.21% :     0.000002s :     10: predicate.specialize_transform
 1.73% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.21% :     0.000004s :     21: predicate.switch_defer_inline
 1.83% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.89% :     0.000013s :     68: predicate.switch_simplify
 1.51% :     0.000003s :     15: predicate.tile_eliminate
 1.29% :     0.000002s :     15: predicate.transpose_eliminate
 1.39% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.95% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.62% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.15% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.87% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000985    13
45.34% :     0.000447s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.96% :     0.000069s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.69% :     0.000470s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.051388    76
 0.15% :     0.000077s :      1: add_recomputation
 0.37% :     0.000188s :      1: auto_monad
 0.05% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000356s :      1: bootstrap
 0.07% :     0.000034s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.07% :     0.000036s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.06% :     0.000029s :      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
14.80% :     0.007604s :      1: jit_opt_a
 0.44% :     0.000228s :      1: jit_opt_after_cconv
 0.16% :     0.000082s :      1: jit_opt_b
 1.01% :     0.000520s :      1: loop_unroll
 1.65% :     0.000846s :      1: mutable_eliminate
 2.61% :     0.001339s :     26: opt.transform.jit_opt_a
 0.17% :     0.000087s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000046s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000029s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.10% :     0.000567s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000006s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000149s :      1: py_interpret_to_execute
 0.04% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000024s :      1: remove_dup_value
 1.56% :     0.000802s :      1: renormalize.infer
 0.95% :     0.000486s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.62% :     0.000320s :      1: rewriter_after_opt_a
 0.21% :     0.000107s :      1: rewriter_before_opt_a
 0.22% :     0.000114s :      1: symbol_engine_optimizer
23.08% :     0.011862s :      1: task_emit
48.87% :     0.025111s :      1: type_inference
 0.17% :     0.000088s :      1: validate


TotalTime = 0.0440521, [33]
[bootstrap]: 0.00030533
[type_inference]: 0.0235841
[event_method]: 2.099e-05
[auto_monad]: 0.00015858
[graph_reusing]: 6.25002e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 0.00012543
[rewriter_before_opt_a]: 9.835e-05
[expand_dump_flag]: 3.38999e-06
[jit_opt_a]: 0.00674855, [2]
    [Cycle 1]: 0.00238411, [27]
        [switch_simplify]: 6.509e-05
        [loop_unroll]: 3.652e-05
        [a_1]: 0.00071838
        [with_stream_mark]: 1.935e-05
        [recompute_prepare]: 1.132e-05
        [updatestate_depend_eliminate]: 6.47001e-06
        [updatestate_assign_eliminate]: 5.96e-06
        [updatestate_loads_eliminate]: 4.97e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 9.59e-06
        [updatestate_useless_node_eliminater]: 1.164e-05
        [accelerated_algorithm]: 9.14e-06
        [meta_shard_fg_expand]: 2.81e-06
        [get_grad_eliminate_]: 8.23999e-06
        [merge_forward]: 6.34999e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.004e-05
        [j_node_and_user_rematch]: 1.445e-05
        [meta_fg_expand]: 3.48e-06
        [replace_old_param]: 1.145e-05
        [inline_without_move]: 8.31002e-06
        [renormalize]: 0.00112275
        [add_forward_monad_depend]: 7.23e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 2.24e-05
        [cse]: 4.235e-05
        [replace_applicator]: 2.008e-05
    [Cycle 2]: 0.00056353, [27]
        [switch_simplify]: 9.39e-06
        [loop_unroll]: 8.16002e-06
        [a_1]: 0.00024384
        [with_stream_mark]: 1.572e-05
        [recompute_prepare]: 8.94e-06
        [updatestate_depend_eliminate]: 5.61e-06
        [updatestate_assign_eliminate]: 5.19998e-06
        [updatestate_loads_eliminate]: 4.18999e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 8.45001e-06
        [updatestate_useless_node_eliminater]: 1.178e-05
        [accelerated_algorithm]: 7.87e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 7.21999e-06
        [merge_forward]: 4.99e-06
        [cell_reuse_recompute_pass]: 2.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.755e-05
        [j_node_and_user_rematch]: 1.31e-05
        [meta_fg_expand]: 3.06001e-06
        [replace_old_param]: 1.066e-05
        [inline_without_move]: 7.6e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.00999e-06
        [auto_monad_grad]: 1.47001e-06
        [auto_monad_eliminator]: 1.111e-05
        [cse]: 2.344e-05
        [replace_applicator]: 8.17998e-06
[py_interpret_to_execute_after_opt_a]: 1.381e-05
[rewriter_after_opt_a]: 0.00028116
[convert_after_rewriter]: 1.139e-05
[order_py_execute_after_rewriter]: 7.2e-06
[mutable_eliminate]: 0.00072078
[jit_opt_b]: 6.91e-05, [1]
    [Cycle 1]: 6.159e-05, [2]
        [frontend_op_eliminate]: 2.451e-05
        [inline_after_opt_a]: 2.45e-05
[cconv]: 3.045e-05
[loop_unroll]: 0.00045009
[jit_opt_after_cconv]: 0.00020283, [1]
    [Cycle 1]: 0.00019639, [11]
        [c_1]: 4.709e-05
        [parameter_eliminate]: 3.26001e-06
        [updatestate_depend_eliminate]: 9.32001e-06
        [updatestate_assign_eliminate]: 4.64002e-06
        [updatestate_loads_eliminate]: 3.95e-06
        [cse]: 3.176e-05
        [call_graph_tuple_transform]: 2.353e-05
        [tuple_list_get_item_eliminator]: 8.23999e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 1.10001e-06
        [switch_simplify]: 8.95999e-06
[remove_dup_value]: 2.019e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 1.069e-05
[add_recomputation]: 7.058e-05
[cse_after_recomputation]: 3.075e-05, [1]
    [Cycle 1]: 2.459e-05, [1]
        [cse]: 1.858e-05
[auto_monad_reorder]: 2.519e-05
[get_jit_bprop_graph]: 2.37999e-06
[rewriter_after_jit_bprop_graph]: 5.59e-06
[opt_after_jit_grad]: 0.00050137
[symbol_engine_optimizer]: 0.00010119, [1]
    [Cycle 1]: 9.408e-05, [6]
        [build]: 1.191e-05
        [elim_shapecalc]: 1.126e-05
        [elim_not_effective]: 1.889e-05
        [opt_reshape]: 8.66002e-06
        [fold_const_symbol]: 1.331e-05
        [renormalize]: 4.00003e-07
[validate]: 5.887e-05
[backend_pass]: 1.42e-06
[task_emit]: 0.0101384
[execute]: 5.45001e-06
Sums
    bootstrap                            :     0.000305s :  0.77%
    type_inference                       :     0.023584s : 59.59%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000159s :  0.40%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000125s :  0.32%
    rewriter_before_opt_a                :     0.000098s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000962s :  2.43%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.03%
    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    :     0.000028s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001123s :  2.84%
    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      :     0.000034s :  0.08%
    jit_opt_a.cse                        :     0.000066s :  0.17%
    jit_opt_a.replace_applicator         :     0.000028s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000281s :  0.71%
    convert_after_rewriter               :     0.000011s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000721s :  1.82%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000030s :  0.08%
    loop_unroll                          :     0.000450s :  1.14%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000032s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000071s :  0.18%
    cse_after_recomputation.cse          :     0.000019s :  0.05%
    auto_monad_reorder                   :     0.000025s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000501s :  1.27%
    symbol_engine_optimizer.build        :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000059s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010138s : 25.62%
    execute                              :     0.000005s :  0.01%


Time group info:
------[substitution.]   0.000287    47
 3.47% :     0.000010s :      2: substitution.depend_value_elim
 0.92% :     0.000003s :      4: substitution.elim_not_effective
 0.66% :     0.000002s :      4: substitution.fold_const_symbol
 2.50% :     0.000007s :      5: substitution.graph_param_transform
74.57% :     0.000214s :      4: substitution.inline
 1.79% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.69% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.78% :     0.000005s :      2: substitution.replace_old_param
 4.68% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.74% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.21% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023504     2
93.09% :     0.021881s :      1: type_inference.infer
 6.91% :     0.001624s :      1: type_inference.specialize
------[replace.]   0.000093     6
67.49% :     0.000063s :      4: replace.inline
32.51% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000223     6
94.52% :     0.000211s :      4: match.inline
 5.48% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1021
 1.37% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.12% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     15: predicate.addn_check_dump
 1.58% :     0.000003s :     15: predicate.addn_zero_filter
 1.99% :     0.000003s :     15: predicate.arithmetic_simplify
 1.37% :     0.000002s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     15: predicate.compare_switch_simplify
 1.32% :     0.000002s :     15: predicate.depend_value_elim
 1.11% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.94% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.52% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     15: predicate.environ_get_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.00% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 5.13% :     0.000009s :     31: predicate.inline
 1.01% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.50% :     0.000003s :     10: predicate.less_batch_normalization
 1.34% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     22: predicate.load_eliminater
 1.33% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.49% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     15: predicate.merge_addn
 1.14% :     0.000002s :     15: predicate.minmaximum_grad
 1.49% :     0.000003s :      5: predicate.mutable_eliminate
 0.48% :     0.000001s :      5: predicate.opt_reshape
 2.41% :     0.000004s :     22: predicate.partial_eliminate
 1.22% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.77% :     0.000003s :     15: predicate.reduce_eliminate
 1.47% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.70% :     0.000003s :     27: predicate.replace_applicator
 0.59% :     0.000001s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.19% :     0.000002s :     15: predicate.reshape_eliminate
 1.51% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :     10: predicate.special_op_eliminate
 1.10% :     0.000002s :     10: predicate.specialize_transform
 1.47% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.41% :     0.000004s :     21: predicate.switch_defer_inline
 1.90% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.72% :     0.000012s :     68: predicate.switch_simplify
 1.30% :     0.000002s :     15: predicate.tile_eliminate
 1.19% :     0.000002s :     15: predicate.transpose_eliminate
 1.52% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.79% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     15: predicate.value_based_eliminate
 0.45% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000939    13
40.23% :     0.000378s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.49% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
55.28% :     0.000519s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046633    76
 0.16% :     0.000074s :      1: add_recomputation
 0.35% :     0.000164s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.70% :     0.000326s :      1: bootstrap
 0.07% :     0.000033s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.05% :     0.000025s :      1: event_method
 0.02% :     0.000009s :      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
14.48% :     0.006753s :      1: jit_opt_a
 0.44% :     0.000206s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 0.98% :     0.000459s :      1: loop_unroll
 1.57% :     0.000732s :      1: mutable_eliminate
 2.73% :     0.001274s :     26: opt.transform.jit_opt_a
 0.18% :     0.000084s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      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
 0.10% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.09% :     0.000511s :      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
 0.28% :     0.000130s :      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
 1.44% :     0.000673s :      1: renormalize.infer
 0.95% :     0.000441s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000286s :      1: rewriter_after_opt_a
 0.22% :     0.000103s :      1: rewriter_before_opt_a
 0.22% :     0.000104s :      1: symbol_engine_optimizer
21.77% :     0.010152s :      1: task_emit
50.61% :     0.023600s :      1: type_inference
 0.18% :     0.000086s :      1: validate


TotalTime = 0.0467636, [33]
[bootstrap]: 0.00030743
[type_inference]: 0.0242663
[event_method]: 2.348e-05
[auto_monad]: 0.00016895
[graph_reusing]: 5.84e-06
[pre_auto_parallel]: 3.18e-06
[py_interpret_to_execute]: 0.00012731
[rewriter_before_opt_a]: 9.939e-05
[expand_dump_flag]: 4.15999e-06
[jit_opt_a]: 0.0073089, [2]
    [Cycle 1]: 0.00281964, [27]
        [switch_simplify]: 7.161e-05
        [loop_unroll]: 3.848e-05
        [a_1]: 0.00079692
        [with_stream_mark]: 2.674e-05
        [recompute_prepare]: 1.079e-05
        [updatestate_depend_eliminate]: 7.03998e-06
        [updatestate_assign_eliminate]: 6.29999e-06
        [updatestate_loads_eliminate]: 5.22999e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 9.42999e-06
        [updatestate_useless_node_eliminater]: 1.341e-05
        [accelerated_algorithm]: 9.62001e-06
        [meta_shard_fg_expand]: 3.07002e-06
        [get_grad_eliminate_]: 9.39998e-06
        [merge_forward]: 6.34001e-06
        [cell_reuse_recompute_pass]: 1.88997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.138e-05
        [j_node_and_user_rematch]: 1.459e-05
        [meta_fg_expand]: 3.73999e-06
        [replace_old_param]: 1.302e-05
        [inline_without_move]: 8.99e-06
        [renormalize]: 0.00142199
        [add_forward_monad_depend]: 8.85999e-06
        [auto_monad_grad]: 3.03e-06
        [auto_monad_eliminator]: 2.627e-05
        [cse]: 4.526e-05
        [replace_applicator]: 1.944e-05
    [Cycle 2]: 0.00051161, [27]
        [switch_simplify]: 9.79999e-06
        [loop_unroll]: 8.40999e-06
        [a_1]: 0.00017978
        [with_stream_mark]: 1.869e-05
        [recompute_prepare]: 8.15999e-06
        [updatestate_depend_eliminate]: 5.16002e-06
        [updatestate_assign_eliminate]: 6.15997e-06
        [updatestate_loads_eliminate]: 3.97998e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 8.54998e-06
        [updatestate_useless_node_eliminater]: 1.067e-05
        [accelerated_algorithm]: 8.38001e-06
        [meta_shard_fg_expand]: 2.66e-06
        [get_grad_eliminate_]: 7.7e-06
        [merge_forward]: 4.99998e-06
        [cell_reuse_recompute_pass]: 3.78001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.891e-05
        [j_node_and_user_rematch]: 1.398e-05
        [meta_fg_expand]: 3.48999e-06
        [replace_old_param]: 1.144e-05
        [inline_without_move]: 7.89002e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 1.52001e-06
        [auto_monad_eliminator]: 1.094e-05
        [cse]: 2.535e-05
        [replace_applicator]: 8.42e-06
[py_interpret_to_execute_after_opt_a]: 1.833e-05
[rewriter_after_opt_a]: 0.0002963
[convert_after_rewriter]: 1.287e-05
[order_py_execute_after_rewriter]: 7.93999e-06
[mutable_eliminate]: 0.00079152
[jit_opt_b]: 7.915e-05, [1]
    [Cycle 1]: 7.002e-05, [2]
        [frontend_op_eliminate]: 2.577e-05
        [inline_after_opt_a]: 2.987e-05
[cconv]: 3.721e-05
[loop_unroll]: 0.00057593
[jit_opt_after_cconv]: 0.00023556, [1]
    [Cycle 1]: 0.00022739, [11]
        [c_1]: 5.527e-05
        [parameter_eliminate]: 4.65999e-06
        [updatestate_depend_eliminate]: 9.81e-06
        [updatestate_assign_eliminate]: 5.43002e-06
        [updatestate_loads_eliminate]: 5.66e-06
        [cse]: 4.462e-05
        [call_graph_tuple_transform]: 2.506e-05
        [tuple_list_get_item_eliminator]: 9.53997e-06
        [none_parameter_eliminate]: 1.99e-06
        [renormalize]: 1.19e-06
        [switch_simplify]: 9.21002e-06
[remove_dup_value]: 2.012e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 1.351e-05
[add_recomputation]: 7.408e-05
[cse_after_recomputation]: 3.58e-05, [1]
    [Cycle 1]: 2.829e-05, [1]
        [cse]: 2.097e-05
[auto_monad_reorder]: 3.002e-05
[get_jit_bprop_graph]: 2.26e-06
[rewriter_after_jit_bprop_graph]: 8.23999e-06
[opt_after_jit_grad]: 0.00067223
[symbol_engine_optimizer]: 0.00012171, [1]
    [Cycle 1]: 0.00011338, [6]
        [build]: 1.629e-05
        [elim_shapecalc]: 1.282e-05
        [elim_not_effective]: 2.382e-05
        [opt_reshape]: 1.059e-05
        [fold_const_symbol]: 1.674e-05
        [renormalize]: 7.59988e-07
[validate]: 6.711e-05
[backend_pass]: 1.47999e-06
[task_emit]: 0.0110659
[execute]: 9.41e-06
Sums
    bootstrap                            :     0.000307s :  0.73%
    type_inference                       :     0.024266s : 57.72%
    event_method                         :     0.000023s :  0.06%
    auto_monad                           :     0.000169s :  0.40%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000127s :  0.30%
    rewriter_before_opt_a                :     0.000099s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000081s :  0.19%
    jit_opt_a.loop_unroll                :     0.000047s :  0.11%
    jit_opt_a.a_1                        :     0.000977s :  2.32%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  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 :     0.000024s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000024s :  0.06%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001422s :  3.38%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000037s :  0.09%
    jit_opt_a.cse                        :     0.000071s :  0.17%
    jit_opt_a.replace_applicator         :     0.000028s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.04%
    rewriter_after_opt_a                 :     0.000296s :  0.70%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000792s :  1.88%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000030s :  0.07%
    cconv                                :     0.000037s :  0.09%
    loop_unroll                          :     0.000576s :  1.37%
    jit_opt_after_cconv.c_1              :     0.000055s :  0.13%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.cse              :     0.000045s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000010s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000014s :  0.03%
    add_recomputation                    :     0.000074s :  0.18%
    cse_after_recomputation.cse          :     0.000021s :  0.05%
    auto_monad_reorder                   :     0.000030s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000672s :  1.60%
    symbol_engine_optimizer.build        :     0.000016s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000024s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000011s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000017s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000067s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011066s : 26.32%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000349    47
 3.39% :     0.000012s :      2: substitution.depend_value_elim
 0.81% :     0.000003s :      4: substitution.elim_not_effective
 0.75% :     0.000003s :      4: substitution.fold_const_symbol
 2.02% :     0.000007s :      5: substitution.graph_param_transform
76.46% :     0.000266s :      4: substitution.inline
 1.47% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.21% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.72% :     0.000006s :      2: substitution.replace_old_param
 3.85% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.61% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.70% :     0.000013s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024186     2
93.18% :     0.022537s :      1: type_inference.infer
 6.82% :     0.001649s :      1: type_inference.specialize
------[replace.]   0.000100     6
65.78% :     0.000066s :      4: replace.inline
34.22% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000275     6
95.56% :     0.000263s :      4: match.inline
 4.44% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000181  1021
 1.35% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.07% :     0.000002s :     15: predicate.addn_check_dump
 1.44% :     0.000003s :     15: predicate.addn_zero_filter
 1.99% :     0.000004s :     15: predicate.arithmetic_simplify
 1.45% :     0.000003s :     15: predicate.cast_eliminate
 0.51% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.44% :     0.000003s :     15: predicate.compare_switch_simplify
 1.25% :     0.000002s :     15: predicate.depend_value_elim
 1.27% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.00% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      5: predicate.elim_not_effective
 0.68% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     15: predicate.environ_get_eliminate
 1.30% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.12% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 5.37% :     0.000010s :     31: predicate.inline
 1.37% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.82% :     0.000003s :     10: predicate.less_batch_normalization
 1.47% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.72% :     0.000003s :     22: predicate.load_eliminater
 1.38% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.34% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.94% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000002s :     15: predicate.merge_addn
 1.06% :     0.000002s :     15: predicate.minmaximum_grad
 1.62% :     0.000003s :      5: predicate.mutable_eliminate
 0.61% :     0.000001s :      5: predicate.opt_reshape
 2.35% :     0.000004s :     22: predicate.partial_eliminate
 1.22% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.75% :     0.000003s :     15: predicate.reduce_eliminate
 1.54% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.67% :     0.000003s :     27: predicate.replace_applicator
 0.60% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.18% :     0.000002s :     15: predicate.reshape_eliminate
 1.20% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.41% :     0.000003s :     15: predicate.same_eliminate
 0.52% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.97% :     0.000002s :     10: predicate.special_op_eliminate
 1.03% :     0.000002s :     10: predicate.specialize_transform
 1.49% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.65% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.43% :     0.000004s :     21: predicate.switch_defer_inline
 1.85% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.34% :     0.000011s :     68: predicate.switch_simplify
 1.16% :     0.000002s :     15: predicate.tile_eliminate
 1.22% :     0.000002s :     15: predicate.transpose_eliminate
 1.46% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 2.11% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.24% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000003s :     15: predicate.value_based_eliminate
 0.39% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001012    13
40.37% :     0.000409s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.68% :     0.000068s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.95% :     0.000536s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049710    76
 0.16% :     0.000078s :      1: add_recomputation
 0.35% :     0.000175s :      1: auto_monad
 0.07% :     0.000033s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.65% :     0.000323s :      1: bootstrap
 0.08% :     0.000041s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.08% :     0.000039s :      1: cse_after_recomputation
 0.03% :     0.000016s :      1: environ_conv
 0.06% :     0.000027s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
14.71% :     0.007313s :      1: jit_opt_a
 0.48% :     0.000239s :      1: jit_opt_after_cconv
 0.17% :     0.000082s :      1: jit_opt_b
 1.18% :     0.000587s :      1: loop_unroll
 1.62% :     0.000804s :      1: mutable_eliminate
 2.63% :     0.001305s :     26: opt.transform.jit_opt_a
 0.19% :     0.000095s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000046s :      4: opt.transform.jit_opt_b
 0.04% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000039s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000060s :      4: opt.transform.symbol_engine_opt
 1.38% :     0.000685s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000131s :      1: py_interpret_to_execute
 0.04% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000023s :      1: remove_dup_value
 1.77% :     0.000878s :      1: renormalize.infer
 1.07% :     0.000532s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000302s :      1: rewriter_after_opt_a
 0.21% :     0.000104s :      1: rewriter_before_opt_a
 0.25% :     0.000125s :      1: symbol_engine_optimizer
22.30% :     0.011088s :      1: task_emit
48.86% :     0.024288s :      1: type_inference
 0.20% :     0.000098s :      1: validate


TotalTime = 0.0465349, [33]
[bootstrap]: 0.000343
[type_inference]: 0.0244492
[event_method]: 2.107e-05
[auto_monad]: 0.00016919
[graph_reusing]: 6.76999e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 0.00013289
[rewriter_before_opt_a]: 9.795e-05
[expand_dump_flag]: 4.17998e-06
[jit_opt_a]: 0.00734249, [2]
    [Cycle 1]: 0.00283428, [27]
        [switch_simplify]: 7.087e-05
        [loop_unroll]: 3.703e-05
        [a_1]: 0.0007737
        [with_stream_mark]: 2.581e-05
        [recompute_prepare]: 1.167e-05
        [updatestate_depend_eliminate]: 6.69999e-06
        [updatestate_assign_eliminate]: 5.76998e-06
        [updatestate_loads_eliminate]: 4.68999e-06
        [parameter_eliminate]: 2.15002e-06
        [specialize_transform]: 9.28002e-06
        [updatestate_useless_node_eliminater]: 1.202e-05
        [accelerated_algorithm]: 8.26002e-06
        [meta_shard_fg_expand]: 3.17997e-06
        [get_grad_eliminate_]: 8.02e-06
        [merge_forward]: 5.64998e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.964e-05
        [j_node_and_user_rematch]: 0.00026157
        [meta_fg_expand]: 4.28999e-06
        [replace_old_param]: 1.634e-05
        [inline_without_move]: 8.95999e-06
        [renormalize]: 0.0012272
        [add_forward_monad_depend]: 7.98999e-06
        [auto_monad_grad]: 3.39001e-06
        [auto_monad_eliminator]: 2.35e-05
        [cse]: 4.196e-05
        [replace_applicator]: 2.151e-05
    [Cycle 2]: 0.00049066, [27]
        [switch_simplify]: 9.32999e-06
        [loop_unroll]: 8.64e-06
        [a_1]: 0.00017247
        [with_stream_mark]: 1.547e-05
        [recompute_prepare]: 8.34998e-06
        [updatestate_depend_eliminate]: 5.46998e-06
        [updatestate_assign_eliminate]: 5.20999e-06
        [updatestate_loads_eliminate]: 4.03999e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 8.28001e-06
        [updatestate_useless_node_eliminater]: 1.118e-05
        [accelerated_algorithm]: 7.7e-06
        [meta_shard_fg_expand]: 2.71999e-06
        [get_grad_eliminate_]: 7.24001e-06
        [merge_forward]: 5.54998e-06
        [cell_reuse_recompute_pass]: 2.89999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.927e-05
        [j_node_and_user_rematch]: 1.316e-05
        [meta_fg_expand]: 3.00002e-06
        [replace_old_param]: 1.019e-05
        [inline_without_move]: 7.16999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 1.39e-06
        [auto_monad_eliminator]: 1.092e-05
        [cse]: 2.133e-05
        [replace_applicator]: 7.59002e-06
[py_interpret_to_execute_after_opt_a]: 1.62e-05
[rewriter_after_opt_a]: 0.00029627
[convert_after_rewriter]: 1.19e-05
[order_py_execute_after_rewriter]: 7.65e-06
[mutable_eliminate]: 0.00077833
[jit_opt_b]: 7.362e-05, [1]
    [Cycle 1]: 6.462e-05, [2]
        [frontend_op_eliminate]: 2.635e-05
        [inline_after_opt_a]: 2.541e-05
[cconv]: 3.286e-05
[loop_unroll]: 0.00052063
[jit_opt_after_cconv]: 0.00025438, [1]
    [Cycle 1]: 0.00024638, [11]
        [c_1]: 7.779e-05
        [parameter_eliminate]: 4.58001e-06
        [updatestate_depend_eliminate]: 8.82e-06
        [updatestate_assign_eliminate]: 5.66998e-06
        [updatestate_loads_eliminate]: 5.05001e-06
        [cse]: 3.906e-05
        [call_graph_tuple_transform]: 2.593e-05
        [tuple_list_get_item_eliminator]: 9.18002e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 9.89996e-07
        [switch_simplify]: 9.52001e-06
[remove_dup_value]: 1.916e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 1.319e-05
[add_recomputation]: 7.447e-05
[cse_after_recomputation]: 3.152e-05, [1]
    [Cycle 1]: 2.464e-05, [1]
        [cse]: 1.814e-05
[auto_monad_reorder]: 2.606e-05
[get_jit_bprop_graph]: 2.16998e-06
[rewriter_after_jit_bprop_graph]: 6.81001e-06
[opt_after_jit_grad]: 0.00052549
[symbol_engine_optimizer]: 0.0001061, [1]
    [Cycle 1]: 9.98e-05, [6]
        [build]: 1.479e-05
        [elim_shapecalc]: 1.168e-05
        [elim_not_effective]: 1.975e-05
        [opt_reshape]: 9.42001e-06
        [fold_const_symbol]: 1.412e-05
        [renormalize]: 5.49975e-07
[validate]: 6.213e-05
[backend_pass]: 1.18001e-06
[task_emit]: 0.0107812
[execute]: 1.094e-05
Sums
    bootstrap                            :     0.000343s :  0.82%
    type_inference                       :     0.024449s : 58.58%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000169s :  0.41%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000133s :  0.32%
    rewriter_before_opt_a                :     0.000098s :  0.23%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.19%
    jit_opt_a.loop_unroll                :     0.000046s :  0.11%
    jit_opt_a.a_1                        :     0.000946s :  2.27%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  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 :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000275s :  0.66%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000027s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001227s :  2.94%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000034s :  0.08%
    jit_opt_a.cse                        :     0.000063s :  0.15%
    jit_opt_a.replace_applicator         :     0.000029s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.04%
    rewriter_after_opt_a                 :     0.000296s :  0.71%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000778s :  1.86%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000033s :  0.08%
    loop_unroll                          :     0.000521s :  1.25%
    jit_opt_after_cconv.c_1              :     0.000078s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000039s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000074s :  0.18%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000525s :  1.26%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000062s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010781s : 25.83%
    execute                              :     0.000011s :  0.03%


Time group info:
------[substitution.]   0.000329    47
 3.28% :     0.000011s :      2: substitution.depend_value_elim
 0.96% :     0.000003s :      4: substitution.elim_not_effective
 0.65% :     0.000002s :      4: substitution.fold_const_symbol
 2.49% :     0.000008s :      5: substitution.graph_param_transform
75.91% :     0.000250s :      4: substitution.inline
 1.95% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.24% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.70% :     0.000006s :      2: substitution.replace_old_param
 4.43% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.35% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.03% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024365     2
94.18% :     0.022947s :      1: type_inference.infer
 5.82% :     0.001418s :      1: type_inference.specialize
------[replace.]   0.000098     6
69.20% :     0.000068s :      4: replace.inline
30.80% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000260     6
94.82% :     0.000247s :      4: match.inline
 5.18% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1021
 1.26% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.07% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.06% :     0.000002s :     15: predicate.addn_check_dump
 1.42% :     0.000003s :     15: predicate.addn_zero_filter
 1.93% :     0.000003s :     15: predicate.arithmetic_simplify
 1.45% :     0.000003s :     15: predicate.cast_eliminate
 0.53% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.06% :     0.000002s :     15: predicate.compare_switch_simplify
 1.33% :     0.000002s :     15: predicate.depend_value_elim
 1.05% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.64% :     0.000003s :     15: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.57% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.45% :     0.000003s :     15: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     15: predicate.environ_get_eliminate
 1.23% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 0.98% :     0.000002s :     10: predicate.get_grad_eliminate
 0.35% :     0.000001s :      5: predicate.graph_param_transform
 5.34% :     0.000009s :     31: predicate.inline
 1.09% :     0.000002s :     10: predicate.inline_without_move
 0.52% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :     10: predicate.less_batch_normalization
 1.36% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     22: predicate.load_eliminater
 1.60% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.78% :     0.000007s :     42: predicate.loop_unroll_before_grad
 2.16% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     15: predicate.merge_addn
 1.50% :     0.000003s :     15: predicate.minmaximum_grad
 1.42% :     0.000003s :      5: predicate.mutable_eliminate
 0.74% :     0.000001s :      5: predicate.opt_reshape
 2.31% :     0.000004s :     22: predicate.partial_eliminate
 1.24% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.60% :     0.000003s :     15: predicate.reduce_eliminate
 1.40% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.78% :     0.000003s :     27: predicate.replace_applicator
 0.60% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.37% :     0.000002s :     15: predicate.reshape_eliminate
 1.40% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     15: predicate.same_eliminate
 0.52% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.47% :     0.000003s :     10: predicate.special_op_eliminate
 1.05% :     0.000002s :     10: predicate.specialize_transform
 1.38% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.62% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.25% :     0.000004s :     21: predicate.switch_defer_inline
 1.83% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.57% :     0.000012s :     68: predicate.switch_simplify
 1.41% :     0.000003s :     15: predicate.tile_eliminate
 1.20% :     0.000002s :     15: predicate.transpose_eliminate
 1.48% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.34% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.71% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.56% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000869    13
46.22% :     0.000402s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.38% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.40% :     0.000403s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049450    76
 0.16% :     0.000077s :      1: add_recomputation
 0.35% :     0.000175s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.74% :     0.000367s :      1: bootstrap
 0.07% :     0.000036s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000016s :      1: environ_conv
 0.05% :     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% :     0.000009s :      1: graph_reusing
14.86% :     0.007346s :      1: jit_opt_a
 0.52% :     0.000258s :      1: jit_opt_after_cconv
 0.15% :     0.000076s :      1: jit_opt_b
 1.07% :     0.000530s :      1: loop_unroll
 1.60% :     0.000789s :      1: mutable_eliminate
 3.06% :     0.001514s :     26: opt.transform.jit_opt_a
 0.24% :     0.000118s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000044s :      4: opt.transform.jit_opt_b
 0.04% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000023s :      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
 1.08% :     0.000534s :      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
 0.28% :     0.000137s :      1: py_interpret_to_execute
 0.04% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.54% :     0.000759s :      1: renormalize.infer
 0.92% :     0.000457s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000301s :      1: rewriter_after_opt_a
 0.21% :     0.000103s :      1: rewriter_before_opt_a
 0.22% :     0.000109s :      1: symbol_engine_optimizer
21.84% :     0.010801s :      1: task_emit
49.49% :     0.024471s :      1: type_inference
 0.18% :     0.000088s :      1: validate


TotalTime = 0.0500149, [33]
[bootstrap]: 0.00034437
[type_inference]: 0.0250487
[event_method]: 2.282e-05
[auto_monad]: 0.00017313
[graph_reusing]: 6.69999e-06
[pre_auto_parallel]: 3.33998e-06
[py_interpret_to_execute]: 0.00013511
[rewriter_before_opt_a]: 9.837e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.00748412, [2]
    [Cycle 1]: 0.00270346, [27]
        [switch_simplify]: 6.921e-05
        [loop_unroll]: 3.732e-05
        [a_1]: 0.00082716
        [with_stream_mark]: 2.524e-05
        [recompute_prepare]: 1.295e-05
        [updatestate_depend_eliminate]: 6.44999e-06
        [updatestate_assign_eliminate]: 6.10002e-06
        [updatestate_loads_eliminate]: 4.82e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 1.087e-05
        [updatestate_useless_node_eliminater]: 1.173e-05
        [accelerated_algorithm]: 8.81002e-06
        [meta_shard_fg_expand]: 3.28e-06
        [get_grad_eliminate_]: 8.38001e-06
        [merge_forward]: 5.09e-06
        [cell_reuse_recompute_pass]: 2.13002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.103e-05
        [j_node_and_user_rematch]: 1.512e-05
        [meta_fg_expand]: 3.65e-06
        [replace_old_param]: 1.318e-05
        [inline_without_move]: 8.84e-06
        [renormalize]: 0.00130542
        [add_forward_monad_depend]: 8.42e-06
        [auto_monad_grad]: 3.44001e-06
        [auto_monad_eliminator]: 2.263e-05
        [cse]: 3.072e-05
        [replace_applicator]: 2.305e-05
    [Cycle 2]: 0.00050505, [27]
        [switch_simplify]: 9.10999e-06
        [loop_unroll]: 8.22e-06
        [a_1]: 0.00017682
        [with_stream_mark]: 1.594e-05
        [recompute_prepare]: 8.63001e-06
        [updatestate_depend_eliminate]: 5.99999e-06
        [updatestate_assign_eliminate]: 5.45001e-06
        [updatestate_loads_eliminate]: 4.3e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 8.44002e-06
        [updatestate_useless_node_eliminater]: 1.146e-05
        [accelerated_algorithm]: 8.10999e-06
        [meta_shard_fg_expand]: 2.91e-06
        [get_grad_eliminate_]: 8.30999e-06
        [merge_forward]: 5.20001e-06
        [cell_reuse_recompute_pass]: 3.18e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.946e-05
        [j_node_and_user_rematch]: 1.376e-05
        [meta_fg_expand]: 3.19001e-06
        [replace_old_param]: 1.192e-05
        [inline_without_move]: 8.11002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 2.26998e-06
        [auto_monad_eliminator]: 1.031e-05
        [cse]: 2.098e-05
        [replace_applicator]: 9.12001e-06
[py_interpret_to_execute_after_opt_a]: 1.923e-05
[rewriter_after_opt_a]: 0.00029866
[convert_after_rewriter]: 1.162e-05
[order_py_execute_after_rewriter]: 7.57998e-06
[mutable_eliminate]: 0.00085459
[jit_opt_b]: 7.651e-05, [1]
    [Cycle 1]: 6.74e-05, [2]
        [frontend_op_eliminate]: 2.613e-05
        [inline_after_opt_a]: 2.664e-05
[cconv]: 3.525e-05
[loop_unroll]: 0.00053532
[jit_opt_after_cconv]: 0.0002235, [1]
    [Cycle 1]: 0.00021549, [11]
        [c_1]: 5.144e-05
        [parameter_eliminate]: 3.56999e-06
        [updatestate_depend_eliminate]: 8.67998e-06
        [updatestate_assign_eliminate]: 5.62999e-06
        [updatestate_loads_eliminate]: 4.38999e-06
        [cse]: 4.165e-05
        [call_graph_tuple_transform]: 2.456e-05
        [tuple_list_get_item_eliminator]: 8.24002e-06
        [none_parameter_eliminate]: 1.94e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 9.69e-06
[remove_dup_value]: 1.859e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 1.332e-05
[add_recomputation]: 7.308e-05
[cse_after_recomputation]: 3.237e-05, [1]
    [Cycle 1]: 2.559e-05, [1]
        [cse]: 1.877e-05
[auto_monad_reorder]: 2.581e-05
[get_jit_bprop_graph]: 2.11998e-06
[rewriter_after_jit_bprop_graph]: 7.63001e-06
[opt_after_jit_grad]: 0.00057497
[symbol_engine_optimizer]: 0.00011246, [1]
    [Cycle 1]: 0.00010474, [6]
        [build]: 1.593e-05
        [elim_shapecalc]: 1.33e-05
        [elim_not_effective]: 2.042e-05
        [opt_reshape]: 9.02999e-06
        [fold_const_symbol]: 1.572e-05
        [renormalize]: 6.89994e-07
[validate]: 6.648e-05
[backend_pass]: 1.03001e-06
[task_emit]: 0.0134237
[execute]: 1.205e-05
Sums
    bootstrap                            :     0.000344s :  0.77%
    type_inference                       :     0.025049s : 55.65%
    event_method                         :     0.000023s :  0.05%
    auto_monad                           :     0.000173s :  0.38%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000135s :  0.30%
    rewriter_before_opt_a                :     0.000098s :  0.22%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000078s :  0.17%
    jit_opt_a.loop_unroll                :     0.000046s :  0.10%
    jit_opt_a.a_1                        :     0.001004s :  2.23%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.06%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001305s :  2.90%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  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                        :     0.000052s :  0.11%
    jit_opt_a.replace_applicator         :     0.000032s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.04%
    rewriter_after_opt_a                 :     0.000299s :  0.66%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000855s :  1.90%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.06%
    cconv                                :     0.000035s :  0.08%
    loop_unroll                          :     0.000535s :  1.19%
    jit_opt_after_cconv.c_1              :     0.000051s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000042s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.05%
    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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000019s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000073s :  0.16%
    cse_after_recomputation.cse          :     0.000019s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000575s :  1.28%
    symbol_engine_optimizer.build        :     0.000016s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000066s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.013424s : 29.82%
    execute                              :     0.000012s :  0.03%


Time group info:
------[substitution.]   0.000381    47
 3.00% :     0.000011s :      2: substitution.depend_value_elim
 0.77% :     0.000003s :      4: substitution.elim_not_effective
 0.54% :     0.000002s :      4: substitution.fold_const_symbol
 1.85% :     0.000007s :      5: substitution.graph_param_transform
78.78% :     0.000300s :      4: substitution.inline
 1.53% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.09% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.85% :     0.000007s :      2: substitution.replace_old_param
 3.39% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 2.98% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.23% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024943     2
93.51% :     0.023324s :      1: type_inference.infer
 6.49% :     0.001619s :      1: type_inference.specialize
------[replace.]   0.000097     6
65.43% :     0.000063s :      4: replace.inline
34.57% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000308     6
96.19% :     0.000297s :      4: match.inline
 3.81% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000180  1021
 1.55% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.01% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     15: predicate.addn_check_dump
 1.52% :     0.000003s :     15: predicate.addn_zero_filter
 2.07% :     0.000004s :     15: predicate.arithmetic_simplify
 1.59% :     0.000003s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     15: predicate.compare_switch_simplify
 1.76% :     0.000003s :     15: predicate.depend_value_elim
 1.10% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.11% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.94% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      5: predicate.elim_not_effective
 0.53% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.35% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.06% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.52% :     0.000003s :     15: predicate.environ_get_eliminate
 1.04% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 0.94% :     0.000002s :     10: predicate.get_grad_eliminate
 0.38% :     0.000001s :      5: predicate.graph_param_transform
 5.78% :     0.000010s :     31: predicate.inline
 1.28% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.45% :     0.000003s :     10: predicate.less_batch_normalization
 1.61% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.11% :     0.000004s :     22: predicate.load_eliminater
 1.18% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.35% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     15: predicate.merge_addn
 1.05% :     0.000002s :     15: predicate.minmaximum_grad
 1.82% :     0.000003s :      5: predicate.mutable_eliminate
 0.49% :     0.000001s :      5: predicate.opt_reshape
 2.32% :     0.000004s :     22: predicate.partial_eliminate
 1.13% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.59% :     0.000003s :     15: predicate.reduce_eliminate
 1.52% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.85% :     0.000002s :     10: predicate.remove_not_recompute_node
 1.61% :     0.000003s :     27: predicate.replace_applicator
 0.54% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.16% :     0.000002s :     15: predicate.reshape_eliminate
 1.16% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.95% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     15: predicate.same_eliminate
 0.77% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.06% :     0.000002s :     10: predicate.special_op_eliminate
 1.19% :     0.000002s :     10: predicate.specialize_transform
 1.41% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.63% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.38% :     0.000004s :     21: predicate.switch_defer_inline
 1.98% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.36% :     0.000011s :     68: predicate.switch_simplify
 1.17% :     0.000002s :     15: predicate.tile_eliminate
 1.23% :     0.000002s :     15: predicate.transpose_eliminate
 1.49% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.51% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.67% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000967    13
45.94% :     0.000444s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.89% :     0.000067s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.17% :     0.000456s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.052860    76
 0.14% :     0.000076s :      1: add_recomputation
 0.34% :     0.000179s :      1: auto_monad
 0.05% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.69% :     0.000366s :      1: bootstrap
 0.07% :     0.000038s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000035s :      1: cse_after_recomputation
 0.03% :     0.000016s :      1: environ_conv
 0.05% :     0.000027s :      1: event_method
 0.03% :     0.000018s :      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
14.17% :     0.007488s :      1: jit_opt_a
 0.43% :     0.000226s :      1: jit_opt_after_cconv
 0.15% :     0.000080s :      1: jit_opt_b
 1.03% :     0.000546s :      1: loop_unroll
 1.64% :     0.000867s :      1: mutable_eliminate
 2.53% :     0.001335s :     26: opt.transform.jit_opt_a
 0.17% :     0.000089s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000044s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000036s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000055s :      4: opt.transform.symbol_engine_opt
 1.11% :     0.000587s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000140s :      1: py_interpret_to_execute
 0.04% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.48% :     0.000780s :      1: renormalize.infer
 0.97% :     0.000514s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.58% :     0.000305s :      1: rewriter_after_opt_a
 0.19% :     0.000103s :      1: rewriter_before_opt_a
 0.22% :     0.000116s :      1: symbol_engine_optimizer
25.44% :     0.013447s :      1: task_emit
47.42% :     0.025067s :      1: type_inference
 0.18% :     0.000094s :      1: validate


TotalTime = 0.048348, [33]
[bootstrap]: 0.00030812
[type_inference]: 0.0234648
[event_method]: 2.028e-05
[auto_monad]: 0.00014924
[graph_reusing]: 5.62999e-06
[pre_auto_parallel]: 2.95998e-06
[py_interpret_to_execute]: 0.00012489
[rewriter_before_opt_a]: 9.859e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00677736, [2]
    [Cycle 1]: 0.00256273, [27]
        [switch_simplify]: 6.078e-05
        [loop_unroll]: 3.98e-05
        [a_1]: 0.00072654
        [with_stream_mark]: 1.917e-05
        [recompute_prepare]: 1.102e-05
        [updatestate_depend_eliminate]: 6.25002e-06
        [updatestate_assign_eliminate]: 5.04e-06
        [updatestate_loads_eliminate]: 4.40999e-06
        [parameter_eliminate]: 1.30001e-06
        [specialize_transform]: 9.56e-06
        [updatestate_useless_node_eliminater]: 1.219e-05
        [accelerated_algorithm]: 3.326e-05
        [meta_shard_fg_expand]: 2.75002e-06
        [get_grad_eliminate_]: 9.08002e-06
        [merge_forward]: 4.60001e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 5.293e-05
        [j_node_and_user_rematch]: 1.556e-05
        [meta_fg_expand]: 3.63e-06
        [replace_old_param]: 1.458e-05
        [inline_without_move]: 8.67998e-06
        [renormalize]: 0.00124068
        [add_forward_monad_depend]: 7.55998e-06
        [auto_monad_grad]: 2.88003e-06
        [auto_monad_eliminator]: 2.023e-05
        [cse]: 3.783e-05
        [replace_applicator]: 1.826e-05
    [Cycle 2]: 0.00049501, [27]
        [switch_simplify]: 9.24998e-06
        [loop_unroll]: 8.33001e-06
        [a_1]: 0.00017541
        [with_stream_mark]: 1.605e-05
        [recompute_prepare]: 8.25e-06
        [updatestate_depend_eliminate]: 4.92e-06
        [updatestate_assign_eliminate]: 4.97e-06
        [updatestate_loads_eliminate]: 4.11001e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 7.79002e-06
        [updatestate_useless_node_eliminater]: 1.117e-05
        [accelerated_algorithm]: 8.70001e-06
        [meta_shard_fg_expand]: 2.80997e-06
        [get_grad_eliminate_]: 8.23001e-06
        [merge_forward]: 4.95999e-06
        [cell_reuse_recompute_pass]: 2.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.904e-05
        [j_node_and_user_rematch]: 1.331e-05
        [meta_fg_expand]: 2.91999e-06
        [replace_old_param]: 1.035e-05
        [inline_without_move]: 7.71999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 1.65001e-06
        [auto_monad_eliminator]: 1.032e-05
        [cse]: 2.145e-05
        [replace_applicator]: 8.73001e-06
[py_interpret_to_execute_after_opt_a]: 1.753e-05
[rewriter_after_opt_a]: 0.00030684
[convert_after_rewriter]: 1.187e-05
[order_py_execute_after_rewriter]: 7.11001e-06
[mutable_eliminate]: 0.00079753
[jit_opt_b]: 7.655e-05, [1]
    [Cycle 1]: 6.76e-05, [2]
        [frontend_op_eliminate]: 2.62e-05
        [inline_after_opt_a]: 2.792e-05
[cconv]: 3.274e-05
[loop_unroll]: 0.00053672
[jit_opt_after_cconv]: 0.00025477, [1]
    [Cycle 1]: 0.00024583, [11]
        [c_1]: 5.337e-05
        [parameter_eliminate]: 5.42001e-06
        [updatestate_depend_eliminate]: 1.214e-05
        [updatestate_assign_eliminate]: 5.29e-06
        [updatestate_loads_eliminate]: 4.47998e-06
        [cse]: 4.609e-05
        [call_graph_tuple_transform]: 2.598e-05
        [tuple_list_get_item_eliminator]: 9.05001e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 9.10019e-07
        [switch_simplify]: 9.22999e-06
[remove_dup_value]: 2.054e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 1.439e-05
[add_recomputation]: 7.549e-05
[cse_after_recomputation]: 3.445e-05, [1]
    [Cycle 1]: 2.592e-05, [1]
        [cse]: 1.964e-05
[auto_monad_reorder]: 2.62e-05
[get_jit_bprop_graph]: 1.94999e-06
[rewriter_after_jit_bprop_graph]: 8.47e-06
[opt_after_jit_grad]: 0.00060828
[symbol_engine_optimizer]: 0.0001096, [1]
    [Cycle 1]: 0.00010268, [6]
        [build]: 1.591e-05
        [elim_shapecalc]: 1.266e-05
        [elim_not_effective]: 2.067e-05
        [opt_reshape]: 9.10999e-06
        [fold_const_symbol]: 1.371e-05
        [renormalize]: 4.19997e-07
[validate]: 6.939e-05
[backend_pass]: 1.07998e-06
[task_emit]: 0.014112
[execute]: 9.20001e-06
Sums
    bootstrap                            :     0.000308s :  0.70%
    type_inference                       :     0.023465s : 53.45%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000149s :  0.34%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000125s :  0.28%
    rewriter_before_opt_a                :     0.000099s :  0.22%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000070s :  0.16%
    jit_opt_a.loop_unroll                :     0.000048s :  0.11%
    jit_opt_a.a_1                        :     0.000902s :  2.05%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000042s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000072s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000025s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001241s :  2.83%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000031s :  0.07%
    jit_opt_a.cse                        :     0.000059s :  0.14%
    jit_opt_a.replace_applicator         :     0.000027s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.04%
    rewriter_after_opt_a                 :     0.000307s :  0.70%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000798s :  1.82%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000028s :  0.06%
    cconv                                :     0.000033s :  0.07%
    loop_unroll                          :     0.000537s :  1.22%
    jit_opt_after_cconv.c_1              :     0.000053s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  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              :     0.000046s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000014s :  0.03%
    add_recomputation                    :     0.000075s :  0.17%
    cse_after_recomputation.cse          :     0.000020s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000608s :  1.39%
    symbol_engine_optimizer.build        :     0.000016s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000069s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.014112s : 32.14%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000291    47
 3.90% :     0.000011s :      2: substitution.depend_value_elim
 0.96% :     0.000003s :      4: substitution.elim_not_effective
 0.76% :     0.000002s :      4: substitution.fold_const_symbol
 2.46% :     0.000007s :      5: substitution.graph_param_transform
74.59% :     0.000217s :      4: substitution.inline
 1.57% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.70% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.61% :     0.000005s :      2: substitution.replace_old_param
 3.82% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
 3.79% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.84% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023392     2
93.82% :     0.021947s :      1: type_inference.infer
 6.18% :     0.001445s :      1: type_inference.specialize
------[replace.]   0.000091     6
67.47% :     0.000061s :      4: replace.inline
32.53% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000224     6
95.56% :     0.000214s :      4: match.inline
 4.44% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000176  1021
 1.31% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.29% :     0.000002s :     15: predicate.addn_zero_filter
 1.96% :     0.000003s :     15: predicate.arithmetic_simplify
 1.45% :     0.000003s :     15: predicate.cast_eliminate
 0.64% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.38% :     0.000002s :     15: predicate.depend_value_elim
 1.10% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.17% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.13% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      5: predicate.elim_not_effective
 0.64% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     15: predicate.environ_get_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.28% :     0.000002s :     10: predicate.get_grad_eliminate
 0.39% :     0.000001s :      5: predicate.graph_param_transform
 4.90% :     0.000009s :     31: predicate.inline
 1.18% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.66% :     0.000003s :     10: predicate.less_batch_normalization
 1.58% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     22: predicate.load_eliminater
 1.53% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.50% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.99% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 1.44% :     0.000003s :      5: predicate.mutable_eliminate
 0.52% :     0.000001s :      5: predicate.opt_reshape
 2.54% :     0.000004s :     22: predicate.partial_eliminate
 1.20% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     15: predicate.reduce_eliminate
 1.55% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.98% :     0.000003s :     27: predicate.replace_applicator
 0.71% :     0.000001s :     10: predicate.replace_old_param
 0.59% :     0.000001s :      5: predicate.reset_defer_inline
 1.16% :     0.000002s :     15: predicate.reshape_eliminate
 1.26% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.11% :     0.000002s :     10: predicate.special_op_eliminate
 1.10% :     0.000002s :     10: predicate.specialize_transform
 1.37% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.22% :     0.000004s :     21: predicate.switch_defer_inline
 1.86% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.65% :     0.000012s :     68: predicate.switch_simplify
 1.40% :     0.000002s :     15: predicate.tile_eliminate
 1.16% :     0.000002s :     15: predicate.transpose_eliminate
 1.71% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.14% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000896    13
45.35% :     0.000407s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.60% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.04% :     0.000440s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.051077    76
 0.15% :     0.000079s :      1: add_recomputation
 0.30% :     0.000155s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.64% :     0.000328s :      1: bootstrap
 0.07% :     0.000036s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.07% :     0.000037s :      1: cse_after_recomputation
 0.03% :     0.000017s :      1: environ_conv
 0.05% :     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
 0.02% :     0.000008s :      1: graph_reusing
13.28% :     0.006781s :      1: jit_opt_a
 0.51% :     0.000258s :      1: jit_opt_after_cconv
 0.16% :     0.000080s :      1: jit_opt_b
 1.07% :     0.000548s :      1: loop_unroll
 1.58% :     0.000809s :      1: mutable_eliminate
 2.50% :     0.001277s :     26: opt.transform.jit_opt_a
 0.18% :     0.000093s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000046s :      4: opt.transform.jit_opt_b
 0.04% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.21% :     0.000620s :      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
 0.25% :     0.000129s :      1: py_interpret_to_execute
 0.04% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000024s :      1: remove_dup_value
 1.49% :     0.000759s :      1: renormalize.infer
 0.93% :     0.000473s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000311s :      1: rewriter_after_opt_a
 0.20% :     0.000103s :      1: rewriter_before_opt_a
 0.22% :     0.000113s :      1: symbol_engine_optimizer
27.67% :     0.014132s :      1: task_emit
45.97% :     0.023481s :      1: type_inference
 0.19% :     0.000098s :      1: validate


TotalTime = 0.0460973, [33]
[bootstrap]: 0.00034963
[type_inference]: 0.0239288
[event_method]: 2.078e-05
[auto_monad]: 0.00015852
[graph_reusing]: 6.34001e-06
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 0.00012896
[rewriter_before_opt_a]: 0.00010155
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00687936, [2]
    [Cycle 1]: 0.00261899, [27]
        [switch_simplify]: 6.587e-05
        [loop_unroll]: 3.75e-05
        [a_1]: 0.00076774
        [with_stream_mark]: 2.489e-05
        [recompute_prepare]: 1.097e-05
        [updatestate_depend_eliminate]: 6.44999e-06
        [updatestate_assign_eliminate]: 5.07e-06
        [updatestate_loads_eliminate]: 4.68001e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 9.05001e-06
        [updatestate_useless_node_eliminater]: 1.105e-05
        [accelerated_algorithm]: 8.82999e-06
        [meta_shard_fg_expand]: 2.68e-06
        [get_grad_eliminate_]: 8.51002e-06
        [merge_forward]: 5.61e-06
        [cell_reuse_recompute_pass]: 1.93002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.021e-05
        [j_node_and_user_rematch]: 1.413e-05
        [meta_fg_expand]: 4.126e-05
        [replace_old_param]: 1.333e-05
        [inline_without_move]: 8.80001e-06
        [renormalize]: 0.0012544
        [add_forward_monad_depend]: 7.33e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 2.333e-05
        [cse]: 4.309e-05
        [replace_applicator]: 2.065e-05
    [Cycle 2]: 0.00048155, [27]
        [switch_simplify]: 8.91002e-06
        [loop_unroll]: 8.23001e-06
        [a_1]: 0.00017003
        [with_stream_mark]: 1.412e-05
        [recompute_prepare]: 8.77999e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 5.05001e-06
        [updatestate_loads_eliminate]: 3.95e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 8.2e-06
        [updatestate_useless_node_eliminater]: 1.069e-05
        [accelerated_algorithm]: 8.22998e-06
        [meta_shard_fg_expand]: 2.76e-06
        [get_grad_eliminate_]: 7.47002e-06
        [merge_forward]: 4.72e-06
        [cell_reuse_recompute_pass]: 3.58e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.689e-05
        [j_node_and_user_rematch]: 1.246e-05
        [meta_fg_expand]: 2.83998e-06
        [replace_old_param]: 1.088e-05
        [inline_without_move]: 7.45e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 1.061e-05
        [cse]: 2.249e-05
        [replace_applicator]: 8.09997e-06
[py_interpret_to_execute_after_opt_a]: 1.477e-05
[rewriter_after_opt_a]: 0.00029379
[convert_after_rewriter]: 1.162e-05
[order_py_execute_after_rewriter]: 7.13998e-06
[mutable_eliminate]: 0.0007268
[jit_opt_b]: 6.9e-05, [1]
    [Cycle 1]: 6.099e-05, [2]
        [frontend_op_eliminate]: 2.361e-05
        [inline_after_opt_a]: 2.521e-05
[cconv]: 3.075e-05
[loop_unroll]: 0.00047039
[jit_opt_after_cconv]: 0.00020919, [1]
    [Cycle 1]: 0.00020164, [11]
        [c_1]: 4.742e-05
        [parameter_eliminate]: 3.5e-06
        [updatestate_depend_eliminate]: 8.74998e-06
        [updatestate_assign_eliminate]: 4.83001e-06
        [updatestate_loads_eliminate]: 4.37e-06
        [cse]: 3.794e-05
        [call_graph_tuple_transform]: 2.356e-05
        [tuple_list_get_item_eliminator]: 8.53001e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 9.10999e-06
[remove_dup_value]: 1.829e-05
[partial_unused_args_eliminate]: 2.21998e-06
[environ_conv]: 1.267e-05
[add_recomputation]: 7.212e-05
[cse_after_recomputation]: 8.166e-05, [1]
    [Cycle 1]: 7.454e-05, [1]
        [cse]: 6.721e-05
[auto_monad_reorder]: 0.00010418
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 8.43999e-06
[opt_after_jit_grad]: 0.00054664
[symbol_engine_optimizer]: 0.00010876, [1]
    [Cycle 1]: 0.00010229, [6]
        [build]: 1.537e-05
        [elim_shapecalc]: 1.226e-05
        [elim_not_effective]: 1.954e-05
        [opt_reshape]: 8.27e-06
        [fold_const_symbol]: 1.532e-05
        [renormalize]: 3.00002e-07
[validate]: 6.215e-05
[backend_pass]: 1.05999e-06
[task_emit]: 0.0114061
[execute]: 9.83002e-06
Sums
    bootstrap                            :     0.000350s :  0.84%
    type_inference                       :     0.023929s : 57.48%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000159s :  0.38%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000129s :  0.31%
    rewriter_before_opt_a                :     0.000102s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.18%
    jit_opt_a.loop_unroll                :     0.000046s :  0.11%
    jit_opt_a.a_1                        :     0.000938s :  2.25%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000044s :  0.11%
    jit_opt_a.replace_old_param          :     0.000024s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001254s :  3.01%
    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      :     0.000034s :  0.08%
    jit_opt_a.cse                        :     0.000066s :  0.16%
    jit_opt_a.replace_applicator         :     0.000029s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000294s :  0.71%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000727s :  1.75%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000031s :  0.07%
    loop_unroll                          :     0.000470s :  1.13%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  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              :     0.000038s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000018s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000072s :  0.17%
    cse_after_recomputation.cse          :     0.000067s :  0.16%
    auto_monad_reorder                   :     0.000104s :  0.25%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000547s :  1.31%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  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  :     0.000000s :  0.00%
    validate                             :     0.000062s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011406s : 27.40%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000324    47
 3.32% :     0.000011s :      2: substitution.depend_value_elim
 0.89% :     0.000003s :      4: substitution.elim_not_effective
 0.68% :     0.000002s :      4: substitution.fold_const_symbol
 2.04% :     0.000007s :      5: substitution.graph_param_transform
77.62% :     0.000252s :      4: substitution.inline
 1.42% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.40% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.66% :     0.000005s :      2: substitution.replace_old_param
 4.01% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.01% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 2.97% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023852     2
93.74% :     0.022358s :      1: type_inference.infer
 6.26% :     0.001493s :      1: type_inference.specialize
------[replace.]   0.000094     6
65.66% :     0.000061s :      4: replace.inline
34.34% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
95.46% :     0.000249s :      4: match.inline
 4.54% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1021
 1.27% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     15: predicate.addn_check_dump
 1.34% :     0.000002s :     15: predicate.addn_zero_filter
 1.91% :     0.000003s :     15: predicate.arithmetic_simplify
 1.70% :     0.000003s :     15: predicate.cast_eliminate
 0.47% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     15: predicate.compare_switch_simplify
 1.39% :     0.000002s :     15: predicate.depend_value_elim
 1.13% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.16% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      5: predicate.elim_not_effective
 0.62% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.18% :     0.000002s :     10: predicate.get_grad_eliminate
 0.29% :     0.000001s :      5: predicate.graph_param_transform
 5.19% :     0.000009s :     31: predicate.inline
 1.22% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :     10: predicate.less_batch_normalization
 1.60% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     22: predicate.load_eliminater
 1.57% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.43% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.06% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     15: predicate.merge_addn
 1.11% :     0.000002s :     15: predicate.minmaximum_grad
 1.79% :     0.000003s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.33% :     0.000004s :     22: predicate.partial_eliminate
 1.16% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     15: predicate.reduce_eliminate
 1.46% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.79% :     0.000003s :     27: predicate.replace_applicator
 0.69% :     0.000001s :     10: predicate.replace_old_param
 0.42% :     0.000001s :      5: predicate.reset_defer_inline
 1.18% :     0.000002s :     15: predicate.reshape_eliminate
 1.27% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     15: predicate.same_eliminate
 0.55% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.03% :     0.000002s :     10: predicate.special_op_eliminate
 1.04% :     0.000002s :     10: predicate.specialize_transform
 1.55% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.19% :     0.000004s :     21: predicate.switch_defer_inline
 1.89% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.40% :     0.000011s :     68: predicate.switch_simplify
 1.44% :     0.000002s :     15: predicate.tile_eliminate
 1.22% :     0.000002s :     15: predicate.transpose_eliminate
 1.54% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.74% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.32% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000944    13
48.32% :     0.000456s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.70% :     0.000063s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.98% :     0.000425s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048794    76
 0.15% :     0.000075s :      1: add_recomputation
 0.34% :     0.000164s :      1: auto_monad
 0.22% :     0.000110s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.76% :     0.000373s :      1: bootstrap
 0.07% :     0.000034s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.17% :     0.000084s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.05% :     0.000025s :      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% :     0.000009s :      1: graph_reusing
14.11% :     0.006883s :      1: jit_opt_a
 0.43% :     0.000212s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 0.98% :     0.000480s :      1: loop_unroll
 1.51% :     0.000737s :      1: mutable_eliminate
 2.57% :     0.001252s :     26: opt.transform.jit_opt_a
 0.17% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      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
 0.11% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.14% :     0.000556s :      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
 0.27% :     0.000134s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.59% :     0.000777s :      1: renormalize.infer
 0.96% :     0.000466s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000298s :      1: rewriter_after_opt_a
 0.22% :     0.000106s :      1: rewriter_before_opt_a
 0.23% :     0.000112s :      1: symbol_engine_optimizer
23.41% :     0.011424s :      1: task_emit
49.07% :     0.023944s :      1: type_inference
 0.18% :     0.000088s :      1: validate


TotalTime = 0.0468034, [33]
[bootstrap]: 0.00031603
[type_inference]: 0.0239148
[event_method]: 2.165e-05
[auto_monad]: 0.00016142
[graph_reusing]: 7.26999e-06
[pre_auto_parallel]: 2.94999e-06
[py_interpret_to_execute]: 0.00014032
[rewriter_before_opt_a]: 0.00010058
[expand_dump_flag]: 3.38999e-06
[jit_opt_a]: 0.00711631, [2]
    [Cycle 1]: 0.00254708, [27]
        [switch_simplify]: 6.773e-05
        [loop_unroll]: 3.706e-05
        [a_1]: 0.00073261
        [with_stream_mark]: 2.175e-05
        [recompute_prepare]: 1.124e-05
        [updatestate_depend_eliminate]: 6.78998e-06
        [updatestate_assign_eliminate]: 6.01e-06
        [updatestate_loads_eliminate]: 5.20999e-06
        [parameter_eliminate]: 1.98997e-06
        [specialize_transform]: 9.08002e-06
        [updatestate_useless_node_eliminater]: 1.161e-05
        [accelerated_algorithm]: 8.25999e-06
        [meta_shard_fg_expand]: 2.86e-06
        [get_grad_eliminate_]: 8.04002e-06
        [merge_forward]: 5.64998e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.98e-05
        [j_node_and_user_rematch]: 1.417e-05
        [meta_fg_expand]: 3.78001e-06
        [replace_old_param]: 1.241e-05
        [inline_without_move]: 8.69e-06
        [renormalize]: 0.00125735
        [add_forward_monad_depend]: 6.89001e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 2.292e-05
        [cse]: 4.274e-05
        [replace_applicator]: 1.941e-05
    [Cycle 2]: 0.0004989, [27]
        [switch_simplify]: 1.017e-05
        [loop_unroll]: 8.75999e-06
        [a_1]: 0.00017968
        [with_stream_mark]: 1.418e-05
        [recompute_prepare]: 8.80001e-06
        [updatestate_depend_eliminate]: 5.59998e-06
        [updatestate_assign_eliminate]: 5.27999e-06
        [updatestate_loads_eliminate]: 4.37998e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 8.47e-06
        [updatestate_useless_node_eliminater]: 1.058e-05
        [accelerated_algorithm]: 7.98001e-06
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 7.65e-06
        [merge_forward]: 5.59e-06
        [cell_reuse_recompute_pass]: 3.3e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.796e-05
        [j_node_and_user_rematch]: 1.253e-05
        [meta_fg_expand]: 2.87002e-06
        [replace_old_param]: 1.078e-05
        [inline_without_move]: 8.10999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 1.15001e-06
        [auto_monad_eliminator]: 1.109e-05
        [cse]: 2.223e-05
        [replace_applicator]: 8.17998e-06
[py_interpret_to_execute_after_opt_a]: 1.344e-05
[rewriter_after_opt_a]: 0.00031274
[convert_after_rewriter]: 1.164e-05
[order_py_execute_after_rewriter]: 7.26999e-06
[mutable_eliminate]: 0.00080072
[jit_opt_b]: 7.605e-05, [1]
    [Cycle 1]: 6.702e-05, [2]
        [frontend_op_eliminate]: 2.634e-05
        [inline_after_opt_a]: 2.78e-05
[cconv]: 3.018e-05
[loop_unroll]: 0.00051948
[jit_opt_after_cconv]: 0.00021565, [1]
    [Cycle 1]: 0.00020816, [11]
        [c_1]: 5.241e-05
        [parameter_eliminate]: 3.23998e-06
        [updatestate_depend_eliminate]: 8.15e-06
        [updatestate_assign_eliminate]: 4.58001e-06
        [updatestate_loads_eliminate]: 4.12998e-06
        [cse]: 3.602e-05
        [call_graph_tuple_transform]: 2.382e-05
        [tuple_list_get_item_eliminator]: 8.44002e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 7.59988e-07
        [switch_simplify]: 9.96e-06
[remove_dup_value]: 2.183e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 1.39e-05
[add_recomputation]: 7.109e-05
[cse_after_recomputation]: 3.344e-05, [1]
    [Cycle 1]: 2.574e-05, [1]
        [cse]: 1.938e-05
[auto_monad_reorder]: 2.639e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 8.65999e-06
[opt_after_jit_grad]: 0.00056139
[symbol_engine_optimizer]: 0.00012218, [1]
    [Cycle 1]: 0.00011472, [6]
        [build]: 1.711e-05
        [elim_shapecalc]: 1.366e-05
        [elim_not_effective]: 2.446e-05
        [opt_reshape]: 9.99001e-06
        [fold_const_symbol]: 1.623e-05
        [renormalize]: 8.99978e-07
[validate]: 6.293e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.0118399
[execute]: 8.27998e-06
Sums
    bootstrap                            :     0.000316s :  0.75%
    type_inference                       :     0.023915s : 56.91%
    event_method                         :     0.000022s :  0.05%
    auto_monad                           :     0.000161s :  0.38%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000140s :  0.33%
    rewriter_before_opt_a                :     0.000101s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000078s :  0.19%
    jit_opt_a.loop_unroll                :     0.000046s :  0.11%
    jit_opt_a.a_1                        :     0.000912s :  2.17%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    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  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001257s :  2.99%
    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      :     0.000034s :  0.08%
    jit_opt_a.cse                        :     0.000065s :  0.15%
    jit_opt_a.replace_applicator         :     0.000028s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000313s :  0.74%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000801s :  1.91%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000028s :  0.07%
    cconv                                :     0.000030s :  0.07%
    loop_unroll                          :     0.000519s :  1.24%
    jit_opt_after_cconv.c_1              :     0.000052s :  0.12%
    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              :     0.000036s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000022s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000014s :  0.03%
    add_recomputation                    :     0.000071s :  0.17%
    cse_after_recomputation.cse          :     0.000019s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000561s :  1.34%
    symbol_engine_optimizer.build        :     0.000017s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000024s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000063s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011840s : 28.18%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000296    47
 3.58% :     0.000011s :      2: substitution.depend_value_elim
 1.03% :     0.000003s :      4: substitution.elim_not_effective
 0.83% :     0.000002s :      4: substitution.fold_const_symbol
 2.38% :     0.000007s :      5: substitution.graph_param_transform
74.03% :     0.000219s :      4: substitution.inline
 1.80% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.64% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.87% :     0.000006s :      2: substitution.replace_old_param
 4.55% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.94% :     0.000012s :      3: substitution.updatestate_pure_node_eliminater
 3.37% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023831     2
93.21% :     0.022214s :      1: type_inference.infer
 6.79% :     0.001617s :      1: type_inference.specialize
------[replace.]   0.000097     6
66.41% :     0.000064s :      4: replace.inline
33.59% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000229     6
94.70% :     0.000217s :      4: match.inline
 5.30% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000177  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.26% :     0.000002s :     15: predicate.addn_zero_filter
 1.96% :     0.000003s :     15: predicate.arithmetic_simplify
 1.69% :     0.000003s :     15: predicate.cast_eliminate
 0.54% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.10% :     0.000002s :     15: predicate.compare_switch_simplify
 1.63% :     0.000003s :     15: predicate.depend_value_elim
 1.11% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.90% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      5: predicate.elim_not_effective
 0.65% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.26% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.48% :     0.000003s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.13% :     0.000002s :     10: predicate.get_grad_eliminate
 0.42% :     0.000001s :      5: predicate.graph_param_transform
 4.99% :     0.000009s :     31: predicate.inline
 0.99% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :     10: predicate.less_batch_normalization
 1.39% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.77% :     0.000003s :     22: predicate.load_eliminater
 1.43% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.40% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.08% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     15: predicate.merge_addn
 1.10% :     0.000002s :     15: predicate.minmaximum_grad
 2.14% :     0.000004s :      5: predicate.mutable_eliminate
 0.59% :     0.000001s :      5: predicate.opt_reshape
 2.31% :     0.000004s :     22: predicate.partial_eliminate
 1.16% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.62% :     0.000003s :     15: predicate.reduce_eliminate
 1.68% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.74% :     0.000003s :     27: predicate.replace_applicator
 0.55% :     0.000001s :     10: predicate.replace_old_param
 0.31% :     0.000001s :      5: predicate.reset_defer_inline
 1.29% :     0.000002s :     15: predicate.reshape_eliminate
 1.20% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.19% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.43% :     0.000003s :     15: predicate.same_eliminate
 0.58% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.99% :     0.000002s :     10: predicate.special_op_eliminate
 1.20% :     0.000002s :     10: predicate.specialize_transform
 1.39% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.14% :     0.000004s :     21: predicate.switch_defer_inline
 1.90% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.60% :     0.000012s :     68: predicate.switch_simplify
 1.34% :     0.000002s :     15: predicate.tile_eliminate
 1.20% :     0.000002s :     15: predicate.transpose_eliminate
 1.66% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.83% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.16% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000003s :     15: predicate.value_based_eliminate
 0.45% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000859    13
44.13% :     0.000379s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.92% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.95% :     0.000438s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049498    76
 0.15% :     0.000074s :      1: add_recomputation
 0.34% :     0.000167s :      1: auto_monad
 0.06% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000333s :      1: bootstrap
 0.07% :     0.000033s :      1: cconv
 0.03% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000036s :      1: cse_after_recomputation
 0.03% :     0.000017s :      1: environ_conv
 0.05% :     0.000026s :      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
14.38% :     0.007120s :      1: jit_opt_a
 0.44% :     0.000219s :      1: jit_opt_after_cconv
 0.16% :     0.000079s :      1: jit_opt_b
 1.07% :     0.000529s :      1: loop_unroll
 1.64% :     0.000813s :      1: mutable_eliminate
 2.48% :     0.001228s :     26: opt.transform.jit_opt_a
 0.18% :     0.000090s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000046s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000060s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000571s :      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
 0.29% :     0.000146s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000025s :      1: remove_dup_value
 1.51% :     0.000747s :      1: renormalize.infer
 1.01% :     0.000501s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.64% :     0.000319s :      1: rewriter_after_opt_a
 0.21% :     0.000106s :      1: rewriter_before_opt_a
 0.25% :     0.000125s :      1: symbol_engine_optimizer
23.96% :     0.011861s :      1: task_emit
48.35% :     0.023935s :      1: type_inference
 0.18% :     0.000089s :      1: validate


TotalTime = 0.046001, [33]
[bootstrap]: 0.00032055
[type_inference]: 0.0242347
[event_method]: 2.303e-05
[auto_monad]: 0.00017047
[graph_reusing]: 6.66999e-06
[pre_auto_parallel]: 3.03e-06
[py_interpret_to_execute]: 0.00014439
[rewriter_before_opt_a]: 0.00010539
[expand_dump_flag]: 3.48e-06
[jit_opt_a]: 0.00714552, [2]
    [Cycle 1]: 0.00262116, [27]
        [switch_simplify]: 7.38e-05
        [loop_unroll]: 3.653e-05
        [a_1]: 0.00079017
        [with_stream_mark]: 3.958e-05
        [recompute_prepare]: 1.349e-05
        [updatestate_depend_eliminate]: 7.26999e-06
        [updatestate_assign_eliminate]: 7.12002e-06
        [updatestate_loads_eliminate]: 5.68002e-06
        [parameter_eliminate]: 2.06998e-06
        [specialize_transform]: 1.137e-05
        [updatestate_useless_node_eliminater]: 2.639e-05
        [accelerated_algorithm]: 9.53002e-06
        [meta_shard_fg_expand]: 2.94001e-06
        [get_grad_eliminate_]: 9.67001e-06
        [merge_forward]: 5.66998e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.119e-05
        [j_node_and_user_rematch]: 1.415e-05
        [meta_fg_expand]: 3.7e-06
        [replace_old_param]: 1.305e-05
        [inline_without_move]: 9.10999e-06
        [renormalize]: 0.00122582
        [add_forward_monad_depend]: 6.86999e-06
        [auto_monad_grad]: 3.26001e-06
        [auto_monad_eliminator]: 2.225e-05
        [cse]: 3.129e-05
        [replace_applicator]: 2.081e-05
    [Cycle 2]: 0.00048949, [27]
        [switch_simplify]: 9.51e-06
        [loop_unroll]: 8.45999e-06
        [a_1]: 0.00017405
        [with_stream_mark]: 1.557e-05
        [recompute_prepare]: 8.35001e-06
        [updatestate_depend_eliminate]: 5.92999e-06
        [updatestate_assign_eliminate]: 5.68002e-06
        [updatestate_loads_eliminate]: 4.24002e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 8.08999e-06
        [updatestate_useless_node_eliminater]: 1.131e-05
        [accelerated_algorithm]: 8.06001e-06
        [meta_shard_fg_expand]: 2.34001e-06
        [get_grad_eliminate_]: 7.46001e-06
        [merge_forward]: 4.36002e-06
        [cell_reuse_recompute_pass]: 3.36001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.78e-05
        [j_node_and_user_rematch]: 1.266e-05
        [meta_fg_expand]: 2.81e-06
        [replace_old_param]: 1.04e-05
        [inline_without_move]: 7.6e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 1.081e-05
        [cse]: 2.148e-05
        [replace_applicator]: 7.77e-06
[py_interpret_to_execute_after_opt_a]: 1.693e-05
[rewriter_after_opt_a]: 0.00030241
[convert_after_rewriter]: 1.308e-05
[order_py_execute_after_rewriter]: 7.68999e-06
[mutable_eliminate]: 0.00080004
[jit_opt_b]: 7.234e-05, [1]
    [Cycle 1]: 6.41e-05, [2]
        [frontend_op_eliminate]: 2.655e-05
        [inline_after_opt_a]: 2.528e-05
[cconv]: 3.289e-05
[loop_unroll]: 0.00048111
[jit_opt_after_cconv]: 0.00023531, [1]
    [Cycle 1]: 0.00022793, [11]
        [c_1]: 4.907e-05
        [parameter_eliminate]: 3.9e-06
        [updatestate_depend_eliminate]: 2.328e-05
        [updatestate_assign_eliminate]: 5.72999e-06
        [updatestate_loads_eliminate]: 4.45e-06
        [cse]: 3.88e-05
        [call_graph_tuple_transform]: 2.497e-05
        [tuple_list_get_item_eliminator]: 8.99e-06
        [none_parameter_eliminate]: 1.82999e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 1.101e-05
[remove_dup_value]: 2.084e-05
[partial_unused_args_eliminate]: 2.38002e-06
[environ_conv]: 1.18e-05
[add_recomputation]: 0.00014585
[cse_after_recomputation]: 3.805e-05, [1]
    [Cycle 1]: 2.987e-05, [1]
        [cse]: 2.169e-05
[auto_monad_reorder]: 2.804e-05
[get_jit_bprop_graph]: 1.88002e-06
[rewriter_after_jit_bprop_graph]: 7.87e-06
[opt_after_jit_grad]: 0.00054265
[symbol_engine_optimizer]: 0.00010888, [1]
    [Cycle 1]: 0.0001017, [6]
        [build]: 1.52e-05
        [elim_shapecalc]: 1.248e-05
        [elim_not_effective]: 1.949e-05
        [opt_reshape]: 8.90999e-06
        [fold_const_symbol]: 1.43e-05
        [renormalize]: 5.00004e-07
[validate]: 6.74e-05
[backend_pass]: 1.10001e-06
[task_emit]: 0.0106376
[execute]: 8.54e-06
Sums
    bootstrap                            :     0.000321s :  0.78%
    type_inference                       :     0.024235s : 58.76%
    event_method                         :     0.000023s :  0.06%
    auto_monad                           :     0.000170s :  0.41%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000144s :  0.35%
    rewriter_before_opt_a                :     0.000105s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000083s :  0.20%
    jit_opt_a.loop_unroll                :     0.000045s :  0.11%
    jit_opt_a.a_1                        :     0.000964s :  2.34%
    jit_opt_a.with_stream_mark           :     0.000055s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000038s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.06%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001226s :  2.97%
    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      :     0.000033s :  0.08%
    jit_opt_a.cse                        :     0.000053s :  0.13%
    jit_opt_a.replace_applicator         :     0.000029s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.04%
    rewriter_after_opt_a                 :     0.000302s :  0.73%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000800s :  1.94%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.06%
    cconv                                :     0.000033s :  0.08%
    loop_unroll                          :     0.000481s :  1.17%
    jit_opt_after_cconv.c_1              :     0.000049s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000023s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000039s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000011s :  0.03%
    remove_dup_value                     :     0.000021s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.03%
    add_recomputation                    :     0.000146s :  0.35%
    cse_after_recomputation.cse          :     0.000022s :  0.05%
    auto_monad_reorder                   :     0.000028s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000543s :  1.32%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000067s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010638s : 25.79%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000345    47
 2.96% :     0.000010s :      2: substitution.depend_value_elim
 0.87% :     0.000003s :      4: substitution.elim_not_effective
 0.67% :     0.000002s :      4: substitution.fold_const_symbol
 2.11% :     0.000007s :      5: substitution.graph_param_transform
74.44% :     0.000257s :      4: substitution.inline
 1.44% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.17% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.48% :     0.000005s :      2: substitution.replace_old_param
 3.60% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 3.28% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 6.97% :     0.000024s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024150     2
93.31% :     0.022534s :      1: type_inference.infer
 6.69% :     0.001617s :      1: type_inference.specialize
------[replace.]   0.000103     6
65.64% :     0.000068s :      4: replace.inline
34.36% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000265     6
95.76% :     0.000254s :      4: match.inline
 4.24% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000181  1021
 1.20% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.02% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     15: predicate.addn_check_dump
 1.65% :     0.000003s :     15: predicate.addn_zero_filter
 2.06% :     0.000004s :     15: predicate.arithmetic_simplify
 1.61% :     0.000003s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     15: predicate.compare_switch_simplify
 1.39% :     0.000003s :     15: predicate.depend_value_elim
 1.09% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.44% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.12% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.89% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.62% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.26% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     15: predicate.environ_get_eliminate
 1.28% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.08% :     0.000002s :     10: predicate.get_grad_eliminate
 0.26% :     0.000000s :      5: predicate.graph_param_transform
 4.93% :     0.000009s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.41% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.62% :     0.000003s :     10: predicate.less_batch_normalization
 1.49% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     22: predicate.load_eliminater
 1.57% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.39% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.21% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.39% :     0.000003s :     15: predicate.merge_addn
 1.09% :     0.000002s :     15: predicate.minmaximum_grad
 1.58% :     0.000003s :      5: predicate.mutable_eliminate
 0.53% :     0.000001s :      5: predicate.opt_reshape
 2.35% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.62% :     0.000003s :     15: predicate.reduce_eliminate
 1.51% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :     10: predicate.remove_not_recompute_node
 2.06% :     0.000004s :     27: predicate.replace_applicator
 0.77% :     0.000001s :     10: predicate.replace_old_param
 0.44% :     0.000001s :      5: predicate.reset_defer_inline
 1.11% :     0.000002s :     15: predicate.reshape_eliminate
 1.20% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.89% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.39% :     0.000003s :     15: predicate.same_eliminate
 0.67% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.09% :     0.000002s :     10: predicate.special_op_eliminate
 1.01% :     0.000002s :     10: predicate.specialize_transform
 1.52% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.18% :     0.000004s :     21: predicate.switch_defer_inline
 1.88% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.46% :     0.000012s :     68: predicate.switch_simplify
 1.23% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.47% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.97% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.69% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000003s :     15: predicate.value_based_eliminate
 0.37% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000951    13
44.86% :     0.000427s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.50% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.64% :     0.000462s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048724    76
 0.31% :     0.000152s :      1: add_recomputation
 0.36% :     0.000177s :      1: auto_monad
 0.06% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.70% :     0.000340s :      1: bootstrap
 0.07% :     0.000036s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.08% :     0.000041s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.06% :     0.000028s :      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% :     0.000009s :      1: graph_reusing
14.67% :     0.007149s :      1: jit_opt_a
 0.49% :     0.000238s :      1: jit_opt_after_cconv
 0.15% :     0.000075s :      1: jit_opt_b
 1.01% :     0.000490s :      1: loop_unroll
 1.67% :     0.000812s :      1: mutable_eliminate
 2.68% :     0.001307s :     26: opt.transform.jit_opt_a
 0.18% :     0.000089s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000044s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.13% :     0.000552s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000150s :      1: py_interpret_to_execute
 0.04% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000024s :      1: remove_dup_value
 1.52% :     0.000743s :      1: renormalize.infer
 0.97% :     0.000472s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.63% :     0.000309s :      1: rewriter_after_opt_a
 0.23% :     0.000110s :      1: rewriter_before_opt_a
 0.23% :     0.000112s :      1: symbol_engine_optimizer
21.87% :     0.010655s :      1: task_emit
49.78% :     0.024254s :      1: type_inference
 0.19% :     0.000095s :      1: validate


TotalTime = 0.0816675, [33]
[bootstrap]: 0.00034164
[type_inference]: 0.0388589
[event_method]: 2.573e-05
[auto_monad]: 0.00018615
[graph_reusing]: 7.18e-06
[pre_auto_parallel]: 3.45e-06
[py_interpret_to_execute]: 0.00016848
[rewriter_before_opt_a]: 0.0001067
[expand_dump_flag]: 4.70001e-06
[jit_opt_a]: 0.0243544, [2]
    [Cycle 1]: 0.00289127, [27]
        [switch_simplify]: 7.705e-05
        [loop_unroll]: 3.757e-05
        [a_1]: 0.00079901
        [with_stream_mark]: 2.385e-05
        [recompute_prepare]: 1.326e-05
        [updatestate_depend_eliminate]: 6.82002e-06
        [updatestate_assign_eliminate]: 7.22002e-06
        [updatestate_loads_eliminate]: 4.90999e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 1.001e-05
        [updatestate_useless_node_eliminater]: 1.261e-05
        [accelerated_algorithm]: 9.09998e-06
        [meta_shard_fg_expand]: 2.84001e-06
        [get_grad_eliminate_]: 8.22e-06
        [merge_forward]: 6.46999e-06
        [cell_reuse_recompute_pass]: 2.29001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.295e-05
        [j_node_and_user_rematch]: 1.442e-05
        [meta_fg_expand]: 3.62002e-06
        [replace_old_param]: 1.528e-05
        [inline_without_move]: 9.18002e-06
        [renormalize]: 0.00145718
        [add_forward_monad_depend]: 9.63002e-06
        [auto_monad_grad]: 2.85002e-06
        [auto_monad_eliminator]: 2.679e-05
        [cse]: 4.764e-05
        [replace_applicator]: 2.505e-05
    [Cycle 2]: 0.00054824, [27]
        [switch_simplify]: 1.093e-05
        [loop_unroll]: 9.49999e-06
        [a_1]: 0.00019618
        [with_stream_mark]: 1.72e-05
        [recompute_prepare]: 1.079e-05
        [updatestate_depend_eliminate]: 5.99999e-06
        [updatestate_assign_eliminate]: 5.62001e-06
        [updatestate_loads_eliminate]: 4.47e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 8.73001e-06
        [updatestate_useless_node_eliminater]: 1.067e-05
        [accelerated_algorithm]: 8.24998e-06
        [meta_shard_fg_expand]: 3.43e-06
        [get_grad_eliminate_]: 8.72e-06
        [merge_forward]: 6.16998e-06
        [cell_reuse_recompute_pass]: 4.44002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.194e-05
        [j_node_and_user_rematch]: 1.424e-05
        [meta_fg_expand]: 3.7e-06
        [replace_old_param]: 1.196e-05
        [inline_without_move]: 7.86001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.19001e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 1.165e-05
        [cse]: 2.39e-05
        [replace_applicator]: 8.92e-06
[py_interpret_to_execute_after_opt_a]: 1.436e-05
[rewriter_after_opt_a]: 0.00032556
[convert_after_rewriter]: 1.333e-05
[order_py_execute_after_rewriter]: 8.77e-06
[mutable_eliminate]: 0.00089608
[jit_opt_b]: 8.477e-05, [1]
    [Cycle 1]: 7.377e-05, [2]
        [frontend_op_eliminate]: 2.839e-05
        [inline_after_opt_a]: 3.085e-05
[cconv]: 4.191e-05
[loop_unroll]: 0.00056119
[jit_opt_after_cconv]: 0.00024615, [1]
    [Cycle 1]: 0.00023879, [11]
        [c_1]: 5.759e-05
        [parameter_eliminate]: 5.59e-06
        [updatestate_depend_eliminate]: 1.158e-05
        [updatestate_assign_eliminate]: 5.52001e-06
        [updatestate_loads_eliminate]: 4.44998e-06
        [cse]: 4.789e-05
        [call_graph_tuple_transform]: 2.646e-05
        [tuple_list_get_item_eliminator]: 9.22999e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 1.00001e-06
        [switch_simplify]: 9.36e-06
[remove_dup_value]: 2.037e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 1.519e-05
[add_recomputation]: 7.797e-05
[cse_after_recomputation]: 3.181e-05, [1]
    [Cycle 1]: 2.497e-05, [1]
        [cse]: 1.848e-05
[auto_monad_reorder]: 2.673e-05
[get_jit_bprop_graph]: 2.01998e-06
[rewriter_after_jit_bprop_graph]: 8.48999e-06
[opt_after_jit_grad]: 0.00059931
[symbol_engine_optimizer]: 0.00011084, [1]
    [Cycle 1]: 0.0001042, [6]
        [build]: 1.563e-05
        [elim_shapecalc]: 1.233e-05
        [elim_not_effective]: 2.066e-05
        [opt_reshape]: 9.79e-06
        [fold_const_symbol]: 1.38e-05
        [renormalize]: 6.40022e-07
[validate]: 7.221e-05
[backend_pass]: 1.33002e-06
[task_emit]: 0.0141793
[execute]: 7.21999e-06
Sums
    bootstrap                            :     0.000342s :  0.57%
    type_inference                       :     0.038859s : 64.78%
    event_method                         :     0.000026s :  0.04%
    auto_monad                           :     0.000186s :  0.31%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000168s :  0.28%
    rewriter_before_opt_a                :     0.000107s :  0.18%
    expand_dump_flag                     :     0.000005s :  0.01%
    jit_opt_a.switch_simplify            :     0.000088s :  0.15%
    jit_opt_a.loop_unroll                :     0.000047s :  0.08%
    jit_opt_a.a_1                        :     0.000995s :  1.66%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.03%
    jit_opt_a.merge_forward              :     0.000013s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000045s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000027s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.03%
    jit_opt_a.renormalize                :     0.001457s :  2.43%
    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      :     0.000038s :  0.06%
    jit_opt_a.cse                        :     0.000072s :  0.12%
    jit_opt_a.replace_applicator         :     0.000034s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.02%
    rewriter_after_opt_a                 :     0.000326s :  0.54%
    convert_after_rewriter               :     0.000013s :  0.02%
    order_py_execute_after_rewriter      :     0.000009s :  0.01%
    mutable_eliminate                    :     0.000896s :  1.49%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000031s :  0.05%
    cconv                                :     0.000042s :  0.07%
    loop_unroll                          :     0.000561s :  0.94%
    jit_opt_after_cconv.c_1              :     0.000058s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000048s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000015s :  0.03%
    add_recomputation                    :     0.000078s :  0.13%
    cse_after_recomputation.cse          :     0.000018s :  0.03%
    auto_monad_reorder                   :     0.000027s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.01%
    opt_after_jit_grad                   :     0.000599s :  1.00%
    symbol_engine_optimizer.build        :     0.000016s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000072s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.014179s : 23.64%
    execute                              :     0.000007s :  0.01%


Time group info:
------[substitution.]   0.000343    47
 3.45% :     0.000012s :      2: substitution.depend_value_elim
 0.92% :     0.000003s :      4: substitution.elim_not_effective
 0.59% :     0.000002s :      4: substitution.fold_const_symbol
 2.34% :     0.000008s :      5: substitution.graph_param_transform
74.91% :     0.000257s :      4: substitution.inline
 1.59% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.65% :     0.000009s :      8: substitution.remove_not_recompute_node
 2.24% :     0.000008s :      2: substitution.replace_old_param
 3.93% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
 3.85% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.53% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.038764     2
95.52% :     0.037029s :      1: type_inference.infer
 4.48% :     0.001735s :      1: type_inference.specialize
------[replace.]   0.000102     6
66.57% :     0.000068s :      4: replace.inline
33.43% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000267     6
95.33% :     0.000254s :      4: match.inline
 4.67% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000195  1021
 1.34% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.08% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.00% :     0.000002s :     15: predicate.addn_check_dump
 1.43% :     0.000003s :     15: predicate.addn_zero_filter
 2.15% :     0.000004s :     15: predicate.arithmetic_simplify
 1.74% :     0.000003s :     15: predicate.cast_eliminate
 0.65% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.02% :     0.000002s :     15: predicate.compare_switch_simplify
 1.50% :     0.000003s :     15: predicate.depend_value_elim
 1.18% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.29% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.11% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.81% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.63% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.05% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.06% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.37% :     0.000003s :     15: predicate.environ_get_eliminate
 1.03% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 1.09% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000001s :      5: predicate.graph_param_transform
 5.68% :     0.000011s :     31: predicate.inline
 0.92% :     0.000002s :     10: predicate.inline_without_move
 0.38% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.32% :     0.000003s :     10: predicate.less_batch_normalization
 1.60% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.68% :     0.000003s :     22: predicate.load_eliminater
 1.26% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.34% :     0.000007s :     42: predicate.loop_unroll_before_grad
 2.11% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     15: predicate.merge_addn
 1.08% :     0.000002s :     15: predicate.minmaximum_grad
 1.88% :     0.000004s :      5: predicate.mutable_eliminate
 0.53% :     0.000001s :      5: predicate.opt_reshape
 2.23% :     0.000004s :     22: predicate.partial_eliminate
 1.24% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.71% :     0.000003s :     15: predicate.reduce_eliminate
 1.49% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.95% :     0.000004s :     27: predicate.replace_applicator
 0.63% :     0.000001s :     10: predicate.replace_old_param
 0.44% :     0.000001s :      5: predicate.reset_defer_inline
 1.13% :     0.000002s :     15: predicate.reshape_eliminate
 1.26% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.01% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.36% :     0.000003s :     15: predicate.same_eliminate
 0.52% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.15% :     0.000002s :     10: predicate.special_op_eliminate
 0.91% :     0.000002s :     10: predicate.specialize_transform
 1.38% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.74% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.20% :     0.000004s :     21: predicate.switch_defer_inline
 1.75% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.51% :     0.000013s :     68: predicate.switch_simplify
 1.08% :     0.000002s :     15: predicate.tile_eliminate
 1.37% :     0.000003s :     15: predicate.transpose_eliminate
 1.58% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.57% :     0.000009s :     27: predicate.tuple_list_get_item_eliminator
 2.18% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.76% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.11% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000004s :     15: predicate.value_based_eliminate
 0.37% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000959    13
44.29% :     0.000425s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.31% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.39% :     0.000483s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.084693    76
 0.10% :     0.000081s :      1: add_recomputation
 0.23% :     0.000192s :      1: auto_monad
 0.03% :     0.000029s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.42% :     0.000358s :      1: bootstrap
 0.05% :     0.000045s :      1: cconv
 0.02% :     0.000016s :      1: convert_after_rewriter
 0.04% :     0.000034s :      1: cse_after_recomputation
 0.02% :     0.000018s :      1: environ_conv
 0.04% :     0.000031s :      1: event_method
 0.01% :     0.000013s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
28.76% :     0.024358s :      1: jit_opt_a
 0.29% :     0.000249s :      1: jit_opt_after_cconv
 0.10% :     0.000088s :      1: jit_opt_b
 0.68% :     0.000572s :      1: loop_unroll
 1.08% :     0.000914s :      1: mutable_eliminate
 1.59% :     0.001344s :     26: opt.transform.jit_opt_a
 0.12% :     0.000098s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000050s :      4: opt.transform.jit_opt_b
 0.03% :     0.000022s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000030s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000052s :      4: opt.transform.symbol_engine_opt
 0.72% :     0.000610s :      1: opt_after_jit_grad
 0.01% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000174s :      1: py_interpret_to_execute
 0.02% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000023s :      1: remove_dup_value
 1.10% :     0.000933s :      1: renormalize.infer
 0.60% :     0.000512s :      1: renormalize.specialize
 0.01% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.39% :     0.000331s :      1: rewriter_after_opt_a
 0.13% :     0.000111s :      1: rewriter_before_opt_a
 0.13% :     0.000114s :      1: symbol_engine_optimizer
16.76% :     0.014197s :      1: task_emit
45.91% :     0.038882s :      1: type_inference
 0.12% :     0.000102s :      1: validate


TotalTime = 0.0798937, [33]
[bootstrap]: 0.0002725
[type_inference]: 0.056814
[event_method]: 2.401e-05
[auto_monad]: 0.00018115
[graph_reusing]: 7.28e-06
[pre_auto_parallel]: 3.61999e-06
[py_interpret_to_execute]: 0.00014589
[rewriter_before_opt_a]: 0.00010351
[expand_dump_flag]: 3.37002e-06
[jit_opt_a]: 0.00751104, [2]
    [Cycle 1]: 0.00286878, [27]
        [switch_simplify]: 7.535e-05
        [loop_unroll]: 3.978e-05
        [a_1]: 0.00093905
        [with_stream_mark]: 2.351e-05
        [recompute_prepare]: 1.2e-05
        [updatestate_depend_eliminate]: 6.66e-06
        [updatestate_assign_eliminate]: 6.39999e-06
        [updatestate_loads_eliminate]: 4.79002e-06
        [parameter_eliminate]: 2.05002e-06
        [specialize_transform]: 1.006e-05
        [updatestate_useless_node_eliminater]: 1.292e-05
        [accelerated_algorithm]: 9.63002e-06
        [meta_shard_fg_expand]: 3.14999e-06
        [get_grad_eliminate_]: 9.84001e-06
        [merge_forward]: 5.67001e-06
        [cell_reuse_recompute_pass]: 2.04999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.216e-05
        [j_node_and_user_rematch]: 1.488e-05
        [meta_fg_expand]: 3.58999e-06
        [replace_old_param]: 1.266e-05
        [inline_without_move]: 8.84003e-06
        [renormalize]: 0.00133314
        [add_forward_monad_depend]: 7.31999e-06
        [auto_monad_grad]: 3.01001e-06
        [auto_monad_eliminator]: 2.335e-05
        [cse]: 4.313e-05
        [replace_applicator]: 2.51e-05
    [Cycle 2]: 0.00051175, [27]
        [switch_simplify]: 9.47999e-06
        [loop_unroll]: 8.67e-06
        [a_1]: 0.00018239
        [with_stream_mark]: 1.546e-05
        [recompute_prepare]: 8.80001e-06
        [updatestate_depend_eliminate]: 5.66003e-06
        [updatestate_assign_eliminate]: 5.84999e-06
        [updatestate_loads_eliminate]: 4.33001e-06
        [parameter_eliminate]: 1.86998e-06
        [specialize_transform]: 8.54002e-06
        [updatestate_useless_node_eliminater]: 1.061e-05
        [accelerated_algorithm]: 8.05e-06
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 8.17e-06
        [merge_forward]: 4.78001e-06
        [cell_reuse_recompute_pass]: 4.1e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.94e-05
        [j_node_and_user_rematch]: 1.307e-05
        [meta_fg_expand]: 3.44001e-06
        [replace_old_param]: 1.099e-05
        [inline_without_move]: 7.70998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 1.74e-06
        [auto_monad_eliminator]: 1.214e-05
        [cse]: 2.324e-05
        [replace_applicator]: 8.85001e-06
[py_interpret_to_execute_after_opt_a]: 1.993e-05
[rewriter_after_opt_a]: 0.00032668
[convert_after_rewriter]: 1.374e-05
[order_py_execute_after_rewriter]: 8.51002e-06
[mutable_eliminate]: 0.00096471
[jit_opt_b]: 8.465e-05, [1]
    [Cycle 1]: 7.428e-05, [2]
        [frontend_op_eliminate]: 2.973e-05
        [inline_after_opt_a]: 2.996e-05
[cconv]: 3.841e-05
[loop_unroll]: 0.00058341
[jit_opt_after_cconv]: 0.00025292, [1]
    [Cycle 1]: 0.00024436, [11]
        [c_1]: 5.779e-05
        [parameter_eliminate]: 5.57001e-06
        [updatestate_depend_eliminate]: 1.3e-05
        [updatestate_assign_eliminate]: 5.60001e-06
        [updatestate_loads_eliminate]: 5.49e-06
        [cse]: 4.921e-05
        [call_graph_tuple_transform]: 2.779e-05
        [tuple_list_get_item_eliminator]: 9.79999e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 1.04e-06
        [switch_simplify]: 9.89999e-06
[remove_dup_value]: 2.311e-05
[partial_unused_args_eliminate]: 3.01999e-06
[environ_conv]: 1.555e-05
[add_recomputation]: 7.65e-05
[cse_after_recomputation]: 3.862e-05, [1]
    [Cycle 1]: 3.133e-05, [1]
        [cse]: 2.323e-05
[auto_monad_reorder]: 2.921e-05
[get_jit_bprop_graph]: 2.47001e-06
[rewriter_after_jit_bprop_graph]: 1.171e-05
[opt_after_jit_grad]: 0.00062371
[symbol_engine_optimizer]: 0.00012119, [1]
    [Cycle 1]: 0.00011355, [6]
        [build]: 1.911e-05
        [elim_shapecalc]: 1.312e-05
        [elim_not_effective]: 2.153e-05
        [opt_reshape]: 9.86998e-06
        [fold_const_symbol]: 1.633e-05
        [renormalize]: 8.49977e-07
[validate]: 7.479e-05
[backend_pass]: 1.20999e-06
[task_emit]: 0.0112282
[execute]: 1.08e-05
Sums
    bootstrap                            :     0.000273s :  0.36%
    type_inference                       :     0.056814s : 75.74%
    event_method                         :     0.000024s :  0.03%
    auto_monad                           :     0.000181s :  0.24%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000004s :  0.00%
    py_interpret_to_execute              :     0.000146s :  0.19%
    rewriter_before_opt_a                :     0.000104s :  0.14%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000085s :  0.11%
    jit_opt_a.loop_unroll                :     0.000048s :  0.06%
    jit_opt_a.a_1                        :     0.001121s :  1.49%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000018s :  0.02%
    jit_opt_a.merge_forward              :     0.000010s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000042s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.03%
    jit_opt_a.inline_without_move        :     0.000017s :  0.02%
    jit_opt_a.renormalize                :     0.001333s :  1.78%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000035s :  0.05%
    jit_opt_a.cse                        :     0.000066s :  0.09%
    jit_opt_a.replace_applicator         :     0.000034s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.03%
    rewriter_after_opt_a                 :     0.000327s :  0.44%
    convert_after_rewriter               :     0.000014s :  0.02%
    order_py_execute_after_rewriter      :     0.000009s :  0.01%
    mutable_eliminate                    :     0.000965s :  1.29%
    jit_opt_b.frontend_op_eliminate      :     0.000030s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000030s :  0.04%
    cconv                                :     0.000038s :  0.05%
    loop_unroll                          :     0.000583s :  0.78%
    jit_opt_after_cconv.c_1              :     0.000058s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000049s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000010s :  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  :     0.000010s :  0.01%
    remove_dup_value                     :     0.000023s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000016s :  0.02%
    add_recomputation                    :     0.000077s :  0.10%
    cse_after_recomputation.cse          :     0.000023s :  0.03%
    auto_monad_reorder                   :     0.000029s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.02%
    opt_after_jit_grad                   :     0.000624s :  0.83%
    symbol_engine_optimizer.build        :     0.000019s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000075s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011228s : 14.97%
    execute                              :     0.000011s :  0.01%


Time group info:
------[substitution.]   0.000333    47
 3.85% :     0.000013s :      2: substitution.depend_value_elim
 0.92% :     0.000003s :      4: substitution.elim_not_effective
 0.80% :     0.000003s :      4: substitution.fold_const_symbol
 2.38% :     0.000008s :      5: substitution.graph_param_transform
74.45% :     0.000248s :      4: substitution.inline
 1.79% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.40% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.79% :     0.000006s :      2: substitution.replace_old_param
 4.58% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.75% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.28% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.056725     2
96.93% :     0.054986s :      1: type_inference.infer
 3.07% :     0.001739s :      1: type_inference.specialize
------[replace.]   0.000238     6
85.80% :     0.000204s :      4: replace.inline
14.20% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000259     6
94.54% :     0.000245s :      4: match.inline
 5.46% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000194  1021
 1.13% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.31% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     15: predicate.addn_check_dump
 1.42% :     0.000003s :     15: predicate.addn_zero_filter
 1.92% :     0.000004s :     15: predicate.arithmetic_simplify
 1.70% :     0.000003s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     15: predicate.compare_switch_simplify
 1.47% :     0.000003s :     15: predicate.depend_value_elim
 1.01% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.00% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.31% :     0.000001s :      5: predicate.elim_not_effective
 0.60% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.05% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     15: predicate.environ_get_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.18% :     0.000002s :     10: predicate.get_grad_eliminate
 0.37% :     0.000001s :      5: predicate.graph_param_transform
 5.85% :     0.000011s :     31: predicate.inline
 0.94% :     0.000002s :     10: predicate.inline_without_move
 0.38% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.31% :     0.000003s :     10: predicate.less_batch_normalization
 1.54% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.87% :     0.000004s :     22: predicate.load_eliminater
 1.84% :     0.000004s :      5: predicate.loop_unroll_after_grad
 3.27% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.98% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000002s :     15: predicate.merge_addn
 1.06% :     0.000002s :     15: predicate.minmaximum_grad
 2.29% :     0.000004s :      5: predicate.mutable_eliminate
 0.64% :     0.000001s :      5: predicate.opt_reshape
 2.13% :     0.000004s :     22: predicate.partial_eliminate
 1.07% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     15: predicate.reduce_eliminate
 1.41% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.53% :     0.000003s :     27: predicate.replace_applicator
 0.62% :     0.000001s :     10: predicate.replace_old_param
 0.27% :     0.000001s :      5: predicate.reset_defer_inline
 1.47% :     0.000003s :     15: predicate.reshape_eliminate
 1.58% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 1.26% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.30% :     0.000003s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.31% :     0.000003s :     10: predicate.special_op_eliminate
 1.02% :     0.000002s :     10: predicate.specialize_transform
 1.36% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.20% :     0.000004s :     21: predicate.switch_defer_inline
 1.79% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.83% :     0.000013s :     68: predicate.switch_simplify
 1.13% :     0.000002s :     15: predicate.tile_eliminate
 1.19% :     0.000002s :     15: predicate.transpose_eliminate
 1.47% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.80% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000937    13
38.68% :     0.000362s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.12% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
56.20% :     0.000526s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.082918    76
 0.10% :     0.000081s :      1: add_recomputation
 0.22% :     0.000186s :      1: auto_monad
 0.04% :     0.000032s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.35% :     0.000288s :      1: bootstrap
 0.05% :     0.000042s :      1: cconv
 0.02% :     0.000017s :      1: convert_after_rewriter
 0.05% :     0.000041s :      1: cse_after_recomputation
 0.02% :     0.000018s :      1: environ_conv
 0.04% :     0.000030s :      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
 9.06% :     0.007516s :      1: jit_opt_a
 0.31% :     0.000256s :      1: jit_opt_after_cconv
 0.11% :     0.000089s :      1: jit_opt_b
 0.72% :     0.000594s :      1: loop_unroll
 1.18% :     0.000980s :      1: mutable_eliminate
 1.76% :     0.001461s :     26: opt.transform.jit_opt_a
 0.12% :     0.000101s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000050s :      4: opt.transform.jit_opt_b
 0.02% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000032s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000057s :      4: opt.transform.symbol_engine_opt
 0.77% :     0.000636s :      1: opt_after_jit_grad
 0.01% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.18% :     0.000151s :      1: py_interpret_to_execute
 0.03% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000026s :      1: remove_dup_value
 0.97% :     0.000803s :      1: renormalize.infer
 0.63% :     0.000519s :      1: renormalize.specialize
 0.02% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.40% :     0.000333s :      1: rewriter_after_opt_a
 0.13% :     0.000108s :      1: rewriter_before_opt_a
 0.15% :     0.000124s :      1: symbol_engine_optimizer
13.57% :     0.011249s :      1: task_emit
68.54% :     0.056835s :      1: type_inference
 0.13% :     0.000104s :      1: validate


TotalTime = 0.048178, [33]
[bootstrap]: 0.00039784
[type_inference]: 0.0261747
[event_method]: 2.688e-05
[auto_monad]: 0.0001975
[graph_reusing]: 7.41999e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 0.00015622
[rewriter_before_opt_a]: 0.00010995
[expand_dump_flag]: 3.9e-06
[jit_opt_a]: 0.00701191, [2]
    [Cycle 1]: 0.00248184, [27]
        [switch_simplify]: 7.196e-05
        [loop_unroll]: 3.645e-05
        [a_1]: 0.00076069
        [with_stream_mark]: 1.69e-05
        [recompute_prepare]: 1.035e-05
        [updatestate_depend_eliminate]: 5.86998e-06
        [updatestate_assign_eliminate]: 5.01002e-06
        [updatestate_loads_eliminate]: 1.694e-05
        [parameter_eliminate]: 1.49998e-06
        [specialize_transform]: 9.87001e-06
        [updatestate_useless_node_eliminater]: 1.221e-05
        [accelerated_algorithm]: 9.08002e-06
        [meta_shard_fg_expand]: 2.30002e-06
        [get_grad_eliminate_]: 8.22998e-06
        [merge_forward]: 4.53001e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.759e-05
        [j_node_and_user_rematch]: 1.427e-05
        [meta_fg_expand]: 3.23e-06
        [replace_old_param]: 1.285e-05
        [inline_without_move]: 8.90999e-06
        [renormalize]: 0.00118528
        [add_forward_monad_depend]: 5.40999e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 1.894e-05
        [cse]: 3.023e-05
        [replace_applicator]: 1.77e-05
    [Cycle 2]: 0.00047658, [27]
        [switch_simplify]: 9.40001e-06
        [loop_unroll]: 8.23999e-06
        [a_1]: 0.000173
        [with_stream_mark]: 1.148e-05
        [recompute_prepare]: 8.2e-06
        [updatestate_depend_eliminate]: 4.90001e-06
        [updatestate_assign_eliminate]: 4.79e-06
        [updatestate_loads_eliminate]: 4.24002e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 8.32e-06
        [updatestate_useless_node_eliminater]: 1.056e-05
        [accelerated_algorithm]: 8.35999e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 7.75e-06
        [merge_forward]: 4.60999e-06
        [cell_reuse_recompute_pass]: 3.03e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.644e-05
        [j_node_and_user_rematch]: 1.302e-05
        [meta_fg_expand]: 3.03e-06
        [replace_old_param]: 1.053e-05
        [inline_without_move]: 7.80998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 1.034e-05
        [cse]: 2.009e-05
        [replace_applicator]: 8.03999e-06
[py_interpret_to_execute_after_opt_a]: 1.225e-05
[rewriter_after_opt_a]: 0.00027989
[convert_after_rewriter]: 1.003e-05
[order_py_execute_after_rewriter]: 6.50002e-06
[mutable_eliminate]: 0.00067269
[jit_opt_b]: 7.031e-05, [1]
    [Cycle 1]: 6.318e-05, [2]
        [frontend_op_eliminate]: 2.559e-05
        [inline_after_opt_a]: 2.554e-05
[cconv]: 1.872e-05
[loop_unroll]: 0.00045522
[jit_opt_after_cconv]: 0.00019485, [1]
    [Cycle 1]: 0.00018761, [11]
        [c_1]: 4.565e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 7.83001e-06
        [updatestate_assign_eliminate]: 4.25e-06
        [updatestate_loads_eliminate]: 4.22e-06
        [cse]: 2.926e-05
        [call_graph_tuple_transform]: 2.118e-05
        [tuple_list_get_item_eliminator]: 8.91997e-06
        [none_parameter_eliminate]: 1.34998e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 9.00001e-06
[remove_dup_value]: 1.397e-05
[partial_unused_args_eliminate]: 1.50999e-06
[environ_conv]: 1.072e-05
[add_recomputation]: 5.492e-05
[cse_after_recomputation]: 3.088e-05, [1]
    [Cycle 1]: 2.477e-05, [1]
        [cse]: 1.813e-05
[auto_monad_reorder]: 2.136e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 6.02001e-06
[opt_after_jit_grad]: 0.00061933
[symbol_engine_optimizer]: 0.0001086, [1]
    [Cycle 1]: 0.00010129, [6]
        [build]: 1.444e-05
        [elim_shapecalc]: 1.204e-05
        [elim_not_effective]: 2.014e-05
        [opt_reshape]: 8.83001e-06
        [fold_const_symbol]: 1.392e-05
        [renormalize]: 6.00005e-07
[validate]: 5.455e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.0111692
[execute]: 1.006e-05
Sums
    bootstrap                            :     0.000398s :  0.92%
    type_inference                       :     0.026175s : 60.28%
    event_method                         :     0.000027s :  0.06%
    auto_monad                           :     0.000198s :  0.45%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000156s :  0.36%
    rewriter_before_opt_a                :     0.000110s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000081s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.10%
    jit_opt_a.a_1                        :     0.000934s :  2.15%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000021s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    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 :     0.000034s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    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        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001185s :  2.73%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.07%
    jit_opt_a.cse                        :     0.000050s :  0.12%
    jit_opt_a.replace_applicator         :     0.000026s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000280s :  0.64%
    convert_after_rewriter               :     0.000010s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000673s :  1.55%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000019s :  0.04%
    loop_unroll                          :     0.000455s :  1.05%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.11%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000014s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000011s :  0.02%
    add_recomputation                    :     0.000055s :  0.13%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000021s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000619s :  1.43%
    symbol_engine_optimizer.build        :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000055s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011169s : 25.72%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000285    47
 3.45% :     0.000010s :      2: substitution.depend_value_elim
 1.09% :     0.000003s :      4: substitution.elim_not_effective
 0.72% :     0.000002s :      4: substitution.fold_const_symbol
 1.93% :     0.000006s :      5: substitution.graph_param_transform
77.72% :     0.000222s :      4: substitution.inline
 1.53% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.53% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.73% :     0.000005s :      2: substitution.replace_old_param
 2.95% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
 3.05% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 3.31% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026074     2
93.09% :     0.024272s :      1: type_inference.infer
 6.91% :     0.001802s :      1: type_inference.specialize
------[replace.]   0.000096     6
67.10% :     0.000064s :      4: replace.inline
32.90% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     6
96.75% :     0.000218s :      4: match.inline
 3.25% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000175  1021
 1.47% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     15: predicate.addn_check_dump
 1.43% :     0.000002s :     15: predicate.addn_zero_filter
 2.21% :     0.000004s :     15: predicate.arithmetic_simplify
 1.56% :     0.000003s :     15: predicate.cast_eliminate
 0.65% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.27% :     0.000002s :     15: predicate.compare_switch_simplify
 1.62% :     0.000003s :     15: predicate.depend_value_elim
 1.15% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.05% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      5: predicate.elim_not_effective
 0.61% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.60% :     0.000003s :     15: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     15: predicate.environ_get_eliminate
 1.43% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.12% :     0.000002s :     10: predicate.get_grad_eliminate
 0.29% :     0.000001s :      5: predicate.graph_param_transform
 4.82% :     0.000008s :     31: predicate.inline
 1.13% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.72% :     0.000003s :     10: predicate.less_batch_normalization
 1.56% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.02% :     0.000004s :     22: predicate.load_eliminater
 1.24% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.53% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.05% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.25% :     0.000002s :     15: predicate.merge_addn
 1.12% :     0.000002s :     15: predicate.minmaximum_grad
 1.57% :     0.000003s :      5: predicate.mutable_eliminate
 0.47% :     0.000001s :      5: predicate.opt_reshape
 2.27% :     0.000004s :     22: predicate.partial_eliminate
 1.23% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.60% :     0.000003s :     15: predicate.reduce_eliminate
 1.40% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.70% :     0.000003s :     27: predicate.replace_applicator
 0.58% :     0.000001s :     10: predicate.replace_old_param
 0.30% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000002s :     15: predicate.reshape_eliminate
 1.29% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     15: predicate.same_eliminate
 0.52% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.04% :     0.000002s :     10: predicate.special_op_eliminate
 1.08% :     0.000002s :     10: predicate.specialize_transform
 1.36% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.13% :     0.000004s :     21: predicate.switch_defer_inline
 2.27% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.46% :     0.000011s :     68: predicate.switch_simplify
 1.19% :     0.000002s :     15: predicate.tile_eliminate
 1.39% :     0.000002s :     15: predicate.transpose_eliminate
 1.41% :     0.000002s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.92% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000003s :     15: predicate.value_based_eliminate
 0.39% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000932    13
44.88% :     0.000418s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.51% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.61% :     0.000472s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050797    76
 0.11% :     0.000058s :      1: add_recomputation
 0.40% :     0.000205s :      1: auto_monad
 0.05% :     0.000024s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.83% :     0.000422s :      1: bootstrap
 0.04% :     0.000021s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.07% :     0.000034s :      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% :     0.000010s :      1: graph_reusing
13.81% :     0.007015s :      1: jit_opt_a
 0.39% :     0.000198s :      1: jit_opt_after_cconv
 0.14% :     0.000073s :      1: jit_opt_b
 0.91% :     0.000464s :      1: loop_unroll
 1.34% :     0.000682s :      1: mutable_eliminate
 2.46% :     0.001250s :     26: opt.transform.jit_opt_a
 0.16% :     0.000081s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000044s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.24% :     0.000628s :      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
 0.32% :     0.000163s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 1.40% :     0.000712s :      1: renormalize.infer
 0.91% :     0.000464s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.56% :     0.000285s :      1: rewriter_after_opt_a
 0.23% :     0.000116s :      1: rewriter_before_opt_a
 0.22% :     0.000112s :      1: symbol_engine_optimizer
22.02% :     0.011187s :      1: task_emit
51.59% :     0.026204s :      1: type_inference
 0.15% :     0.000078s :      1: validate


TotalTime = 0.0509304, [33]
[bootstrap]: 0.00033566
[type_inference]: 0.0269727
[event_method]: 2.254e-05
[auto_monad]: 0.00016946
[graph_reusing]: 6.76999e-06
[pre_auto_parallel]: 3.13e-06
[py_interpret_to_execute]: 0.00014052
[rewriter_before_opt_a]: 0.00010391
[expand_dump_flag]: 3.30003e-06
[jit_opt_a]: 0.00777739, [2]
    [Cycle 1]: 0.00292803, [27]
        [switch_simplify]: 7.755e-05
        [loop_unroll]: 3.643e-05
        [a_1]: 0.00081723
        [with_stream_mark]: 2.738e-05
        [recompute_prepare]: 1.193e-05
        [updatestate_depend_eliminate]: 7.06001e-06
        [updatestate_assign_eliminate]: 5.99999e-06
        [updatestate_loads_eliminate]: 5.17e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 9.91e-06
        [updatestate_useless_node_eliminater]: 1.175e-05
        [accelerated_algorithm]: 9.17001e-06
        [meta_shard_fg_expand]: 3.17002e-06
        [get_grad_eliminate_]: 9.04e-06
        [merge_forward]: 5.46e-06
        [cell_reuse_recompute_pass]: 1.97999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.148e-05
        [j_node_and_user_rematch]: 1.465e-05
        [meta_fg_expand]: 3.73001e-06
        [replace_old_param]: 1.284e-05
        [inline_without_move]: 8.42e-06
        [renormalize]: 0.00149843
        [add_forward_monad_depend]: 1.049e-05
        [auto_monad_grad]: 2.71999e-06
        [auto_monad_eliminator]: 2.681e-05
        [cse]: 4.33e-05
        [replace_applicator]: 2.325e-05
    [Cycle 2]: 0.00052922, [27]
        [switch_simplify]: 1.023e-05
        [loop_unroll]: 8.31002e-06
        [a_1]: 0.00018539
        [with_stream_mark]: 1.91e-05
        [recompute_prepare]: 8.72e-06
        [updatestate_depend_eliminate]: 5.94e-06
        [updatestate_assign_eliminate]: 5.42999e-06
        [updatestate_loads_eliminate]: 4.90999e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 8.32e-06
        [updatestate_useless_node_eliminater]: 1.117e-05
        [accelerated_algorithm]: 8.33999e-06
        [meta_shard_fg_expand]: 2.66999e-06
        [get_grad_eliminate_]: 7.78001e-06
        [merge_forward]: 6.22001e-06
        [cell_reuse_recompute_pass]: 3.85e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.864e-05
        [j_node_and_user_rematch]: 1.318e-05
        [meta_fg_expand]: 3.38e-06
        [replace_old_param]: 1.185e-05
        [inline_without_move]: 8.84e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.73002e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 1.166e-05
        [cse]: 2.463e-05
        [replace_applicator]: 8.97e-06
[py_interpret_to_execute_after_opt_a]: 2.079e-05
[rewriter_after_opt_a]: 0.0003255
[convert_after_rewriter]: 1.33e-05
[order_py_execute_after_rewriter]: 7.93001e-06
[mutable_eliminate]: 0.00091178
[jit_opt_b]: 8.221e-05, [1]
    [Cycle 1]: 7.132e-05, [2]
        [frontend_op_eliminate]: 2.834e-05
        [inline_after_opt_a]: 2.949e-05
[cconv]: 3.727e-05
[loop_unroll]: 0.00054126
[jit_opt_after_cconv]: 0.00023979, [1]
    [Cycle 1]: 0.00023159, [11]
        [c_1]: 5.676e-05
        [parameter_eliminate]: 5.19998e-06
        [updatestate_depend_eliminate]: 1.178e-05
        [updatestate_assign_eliminate]: 5.54e-06
        [updatestate_loads_eliminate]: 4.47998e-06
        [cse]: 4.516e-05
        [call_graph_tuple_transform]: 2.457e-05
        [tuple_list_get_item_eliminator]: 9.42001e-06
        [none_parameter_eliminate]: 2.02999e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 9.76e-06
[remove_dup_value]: 2.241e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 6.881e-05
[add_recomputation]: 7.956e-05
[cse_after_recomputation]: 3.706e-05, [1]
    [Cycle 1]: 2.931e-05, [1]
        [cse]: 2.225e-05
[auto_monad_reorder]: 3.008e-05
[get_jit_bprop_graph]: 2.65002e-06
[rewriter_after_jit_bprop_graph]: 9.89001e-06
[opt_after_jit_grad]: 0.00068244
[symbol_engine_optimizer]: 0.00012398, [1]
    [Cycle 1]: 0.00011622, [6]
        [build]: 1.933e-05
        [elim_shapecalc]: 1.316e-05
        [elim_not_effective]: 2.346e-05
        [opt_reshape]: 1.062e-05
        [fold_const_symbol]: 1.478e-05
        [renormalize]: 1.36998e-06
[validate]: 7.631e-05
[backend_pass]: 1.55999e-06
[task_emit]: 0.0117801
[execute]: 9.52999e-06
Sums
    bootstrap                            :     0.000336s :  0.73%
    type_inference                       :     0.026973s : 58.85%
    event_method                         :     0.000023s :  0.05%
    auto_monad                           :     0.000169s :  0.37%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000141s :  0.31%
    rewriter_before_opt_a                :     0.000104s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000088s :  0.19%
    jit_opt_a.loop_unroll                :     0.000045s :  0.10%
    jit_opt_a.a_1                        :     0.001003s :  2.19%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  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 :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000012s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001499s :  3.27%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000038s :  0.08%
    jit_opt_a.cse                        :     0.000068s :  0.15%
    jit_opt_a.replace_applicator         :     0.000032s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.05%
    rewriter_after_opt_a                 :     0.000326s :  0.71%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000912s :  1.99%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000029s :  0.06%
    cconv                                :     0.000037s :  0.08%
    loop_unroll                          :     0.000541s :  1.18%
    jit_opt_after_cconv.c_1              :     0.000057s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000045s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000022s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000069s :  0.15%
    add_recomputation                    :     0.000080s :  0.17%
    cse_after_recomputation.cse          :     0.000022s :  0.05%
    auto_monad_reorder                   :     0.000030s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.02%
    opt_after_jit_grad                   :     0.000682s :  1.49%
    symbol_engine_optimizer.build        :     0.000019s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000011s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000076s :  0.17%
    backend_pass                         :     0.000002s :  0.00%
    task_emit                            :     0.011780s : 25.70%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000359    47
 3.14% :     0.000011s :      2: substitution.depend_value_elim
 0.86% :     0.000003s :      4: substitution.elim_not_effective
 0.70% :     0.000002s :      4: substitution.fold_const_symbol
 2.02% :     0.000007s :      5: substitution.graph_param_transform
76.36% :     0.000274s :      4: substitution.inline
 1.70% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.32% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.89% :     0.000007s :      2: substitution.replace_old_param
 4.11% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.63% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.27% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026888     2
93.80% :     0.025222s :      1: type_inference.infer
 6.20% :     0.001666s :      1: type_inference.specialize
------[replace.]   0.000103     6
66.11% :     0.000068s :      4: replace.inline
33.89% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000284     6
95.24% :     0.000271s :      4: match.inline
 4.76% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000193  1021
 1.20% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.51% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.08% :     0.000002s :     15: predicate.addn_check_dump
 1.58% :     0.000003s :     15: predicate.addn_zero_filter
 2.28% :     0.000004s :     15: predicate.arithmetic_simplify
 1.64% :     0.000003s :     15: predicate.cast_eliminate
 0.43% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.03% :     0.000002s :     15: predicate.compare_switch_simplify
 1.45% :     0.000003s :     15: predicate.depend_value_elim
 1.05% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.35% :     0.000003s :     15: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.04% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.27% :     0.000001s :      5: predicate.elim_not_effective
 0.81% :     0.000002s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.03% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     15: predicate.environ_get_eliminate
 1.00% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      5: predicate.fold_const_symbol
 1.11% :     0.000002s :     10: predicate.get_grad_eliminate
 0.26% :     0.000001s :      5: predicate.graph_param_transform
 5.78% :     0.000011s :     31: predicate.inline
 1.26% :     0.000002s :     10: predicate.inline_without_move
 0.40% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.30% :     0.000003s :     10: predicate.less_batch_normalization
 1.63% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.91% :     0.000004s :     22: predicate.load_eliminater
 1.52% :     0.000003s :      5: predicate.loop_unroll_after_grad
 2.97% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.10% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000002s :     15: predicate.merge_addn
 1.14% :     0.000002s :     15: predicate.minmaximum_grad
 2.37% :     0.000005s :      5: predicate.mutable_eliminate
 0.56% :     0.000001s :      5: predicate.opt_reshape
 2.20% :     0.000004s :     22: predicate.partial_eliminate
 1.33% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.74% :     0.000003s :     15: predicate.reduce_eliminate
 1.53% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.70% :     0.000003s :     27: predicate.replace_applicator
 0.61% :     0.000001s :     10: predicate.replace_old_param
 0.42% :     0.000001s :      5: predicate.reset_defer_inline
 1.22% :     0.000002s :     15: predicate.reshape_eliminate
 1.15% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.03% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     15: predicate.same_eliminate
 0.49% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :     10: predicate.special_op_eliminate
 0.97% :     0.000002s :     10: predicate.specialize_transform
 1.66% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.10% :     0.000004s :     21: predicate.switch_defer_inline
 2.04% :     0.000004s :     21: predicate.switch_layer_defer_inline
 6.22% :     0.000012s :     68: predicate.switch_simplify
 1.50% :     0.000003s :     15: predicate.tile_eliminate
 1.13% :     0.000002s :     15: predicate.transpose_eliminate
 1.36% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.74% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.17% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001095    13
42.55% :     0.000466s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.21% :     0.000068s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.24% :     0.000561s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.053990    76
 0.15% :     0.000083s :      1: add_recomputation
 0.32% :     0.000175s :      1: auto_monad
 0.06% :     0.000033s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.66% :     0.000358s :      1: bootstrap
 0.07% :     0.000040s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.07% :     0.000040s :      1: cse_after_recomputation
 0.13% :     0.000072s :      1: environ_conv
 0.05% :     0.000027s :      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% :     0.000010s :      1: graph_reusing
14.41% :     0.007782s :      1: jit_opt_a
 0.45% :     0.000243s :      1: jit_opt_after_cconv
 0.16% :     0.000086s :      1: jit_opt_b
 1.02% :     0.000553s :      1: loop_unroll
 1.72% :     0.000930s :      1: mutable_eliminate
 2.48% :     0.001339s :     26: opt.transform.jit_opt_a
 0.18% :     0.000097s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000049s :      4: opt.transform.jit_opt_b
 0.03% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000030s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000039s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000058s :      4: opt.transform.symbol_engine_opt
 1.29% :     0.000699s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000146s :      1: py_interpret_to_execute
 0.04% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000025s :      1: remove_dup_value
 1.72% :     0.000931s :      1: renormalize.infer
 1.03% :     0.000554s :      1: renormalize.specialize
 0.02% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.61% :     0.000331s :      1: rewriter_after_opt_a
 0.20% :     0.000108s :      1: rewriter_before_opt_a
 0.24% :     0.000127s :      1: symbol_engine_optimizer
21.86% :     0.011802s :      1: task_emit
49.99% :     0.026992s :      1: type_inference
 0.20% :     0.000110s :      1: validate


TotalTime = 0.0504272, [33]
[bootstrap]: 0.00032461
[type_inference]: 0.0266626
[event_method]: 2.437e-05
[auto_monad]: 0.00017521
[graph_reusing]: 6.48e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 0.00014806
[rewriter_before_opt_a]: 0.00010402
[expand_dump_flag]: 4.10998e-06
[jit_opt_a]: 0.00789183, [2]
    [Cycle 1]: 0.00299592, [27]
        [switch_simplify]: 7.885e-05
        [loop_unroll]: 3.749e-05
        [a_1]: 0.00084243
        [with_stream_mark]: 2.912e-05
        [recompute_prepare]: 1.355e-05
        [updatestate_depend_eliminate]: 7.24001e-06
        [updatestate_assign_eliminate]: 7.20998e-06
        [updatestate_loads_eliminate]: 4.67998e-06
        [parameter_eliminate]: 2.31e-06
        [specialize_transform]: 1.132e-05
        [updatestate_useless_node_eliminater]: 1.375e-05
        [accelerated_algorithm]: 9.05001e-06
        [meta_shard_fg_expand]: 2.71999e-06
        [get_grad_eliminate_]: 8.73001e-06
        [merge_forward]: 6.51e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.071e-05
        [j_node_and_user_rematch]: 1.513e-05
        [meta_fg_expand]: 3.88999e-06
        [replace_old_param]: 1.464e-05
        [inline_without_move]: 8.74e-06
        [renormalize]: 0.00150547
        [add_forward_monad_depend]: 1.028e-05
        [auto_monad_grad]: 2.81e-06
        [auto_monad_eliminator]: 2.966e-05
        [cse]: 4.562e-05
        [replace_applicator]: 2.489e-05
    [Cycle 2]: 0.0005266, [27]
        [switch_simplify]: 1.036e-05
        [loop_unroll]: 8.33999e-06
        [a_1]: 0.00017888
        [with_stream_mark]: 1.926e-05
        [recompute_prepare]: 8.64e-06
        [updatestate_depend_eliminate]: 6.73e-06
        [updatestate_assign_eliminate]: 5.41998e-06
        [updatestate_loads_eliminate]: 4.85999e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 9.17001e-06
        [updatestate_useless_node_eliminater]: 1.163e-05
        [accelerated_algorithm]: 8.79e-06
        [meta_shard_fg_expand]: 3.14001e-06
        [get_grad_eliminate_]: 7.56001e-06
        [merge_forward]: 6.36998e-06
        [cell_reuse_recompute_pass]: 3.44001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.968e-05
        [j_node_and_user_rematch]: 1.377e-05
        [meta_fg_expand]: 2.96999e-06
        [replace_old_param]: 1.175e-05
        [inline_without_move]: 7.41001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.91e-06
        [auto_monad_grad]: 1.83002e-06
        [auto_monad_eliminator]: 1.324e-05
        [cse]: 2.667e-05
        [replace_applicator]: 9.56e-06
[py_interpret_to_execute_after_opt_a]: 2.099e-05
[rewriter_after_opt_a]: 0.00031585
[convert_after_rewriter]: 1.358e-05
[order_py_execute_after_rewriter]: 7.63001e-06
[mutable_eliminate]: 0.00081402
[jit_opt_b]: 7.651e-05, [1]
    [Cycle 1]: 6.678e-05, [2]
        [frontend_op_eliminate]: 2.65e-05
        [inline_after_opt_a]: 2.732e-05
[cconv]: 3.371e-05
[loop_unroll]: 0.00048053
[jit_opt_after_cconv]: 0.00022843, [1]
    [Cycle 1]: 0.00022062, [11]
        [c_1]: 5.297e-05
        [parameter_eliminate]: 4.33999e-06
        [updatestate_depend_eliminate]: 1.14e-05
        [updatestate_assign_eliminate]: 5.07e-06
        [updatestate_loads_eliminate]: 4.23001e-06
        [cse]: 4.175e-05
        [call_graph_tuple_transform]: 2.537e-05
        [tuple_list_get_item_eliminator]: 8.56002e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 9.64999e-06
[remove_dup_value]: 1.957e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 1.318e-05
[add_recomputation]: 7.086e-05
[cse_after_recomputation]: 3.203e-05, [1]
    [Cycle 1]: 2.548e-05, [1]
        [cse]: 1.918e-05
[auto_monad_reorder]: 2.612e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 7.85998e-06
[opt_after_jit_grad]: 0.00050855
[symbol_engine_optimizer]: 0.00016384, [1]
    [Cycle 1]: 0.00015662, [6]
        [build]: 1.609e-05
        [elim_shapecalc]: 1.184e-05
        [elim_not_effective]: 7.232e-05
        [opt_reshape]: 9.66e-06
        [fold_const_symbol]: 1.48e-05
        [renormalize]: 5.39992e-07
[validate]: 6.846e-05
[backend_pass]: 1.10001e-06
[task_emit]: 0.0118978
[execute]: 1.046e-05
Sums
    bootstrap                            :     0.000325s :  0.72%
    type_inference                       :     0.026663s : 58.87%
    event_method                         :     0.000024s :  0.05%
    auto_monad                           :     0.000175s :  0.39%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000148s :  0.33%
    rewriter_before_opt_a                :     0.000104s :  0.23%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000089s :  0.20%
    jit_opt_a.loop_unroll                :     0.000046s :  0.10%
    jit_opt_a.a_1                        :     0.001021s :  2.25%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000020s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000013s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000026s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001506s :  3.32%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000043s :  0.09%
    jit_opt_a.cse                        :     0.000072s :  0.16%
    jit_opt_a.replace_applicator         :     0.000034s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.05%
    rewriter_after_opt_a                 :     0.000316s :  0.70%
    convert_after_rewriter               :     0.000014s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000814s :  1.80%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.06%
    cconv                                :     0.000034s :  0.07%
    loop_unroll                          :     0.000481s :  1.06%
    jit_opt_after_cconv.c_1              :     0.000053s :  0.12%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000042s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000020s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000071s :  0.16%
    cse_after_recomputation.cse          :     0.000019s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000509s :  1.12%
    symbol_engine_optimizer.build        :     0.000016s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000072s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000068s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011898s : 26.27%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000372    47
 3.04% :     0.000011s :      2: substitution.depend_value_elim
 0.74% :     0.000003s :      4: substitution.elim_not_effective
 0.64% :     0.000002s :      4: substitution.fold_const_symbol
 2.14% :     0.000008s :      5: substitution.graph_param_transform
77.15% :     0.000287s :      4: substitution.inline
 1.61% :     0.000006s :      8: substitution.j_node_and_user_rematch
 2.16% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.92% :     0.000007s :      2: substitution.replace_old_param
 3.95% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.37% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.29% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026574     2
93.78% :     0.024920s :      1: type_inference.infer
 6.22% :     0.001654s :      1: type_inference.specialize
------[replace.]   0.000112     6
63.41% :     0.000071s :      4: replace.inline
36.59% :     0.000041s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000297     6
95.44% :     0.000284s :      4: match.inline
 4.56% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000184  1021
 1.28% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.07% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.06% :     0.000002s :     15: predicate.addn_check_dump
 1.29% :     0.000002s :     15: predicate.addn_zero_filter
 2.44% :     0.000004s :     15: predicate.arithmetic_simplify
 1.30% :     0.000002s :     15: predicate.cast_eliminate
 0.48% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.03% :     0.000002s :     15: predicate.compare_switch_simplify
 1.48% :     0.000003s :     15: predicate.depend_value_elim
 1.04% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.99% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.39% :     0.000003s :     15: predicate.environ_get_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 0.95% :     0.000002s :     10: predicate.get_grad_eliminate
 0.34% :     0.000001s :      5: predicate.graph_param_transform
 5.53% :     0.000010s :     31: predicate.inline
 0.91% :     0.000002s :     10: predicate.inline_without_move
 0.50% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.54% :     0.000003s :     10: predicate.less_batch_normalization
 1.59% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 2.46% :     0.000005s :     22: predicate.load_eliminater
 1.58% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.20% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.36% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000002s :     15: predicate.merge_addn
 1.09% :     0.000002s :     15: predicate.minmaximum_grad
 1.49% :     0.000003s :      5: predicate.mutable_eliminate
 0.51% :     0.000001s :      5: predicate.opt_reshape
 2.15% :     0.000004s :     22: predicate.partial_eliminate
 1.41% :     0.000003s :     15: predicate.print_const_string_wrapper
 1.52% :     0.000003s :     15: predicate.reduce_eliminate
 1.45% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.66% :     0.000003s :     27: predicate.replace_applicator
 0.63% :     0.000001s :     10: predicate.replace_old_param
 0.30% :     0.000001s :      5: predicate.reset_defer_inline
 1.20% :     0.000002s :     15: predicate.reshape_eliminate
 1.17% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.18% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.07% :     0.000002s :     10: predicate.special_op_eliminate
 1.07% :     0.000002s :     10: predicate.specialize_transform
 1.48% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.60% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.02% :     0.000004s :     21: predicate.switch_defer_inline
 1.88% :     0.000003s :     21: predicate.switch_layer_defer_inline
 7.10% :     0.000013s :     68: predicate.switch_simplify
 1.21% :     0.000002s :     15: predicate.tile_eliminate
 1.46% :     0.000003s :     15: predicate.transpose_eliminate
 1.39% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.72% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000965    13
44.93% :     0.000434s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.04% :     0.000068s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.03% :     0.000464s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.053547    76
 0.14% :     0.000074s :      1: add_recomputation
 0.34% :     0.000181s :      1: auto_monad
 0.05% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.64% :     0.000345s :      1: bootstrap
 0.07% :     0.000037s :      1: cconv
 0.03% :     0.000017s :      1: convert_after_rewriter
 0.06% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.05% :     0.000029s :      1: event_method
 0.03% :     0.000016s :      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
14.75% :     0.007896s :      1: jit_opt_a
 0.43% :     0.000232s :      1: jit_opt_after_cconv
 0.15% :     0.000079s :      1: jit_opt_b
 0.92% :     0.000490s :      1: loop_unroll
 1.54% :     0.000827s :      1: mutable_eliminate
 2.56% :     0.001369s :     26: opt.transform.jit_opt_a
 0.17% :     0.000092s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000046s :      4: opt.transform.jit_opt_b
 0.03% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000105s :      4: opt.transform.symbol_engine_opt
 0.97% :     0.000518s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000153s :      1: py_interpret_to_execute
 0.05% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000022s :      1: remove_dup_value
 1.59% :     0.000850s :      1: renormalize.infer
 1.20% :     0.000641s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.60% :     0.000323s :      1: rewriter_after_opt_a
 0.20% :     0.000108s :      1: rewriter_before_opt_a
 0.31% :     0.000167s :      1: symbol_engine_optimizer
22.26% :     0.011919s :      1: task_emit
49.83% :     0.026682s :      1: type_inference
 0.18% :     0.000099s :      1: validate


TotalTime = 0.0461736, [33]
[bootstrap]: 0.00034019
[type_inference]: 0.0242147
[event_method]: 2.115e-05
[auto_monad]: 0.00016666
[graph_reusing]: 6.12001e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 0.0001293
[rewriter_before_opt_a]: 9.673e-05
[expand_dump_flag]: 3.40998e-06
[jit_opt_a]: 0.00736939, [2]
    [Cycle 1]: 0.00276332, [27]
        [switch_simplify]: 7.231e-05
        [loop_unroll]: 3.6e-05
        [a_1]: 0.00079586
        [with_stream_mark]: 2.871e-05
        [recompute_prepare]: 1.145e-05
        [updatestate_depend_eliminate]: 7.22002e-06
        [updatestate_assign_eliminate]: 5.72999e-06
        [updatestate_loads_eliminate]: 4.87e-06
        [parameter_eliminate]: 1.67001e-06
        [specialize_transform]: 9.54999e-06
        [updatestate_useless_node_eliminater]: 1.22e-05
        [accelerated_algorithm]: 2.579e-05
        [meta_shard_fg_expand]: 2.94999e-06
        [get_grad_eliminate_]: 8.70999e-06
        [merge_forward]: 5.56e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 8.065e-05
        [j_node_and_user_rematch]: 1.657e-05
        [meta_fg_expand]: 4.03999e-06
        [replace_old_param]: 1.38e-05
        [inline_without_move]: 8.89e-06
        [renormalize]: 0.00129935
        [add_forward_monad_depend]: 8.15999e-06
        [auto_monad_grad]: 2.81e-06
        [auto_monad_eliminator]: 2.393e-05
        [cse]: 4.184e-05
        [replace_applicator]: 2.27e-05
    [Cycle 2]: 0.00049606, [27]
        [switch_simplify]: 9.39e-06
        [loop_unroll]: 8.47e-06
        [a_1]: 0.00017826
        [with_stream_mark]: 1.47e-05
        [recompute_prepare]: 9.15999e-06
        [updatestate_depend_eliminate]: 5.75001e-06
        [updatestate_assign_eliminate]: 6.02001e-06
        [updatestate_loads_eliminate]: 4.45999e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 7.95e-06
        [updatestate_useless_node_eliminater]: 1.062e-05
        [accelerated_algorithm]: 8.1e-06
        [meta_shard_fg_expand]: 2.24999e-06
        [get_grad_eliminate_]: 7.28e-06
        [merge_forward]: 4.67e-06
        [cell_reuse_recompute_pass]: 3.76001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.848e-05
        [j_node_and_user_rematch]: 1.34e-05
        [meta_fg_expand]: 2.90002e-06
        [replace_old_param]: 1.117e-05
        [inline_without_move]: 7.35e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 1.074e-05
        [cse]: 2.153e-05
        [replace_applicator]: 8.48001e-06
[py_interpret_to_execute_after_opt_a]: 1.924e-05
[rewriter_after_opt_a]: 0.00031172
[convert_after_rewriter]: 1.257e-05
[order_py_execute_after_rewriter]: 7.98999e-06
[mutable_eliminate]: 0.00082251
[jit_opt_b]: 7.382e-05, [1]
    [Cycle 1]: 6.553e-05, [2]
        [frontend_op_eliminate]: 2.69e-05
        [inline_after_opt_a]: 2.58e-05
[cconv]: 3.215e-05
[loop_unroll]: 0.00048527
[jit_opt_after_cconv]: 0.00024159, [1]
    [Cycle 1]: 0.00023361, [11]
        [c_1]: 5.936e-05
        [parameter_eliminate]: 3.45e-06
        [updatestate_depend_eliminate]: 9.54e-06
        [updatestate_assign_eliminate]: 5.57999e-06
        [updatestate_loads_eliminate]: 4.77e-06
        [cse]: 4.502e-05
        [call_graph_tuple_transform]: 2.622e-05
        [tuple_list_get_item_eliminator]: 8.43001e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 1.002e-05
[remove_dup_value]: 1.941e-05
[partial_unused_args_eliminate]: 2.11003e-06
[environ_conv]: 1.31e-05
[add_recomputation]: 7.504e-05
[cse_after_recomputation]: 3.311e-05, [1]
    [Cycle 1]: 2.637e-05, [1]
        [cse]: 1.973e-05
[auto_monad_reorder]: 2.714e-05
[get_jit_bprop_graph]: 2.36998e-06
[rewriter_after_jit_bprop_graph]: 8.84e-06
[opt_after_jit_grad]: 0.00054382
[symbol_engine_optimizer]: 0.00010802, [1]
    [Cycle 1]: 0.00010099, [6]
        [build]: 1.283e-05
        [elim_shapecalc]: 1.242e-05
        [elim_not_effective]: 1.997e-05
        [opt_reshape]: 9.14e-06
        [fold_const_symbol]: 1.532e-05
        [renormalize]: 5.8001e-07
[validate]: 6.068e-05
[backend_pass]: 1.04998e-06
[task_emit]: 0.0106247
[execute]: 8.35001e-06
Sums
    bootstrap                            :     0.000340s :  0.82%
    type_inference                       :     0.024215s : 58.62%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000167s :  0.40%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000129s :  0.31%
    rewriter_before_opt_a                :     0.000097s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000082s :  0.20%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000974s :  2.36%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000034s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000099s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001299s :  3.15%
    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      :     0.000035s :  0.08%
    jit_opt_a.cse                        :     0.000063s :  0.15%
    jit_opt_a.replace_applicator         :     0.000031s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.05%
    rewriter_after_opt_a                 :     0.000312s :  0.75%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000823s :  1.99%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000032s :  0.08%
    loop_unroll                          :     0.000485s :  1.17%
    jit_opt_after_cconv.c_1              :     0.000059s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000045s :  0.11%
    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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000075s :  0.18%
    cse_after_recomputation.cse          :     0.000020s :  0.05%
    auto_monad_reorder                   :     0.000027s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000544s :  1.32%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000061s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010625s : 25.72%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000401    47
 2.78% :     0.000011s :      2: substitution.depend_value_elim
 0.73% :     0.000003s :      4: substitution.elim_not_effective
 0.58% :     0.000002s :      4: substitution.fold_const_symbol
 1.87% :     0.000008s :      5: substitution.graph_param_transform
65.60% :     0.000263s :      4: substitution.inline
 1.34% :     0.000005s :      8: substitution.j_node_and_user_rematch
15.89% :     0.000064s :      8: substitution.remove_not_recompute_node
 1.69% :     0.000007s :      2: substitution.replace_old_param
 3.29% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.20% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.04% :     0.000012s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024133     2
93.69% :     0.022611s :      1: type_inference.infer
 6.31% :     0.001522s :      1: type_inference.specialize
------[replace.]   0.000110     6
67.71% :     0.000075s :      4: replace.inline
32.29% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000272     6
95.60% :     0.000260s :      4: match.inline
 4.40% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1021
 1.60% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.15% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.24% :     0.000002s :     15: predicate.addn_zero_filter
 2.50% :     0.000004s :     15: predicate.arithmetic_simplify
 1.48% :     0.000003s :     15: predicate.cast_eliminate
 0.53% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     15: predicate.compare_switch_simplify
 1.30% :     0.000002s :     15: predicate.depend_value_elim
 1.29% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.13% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.89% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.60% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.06% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     15: predicate.environ_get_eliminate
 1.10% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      5: predicate.fold_const_symbol
 0.99% :     0.000002s :     10: predicate.get_grad_eliminate
 0.39% :     0.000001s :      5: predicate.graph_param_transform
 5.18% :     0.000009s :     31: predicate.inline
 1.08% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.46% :     0.000003s :     10: predicate.less_batch_normalization
 1.41% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.94% :     0.000003s :     22: predicate.load_eliminater
 1.30% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.30% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.97% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     15: predicate.merge_addn
 1.06% :     0.000002s :     15: predicate.minmaximum_grad
 1.68% :     0.000003s :      5: predicate.mutable_eliminate
 0.50% :     0.000001s :      5: predicate.opt_reshape
 2.42% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.68% :     0.000003s :     15: predicate.reduce_eliminate
 1.40% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.68% :     0.000003s :     27: predicate.replace_applicator
 0.70% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.10% :     0.000002s :     15: predicate.reshape_eliminate
 1.27% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.98% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     15: predicate.same_eliminate
 0.65% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.08% :     0.000002s :     10: predicate.special_op_eliminate
 1.01% :     0.000002s :     10: predicate.specialize_transform
 1.45% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.18% :     0.000004s :     21: predicate.switch_defer_inline
 1.86% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.72% :     0.000012s :     68: predicate.switch_simplify
 1.28% :     0.000002s :     15: predicate.tile_eliminate
 1.26% :     0.000002s :     15: predicate.transpose_eliminate
 1.37% :     0.000002s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.39% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 2.25% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.33% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000873    13
42.97% :     0.000375s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.58% :     0.000066s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.45% :     0.000432s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049053    76
 0.16% :     0.000078s :      1: add_recomputation
 0.35% :     0.000172s :      1: auto_monad
 0.06% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.73% :     0.000358s :      1: bootstrap
 0.07% :     0.000035s :      1: cconv
 0.03% :     0.000016s :      1: convert_after_rewriter
 0.07% :     0.000035s :      1: cse_after_recomputation
 0.03% :     0.000016s :      1: environ_conv
 0.05% :     0.000025s :      1: event_method
 0.03% :     0.000013s :      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
15.03% :     0.007373s :      1: jit_opt_a
 0.50% :     0.000245s :      1: jit_opt_after_cconv
 0.16% :     0.000077s :      1: jit_opt_b
 1.10% :     0.000539s :      1: loop_unroll
 1.70% :     0.000833s :      1: mutable_eliminate
 2.81% :     0.001378s :     26: opt.transform.jit_opt_a
 0.20% :     0.000099s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000045s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.13% :     0.000554s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000134s :      1: py_interpret_to_execute
 0.04% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000022s :      1: remove_dup_value
 1.64% :     0.000806s :      1: renormalize.infer
 0.98% :     0.000482s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.65% :     0.000317s :      1: rewriter_after_opt_a
 0.21% :     0.000101s :      1: rewriter_before_opt_a
 0.23% :     0.000111s :      1: symbol_engine_optimizer
21.69% :     0.010640s :      1: task_emit
49.40% :     0.024232s :      1: type_inference
 0.18% :     0.000086s :      1: validate


TotalTime = 0.061466, [33]
[bootstrap]: 0.00036237
[type_inference]: 0.025601
[event_method]: 2.232e-05
[auto_monad]: 0.0001894
[graph_reusing]: 7.03e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 0.00014585
[rewriter_before_opt_a]: 0.00010603
[expand_dump_flag]: 3.98999e-06
[jit_opt_a]: 0.00764688, [2]
    [Cycle 1]: 0.0027729, [27]
        [switch_simplify]: 7.828e-05
        [loop_unroll]: 3.697e-05
        [a_1]: 0.00082432
        [with_stream_mark]: 3.017e-05
        [recompute_prepare]: 1.33e-05
        [updatestate_depend_eliminate]: 7.23999e-06
        [updatestate_assign_eliminate]: 6.44999e-06
        [updatestate_loads_eliminate]: 4.93001e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 1.067e-05
        [updatestate_useless_node_eliminater]: 1.189e-05
        [accelerated_algorithm]: 9.36e-06
        [meta_shard_fg_expand]: 3.28e-06
        [get_grad_eliminate_]: 8.47e-06
        [merge_forward]: 6.11e-06
        [cell_reuse_recompute_pass]: 1.46998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.175e-05
        [j_node_and_user_rematch]: 1.445e-05
        [meta_fg_expand]: 3.94002e-06
        [replace_old_param]: 1.38e-05
        [inline_without_move]: 1.016e-05
        [renormalize]: 0.00133503
        [add_forward_monad_depend]: 7.8e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 2.371e-05
        [cse]: 4.475e-05
        [replace_applicator]: 2.23e-05
    [Cycle 2]: 0.00054231, [27]
        [switch_simplify]: 9.10001e-06
        [loop_unroll]: 8.33999e-06
        [a_1]: 0.00017747
        [with_stream_mark]: 1.666e-05
        [recompute_prepare]: 8.25e-06
        [updatestate_depend_eliminate]: 5.57001e-06
        [updatestate_assign_eliminate]: 5.20999e-06
        [updatestate_loads_eliminate]: 4.08001e-06
        [parameter_eliminate]: 1.48002e-06
        [specialize_transform]: 8.34002e-06
        [updatestate_useless_node_eliminater]: 1.192e-05
        [accelerated_algorithm]: 7.91001e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 8.12e-06
        [merge_forward]: 4.89e-06
        [cell_reuse_recompute_pass]: 4.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.911e-05
        [j_node_and_user_rematch]: 1.298e-05
        [meta_fg_expand]: 3.08e-06
        [replace_old_param]: 1.057e-05
        [inline_without_move]: 7.4e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 1.79e-06
        [auto_monad_eliminator]: 1.074e-05
        [cse]: 6.035e-05
        [replace_applicator]: 9.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.867e-05
[rewriter_after_opt_a]: 0.00031234
[convert_after_rewriter]: 1.249e-05
[order_py_execute_after_rewriter]: 6.86999e-06
[mutable_eliminate]: 0.00083756
[jit_opt_b]: 7.489e-05, [1]
    [Cycle 1]: 6.66e-05, [2]
        [frontend_op_eliminate]: 2.571e-05
        [inline_after_opt_a]: 2.655e-05
[cconv]: 3.355e-05
[loop_unroll]: 0.00047482
[jit_opt_after_cconv]: 0.00022345, [1]
    [Cycle 1]: 0.00021559, [11]
        [c_1]: 5.125e-05
        [parameter_eliminate]: 3.03e-06
        [updatestate_depend_eliminate]: 1.078e-05
        [updatestate_assign_eliminate]: 5.83997e-06
        [updatestate_loads_eliminate]: 4.30999e-06
        [cse]: 4.127e-05
        [call_graph_tuple_transform]: 2.34e-05
        [tuple_list_get_item_eliminator]: 8.69e-06
        [none_parameter_eliminate]: 1.97001e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 8.15e-06
[remove_dup_value]: 2.094e-05
[partial_unused_args_eliminate]: 3.13998e-06
[environ_conv]: 1.381e-05
[add_recomputation]: 6.8e-05
[cse_after_recomputation]: 3.158e-05, [1]
    [Cycle 1]: 2.501e-05, [1]
        [cse]: 1.854e-05
[auto_monad_reorder]: 2.802e-05
[get_jit_bprop_graph]: 2.67001e-06
[rewriter_after_jit_bprop_graph]: 7.5e-06
[opt_after_jit_grad]: 0.00052753
[symbol_engine_optimizer]: 0.00011119, [1]
    [Cycle 1]: 0.00010418, [6]
        [build]: 1.746e-05
        [elim_shapecalc]: 1.236e-05
        [elim_not_effective]: 2.014e-05
        [opt_reshape]: 9.51998e-06
        [fold_const_symbol]: 1.386e-05
        [renormalize]: 6.30011e-07
[validate]: 6.761e-05
[backend_pass]: 1.22e-06
[task_emit]: 0.0242288
[execute]: 8.89998e-06
Sums
    bootstrap                            :     0.000362s :  0.64%
    type_inference                       :     0.025601s : 45.40%
    event_method                         :     0.000022s :  0.04%
    auto_monad                           :     0.000189s :  0.34%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000146s :  0.26%
    rewriter_before_opt_a                :     0.000106s :  0.19%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000087s :  0.15%
    jit_opt_a.loop_unroll                :     0.000045s :  0.08%
    jit_opt_a.a_1                        :     0.001002s :  1.78%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000019s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.03%
    jit_opt_a.merge_forward              :     0.000011s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.04%
    jit_opt_a.inline_without_move        :     0.000018s :  0.03%
    jit_opt_a.renormalize                :     0.001335s :  2.37%
    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      :     0.000034s :  0.06%
    jit_opt_a.cse                        :     0.000105s :  0.19%
    jit_opt_a.replace_applicator         :     0.000031s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.03%
    rewriter_after_opt_a                 :     0.000312s :  0.55%
    convert_after_rewriter               :     0.000012s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000838s :  1.49%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.05%
    cconv                                :     0.000034s :  0.06%
    loop_unroll                          :     0.000475s :  0.84%
    jit_opt_after_cconv.c_1              :     0.000051s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000041s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000008s :  0.01%
    remove_dup_value                     :     0.000021s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000014s :  0.02%
    add_recomputation                    :     0.000068s :  0.12%
    cse_after_recomputation.cse          :     0.000019s :  0.03%
    auto_monad_reorder                   :     0.000028s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000528s :  0.94%
    symbol_engine_optimizer.build        :     0.000017s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000068s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.024229s : 42.96%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000358    47
 3.05% :     0.000011s :      2: substitution.depend_value_elim
 0.92% :     0.000003s :      4: substitution.elim_not_effective
 0.61% :     0.000002s :      4: substitution.fold_const_symbol
 2.02% :     0.000007s :      5: substitution.graph_param_transform
77.86% :     0.000279s :      4: substitution.inline
 1.37% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.19% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.72% :     0.000006s :      2: substitution.replace_old_param
 3.64% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.60% :     0.000013s :      3: substitution.updatestate_pure_node_eliminater
 3.01% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.025516     2
93.72% :     0.023914s :      1: type_inference.infer
 6.28% :     0.001602s :      1: type_inference.specialize
------[replace.]   0.000102     6
67.39% :     0.000069s :      4: replace.inline
32.61% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000288     6
95.87% :     0.000276s :      4: match.inline
 4.13% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000182  1021
 1.33% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.00% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     15: predicate.addn_check_dump
 1.23% :     0.000002s :     15: predicate.addn_zero_filter
 2.18% :     0.000004s :     15: predicate.arithmetic_simplify
 1.43% :     0.000003s :     15: predicate.cast_eliminate
 0.46% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     15: predicate.compare_switch_simplify
 1.30% :     0.000002s :     15: predicate.depend_value_elim
 1.09% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.12% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.30% :     0.000001s :      5: predicate.elim_not_effective
 0.58% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     15: predicate.environ_get_eliminate
 1.41% :     0.000003s :     15: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      5: predicate.fold_const_symbol
 0.95% :     0.000002s :     10: predicate.get_grad_eliminate
 0.36% :     0.000001s :      5: predicate.graph_param_transform
 5.55% :     0.000010s :     31: predicate.inline
 1.00% :     0.000002s :     10: predicate.inline_without_move
 0.39% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.51% :     0.000003s :     10: predicate.less_batch_normalization
 1.49% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.95% :     0.000004s :     22: predicate.load_eliminater
 1.55% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.17% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.80% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     15: predicate.merge_addn
 1.11% :     0.000002s :     15: predicate.minmaximum_grad
 2.25% :     0.000004s :      5: predicate.mutable_eliminate
 0.53% :     0.000001s :      5: predicate.opt_reshape
 2.37% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.78% :     0.000003s :     15: predicate.reduce_eliminate
 1.37% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.76% :     0.000003s :     27: predicate.replace_applicator
 0.54% :     0.000001s :     10: predicate.replace_old_param
 0.41% :     0.000001s :      5: predicate.reset_defer_inline
 1.22% :     0.000002s :     15: predicate.reshape_eliminate
 1.23% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.05% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     15: predicate.same_eliminate
 0.57% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.29% :     0.000002s :     10: predicate.special_op_eliminate
 1.04% :     0.000002s :     10: predicate.specialize_transform
 1.40% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.14% :     0.000004s :     21: predicate.switch_defer_inline
 1.81% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.53% :     0.000012s :     68: predicate.switch_simplify
 1.35% :     0.000002s :     15: predicate.tile_eliminate
 1.18% :     0.000002s :     15: predicate.transpose_eliminate
 1.51% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.55% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.96% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.94% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.52% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000003s :     15: predicate.value_based_eliminate
 0.42% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000936    13
44.95% :     0.000421s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 7.28% :     0.000068s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.77% :     0.000447s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.064330    76
 0.11% :     0.000071s :      1: add_recomputation
 0.31% :     0.000196s :      1: auto_monad
 0.05% :     0.000031s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.59% :     0.000381s :      1: bootstrap
 0.06% :     0.000036s :      1: cconv
 0.02% :     0.000016s :      1: convert_after_rewriter
 0.05% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000016s :      1: environ_conv
 0.04% :     0.000029s :      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
 0.02% :     0.000010s :      1: graph_reusing
11.89% :     0.007651s :      1: jit_opt_a
 0.35% :     0.000226s :      1: jit_opt_after_cconv
 0.12% :     0.000078s :      1: jit_opt_b
 0.75% :     0.000484s :      1: loop_unroll
 1.32% :     0.000850s :      1: mutable_eliminate
 2.08% :     0.001337s :     26: opt.transform.jit_opt_a
 0.14% :     0.000087s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000045s :      4: opt.transform.jit_opt_b
 0.03% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000052s :      4: opt.transform.symbol_engine_opt
 0.84% :     0.000538s :      1: opt_after_jit_grad
 0.01% :     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
 0.23% :     0.000150s :      1: py_interpret_to_execute
 0.03% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000023s :      1: remove_dup_value
 1.29% :     0.000830s :      1: renormalize.infer
 0.77% :     0.000494s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.50% :     0.000319s :      1: rewriter_after_opt_a
 0.17% :     0.000111s :      1: rewriter_before_opt_a
 0.18% :     0.000114s :      1: symbol_engine_optimizer
37.69% :     0.024248s :      1: task_emit
39.83% :     0.025622s :      1: type_inference
 0.15% :     0.000096s :      1: validate


TotalTime = 0.0483766, [33]
[bootstrap]: 0.00035181
[type_inference]: 0.0241936
[event_method]: 2.389e-05
[auto_monad]: 0.00017236
[graph_reusing]: 6.61999e-06
[pre_auto_parallel]: 3.66001e-06
[py_interpret_to_execute]: 0.00013596
[rewriter_before_opt_a]: 0.00010185
[expand_dump_flag]: 4.05998e-06
[jit_opt_a]: 0.00704402, [2]
    [Cycle 1]: 0.00263969, [27]
        [switch_simplify]: 6.921e-05
        [loop_unroll]: 3.574e-05
        [a_1]: 0.0007702
        [with_stream_mark]: 2.503e-05
        [recompute_prepare]: 1.103e-05
        [updatestate_depend_eliminate]: 6.44999e-06
        [updatestate_assign_eliminate]: 5.91e-06
        [updatestate_loads_eliminate]: 4.90001e-06
        [parameter_eliminate]: 2.21e-06
        [specialize_transform]: 9.17001e-06
        [updatestate_useless_node_eliminater]: 1.265e-05
        [accelerated_algorithm]: 8.95001e-06
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 8.27e-06
        [merge_forward]: 5.35999e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.052e-05
        [j_node_and_user_rematch]: 1.456e-05
        [meta_fg_expand]: 3.83001e-06
        [replace_old_param]: 1.312e-05
        [inline_without_move]: 8.35001e-06
        [renormalize]: 0.00129993
        [add_forward_monad_depend]: 6.67002e-06
        [auto_monad_grad]: 3.46001e-06
        [auto_monad_eliminator]: 2.299e-05
        [cse]: 4.467e-05
        [replace_applicator]: 1.971e-05
    [Cycle 2]: 0.00047985, [27]
        [switch_simplify]: 8.92e-06
        [loop_unroll]: 8.02e-06
        [a_1]: 0.00017272
        [with_stream_mark]: 1.474e-05
        [recompute_prepare]: 8.02e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 5.12e-06
        [updatestate_loads_eliminate]: 4e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 7.98001e-06
        [updatestate_useless_node_eliminater]: 1.054e-05
        [accelerated_algorithm]: 8.18999e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 7.65998e-06
        [merge_forward]: 4.47998e-06
        [cell_reuse_recompute_pass]: 2.59001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.687e-05
        [j_node_and_user_rematch]: 1.291e-05
        [meta_fg_expand]: 2.83998e-06
        [replace_old_param]: 1.069e-05
        [inline_without_move]: 8.11002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 1.019e-05
        [cse]: 2.026e-05
        [replace_applicator]: 8.2e-06
[py_interpret_to_execute_after_opt_a]: 1.494e-05
[rewriter_after_opt_a]: 0.00030157
[convert_after_rewriter]: 1.164e-05
[order_py_execute_after_rewriter]: 6.58998e-06
[mutable_eliminate]: 0.00079196
[jit_opt_b]: 7.222e-05, [1]
    [Cycle 1]: 6.38e-05, [2]
        [frontend_op_eliminate]: 2.434e-05
        [inline_after_opt_a]: 2.64e-05
[cconv]: 2.964e-05
[loop_unroll]: 0.00048168
[jit_opt_after_cconv]: 0.00021834, [1]
    [Cycle 1]: 0.00021109, [11]
        [c_1]: 5.064e-05
        [parameter_eliminate]: 3.30998e-06
        [updatestate_depend_eliminate]: 9.62999e-06
        [updatestate_assign_eliminate]: 4.73001e-06
        [updatestate_loads_eliminate]: 4.02e-06
        [cse]: 3.96e-05
        [call_graph_tuple_transform]: 2.423e-05
        [tuple_list_get_item_eliminator]: 8.90001e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 9.20001e-07
        [switch_simplify]: 8.57e-06
[remove_dup_value]: 2.317e-05
[partial_unused_args_eliminate]: 1.98002e-06
[environ_conv]: 1.357e-05
[add_recomputation]: 6.992e-05
[cse_after_recomputation]: 3.072e-05, [1]
    [Cycle 1]: 2.463e-05, [1]
        [cse]: 1.86e-05
[auto_monad_reorder]: 2.716e-05
[get_jit_bprop_graph]: 2.34001e-06
[rewriter_after_jit_bprop_graph]: 7.73001e-06
[opt_after_jit_grad]: 0.00054562
[symbol_engine_optimizer]: 0.00011512, [1]
    [Cycle 1]: 0.00010791, [6]
        [build]: 1.829e-05
        [elim_shapecalc]: 1.293e-05
        [elim_not_effective]: 2.12e-05
        [opt_reshape]: 8.84e-06
        [fold_const_symbol]: 1.421e-05
        [renormalize]: 1.00999e-06
[validate]: 6.775e-05
[backend_pass]: 1.03001e-06
[task_emit]: 0.0132458
[execute]: 8.92999e-06
Sums
    bootstrap                            :     0.000352s :  0.80%
    type_inference                       :     0.024194s : 55.30%
    event_method                         :     0.000024s :  0.05%
    auto_monad                           :     0.000172s :  0.39%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000136s :  0.31%
    rewriter_before_opt_a                :     0.000102s :  0.23%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000078s :  0.18%
    jit_opt_a.loop_unroll                :     0.000044s :  0.10%
    jit_opt_a.a_1                        :     0.000943s :  2.16%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000024s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001300s :  2.97%
    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      :     0.000033s :  0.08%
    jit_opt_a.cse                        :     0.000065s :  0.15%
    jit_opt_a.replace_applicator         :     0.000028s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.03%
    rewriter_after_opt_a                 :     0.000302s :  0.69%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000792s :  1.81%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.06%
    cconv                                :     0.000030s :  0.07%
    loop_unroll                          :     0.000482s :  1.10%
    jit_opt_after_cconv.c_1              :     0.000051s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  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              :     0.000040s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000023s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000014s :  0.03%
    add_recomputation                    :     0.000070s :  0.16%
    cse_after_recomputation.cse          :     0.000019s :  0.04%
    auto_monad_reorder                   :     0.000027s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000546s :  1.25%
    symbol_engine_optimizer.build        :     0.000018s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000068s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.013246s : 30.28%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000330    47
 2.88% :     0.000009s :      2: substitution.depend_value_elim
 0.90% :     0.000003s :      4: substitution.elim_not_effective
 0.66% :     0.000002s :      4: substitution.fold_const_symbol
 2.45% :     0.000008s :      5: substitution.graph_param_transform
77.46% :     0.000256s :      4: substitution.inline
 1.47% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.23% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.51% :     0.000005s :      2: substitution.replace_old_param
 4.42% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.01% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 3.00% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024109     2
93.02% :     0.022425s :      1: type_inference.infer
 6.98% :     0.001683s :      1: type_inference.specialize
------[replace.]   0.000097     6
66.28% :     0.000064s :      4: replace.inline
33.72% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     6
94.96% :     0.000253s :      4: match.inline
 5.04% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000176  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     15: predicate.addn_check_dump
 1.31% :     0.000002s :     15: predicate.addn_zero_filter
 1.89% :     0.000003s :     15: predicate.arithmetic_simplify
 1.40% :     0.000002s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.23% :     0.000002s :     15: predicate.compare_switch_simplify
 1.34% :     0.000002s :     15: predicate.depend_value_elim
 1.12% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.05% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      5: predicate.elim_not_effective
 0.70% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000003s :     15: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.51% :     0.000003s :     15: predicate.environ_get_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.03% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 5.88% :     0.000010s :     31: predicate.inline
 0.99% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :     10: predicate.less_batch_normalization
 1.56% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.90% :     0.000003s :     22: predicate.load_eliminater
 1.42% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.32% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.05% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000002s :     15: predicate.merge_addn
 1.10% :     0.000002s :     15: predicate.minmaximum_grad
 1.78% :     0.000003s :      5: predicate.mutable_eliminate
 0.53% :     0.000001s :      5: predicate.opt_reshape
 2.19% :     0.000004s :     22: predicate.partial_eliminate
 1.19% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.74% :     0.000003s :     15: predicate.reduce_eliminate
 1.41% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.64% :     0.000003s :     27: predicate.replace_applicator
 0.65% :     0.000001s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.16% :     0.000002s :     15: predicate.reshape_eliminate
 1.47% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     15: predicate.same_eliminate
 0.56% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.13% :     0.000002s :     10: predicate.special_op_eliminate
 1.05% :     0.000002s :     10: predicate.specialize_transform
 1.38% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.15% :     0.000004s :     21: predicate.switch_defer_inline
 1.85% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.63% :     0.000012s :     68: predicate.switch_simplify
 1.20% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000002s :     15: predicate.transpose_eliminate
 1.44% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.29% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.74% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.37% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000003s :     15: predicate.value_based_eliminate
 0.39% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000992    13
39.54% :     0.000392s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.50% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
53.96% :     0.000535s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.051132    76
 0.14% :     0.000073s :      1: add_recomputation
 0.35% :     0.000178s :      1: auto_monad
 0.06% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.72% :     0.000367s :      1: bootstrap
 0.06% :     0.000032s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.06% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000016s :      1: environ_conv
 0.06% :     0.000029s :      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% :     0.000009s :      1: graph_reusing
13.78% :     0.007048s :      1: jit_opt_a
 0.43% :     0.000221s :      1: jit_opt_after_cconv
 0.15% :     0.000075s :      1: jit_opt_b
 0.96% :     0.000492s :      1: loop_unroll
 1.57% :     0.000803s :      1: mutable_eliminate
 2.46% :     0.001258s :     26: opt.transform.jit_opt_a
 0.17% :     0.000088s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.09% :     0.000556s :      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% :     0.000006s :      1: pre_auto_parallel
 0.28% :     0.000141s :      1: py_interpret_to_execute
 0.03% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000026s :      1: remove_dup_value
 1.54% :     0.000789s :      1: renormalize.infer
 0.98% :     0.000500s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.60% :     0.000306s :      1: rewriter_after_opt_a
 0.21% :     0.000106s :      1: rewriter_before_opt_a
 0.23% :     0.000118s :      1: symbol_engine_optimizer
25.94% :     0.013264s :      1: task_emit
47.35% :     0.024213s :      1: type_inference
 0.19% :     0.000096s :      1: validate


TotalTime = 0.0514422, [33]
[bootstrap]: 0.00034987
[type_inference]: 0.0260054
[event_method]: 2.56e-05
[auto_monad]: 0.00017845
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 3.14999e-06
[py_interpret_to_execute]: 0.00014223
[rewriter_before_opt_a]: 0.00010263
[expand_dump_flag]: 3.46001e-06
[jit_opt_a]: 0.00827074, [2]
    [Cycle 1]: 0.00307144, [27]
        [switch_simplify]: 7.874e-05
        [loop_unroll]: 4.113e-05
        [a_1]: 0.0008463
        [with_stream_mark]: 2.761e-05
        [recompute_prepare]: 1.263e-05
        [updatestate_depend_eliminate]: 7.01999e-06
        [updatestate_assign_eliminate]: 5.70001e-06
        [updatestate_loads_eliminate]: 5.38002e-06
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 1.12e-05
        [updatestate_useless_node_eliminater]: 1.349e-05
        [accelerated_algorithm]: 1.003e-05
        [meta_shard_fg_expand]: 3.94002e-06
        [get_grad_eliminate_]: 9.27001e-06
        [merge_forward]: 5.47999e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.139e-05
        [j_node_and_user_rematch]: 1.649e-05
        [meta_fg_expand]: 4.30999e-06
        [replace_old_param]: 1.359e-05
        [inline_without_move]: 9.51e-06
        [renormalize]: 0.0015773
        [add_forward_monad_depend]: 8.23001e-06
        [auto_monad_grad]: 3.11999e-06
        [auto_monad_eliminator]: 2.518e-05
        [cse]: 4.593e-05
        [replace_applicator]: 2.288e-05
    [Cycle 2]: 0.00050633, [27]
        [switch_simplify]: 1.029e-05
        [loop_unroll]: 8.23999e-06
        [a_1]: 0.00017947
        [with_stream_mark]: 1.779e-05
        [recompute_prepare]: 8.13999e-06
        [updatestate_depend_eliminate]: 5.88002e-06
        [updatestate_assign_eliminate]: 5.08002e-06
        [updatestate_loads_eliminate]: 4.13001e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 8.54e-06
        [updatestate_useless_node_eliminater]: 1.089e-05
        [accelerated_algorithm]: 8.66002e-06
        [meta_shard_fg_expand]: 2.77002e-06
        [get_grad_eliminate_]: 7.82002e-06
        [merge_forward]: 5.52001e-06
        [cell_reuse_recompute_pass]: 3.78999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.834e-05
        [j_node_and_user_rematch]: 1.401e-05
        [meta_fg_expand]: 2.93e-06
        [replace_old_param]: 1.07e-05
        [inline_without_move]: 7.69002e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 1.24003e-06
        [auto_monad_grad]: 1.32999e-06
        [auto_monad_eliminator]: 1.071e-05
        [cse]: 2.141e-05
        [replace_applicator]: 9.14e-06
[py_interpret_to_execute_after_opt_a]: 1.846e-05
[rewriter_after_opt_a]: 0.00030327
[convert_after_rewriter]: 1.218e-05
[order_py_execute_after_rewriter]: 7.41999e-06
[mutable_eliminate]: 0.00087033
[jit_opt_b]: 7.658e-05, [1]
    [Cycle 1]: 6.723e-05, [2]
        [frontend_op_eliminate]: 2.589e-05
        [inline_after_opt_a]: 2.796e-05
[cconv]: 3.502e-05
[loop_unroll]: 0.00053906
[jit_opt_after_cconv]: 0.00023798, [1]
    [Cycle 1]: 0.00022966, [11]
        [c_1]: 5.832e-05
        [parameter_eliminate]: 5.09998e-06
        [updatestate_depend_eliminate]: 1.188e-05
        [updatestate_assign_eliminate]: 5.35999e-06
        [updatestate_loads_eliminate]: 4.65999e-06
        [cse]: 4.289e-05
        [call_graph_tuple_transform]: 2.428e-05
        [tuple_list_get_item_eliminator]: 8.28999e-06
        [none_parameter_eliminate]: 1.79e-06
        [renormalize]: 1.05001e-06
        [switch_simplify]: 1.086e-05
[remove_dup_value]: 2.21e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 1.433e-05
[add_recomputation]: 7.353e-05
[cse_after_recomputation]: 3.145e-05, [1]
    [Cycle 1]: 2.471e-05, [1]
        [cse]: 1.811e-05
[auto_monad_reorder]: 2.618e-05
[get_jit_bprop_graph]: 2.54001e-06
[rewriter_after_jit_bprop_graph]: 7.80998e-06
[opt_after_jit_grad]: 0.00063547
[symbol_engine_optimizer]: 0.00011363, [1]
    [Cycle 1]: 0.00010631, [6]
        [build]: 1.772e-05
        [elim_shapecalc]: 1.281e-05
        [elim_not_effective]: 1.913e-05
        [opt_reshape]: 9.15001e-06
        [fold_const_symbol]: 1.715e-05
        [renormalize]: 8.09989e-07
[validate]: 7.077e-05
[backend_pass]: 1.19998e-06
[task_emit]: 0.0129739
[execute]: 9.27001e-06
Sums
    bootstrap                            :     0.000350s :  0.76%
    type_inference                       :     0.026005s : 56.56%
    event_method                         :     0.000026s :  0.06%
    auto_monad                           :     0.000178s :  0.39%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000142s :  0.31%
    rewriter_before_opt_a                :     0.000103s :  0.22%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000089s :  0.19%
    jit_opt_a.loop_unroll                :     0.000049s :  0.11%
    jit_opt_a.a_1                        :     0.001026s :  2.23%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000020s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000024s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001577s :  3.43%
    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      :     0.000036s :  0.08%
    jit_opt_a.cse                        :     0.000067s :  0.15%
    jit_opt_a.replace_applicator         :     0.000032s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.04%
    rewriter_after_opt_a                 :     0.000303s :  0.66%
    convert_after_rewriter               :     0.000012s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000870s :  1.89%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000028s :  0.06%
    cconv                                :     0.000035s :  0.08%
    loop_unroll                          :     0.000539s :  1.17%
    jit_opt_after_cconv.c_1              :     0.000058s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000043s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.05%
    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      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000011s :  0.02%
    remove_dup_value                     :     0.000022s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000014s :  0.03%
    add_recomputation                    :     0.000074s :  0.16%
    cse_after_recomputation.cse          :     0.000018s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000635s :  1.38%
    symbol_engine_optimizer.build        :     0.000018s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000017s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000071s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012974s : 28.21%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000381    47
 3.23% :     0.000012s :      2: substitution.depend_value_elim
 0.67% :     0.000003s :      4: substitution.elim_not_effective
 0.70% :     0.000003s :      4: substitution.fold_const_symbol
 1.86% :     0.000007s :      5: substitution.graph_param_transform
77.82% :     0.000297s :      4: substitution.inline
 1.38% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.03% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.77% :     0.000007s :      2: substitution.replace_old_param
 3.41% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.77% :     0.000014s :      3: substitution.updatestate_pure_node_eliminater
 3.35% :     0.000013s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.025921     2
93.76% :     0.024303s :      1: type_inference.infer
 6.24% :     0.001618s :      1: type_inference.specialize
------[replace.]   0.000104     6
68.49% :     0.000071s :      4: replace.inline
31.51% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000306     6
96.11% :     0.000294s :      4: match.inline
 3.89% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000189  1021
 1.27% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.37% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.24% :     0.000002s :     15: predicate.addn_zero_filter
 2.24% :     0.000004s :     15: predicate.arithmetic_simplify
 1.48% :     0.000003s :     15: predicate.cast_eliminate
 0.49% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     15: predicate.compare_switch_simplify
 1.80% :     0.000003s :     15: predicate.depend_value_elim
 1.03% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.09% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.20% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      5: predicate.elim_not_effective
 0.50% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     15: predicate.environ_get_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.02% :     0.000002s :     10: predicate.get_grad_eliminate
 0.37% :     0.000001s :      5: predicate.graph_param_transform
 5.28% :     0.000010s :     31: predicate.inline
 1.08% :     0.000002s :     10: predicate.inline_without_move
 0.39% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.32% :     0.000003s :     10: predicate.less_batch_normalization
 1.43% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.74% :     0.000003s :     22: predicate.load_eliminater
 1.94% :     0.000004s :      5: predicate.loop_unroll_after_grad
 3.81% :     0.000007s :     42: predicate.loop_unroll_before_grad
 2.15% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000002s :     15: predicate.merge_addn
 1.08% :     0.000002s :     15: predicate.minmaximum_grad
 1.95% :     0.000004s :      5: predicate.mutable_eliminate
 0.55% :     0.000001s :      5: predicate.opt_reshape
 2.20% :     0.000004s :     22: predicate.partial_eliminate
 1.25% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.62% :     0.000003s :     15: predicate.reduce_eliminate
 1.29% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.69% :     0.000003s :     27: predicate.replace_applicator
 0.64% :     0.000001s :     10: predicate.replace_old_param
 0.41% :     0.000001s :      5: predicate.reset_defer_inline
 1.15% :     0.000002s :     15: predicate.reshape_eliminate
 1.10% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.81% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.33% :     0.000003s :     15: predicate.same_eliminate
 0.80% :     0.000002s :     10: predicate.set_cell_output_no_recompute
 1.07% :     0.000002s :     10: predicate.special_op_eliminate
 1.19% :     0.000002s :     10: predicate.specialize_transform
 1.52% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.05% :     0.000004s :     21: predicate.switch_defer_inline
 1.76% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.46% :     0.000012s :     68: predicate.switch_simplify
 1.25% :     0.000002s :     15: predicate.tile_eliminate
 1.22% :     0.000002s :     15: predicate.transpose_eliminate
 1.45% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 4.13% :     0.000008s :     27: predicate.tuple_list_get_item_eliminator
 2.10% :     0.000004s :     15: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.93% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     15: predicate.value_based_eliminate
 0.41% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.81% :     0.000002s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001051    13
45.82% :     0.000482s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.87% :     0.000072s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.31% :     0.000497s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.054605    76
 0.14% :     0.000076s :      1: add_recomputation
 0.34% :     0.000184s :      1: auto_monad
 0.05% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.68% :     0.000370s :      1: bootstrap
 0.07% :     0.000039s :      1: cconv
 0.03% :     0.000015s :      1: convert_after_rewriter
 0.06% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000017s :      1: environ_conv
 0.06% :     0.000031s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000006s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
15.15% :     0.008275s :      1: jit_opt_a
 0.44% :     0.000241s :      1: jit_opt_after_cconv
 0.15% :     0.000080s :      1: jit_opt_b
 1.00% :     0.000548s :      1: loop_unroll
 1.62% :     0.000887s :      1: mutable_eliminate
 2.51% :     0.001373s :     26: opt.transform.jit_opt_a
 0.18% :     0.000097s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000046s :      4: opt.transform.jit_opt_b
 0.04% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000054s :      4: opt.transform.symbol_engine_opt
 1.18% :     0.000647s :      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
 0.27% :     0.000147s :      1: py_interpret_to_execute
 0.04% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000025s :      1: remove_dup_value
 1.87% :     0.001019s :      1: renormalize.infer
 1.00% :     0.000545s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.56% :     0.000308s :      1: rewriter_after_opt_a
 0.20% :     0.000107s :      1: rewriter_before_opt_a
 0.21% :     0.000116s :      1: symbol_engine_optimizer
23.79% :     0.012993s :      1: task_emit
47.67% :     0.026029s :      1: type_inference
 0.18% :     0.000100s :      1: validate


TotalTime = 0.0422786, [33]
[bootstrap]: 0.00028161
[type_inference]: 0.0222563
[event_method]: 2.068e-05
[auto_monad]: 0.00014336
[graph_reusing]: 4.87998e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 0.00012103
[rewriter_before_opt_a]: 0.00011589
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00627685, [2]
    [Cycle 1]: 0.00228958, [27]
        [switch_simplify]: 5.708e-05
        [loop_unroll]: 3.698e-05
        [a_1]: 0.00076149
        [with_stream_mark]: 1.791e-05
        [recompute_prepare]: 1.107e-05
        [updatestate_depend_eliminate]: 5.94999e-06
        [updatestate_assign_eliminate]: 4.52e-06
        [updatestate_loads_eliminate]: 3.97998e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 9.21998e-06
        [updatestate_useless_node_eliminater]: 1.148e-05
        [accelerated_algorithm]: 8.63001e-06
        [meta_shard_fg_expand]: 2.39999e-06
        [get_grad_eliminate_]: 8.06001e-06
        [merge_forward]: 4.85001e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.799e-05
        [j_node_and_user_rematch]: 1.381e-05
        [meta_fg_expand]: 3.75e-06
        [replace_old_param]: 1.148e-05
        [inline_without_move]: 8.03999e-06
        [renormalize]: 0.00105407
        [add_forward_monad_depend]: 4.97e-06
        [auto_monad_grad]: 1.24998e-06
        [auto_monad_eliminator]: 1.437e-05
        [cse]: 2.727e-05
        [replace_applicator]: 1.396e-05
    [Cycle 2]: 0.00046421, [27]
        [switch_simplify]: 8.85999e-06
        [loop_unroll]: 8.15e-06
        [a_1]: 0.00016877
        [with_stream_mark]: 1.058e-05
        [recompute_prepare]: 8.02e-06
        [updatestate_depend_eliminate]: 4.82e-06
        [updatestate_assign_eliminate]: 4.50001e-06
        [updatestate_loads_eliminate]: 3.78999e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 8.1e-06
        [updatestate_useless_node_eliminater]: 1.099e-05
        [accelerated_algorithm]: 7.87e-06
        [meta_shard_fg_expand]: 1.79e-06
        [get_grad_eliminate_]: 7.66999e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 2.24001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.593e-05
        [j_node_and_user_rematch]: 1.268e-05
        [meta_fg_expand]: 2.86999e-06
        [replace_old_param]: 9.79e-06
        [inline_without_move]: 7.50998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 1.011e-05
        [cse]: 2.006e-05
        [replace_applicator]: 7.8e-06
[py_interpret_to_execute_after_opt_a]: 1.086e-05
[rewriter_after_opt_a]: 0.00026481
[convert_after_rewriter]: 1.044e-05
[order_py_execute_after_rewriter]: 6.37001e-06
[mutable_eliminate]: 0.00061541
[jit_opt_b]: 6.487e-05, [1]
    [Cycle 1]: 5.886e-05, [2]
        [frontend_op_eliminate]: 2.347e-05
        [inline_after_opt_a]: 2.394e-05
[cconv]: 1.664e-05
[loop_unroll]: 0.00042633
[jit_opt_after_cconv]: 0.000203, [1]
    [Cycle 1]: 0.00019614, [11]
        [c_1]: 4.25e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 6.49001e-06
        [updatestate_assign_eliminate]: 4.25999e-06
        [updatestate_loads_eliminate]: 3.96001e-06
        [cse]: 2.793e-05
        [call_graph_tuple_transform]: 2.056e-05
        [tuple_list_get_item_eliminator]: 8.25e-06
        [none_parameter_eliminate]: 1.474e-05
        [renormalize]: 7.29982e-07
        [switch_simplify]: 9.82999e-06
[remove_dup_value]: 1.495e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 8.39002e-06
[add_recomputation]: 5.295e-05
[cse_after_recomputation]: 3.133e-05, [1]
    [Cycle 1]: 2.561e-05, [1]
        [cse]: 1.927e-05
[auto_monad_reorder]: 2.102e-05
[get_jit_bprop_graph]: 1.80001e-06
[rewriter_after_jit_bprop_graph]: 6.09001e-06
[opt_after_jit_grad]: 0.00050423
[symbol_engine_optimizer]: 0.00010312, [1]
    [Cycle 1]: 9.695e-05, [6]
        [build]: 1.483e-05
        [elim_shapecalc]: 1.142e-05
        [elim_not_effective]: 1.856e-05
        [opt_reshape]: 8.94998e-06
        [fold_const_symbol]: 1.355e-05
        [renormalize]: 5.00004e-07
[validate]: 4.863e-05
[backend_pass]: 8.60018e-07
[task_emit]: 0.0104256
[execute]: 6.12999e-06
Sums
    bootstrap                            :     0.000282s :  0.74%
    type_inference                       :     0.022256s : 58.36%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000143s :  0.38%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000121s :  0.32%
    rewriter_before_opt_a                :     0.000116s :  0.30%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.17%
    jit_opt_a.loop_unroll                :     0.000045s :  0.12%
    jit_opt_a.a_1                        :     0.000930s :  2.44%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001054s :  2.76%
    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      :     0.000024s :  0.06%
    jit_opt_a.cse                        :     0.000047s :  0.12%
    jit_opt_a.replace_applicator         :     0.000022s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000265s :  0.69%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000615s :  1.61%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000017s :  0.04%
    loop_unroll                          :     0.000426s :  1.12%
    jit_opt_after_cconv.c_1              :     0.000042s :  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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000015s :  0.04%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000010s :  0.03%
    remove_dup_value                     :     0.000015s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.02%
    add_recomputation                    :     0.000053s :  0.14%
    cse_after_recomputation.cse          :     0.000019s :  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                   :     0.000504s :  1.32%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010426s : 27.34%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000322    47
 3.00% :     0.000010s :      2: substitution.depend_value_elim
 0.83% :     0.000003s :      4: substitution.elim_not_effective
 0.63% :     0.000002s :      4: substitution.fold_const_symbol
 1.78% :     0.000006s :      5: substitution.graph_param_transform
80.75% :     0.000260s :      4: substitution.inline
 1.43% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.09% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.15% :     0.000004s :      2: substitution.replace_old_param
 3.15% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
 2.50% :     0.000008s :      3: substitution.updatestate_pure_node_eliminater
 2.68% :     0.000009s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022192     2
93.64% :     0.020780s :      1: type_inference.infer
 6.36% :     0.001412s :      1: type_inference.specialize
------[replace.]   0.000092     6
65.80% :     0.000061s :      4: replace.inline
34.20% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     6
96.64% :     0.000257s :      4: match.inline
 3.36% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1021
 1.46% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.29% :     0.000002s :     15: predicate.addn_zero_filter
 2.03% :     0.000003s :     15: predicate.arithmetic_simplify
 1.37% :     0.000002s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     15: predicate.compare_switch_simplify
 1.41% :     0.000002s :     15: predicate.depend_value_elim
 1.16% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     15: predicate.environ_get_eliminate
 1.18% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      5: predicate.fold_const_symbol
 1.13% :     0.000002s :     10: predicate.get_grad_eliminate
 0.28% :     0.000000s :      5: predicate.graph_param_transform
 4.94% :     0.000008s :     31: predicate.inline
 1.08% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :     10: predicate.less_batch_normalization
 1.49% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     22: predicate.load_eliminater
 1.23% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.49% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.19% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     15: predicate.merge_addn
 1.11% :     0.000002s :     15: predicate.minmaximum_grad
 1.71% :     0.000003s :      5: predicate.mutable_eliminate
 0.53% :     0.000001s :      5: predicate.opt_reshape
 2.37% :     0.000004s :     22: predicate.partial_eliminate
 1.23% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.74% :     0.000003s :     15: predicate.reduce_eliminate
 1.61% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.69% :     0.000003s :     27: predicate.replace_applicator
 0.74% :     0.000001s :     10: predicate.replace_old_param
 0.30% :     0.000000s :      5: predicate.reset_defer_inline
 1.29% :     0.000002s :     15: predicate.reshape_eliminate
 1.31% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.56% :     0.000003s :     15: predicate.same_eliminate
 0.61% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.06% :     0.000002s :     10: predicate.special_op_eliminate
 1.18% :     0.000002s :     10: predicate.specialize_transform
 1.45% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.31% :     0.000004s :     21: predicate.switch_defer_inline
 1.89% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.46% :     0.000011s :     68: predicate.switch_simplify
 1.27% :     0.000002s :     15: predicate.tile_eliminate
 1.27% :     0.000002s :     15: predicate.transpose_eliminate
 1.58% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.56% :     0.000003s :     15: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.80% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.31% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000003s :     15: predicate.value_based_eliminate
 0.46% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000792    13
43.40% :     0.000344s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.34% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.26% :     0.000406s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044739    76
 0.13% :     0.000056s :      1: add_recomputation
 0.33% :     0.000149s :      1: auto_monad
 0.05% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000295s :      1: bootstrap
 0.04% :     0.000019s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000034s :      1: cse_after_recomputation
 0.02% :     0.000011s :      1: environ_conv
 0.05% :     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
 0.02% :     0.000008s :      1: graph_reusing
14.04% :     0.006280s :      1: jit_opt_a
 0.46% :     0.000206s :      1: jit_opt_after_cconv
 0.15% :     0.000067s :      1: jit_opt_b
 0.97% :     0.000434s :      1: loop_unroll
 1.39% :     0.000623s :      1: mutable_eliminate
 2.74% :     0.001224s :     26: opt.transform.jit_opt_a
 0.17% :     0.000077s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000041s :      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
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.14% :     0.000512s :      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
 0.28% :     0.000125s :      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
 1.44% :     0.000646s :      1: renormalize.infer
 0.90% :     0.000401s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.60% :     0.000269s :      1: rewriter_after_opt_a
 0.27% :     0.000121s :      1: rewriter_before_opt_a
 0.24% :     0.000106s :      1: symbol_engine_optimizer
23.33% :     0.010440s :      1: task_emit
49.78% :     0.022269s :      1: type_inference
 0.16% :     0.000071s :      1: validate


TotalTime = 0.0440975, [33]
[bootstrap]: 0.00031139
[type_inference]: 0.023707
[event_method]: 1.984e-05
[auto_monad]: 0.00014086
[graph_reusing]: 4.57e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 0.00011674
[rewriter_before_opt_a]: 9.079e-05
[expand_dump_flag]: 2.51998e-06
[jit_opt_a]: 0.00641398, [2]
    [Cycle 1]: 0.00236161, [27]
        [switch_simplify]: 5.467e-05
        [loop_unroll]: 3.655e-05
        [a_1]: 0.00067871
        [with_stream_mark]: 1.855e-05
        [recompute_prepare]: 1.123e-05
        [updatestate_depend_eliminate]: 6.26e-06
        [updatestate_assign_eliminate]: 5.00999e-06
        [updatestate_loads_eliminate]: 4.19002e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 9.89999e-06
        [updatestate_useless_node_eliminater]: 1.199e-05
        [accelerated_algorithm]: 8.41002e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 8.67998e-06
        [merge_forward]: 4.70001e-06
        [cell_reuse_recompute_pass]: 2.35002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.762e-05
        [j_node_and_user_rematch]: 1.408e-05
        [meta_fg_expand]: 3.30998e-06
        [replace_old_param]: 1.14e-05
        [inline_without_move]: 8.02e-06
        [renormalize]: 0.00115567
        [add_forward_monad_depend]: 8.39002e-06
        [auto_monad_grad]: 2.89001e-06
        [auto_monad_eliminator]: 2.397e-05
        [cse]: 4.15e-05
        [replace_applicator]: 2.127e-05
    [Cycle 2]: 0.00048806, [27]
        [switch_simplify]: 9.06002e-06
        [loop_unroll]: 7.76001e-06
        [a_1]: 0.00017375
        [with_stream_mark]: 1.576e-05
        [recompute_prepare]: 8.60001e-06
        [updatestate_depend_eliminate]: 5.03002e-06
        [updatestate_assign_eliminate]: 5.51e-06
        [updatestate_loads_eliminate]: 4.17998e-06
        [parameter_eliminate]: 1.72001e-06
        [specialize_transform]: 8.27e-06
        [updatestate_useless_node_eliminater]: 1.113e-05
        [accelerated_algorithm]: 8.15e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 7.46999e-06
        [merge_forward]: 4.60999e-06
        [cell_reuse_recompute_pass]: 3.97e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.783e-05
        [j_node_and_user_rematch]: 1.268e-05
        [meta_fg_expand]: 2.72001e-06
        [replace_old_param]: 1.041e-05
        [inline_without_move]: 7.93001e-06
        [renormalize]: 9.99717e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.36002e-06
        [auto_monad_eliminator]: 9.82999e-06
        [cse]: 2.137e-05
        [replace_applicator]: 8.01001e-06
[py_interpret_to_execute_after_opt_a]: 1.589e-05
[rewriter_after_opt_a]: 0.00029246
[convert_after_rewriter]: 1.341e-05
[order_py_execute_after_rewriter]: 8.05e-06
[mutable_eliminate]: 0.00075909
[jit_opt_b]: 6.876e-05, [1]
    [Cycle 1]: 6.068e-05, [2]
        [frontend_op_eliminate]: 2.433e-05
        [inline_after_opt_a]: 2.442e-05
[cconv]: 2.936e-05
[loop_unroll]: 0.00046908
[jit_opt_after_cconv]: 0.0002047, [1]
    [Cycle 1]: 0.00019774, [11]
        [c_1]: 4.772e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 8.47e-06
        [updatestate_assign_eliminate]: 4.53999e-06
        [updatestate_loads_eliminate]: 4.3e-06
        [cse]: 3.187e-05
        [call_graph_tuple_transform]: 2.451e-05
        [tuple_list_get_item_eliminator]: 8.55999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 8.55999e-06
[remove_dup_value]: 2.007e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 1.135e-05
[add_recomputation]: 6.677e-05
[cse_after_recomputation]: 2.95e-05, [1]
    [Cycle 1]: 2.396e-05, [1]
        [cse]: 1.742e-05
[auto_monad_reorder]: 2.747e-05
[get_jit_bprop_graph]: 1.97001e-06
[rewriter_after_jit_bprop_graph]: 6.27001e-06
[opt_after_jit_grad]: 0.00050987
[symbol_engine_optimizer]: 0.00010505, [1]
    [Cycle 1]: 9.815e-05, [6]
        [build]: 1.478e-05
        [elim_shapecalc]: 1.243e-05
        [elim_not_effective]: 1.906e-05
        [opt_reshape]: 8.55001e-06
        [fold_const_symbol]: 1.356e-05
        [renormalize]: 4.49974e-07
[validate]: 5.873e-05
[backend_pass]: 1.45001e-06
[task_emit]: 0.0101345
[execute]: 7.53e-06
Sums
    bootstrap                            :     0.000311s :  0.79%
    type_inference                       :     0.023707s : 59.79%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000141s :  0.36%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000117s :  0.29%
    rewriter_before_opt_a                :     0.000091s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.16%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000852s :  2.15%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    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 :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001156s :  2.91%
    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      :     0.000034s :  0.09%
    jit_opt_a.cse                        :     0.000063s :  0.16%
    jit_opt_a.replace_applicator         :     0.000029s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.04%
    rewriter_after_opt_a                 :     0.000292s :  0.74%
    convert_after_rewriter               :     0.000013s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000759s :  1.91%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000029s :  0.07%
    loop_unroll                          :     0.000469s :  1.18%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.12%
    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              :     0.000032s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000020s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000011s :  0.03%
    add_recomputation                    :     0.000067s :  0.17%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000027s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000510s :  1.29%
    symbol_engine_optimizer.build        :     0.000015s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000059s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010134s : 25.56%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000255    47
 4.49% :     0.000011s :      2: substitution.depend_value_elim
 1.13% :     0.000003s :      4: substitution.elim_not_effective
 0.80% :     0.000002s :      4: substitution.fold_const_symbol
 2.94% :     0.000008s :      5: substitution.graph_param_transform
71.74% :     0.000183s :      4: substitution.inline
 1.93% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.81% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.87% :     0.000005s :      2: substitution.replace_old_param
 3.73% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
 4.38% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 4.18% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.023636     2
93.85% :     0.022182s :      1: type_inference.infer
 6.15% :     0.001454s :      1: type_inference.specialize
------[replace.]   0.000086     6
64.44% :     0.000056s :      4: replace.inline
35.56% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
95.56% :     0.000181s :      4: match.inline
 4.44% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1021
 1.25% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     15: predicate.addn_check_dump
 1.29% :     0.000002s :     15: predicate.addn_zero_filter
 1.97% :     0.000003s :     15: predicate.arithmetic_simplify
 1.97% :     0.000003s :     15: predicate.cast_eliminate
 0.45% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     15: predicate.compare_switch_simplify
 1.34% :     0.000002s :     15: predicate.depend_value_elim
 1.17% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.07% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      5: predicate.elim_not_effective
 0.59% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.32% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     15: predicate.environ_get_eliminate
 1.30% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      5: predicate.fold_const_symbol
 1.24% :     0.000002s :     10: predicate.get_grad_eliminate
 0.29% :     0.000001s :      5: predicate.graph_param_transform
 5.47% :     0.000009s :     31: predicate.inline
 1.17% :     0.000002s :     10: predicate.inline_without_move
 0.45% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :     10: predicate.less_batch_normalization
 1.64% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     22: predicate.load_eliminater
 1.34% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.34% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.89% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000002s :     15: predicate.merge_addn
 1.16% :     0.000002s :     15: predicate.minmaximum_grad
 1.85% :     0.000003s :      5: predicate.mutable_eliminate
 0.48% :     0.000001s :      5: predicate.opt_reshape
 2.27% :     0.000004s :     22: predicate.partial_eliminate
 1.19% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.60% :     0.000003s :     15: predicate.reduce_eliminate
 1.39% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.99% :     0.000003s :     27: predicate.replace_applicator
 0.55% :     0.000001s :     10: predicate.replace_old_param
 0.34% :     0.000001s :      5: predicate.reset_defer_inline
 1.25% :     0.000002s :     15: predicate.reshape_eliminate
 1.27% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     15: predicate.same_eliminate
 0.74% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.04% :     0.000002s :     10: predicate.special_op_eliminate
 1.05% :     0.000002s :     10: predicate.specialize_transform
 1.38% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.20% :     0.000004s :     21: predicate.switch_defer_inline
 1.86% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.35% :     0.000011s :     68: predicate.switch_simplify
 1.23% :     0.000002s :     15: predicate.tile_eliminate
 1.25% :     0.000002s :     15: predicate.transpose_eliminate
 1.42% :     0.000002s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.81% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.96% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.38% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000003s :     15: predicate.value_based_eliminate
 0.38% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000822    13
43.54% :     0.000358s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.00% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.46% :     0.000423s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046596    76
 0.15% :     0.000070s :      1: add_recomputation
 0.32% :     0.000147s :      1: auto_monad
 0.07% :     0.000030s :      1: auto_monad_reorder
 0.02% :     0.000007s :      1: backend_pass
 0.70% :     0.000328s :      1: bootstrap
 0.07% :     0.000032s :      1: cconv
 0.04% :     0.000017s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000024s :      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% :     0.000007s :      1: graph_reusing
13.77% :     0.006418s :      1: jit_opt_a
 0.45% :     0.000208s :      1: jit_opt_after_cconv
 0.15% :     0.000072s :      1: jit_opt_b
 1.03% :     0.000478s :      1: loop_unroll
 1.65% :     0.000771s :      1: mutable_eliminate
 2.48% :     0.001156s :     26: opt.transform.jit_opt_a
 0.18% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.11% :     0.000519s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000121s :      1: py_interpret_to_execute
 0.04% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.50% :     0.000701s :      1: renormalize.infer
 0.96% :     0.000445s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.64% :     0.000299s :      1: rewriter_after_opt_a
 0.21% :     0.000096s :      1: rewriter_before_opt_a
 0.23% :     0.000108s :      1: symbol_engine_optimizer
21.79% :     0.010154s :      1: task_emit
50.91% :     0.023721s :      1: type_inference
 0.61% :     0.000285s :      1: validate


TotalTime = 0.0697085, [33]
[bootstrap]: 0.00028973
[type_inference]: 0.0479831
[event_method]: 1.959e-05
[auto_monad]: 0.00021049
[graph_reusing]: 5.47001e-06
[pre_auto_parallel]: 2.96999e-06
[py_interpret_to_execute]: 0.00012654
[rewriter_before_opt_a]: 9.426e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.00688878, [2]
    [Cycle 1]: 0.00249895, [27]
        [switch_simplify]: 6.506e-05
        [loop_unroll]: 3.57e-05
        [a_1]: 0.00075131
        [with_stream_mark]: 2.081e-05
        [recompute_prepare]: 1.1e-05
        [updatestate_depend_eliminate]: 5.79e-06
        [updatestate_assign_eliminate]: 6.02001e-06
        [updatestate_loads_eliminate]: 4.99998e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 9.32001e-06
        [updatestate_useless_node_eliminater]: 1.151e-05
        [accelerated_algorithm]: 8.32e-06
        [meta_shard_fg_expand]: 3.05002e-06
        [get_grad_eliminate_]: 8.52e-06
        [merge_forward]: 5.79e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.004e-05
        [j_node_and_user_rematch]: 1.327e-05
        [meta_fg_expand]: 3.49001e-06
        [replace_old_param]: 1.172e-05
        [inline_without_move]: 8.08999e-06
        [renormalize]: 0.00122805
        [add_forward_monad_depend]: 7.45e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 2.035e-05
        [cse]: 3.075e-05
        [replace_applicator]: 2.153e-05
    [Cycle 2]: 0.00048707, [27]
        [switch_simplify]: 9.00999e-06
        [loop_unroll]: 7.70998e-06
        [a_1]: 0.00017373
        [with_stream_mark]: 1.515e-05
        [recompute_prepare]: 8.08999e-06
        [updatestate_depend_eliminate]: 4.84e-06
        [updatestate_assign_eliminate]: 4.78001e-06
        [updatestate_loads_eliminate]: 4.03001e-06
        [parameter_eliminate]: 1.40001e-06
        [specialize_transform]: 8.13999e-06
        [updatestate_useless_node_eliminater]: 1.112e-05
        [accelerated_algorithm]: 7.87e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 7.34002e-06
        [merge_forward]: 6.02999e-06
        [cell_reuse_recompute_pass]: 3.57002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.813e-05
        [j_node_and_user_rematch]: 1.286e-05
        [meta_fg_expand]: 2.87002e-06
        [replace_old_param]: 1.072e-05
        [inline_without_move]: 8.1e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.62001e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 1.062e-05
        [cse]: 2.134e-05
        [replace_applicator]: 8.02e-06
[py_interpret_to_execute_after_opt_a]: 1.713e-05
[rewriter_after_opt_a]: 0.00031771
[convert_after_rewriter]: 1.12e-05
[order_py_execute_after_rewriter]: 7.28e-06
[mutable_eliminate]: 0.00083946
[jit_opt_b]: 6.977e-05, [1]
    [Cycle 1]: 6.226e-05, [2]
        [frontend_op_eliminate]: 2.502e-05
        [inline_after_opt_a]: 2.503e-05
[cconv]: 2.934e-05
[loop_unroll]: 0.00048509
[jit_opt_after_cconv]: 0.00021147, [1]
    [Cycle 1]: 0.0002036, [11]
        [c_1]: 5.197e-05
        [parameter_eliminate]: 3.17997e-06
        [updatestate_depend_eliminate]: 8.45001e-06
        [updatestate_assign_eliminate]: 4.72998e-06
        [updatestate_loads_eliminate]: 4.35e-06
        [cse]: 3.462e-05
        [call_graph_tuple_transform]: 2.402e-05
        [tuple_list_get_item_eliminator]: 8.55001e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 7.30011e-07
        [switch_simplify]: 9.50001e-06
[remove_dup_value]: 2.012e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 1.205e-05
[add_recomputation]: 6.651e-05
[cse_after_recomputation]: 3.022e-05, [1]
    [Cycle 1]: 2.383e-05, [1]
        [cse]: 1.777e-05
[auto_monad_reorder]: 2.654e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 8.12998e-06
[opt_after_jit_grad]: 0.00053851
[symbol_engine_optimizer]: 0.00011087, [1]
    [Cycle 1]: 0.0001034, [6]
        [build]: 1.5e-05
        [elim_shapecalc]: 1.2e-05
        [elim_not_effective]: 2.002e-05
        [opt_reshape]: 9.81e-06
        [fold_const_symbol]: 1.443e-05
        [renormalize]: 8.60018e-07
[validate]: 6.179e-05
[backend_pass]: 1.08001e-06
[task_emit]: 0.0109832
[execute]: 8.77999e-06
Sums
    bootstrap                            :     0.000290s :  0.44%
    type_inference                       :     0.047983s : 73.65%
    event_method                         :     0.000020s :  0.03%
    auto_monad                           :     0.000210s :  0.32%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000127s :  0.19%
    rewriter_before_opt_a                :     0.000094s :  0.14%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.11%
    jit_opt_a.loop_unroll                :     0.000043s :  0.07%
    jit_opt_a.a_1                        :     0.000925s :  1.42%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000017s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.02%
    jit_opt_a.merge_forward              :     0.000012s :  0.02%
    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    :     0.000026s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000022s :  0.03%
    jit_opt_a.inline_without_move        :     0.000016s :  0.02%
    jit_opt_a.renormalize                :     0.001228s :  1.88%
    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      :     0.000031s :  0.05%
    jit_opt_a.cse                        :     0.000052s :  0.08%
    jit_opt_a.replace_applicator         :     0.000030s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.03%
    rewriter_after_opt_a                 :     0.000318s :  0.49%
    convert_after_rewriter               :     0.000011s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.01%
    mutable_eliminate                    :     0.000839s :  1.29%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.04%
    cconv                                :     0.000029s :  0.05%
    loop_unroll                          :     0.000485s :  0.74%
    jit_opt_after_cconv.c_1              :     0.000052s :  0.08%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.01%
    remove_dup_value                     :     0.000020s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000012s :  0.02%
    add_recomputation                    :     0.000067s :  0.10%
    cse_after_recomputation.cse          :     0.000018s :  0.03%
    auto_monad_reorder                   :     0.000027s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.01%
    opt_after_jit_grad                   :     0.000539s :  0.83%
    symbol_engine_optimizer.build        :     0.000015s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000062s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010983s : 16.86%
    execute                              :     0.000009s :  0.01%


Time group info:
------[substitution.]   0.000318    47
 3.51% :     0.000011s :      2: substitution.depend_value_elim
 0.87% :     0.000003s :      4: substitution.elim_not_effective
 0.66% :     0.000002s :      4: substitution.fold_const_symbol
 2.18% :     0.000007s :      5: substitution.graph_param_transform
76.70% :     0.000244s :      4: substitution.inline
 1.55% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.40% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.58% :     0.000005s :      2: substitution.replace_old_param
 3.97% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
 3.11% :     0.000010s :      3: substitution.updatestate_pure_node_eliminater
 3.47% :     0.000011s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.047916     2
96.90% :     0.046432s :      1: type_inference.infer
 3.10% :     0.001484s :      1: type_inference.specialize
------[replace.]   0.000093     6
65.84% :     0.000061s :      4: replace.inline
34.16% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     6
95.47% :     0.000240s :      4: match.inline
 4.53% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1021
 1.47% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.18% :     0.000002s :     15: predicate.addn_check_dump
 1.21% :     0.000002s :     15: predicate.addn_zero_filter
 2.07% :     0.000004s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.52% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.10% :     0.000002s :     15: predicate.compare_switch_simplify
 1.38% :     0.000002s :     15: predicate.depend_value_elim
 1.25% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.14% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      5: predicate.elim_not_effective
 0.62% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.11% :     0.000002s :     10: predicate.get_grad_eliminate
 0.41% :     0.000001s :      5: predicate.graph_param_transform
 5.15% :     0.000009s :     31: predicate.inline
 1.31% :     0.000002s :     10: predicate.inline_without_move
 0.42% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :     10: predicate.less_batch_normalization
 1.38% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     22: predicate.load_eliminater
 1.33% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.21% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.09% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     15: predicate.merge_addn
 1.16% :     0.000002s :     15: predicate.minmaximum_grad
 1.51% :     0.000003s :      5: predicate.mutable_eliminate
 0.61% :     0.000001s :      5: predicate.opt_reshape
 2.48% :     0.000004s :     22: predicate.partial_eliminate
 1.18% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.78% :     0.000003s :     15: predicate.reduce_eliminate
 1.42% :     0.000002s :     17: predicate.redundant_stop_gradient_eliminater
 0.88% :     0.000002s :     10: predicate.remove_not_recompute_node
 1.72% :     0.000003s :     27: predicate.replace_applicator
 0.67% :     0.000001s :     10: predicate.replace_old_param
 0.39% :     0.000001s :      5: predicate.reset_defer_inline
 1.15% :     0.000002s :     15: predicate.reshape_eliminate
 1.27% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 1.02% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     15: predicate.same_eliminate
 0.62% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.09% :     0.000002s :     10: predicate.special_op_eliminate
 1.07% :     0.000002s :     10: predicate.specialize_transform
 1.36% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.13% :     0.000004s :     21: predicate.switch_defer_inline
 1.89% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.47% :     0.000011s :     68: predicate.switch_simplify
 1.33% :     0.000002s :     15: predicate.tile_eliminate
 1.38% :     0.000002s :     15: predicate.transpose_eliminate
 1.81% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.76% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.41% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000003s :     15: predicate.value_based_eliminate
 0.44% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000915    13
46.77% :     0.000428s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.19% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.04% :     0.000430s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.072376    76
 0.10% :     0.000070s :      1: add_recomputation
 0.30% :     0.000216s :      1: auto_monad
 0.04% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.42% :     0.000302s :      1: bootstrap
 0.04% :     0.000032s :      1: cconv
 0.02% :     0.000014s :      1: convert_after_rewriter
 0.04% :     0.000032s :      1: cse_after_recomputation
 0.02% :     0.000014s :      1: environ_conv
 0.03% :     0.000023s :      1: event_method
 0.02% :     0.000014s :      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
 9.52% :     0.006892s :      1: jit_opt_a
 0.30% :     0.000215s :      1: jit_opt_after_cconv
 0.10% :     0.000072s :      1: jit_opt_b
 0.68% :     0.000493s :      1: loop_unroll
 1.17% :     0.000848s :      1: mutable_eliminate
 1.71% :     0.001236s :     26: opt.transform.jit_opt_a
 0.12% :     0.000090s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000043s :      4: opt.transform.jit_opt_b
 0.02% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000053s :      4: opt.transform.symbol_engine_opt
 0.76% :     0.000549s :      1: opt_after_jit_grad
 0.01% :     0.000010s :      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% :     0.000130s :      1: py_interpret_to_execute
 0.03% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000023s :      1: remove_dup_value
 1.03% :     0.000748s :      1: renormalize.infer
 0.65% :     0.000470s :      1: renormalize.specialize
 0.01% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.45% :     0.000324s :      1: rewriter_after_opt_a
 0.14% :     0.000098s :      1: rewriter_before_opt_a
 0.16% :     0.000114s :      1: symbol_engine_optimizer
15.20% :     0.011003s :      1: task_emit
66.31% :     0.047994s :      1: type_inference
 0.12% :     0.000088s :      1: validate


TotalTime = 0.0448794, [33]
[bootstrap]: 0.00031754
[type_inference]: 0.0245974
[event_method]: 2.093e-05
[auto_monad]: 0.00015162
[graph_reusing]: 4.68001e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 0.00013418
[rewriter_before_opt_a]: 9.268e-05
[expand_dump_flag]: 2.53e-06
[jit_opt_a]: 0.00635371, [2]
    [Cycle 1]: 0.00226824, [27]
        [switch_simplify]: 6.162e-05
        [loop_unroll]: 3.604e-05
        [a_1]: 0.00073682
        [with_stream_mark]: 1.68e-05
        [recompute_prepare]: 1.098e-05
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 4.84e-06
        [updatestate_loads_eliminate]: 4.53001e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 9.71e-06
        [updatestate_useless_node_eliminater]: 1.185e-05
        [accelerated_algorithm]: 8.63001e-06
        [meta_shard_fg_expand]: 2.58e-06
        [get_grad_eliminate_]: 8.07998e-06
        [merge_forward]: 4.25999e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.657e-05
        [j_node_and_user_rematch]: 1.308e-05
        [meta_fg_expand]: 3.19001e-06
        [replace_old_param]: 1.123e-05
        [inline_without_move]: 8.57e-06
        [renormalize]: 0.00104559
        [add_forward_monad_depend]: 4.89e-06
        [auto_monad_grad]: 1.54e-06
        [auto_monad_eliminator]: 1.677e-05
        [cse]: 2.907e-05
        [replace_applicator]: 1.539e-05
    [Cycle 2]: 0.00046019, [27]
        [switch_simplify]: 8.92999e-06
        [loop_unroll]: 7.97998e-06
        [a_1]: 0.00016481
        [with_stream_mark]: 1.157e-05
        [recompute_prepare]: 8.3e-06
        [updatestate_depend_eliminate]: 5.06997e-06
        [updatestate_assign_eliminate]: 4.77e-06
        [updatestate_loads_eliminate]: 3.83999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 8.27e-06
        [updatestate_useless_node_eliminater]: 1.099e-05
        [accelerated_algorithm]: 8.28001e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 7.51001e-06
        [merge_forward]: 4.08999e-06
        [cell_reuse_recompute_pass]: 1.70001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.589e-05
        [j_node_and_user_rematch]: 1.257e-05
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 9.79999e-06
        [inline_without_move]: 8.13999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 9.48997e-06
        [cse]: 1.897e-05
        [replace_applicator]: 7.54002e-06
[py_interpret_to_execute_after_opt_a]: 1.067e-05
[rewriter_after_opt_a]: 0.00027092
[convert_after_rewriter]: 9.78002e-06
[order_py_execute_after_rewriter]: 6.91001e-06
[mutable_eliminate]: 0.00059849
[jit_opt_b]: 6.647e-05, [1]
    [Cycle 1]: 5.977e-05, [2]
        [frontend_op_eliminate]: 2.436e-05
        [inline_after_opt_a]: 2.379e-05
[cconv]: 1.706e-05
[loop_unroll]: 0.0004365
[jit_opt_after_cconv]: 0.00018879, [1]
    [Cycle 1]: 0.00018235, [11]
        [c_1]: 4.564e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 7.59002e-06
        [updatestate_assign_eliminate]: 4.25e-06
        [updatestate_loads_eliminate]: 3.98999e-06
        [cse]: 2.643e-05
        [call_graph_tuple_transform]: 2.156e-05
        [tuple_list_get_item_eliminator]: 8.25e-06
        [none_parameter_eliminate]: 9.20001e-07
        [renormalize]: 6.29982e-07
        [switch_simplify]: 8.38001e-06
[remove_dup_value]: 1.247e-05
[partial_unused_args_eliminate]: 1.53002e-06
[environ_conv]: 7.77e-06
[add_recomputation]: 0.00014723
[cse_after_recomputation]: 3.805e-05, [1]
    [Cycle 1]: 3.16e-05, [1]
        [cse]: 2.351e-05
[auto_monad_reorder]: 2.048e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 5.61e-06
[opt_after_jit_grad]: 0.00054579
[symbol_engine_optimizer]: 0.00011067, [1]
    [Cycle 1]: 0.00010363, [6]
        [build]: 1.104e-05
        [elim_shapecalc]: 1.325e-05
        [elim_not_effective]: 2.127e-05
        [opt_reshape]: 1.009e-05
        [fold_const_symbol]: 1.524e-05
        [renormalize]: 4.50003e-07
[validate]: 5.043e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.0104103
[execute]: 8.22e-06
Sums
    bootstrap                            :     0.000318s :  0.78%
    type_inference                       :     0.024597s : 60.59%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000152s :  0.37%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000134s :  0.33%
    rewriter_before_opt_a                :     0.000093s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.17%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000902s :  2.22%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  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 :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.04%
    jit_opt_a.renormalize                :     0.001046s :  2.58%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.06%
    jit_opt_a.cse                        :     0.000048s :  0.12%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000271s :  0.67%
    convert_after_rewriter               :     0.000010s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000598s :  1.47%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.06%
    cconv                                :     0.000017s :  0.04%
    loop_unroll                          :     0.000437s :  1.08%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.11%
    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 :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  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  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000012s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000008s :  0.02%
    add_recomputation                    :     0.000147s :  0.36%
    cse_after_recomputation.cse          :     0.000024s :  0.06%
    auto_monad_reorder                   :     0.000020s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000546s :  1.34%
    symbol_engine_optimizer.build        :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000050s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010410s : 25.64%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000275    47
 3.05% :     0.000008s :      2: substitution.depend_value_elim
 0.88% :     0.000002s :      4: substitution.elim_not_effective
 0.73% :     0.000002s :      4: substitution.fold_const_symbol
 1.92% :     0.000005s :      5: substitution.graph_param_transform
78.64% :     0.000216s :      4: substitution.inline
 1.43% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.41% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.57% :     0.000004s :      2: substitution.replace_old_param
 3.23% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
 3.17% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.95% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.024527     2
94.08% :     0.023075s :      1: type_inference.infer
 5.92% :     0.001451s :      1: type_inference.specialize
------[replace.]   0.000092     6
64.88% :     0.000060s :      4: replace.inline
35.12% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
96.51% :     0.000213s :      4: match.inline
 3.49% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1021
 1.36% :     0.000002s :     15: predicate.accumulaten_eliminater
 1.59% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.30% :     0.000002s :     15: predicate.addn_zero_filter
 2.24% :     0.000004s :     15: predicate.arithmetic_simplify
 1.34% :     0.000002s :     15: predicate.cast_eliminate
 0.48% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     15: predicate.compare_switch_simplify
 1.43% :     0.000002s :     15: predicate.depend_value_elim
 1.19% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      5: predicate.elim_not_effective
 0.65% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     15: predicate.environ_get_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.32% :     0.000001s :      5: predicate.fold_const_symbol
 1.12% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000000s :      5: predicate.graph_param_transform
 4.97% :     0.000008s :     31: predicate.inline
 1.00% :     0.000002s :     10: predicate.inline_without_move
 0.44% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :     10: predicate.less_batch_normalization
 1.54% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.90% :     0.000003s :     22: predicate.load_eliminater
 1.18% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.39% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     15: predicate.merge_addn
 1.15% :     0.000002s :     15: predicate.minmaximum_grad
 1.37% :     0.000002s :      5: predicate.mutable_eliminate
 0.65% :     0.000001s :      5: predicate.opt_reshape
 2.31% :     0.000004s :     22: predicate.partial_eliminate
 1.21% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.99% :     0.000003s :     15: predicate.reduce_eliminate
 1.56% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.69% :     0.000003s :     27: predicate.replace_applicator
 0.56% :     0.000001s :     10: predicate.replace_old_param
 0.31% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000002s :     15: predicate.reshape_eliminate
 1.20% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     15: predicate.same_eliminate
 0.59% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :     10: predicate.special_op_eliminate
 1.18% :     0.000002s :     10: predicate.specialize_transform
 1.50% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.61% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.35% :     0.000004s :     21: predicate.switch_defer_inline
 1.91% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.75% :     0.000011s :     68: predicate.switch_simplify
 1.41% :     0.000002s :     15: predicate.tile_eliminate
 1.22% :     0.000002s :     15: predicate.transpose_eliminate
 1.62% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.90% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.34% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000003s :     15: predicate.value_based_eliminate
 0.48% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000845    13
43.34% :     0.000366s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 8.31% :     0.000070s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.34% :     0.000409s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047313    76
 0.32% :     0.000152s :      1: add_recomputation
 0.33% :     0.000158s :      1: auto_monad
 0.05% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.70% :     0.000332s :      1: bootstrap
 0.04% :     0.000020s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.09% :     0.000041s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.05% :     0.000025s :      1: event_method
 0.03% :     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
13.44% :     0.006357s :      1: jit_opt_a
 0.40% :     0.000191s :      1: jit_opt_after_cconv
 0.15% :     0.000069s :      1: jit_opt_b
 0.94% :     0.000445s :      1: loop_unroll
 1.28% :     0.000607s :      1: mutable_eliminate
 2.54% :     0.001200s :     26: opt.transform.jit_opt_a
 0.17% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000042s :      4: opt.transform.jit_opt_b
 0.03% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      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
 1.17% :     0.000556s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000139s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000016s :      1: remove_dup_value
 1.34% :     0.000632s :      1: renormalize.infer
 0.86% :     0.000405s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.58% :     0.000275s :      1: rewriter_after_opt_a
 0.21% :     0.000098s :      1: rewriter_before_opt_a
 0.24% :     0.000114s :      1: symbol_engine_optimizer
22.04% :     0.010429s :      1: task_emit
52.02% :     0.024614s :      1: type_inference
 0.16% :     0.000075s :      1: validate


TotalTime = 0.0435836, [33]
[bootstrap]: 0.00032721
[type_inference]: 0.0228671
[event_method]: 2.04e-05
[auto_monad]: 0.00015192
[graph_reusing]: 5.49998e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 0.00013138
[rewriter_before_opt_a]: 9.573e-05
[expand_dump_flag]: 4.57998e-06
[jit_opt_a]: 0.00647492, [2]
    [Cycle 1]: 0.00235069, [27]
        [switch_simplify]: 5.623e-05
        [loop_unroll]: 3.669e-05
        [a_1]: 0.0007057
        [with_stream_mark]: 2.973e-05
        [recompute_prepare]: 1.173e-05
        [updatestate_depend_eliminate]: 6.97002e-06
        [updatestate_assign_eliminate]: 5.00001e-06
        [updatestate_loads_eliminate]: 4.22e-06
        [parameter_eliminate]: 1.53002e-06
        [specialize_transform]: 9.76998e-06
        [updatestate_useless_node_eliminater]: 1.275e-05
        [accelerated_algorithm]: 8.52998e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 8.31002e-06
        [merge_forward]: 4.93001e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.798e-05
        [j_node_and_user_rematch]: 1.338e-05
        [meta_fg_expand]: 3.63e-06
        [replace_old_param]: 1.195e-05
        [inline_without_move]: 8.24002e-06
        [renormalize]: 0.00113274
        [add_forward_monad_depend]: 5.64e-06
        [auto_monad_grad]: 2.16998e-06
        [auto_monad_eliminator]: 1.961e-05
        [cse]: 2.826e-05
        [replace_applicator]: 1.58e-05
    [Cycle 2]: 0.00046773, [27]
        [switch_simplify]: 9.09e-06
        [loop_unroll]: 7.79002e-06
        [a_1]: 0.0001707
        [with_stream_mark]: 1.225e-05
        [recompute_prepare]: 7.65998e-06
        [updatestate_depend_eliminate]: 5.13002e-06
        [updatestate_assign_eliminate]: 4.97e-06
        [updatestate_loads_eliminate]: 3.94002e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 8.31002e-06
        [updatestate_useless_node_eliminater]: 1.111e-05
        [accelerated_algorithm]: 7.72998e-06
        [meta_shard_fg_expand]: 1.63002e-06
        [get_grad_eliminate_]: 7.41999e-06
        [merge_forward]: 4.18001e-06
        [cell_reuse_recompute_pass]: 2.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.604e-05
        [j_node_and_user_rematch]: 1.232e-05
        [meta_fg_expand]: 2.68e-06
        [replace_old_param]: 9.59e-06
        [inline_without_move]: 7.35998e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 1.049e-05
        [cse]: 1.94e-05
        [replace_applicator]: 7.61999e-06
[py_interpret_to_execute_after_opt_a]: 1.374e-05
[rewriter_after_opt_a]: 0.00042569
[convert_after_rewriter]: 1.377e-05
[order_py_execute_after_rewriter]: 8.57998e-06
[mutable_eliminate]: 0.00074023
[jit_opt_b]: 7.122e-05, [1]
    [Cycle 1]: 6.36e-05, [2]
        [frontend_op_eliminate]: 2.477e-05
        [inline_after_opt_a]: 2.601e-05
[cconv]: 1.954e-05
[loop_unroll]: 0.00047665
[jit_opt_after_cconv]: 0.00020432, [1]
    [Cycle 1]: 0.00019744, [11]
        [c_1]: 5.026e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 7.92e-06
        [updatestate_assign_eliminate]: 4.53999e-06
        [updatestate_loads_eliminate]: 4.32e-06
        [cse]: 2.988e-05
        [call_graph_tuple_transform]: 2.146e-05
        [tuple_list_get_item_eliminator]: 9.54e-06
        [none_parameter_eliminate]: 1.17e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 9.84999e-06
[remove_dup_value]: 1.299e-05
[partial_unused_args_eliminate]: 1.53002e-06
[environ_conv]: 1.012e-05
[add_recomputation]: 5.424e-05
[cse_after_recomputation]: 3.159e-05, [1]
    [Cycle 1]: 2.483e-05, [1]
        [cse]: 1.747e-05
[auto_monad_reorder]: 2.064e-05
[get_jit_bprop_graph]: 1.58002e-06
[rewriter_after_jit_bprop_graph]: 5.35001e-06
[opt_after_jit_grad]: 0.00053783
[symbol_engine_optimizer]: 0.00010864, [1]
    [Cycle 1]: 0.00010115, [6]
        [build]: 1.307e-05
        [elim_shapecalc]: 1.292e-05
        [elim_not_effective]: 2.034e-05
        [opt_reshape]: 9.61e-06
        [fold_const_symbol]: 1.413e-05
        [renormalize]: 4.09986e-07
[validate]: 5.363e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.0104608
[execute]: 8.21002e-06
Sums
    bootstrap                            :     0.000327s :  0.83%
    type_inference                       :     0.022867s : 58.23%
    event_method                         :     0.000020s :  0.05%
    auto_monad                           :     0.000152s :  0.39%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000131s :  0.33%
    rewriter_before_opt_a                :     0.000096s :  0.24%
    expand_dump_flag                     :     0.000005s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.17%
    jit_opt_a.loop_unroll                :     0.000044s :  0.11%
    jit_opt_a.a_1                        :     0.000876s :  2.23%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    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 :     0.000034s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001133s :  2.88%
    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      :     0.000030s :  0.08%
    jit_opt_a.cse                        :     0.000048s :  0.12%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000426s :  1.08%
    convert_after_rewriter               :     0.000014s :  0.04%
    order_py_execute_after_rewriter      :     0.000009s :  0.02%
    mutable_eliminate                    :     0.000740s :  1.88%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.07%
    cconv                                :     0.000020s :  0.05%
    loop_unroll                          :     0.000477s :  1.21%
    jit_opt_after_cconv.c_1              :     0.000050s :  0.13%
    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              :     0.000030s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000010s :  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  :     0.000010s :  0.03%
    remove_dup_value                     :     0.000013s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000010s :  0.03%
    add_recomputation                    :     0.000054s :  0.14%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000021s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000538s :  1.37%
    symbol_engine_optimizer.build        :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000054s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010461s : 26.64%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000270    47
 3.85% :     0.000010s :      2: substitution.depend_value_elim
 0.93% :     0.000002s :      4: substitution.elim_not_effective
 0.75% :     0.000002s :      4: substitution.fold_const_symbol
 2.05% :     0.000006s :      5: substitution.graph_param_transform
75.29% :     0.000203s :      4: substitution.inline
 1.55% :     0.000004s :      8: substitution.j_node_and_user_rematch
 2.51% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.41% :     0.000004s :      2: substitution.replace_old_param
 4.16% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
 3.90% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.60% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.022800     2
93.68% :     0.021360s :      1: type_inference.infer
 6.32% :     0.001440s :      1: type_inference.specialize
------[replace.]   0.000091     6
65.49% :     0.000060s :      4: replace.inline
34.51% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
95.25% :     0.000200s :      4: match.inline
 4.75% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1021
 1.59% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.09% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     15: predicate.addn_check_dump
 1.53% :     0.000003s :     15: predicate.addn_zero_filter
 2.19% :     0.000004s :     15: predicate.arithmetic_simplify
 1.31% :     0.000002s :     15: predicate.cast_eliminate
 0.53% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     15: predicate.compare_switch_simplify
 1.32% :     0.000002s :     15: predicate.depend_value_elim
 1.16% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     15: predicate.dict_set_item_eliminator
 1.04% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      5: predicate.elim_not_effective
 0.64% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     15: predicate.environ_get_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      5: predicate.fold_const_symbol
 1.04% :     0.000002s :     10: predicate.get_grad_eliminate
 0.25% :     0.000000s :      5: predicate.graph_param_transform
 5.04% :     0.000009s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.43% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :     10: predicate.less_batch_normalization
 1.50% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.92% :     0.000003s :     22: predicate.load_eliminater
 1.39% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.43% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     15: predicate.merge_addn
 1.10% :     0.000002s :     15: predicate.minmaximum_grad
 2.00% :     0.000003s :      5: predicate.mutable_eliminate
 0.72% :     0.000001s :      5: predicate.opt_reshape
 2.33% :     0.000004s :     22: predicate.partial_eliminate
 1.29% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     15: predicate.reduce_eliminate
 1.59% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.84% :     0.000003s :     27: predicate.replace_applicator
 0.75% :     0.000001s :     10: predicate.replace_old_param
 0.29% :     0.000001s :      5: predicate.reset_defer_inline
 1.23% :     0.000002s :     15: predicate.reshape_eliminate
 1.38% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     15: predicate.same_eliminate
 0.61% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.27% :     0.000002s :     10: predicate.special_op_eliminate
 1.28% :     0.000002s :     10: predicate.specialize_transform
 1.58% :     0.000003s :     15: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     15: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.21% :     0.000004s :     21: predicate.switch_defer_inline
 1.88% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.11% :     0.000010s :     68: predicate.switch_simplify
 1.45% :     0.000002s :     15: predicate.tile_eliminate
 1.25% :     0.000002s :     15: predicate.transpose_eliminate
 1.56% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.45% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 1.84% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 3.17% :     0.000005s :     32: predicate.updatestate_useless_node_eliminater
 2.05% :     0.000003s :     15: predicate.value_based_eliminate
 0.61% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000841    13
40.23% :     0.000338s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 5.36% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
54.41% :     0.000457s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046092    76
 0.13% :     0.000058s :      1: add_recomputation
 0.34% :     0.000157s :      1: auto_monad
 0.05% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.74% :     0.000342s :      1: bootstrap
 0.05% :     0.000022s :      1: cconv
 0.04% :     0.000017s :      1: convert_after_rewriter
 0.07% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.05% :     0.000025s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
14.05% :     0.006478s :      1: jit_opt_a
 0.45% :     0.000207s :      1: jit_opt_after_cconv
 0.16% :     0.000074s :      1: jit_opt_b
 1.05% :     0.000486s :      1: loop_unroll
 1.63% :     0.000749s :      1: mutable_eliminate
 2.54% :     0.001173s :     26: opt.transform.jit_opt_a
 0.19% :     0.000087s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000043s :      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
 0.07% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000547s :      1: opt_after_jit_grad
 0.02% :     0.000011s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000136s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000016s :      1: remove_dup_value
 1.50% :     0.000691s :      1: renormalize.infer
 0.94% :     0.000433s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.94% :     0.000433s :      1: rewriter_after_opt_a
 0.22% :     0.000100s :      1: rewriter_before_opt_a
 0.24% :     0.000112s :      1: symbol_engine_optimizer
22.73% :     0.010476s :      1: task_emit
49.64% :     0.022880s :      1: type_inference
 0.17% :     0.000077s :      1: validate


TotalTime = 0.0537389, [33]
[bootstrap]: 0.00029657
[type_inference]: 0.0298441
[event_method]: 1.969e-05
[auto_monad]: 0.00015643
[graph_reusing]: 9.47999e-06
[pre_auto_parallel]: 3.13e-06
[py_interpret_to_execute]: 0.00013122
[rewriter_before_opt_a]: 0.0001004
[expand_dump_flag]: 3.34001e-06
[jit_opt_a]: 0.00792933, [2]
    [Cycle 1]: 0.0033437, [27]
        [switch_simplify]: 0.00093637
        [loop_unroll]: 3.973e-05
        [a_1]: 0.00079455
        [with_stream_mark]: 1.986e-05
        [recompute_prepare]: 1.012e-05
        [updatestate_depend_eliminate]: 5.21998e-06
        [updatestate_assign_eliminate]: 4.80999e-06
        [updatestate_loads_eliminate]: 4.33999e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 9.22001e-06
        [updatestate_useless_node_eliminater]: 1.142e-05
        [accelerated_algorithm]: 9.16002e-06
        [meta_shard_fg_expand]: 2.91e-06
        [get_grad_eliminate_]: 9.15999e-06
        [merge_forward]: 5.12e-06
        [cell_reuse_recompute_pass]: 1.68002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.876e-05
        [j_node_and_user_rematch]: 1.353e-05
        [meta_fg_expand]: 3.45e-06
        [replace_old_param]: 1.208e-05
        [inline_without_move]: 8.49998e-06
        [renormalize]: 0.00112217
        [add_forward_monad_depend]: 5.39e-06
        [auto_monad_grad]: 2.30002e-06
        [auto_monad_eliminator]: 1.939e-05
        [cse]: 7.535e-05
        [replace_applicator]: 1.804e-05
    [Cycle 2]: 0.00047515, [27]
        [switch_simplify]: 9.17001e-06
        [loop_unroll]: 8.52998e-06
        [a_1]: 0.00017167
        [with_stream_mark]: 1.324e-05
        [recompute_prepare]: 8.77e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 5.23002e-06
        [updatestate_loads_eliminate]: 4.01001e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 7.9e-06
        [updatestate_useless_node_eliminater]: 1.014e-05
        [accelerated_algorithm]: 7.71999e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 7.33e-06
        [merge_forward]: 4.44002e-06
        [cell_reuse_recompute_pass]: 1.74998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.631e-05
        [j_node_and_user_rematch]: 1.24e-05
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 9.87001e-06
        [inline_without_move]: 7.66999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 1.116e-05
        [cse]: 2.134e-05
        [replace_applicator]: 9.42001e-06
[py_interpret_to_execute_after_opt_a]: 1.464e-05
[rewriter_after_opt_a]: 0.00028563
[convert_after_rewriter]: 1.03e-05
[order_py_execute_after_rewriter]: 7.7e-06
[mutable_eliminate]: 0.00072565
[jit_opt_b]: 7.424e-05, [1]
    [Cycle 1]: 6.63e-05, [2]
        [frontend_op_eliminate]: 2.641e-05
        [inline_after_opt_a]: 2.687e-05
[cconv]: 2.615e-05
[loop_unroll]: 0.0004869
[jit_opt_after_cconv]: 0.00020587, [1]
    [Cycle 1]: 0.00019874, [11]
        [c_1]: 4.832e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 7.75998e-06
        [updatestate_assign_eliminate]: 5.23002e-06
        [updatestate_loads_eliminate]: 4.57e-06
        [cse]: 3.035e-05
        [call_graph_tuple_transform]: 2.493e-05
        [tuple_list_get_item_eliminator]: 8.60999e-06
        [none_parameter_eliminate]: 1.30001e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 8.97e-06
[remove_dup_value]: 1.812e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 1.001e-05
[add_recomputation]: 6.415e-05
[cse_after_recomputation]: 3.34e-05, [1]
    [Cycle 1]: 2.748e-05, [1]
        [cse]: 2.084e-05
[auto_monad_reorder]: 2.691e-05
[get_jit_bprop_graph]: 1.93002e-06
[rewriter_after_jit_bprop_graph]: 4.74e-06
[opt_after_jit_grad]: 0.00050435
[symbol_engine_optimizer]: 0.00010976, [1]
    [Cycle 1]: 0.00010285, [6]
        [build]: 1.438e-05
        [elim_shapecalc]: 1.334e-05
        [elim_not_effective]: 1.984e-05
        [opt_reshape]: 9.27999e-06
        [fold_const_symbol]: 1.421e-05
        [renormalize]: 1.03001e-06
[validate]: 5.703e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.0123407
[execute]: 8.07e-06
Sums
    bootstrap                            :     0.000297s :  0.61%
    type_inference                       :     0.029844s : 60.94%
    event_method                         :     0.000020s :  0.04%
    auto_monad                           :     0.000156s :  0.32%
    graph_reusing                        :     0.000009s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000131s :  0.27%
    rewriter_before_opt_a                :     0.000100s :  0.21%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000946s :  1.93%
    jit_opt_a.loop_unroll                :     0.000048s :  0.10%
    jit_opt_a.a_1                        :     0.000966s :  1.97%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.00%
    jit_opt_a.specialize_transform       :     0.000017s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000022s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.03%
    jit_opt_a.merge_forward              :     0.000010s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000022s :  0.04%
    jit_opt_a.inline_without_move        :     0.000016s :  0.03%
    jit_opt_a.renormalize                :     0.001122s :  2.29%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000031s :  0.06%
    jit_opt_a.cse                        :     0.000097s :  0.20%
    jit_opt_a.replace_applicator         :     0.000027s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.03%
    rewriter_after_opt_a                 :     0.000286s :  0.58%
    convert_after_rewriter               :     0.000010s :  0.02%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000726s :  1.48%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.05%
    cconv                                :     0.000026s :  0.05%
    loop_unroll                          :     0.000487s :  0.99%
    jit_opt_after_cconv.c_1              :     0.000048s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.00%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000018s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000010s :  0.02%
    add_recomputation                    :     0.000064s :  0.13%
    cse_after_recomputation.cse          :     0.000021s :  0.04%
    auto_monad_reorder                   :     0.000027s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000504s :  1.03%
    symbol_engine_optimizer.build        :     0.000014s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000057s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012341s : 25.20%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000330    47
 2.87% :     0.000009s :      2: substitution.depend_value_elim
 0.97% :     0.000003s :      4: substitution.elim_not_effective
 0.68% :     0.000002s :      4: substitution.fold_const_symbol
 2.20% :     0.000007s :      5: substitution.graph_param_transform
79.12% :     0.000261s :      4: substitution.inline
 1.42% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.29% :     0.000008s :      8: substitution.remove_not_recompute_node
 1.39% :     0.000005s :      2: substitution.replace_old_param
 3.77% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
 2.75% :     0.000009s :      3: substitution.updatestate_pure_node_eliminater
 2.55% :     0.000008s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.029769     2
94.47% :     0.028122s :      1: type_inference.infer
 5.53% :     0.001647s :      1: type_inference.specialize
------[replace.]   0.000095     6
67.31% :     0.000064s :      4: replace.inline
32.69% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000269     6
95.79% :     0.000258s :      4: match.inline
 4.21% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000203  1021
 1.23% :     0.000002s :     15: predicate.accumulaten_eliminater
 0.92% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 0.97% :     0.000002s :     15: predicate.addn_check_dump
 1.35% :     0.000003s :     15: predicate.addn_zero_filter
 1.94% :     0.000004s :     15: predicate.arithmetic_simplify
 1.14% :     0.000002s :     15: predicate.cast_eliminate
 0.50% :     0.000001s :      5: predicate.check_bprop_eliminate
 0.95% :     0.000002s :     15: predicate.compare_switch_simplify
 1.46% :     0.000003s :     15: predicate.depend_value_elim
 1.01% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.10% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.76% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.30% :     0.000001s :      5: predicate.elim_not_effective
 0.51% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000002s :     15: predicate.environ_add_const_eliminate
 0.99% :     0.000002s :     15: predicate.environ_get_add_eliminate
 0.99% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     15: predicate.environ_get_eliminate
 1.03% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      5: predicate.fold_const_symbol
 1.15% :     0.000002s :     10: predicate.get_grad_eliminate
 0.27% :     0.000001s :      5: predicate.graph_param_transform
 4.48% :     0.000009s :     31: predicate.inline
 0.88% :     0.000002s :     10: predicate.inline_without_move
 0.36% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.15% :     0.000002s :     10: predicate.less_batch_normalization
 1.18% :     0.000002s :     17: predicate.list_to_tuple_eliminator_
 1.68% :     0.000003s :     22: predicate.load_eliminater
 1.26% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.10% :     0.000006s :     42: predicate.loop_unroll_before_grad
 1.65% :     0.000003s :     20: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000002s :     15: predicate.merge_addn
 0.97% :     0.000002s :     15: predicate.minmaximum_grad
 1.51% :     0.000003s :      5: predicate.mutable_eliminate
 0.40% :     0.000001s :      5: predicate.opt_reshape
 1.85% :     0.000004s :     22: predicate.partial_eliminate
 1.11% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.35% :     0.000003s :     15: predicate.reduce_eliminate
 1.37% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :     10: predicate.remove_not_recompute_node
 1.50% :     0.000003s :     27: predicate.replace_applicator
 0.47% :     0.000001s :     10: predicate.replace_old_param
 0.28% :     0.000001s :      5: predicate.reset_defer_inline
 1.09% :     0.000002s :     15: predicate.reshape_eliminate
 1.22% :     0.000002s :     15: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.14% :     0.000002s :     15: predicate.same_eliminate
 0.54% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 0.94% :     0.000002s :     10: predicate.special_op_eliminate
 0.95% :     0.000002s :     10: predicate.specialize_transform
 1.23% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 1.87% :     0.000004s :     21: predicate.switch_defer_inline
 1.75% :     0.000004s :     21: predicate.switch_layer_defer_inline
19.39% :     0.000039s :     68: predicate.switch_simplify
 1.12% :     0.000002s :     15: predicate.tile_eliminate
 1.12% :     0.000002s :     15: predicate.transpose_eliminate
 1.31% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 2.98% :     0.000006s :     27: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.27% :     0.000003s :     17: predicate.tuple_to_list_eliminator_
 1.51% :     0.000003s :     22: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000003s :     15: predicate.value_based_eliminate
 0.34% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000849    13
41.80% :     0.000355s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 6.68% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.53% :     0.000438s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.057212    76
 0.12% :     0.000068s :      1: add_recomputation
 0.28% :     0.000162s :      1: auto_monad
 0.05% :     0.000030s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.54% :     0.000309s :      1: bootstrap
 0.05% :     0.000029s :      1: cconv
 0.02% :     0.000013s :      1: convert_after_rewriter
 0.06% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000012s :      1: environ_conv
 0.04% :     0.000024s :      1: event_method
 0.02% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000013s :      1: graph_reusing
13.87% :     0.007933s :      1: jit_opt_a
 0.37% :     0.000210s :      1: jit_opt_after_cconv
 0.13% :     0.000077s :      1: jit_opt_b
 0.87% :     0.000496s :      1: loop_unroll
 1.28% :     0.000734s :      1: mutable_eliminate
 3.76% :     0.002149s :     26: opt.transform.jit_opt_a
 0.15% :     0.000086s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000046s :      4: opt.transform.jit_opt_b
 0.03% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000053s :      4: opt.transform.symbol_engine_opt
 0.90% :     0.000513s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000136s :      1: py_interpret_to_execute
 0.03% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.19% :     0.000683s :      1: renormalize.infer
 0.75% :     0.000431s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.51% :     0.000291s :      1: rewriter_after_opt_a
 0.18% :     0.000105s :      1: rewriter_before_opt_a
 0.20% :     0.000112s :      1: symbol_engine_optimizer
21.59% :     0.012354s :      1: task_emit
52.20% :     0.029863s :      1: type_inference
 0.14% :     0.000081s :      1: validate


TotalTime = 0.0489076, [33]
[bootstrap]: 0.0003382
[type_inference]: 0.026437
[event_method]: 2.158e-05
[auto_monad]: 0.00018687
[graph_reusing]: 7.25e-06
[pre_auto_parallel]: 3.28e-06
[py_interpret_to_execute]: 0.00015629
[rewriter_before_opt_a]: 0.00010621
[expand_dump_flag]: 3.83999e-06
[jit_opt_a]: 0.00732906, [2]
    [Cycle 1]: 0.00257742, [27]
        [switch_simplify]: 7.589e-05
        [loop_unroll]: 3.609e-05
        [a_1]: 0.00075982
        [with_stream_mark]: 2.379e-05
        [recompute_prepare]: 1.406e-05
        [updatestate_depend_eliminate]: 6.79999e-06
        [updatestate_assign_eliminate]: 6.05002e-06
        [updatestate_loads_eliminate]: 4.74e-06
        [parameter_eliminate]: 2.05002e-06
        [specialize_transform]: 1.023e-05
        [updatestate_useless_node_eliminater]: 1.233e-05
        [accelerated_algorithm]: 8.69e-06
        [meta_shard_fg_expand]: 3.24001e-06
        [get_grad_eliminate_]: 8.22e-06
        [merge_forward]: 5.41998e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.029e-05
        [j_node_and_user_rematch]: 1.479e-05
        [meta_fg_expand]: 3.51999e-06
        [replace_old_param]: 1.217e-05
        [inline_without_move]: 8.30999e-06
        [renormalize]: 0.00124013
        [add_forward_monad_depend]: 1.043e-05
        [auto_monad_grad]: 2.58998e-06
        [auto_monad_eliminator]: 2.237e-05
        [cse]: 4.153e-05
        [replace_applicator]: 2.194e-05
    [Cycle 2]: 0.00058053, [27]
        [switch_simplify]: 9.81e-06
        [loop_unroll]: 8.39998e-06
        [a_1]: 0.00017499
        [with_stream_mark]: 1.576e-05
        [recompute_prepare]: 1.048e-05
        [updatestate_depend_eliminate]: 5.21998e-06
        [updatestate_assign_eliminate]: 5.17e-06
        [updatestate_loads_eliminate]: 4.59002e-06
        [parameter_eliminate]: 1.70001e-06
        [specialize_transform]: 9.34e-06
        [updatestate_useless_node_eliminater]: 1.13e-05
        [accelerated_algorithm]: 7.53e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 7.58001e-06
        [merge_forward]: 5.25999e-06
        [cell_reuse_recompute_pass]: 3.23998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.861e-05
        [j_node_and_user_rematch]: 1.285e-05
        [meta_fg_expand]: 2.71999e-06
        [replace_old_param]: 1.164e-05
        [inline_without_move]: 7.25e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 2.53003e-06
        [auto_monad_grad]: 2.44999e-06
        [auto_monad_eliminator]: 1.332e-05
        [cse]: 2.57e-05
        [replace_applicator]: 9.59e-06
[py_interpret_to_execute_after_opt_a]: 1.581e-05
[rewriter_after_opt_a]: 0.0003123
[convert_after_rewriter]: 1.353e-05
[order_py_execute_after_rewriter]: 7.68001e-06
[mutable_eliminate]: 0.00078884
[jit_opt_b]: 7.718e-05, [1]
    [Cycle 1]: 6.843e-05, [2]
        [frontend_op_eliminate]: 2.672e-05
        [inline_after_opt_a]: 2.74e-05
[cconv]: 3.117e-05
[loop_unroll]: 0.00057065
[jit_opt_after_cconv]: 0.00021584, [1]
    [Cycle 1]: 0.0002087, [11]
        [c_1]: 5.295e-05
        [parameter_eliminate]: 3.88999e-06
        [updatestate_depend_eliminate]: 9.56e-06
        [updatestate_assign_eliminate]: 4.74e-06
        [updatestate_loads_eliminate]: 4.3e-06
        [cse]: 3.253e-05
        [call_graph_tuple_transform]: 2.561e-05
        [tuple_list_get_item_eliminator]: 8.60999e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 9.92999e-06
[remove_dup_value]: 1.97e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 1.261e-05
[add_recomputation]: 6.825e-05
[cse_after_recomputation]: 3.39e-05, [1]
    [Cycle 1]: 2.669e-05, [1]
        [cse]: 1.957e-05
[auto_monad_reorder]: 2.573e-05
[get_jit_bprop_graph]: 3.31999e-06
[rewriter_after_jit_bprop_graph]: 6.47001e-06
[opt_after_jit_grad]: 0.00058115
[symbol_engine_optimizer]: 0.00011373, [1]
    [Cycle 1]: 0.00010587, [6]
        [build]: 1.597e-05
        [elim_shapecalc]: 1.315e-05
        [elim_not_effective]: 2.038e-05
        [opt_reshape]: 9.64999e-06
        [fold_const_symbol]: 1.426e-05
        [renormalize]: 1.04e-06
[validate]: 6.217e-05
[backend_pass]: 1.08001e-06
[task_emit]: 0.0110881
[execute]: 8.97e-06
Sums
    bootstrap                            :     0.000338s :  0.77%
    type_inference                       :     0.026437s : 60.15%
    event_method                         :     0.000022s :  0.05%
    auto_monad                           :     0.000187s :  0.43%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000156s :  0.36%
    rewriter_before_opt_a                :     0.000106s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000086s :  0.19%
    jit_opt_a.loop_unroll                :     0.000044s :  0.10%
    jit_opt_a.a_1                        :     0.000935s :  2.13%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000025s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000020s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.05%
    jit_opt_a.inline_without_move        :     0.000016s :  0.04%
    jit_opt_a.renormalize                :     0.001240s :  2.82%
    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      :     0.000036s :  0.08%
    jit_opt_a.cse                        :     0.000067s :  0.15%
    jit_opt_a.replace_applicator         :     0.000032s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.04%
    rewriter_after_opt_a                 :     0.000312s :  0.71%
    convert_after_rewriter               :     0.000014s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.000789s :  1.79%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.06%
    cconv                                :     0.000031s :  0.07%
    loop_unroll                          :     0.000571s :  1.30%
    jit_opt_after_cconv.c_1              :     0.000053s :  0.12%
    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 :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  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  :     0.000010s :  0.02%
    remove_dup_value                     :     0.000020s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.03%
    add_recomputation                    :     0.000068s :  0.16%
    cse_after_recomputation.cse          :     0.000020s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.06%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000581s :  1.32%
    symbol_engine_optimizer.build        :     0.000016s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000062s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011088s : 25.23%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000306    47
 3.56% :     0.000011s :      2: substitution.depend_value_elim
 0.98% :     0.000003s :      4: substitution.elim_not_effective
 0.75% :     0.000002s :      4: substitution.fold_const_symbol
 2.39% :     0.000007s :      5: substitution.graph_param_transform
74.97% :     0.000230s :      4: substitution.inline
 1.72% :     0.000005s :      8: substitution.j_node_and_user_rematch
 2.28% :     0.000007s :      8: substitution.remove_not_recompute_node
 1.83% :     0.000006s :      2: substitution.replace_old_param
 4.81% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
 3.61% :     0.000011s :      3: substitution.updatestate_pure_node_eliminater
 3.11% :     0.000010s :      5: substitution.updatestate_useless_node_eliminater
------[type_inference.]   0.026348     2
93.66% :     0.024677s :      1: type_inference.infer
 6.34% :     0.001671s :      1: type_inference.specialize
------[replace.]   0.000099     6
64.69% :     0.000064s :      4: replace.inline
35.31% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     6
94.36% :     0.000227s :      4: match.inline
 5.64% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000181  1021
 1.45% :     0.000003s :     15: predicate.accumulaten_eliminater
 1.09% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     15: predicate.addn_check_dump
 1.31% :     0.000002s :     15: predicate.addn_zero_filter
 2.20% :     0.000004s :     15: predicate.arithmetic_simplify
 1.32% :     0.000002s :     15: predicate.cast_eliminate
 0.47% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.28% :     0.000002s :     15: predicate.compare_switch_simplify
 1.29% :     0.000002s :     15: predicate.depend_value_elim
 1.15% :     0.000002s :     15: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     15: predicate.dict_get_item_eliminator
 1.09% :     0.000002s :     15: predicate.dict_set_item_eliminator
 0.91% :     0.000002s :      5: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      5: predicate.elim_not_effective
 0.61% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     15: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     15: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     15: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     15: predicate.environ_get_eliminate
 1.15% :     0.000002s :     15: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      5: predicate.fold_const_symbol
 1.07% :     0.000002s :     10: predicate.get_grad_eliminate
 0.39% :     0.000001s :      5: predicate.graph_param_transform
 5.19% :     0.000009s :     31: predicate.inline
 1.03% :     0.000002s :     10: predicate.inline_without_move
 0.40% :     0.000001s :     10: predicate.j_node_and_user_rematch
 1.09% :     0.000002s :     10: predicate.less_batch_normalization
 1.52% :     0.000003s :     17: predicate.list_to_tuple_eliminator_
 1.98% :     0.000004s :     22: predicate.load_eliminater
 1.57% :     0.000003s :      5: predicate.loop_unroll_after_grad
 3.30% :     0.000006s :     42: predicate.loop_unroll_before_grad
 2.24% :     0.000004s :     20: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     15: predicate.merge_addn
 1.13% :     0.000002s :     15: predicate.minmaximum_grad
 1.78% :     0.000003s :      5: predicate.mutable_eliminate
 0.71% :     0.000001s :      5: predicate.opt_reshape
 2.34% :     0.000004s :     22: predicate.partial_eliminate
 1.17% :     0.000002s :     15: predicate.print_const_string_wrapper
 1.50% :     0.000003s :     15: predicate.reduce_eliminate
 1.62% :     0.000003s :     17: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     10: predicate.remove_not_recompute_node
 2.18% :     0.000004s :     27: predicate.replace_applicator
 0.72% :     0.000001s :     10: predicate.replace_old_param
 0.33% :     0.000001s :      5: predicate.reset_defer_inline
 1.14% :     0.000002s :     15: predicate.reshape_eliminate
 1.50% :     0.000003s :     15: predicate.row_tensor_add_zeros_like
 0.90% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     15: predicate.same_eliminate
 0.81% :     0.000001s :     10: predicate.set_cell_output_no_recompute
 1.04% :     0.000002s :     10: predicate.special_op_eliminate
 1.06% :     0.000002s :     10: predicate.specialize_transform
 1.38% :     0.000002s :     15: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000003s :     15: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.15% :     0.000004s :     21: predicate.switch_defer_inline
 1.92% :     0.000003s :     21: predicate.switch_layer_defer_inline
 6.79% :     0.000012s :     68: predicate.switch_simplify
 1.22% :     0.000002s :     15: predicate.tile_eliminate
 1.20% :     0.000002s :     15: predicate.transpose_eliminate
 1.47% :     0.000003s :     15: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     15: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000007s :     27: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000003s :     15: predicate.tuple_list_set_item_eliminator
 1.34% :     0.000002s :     17: predicate.tuple_to_list_eliminator_
 2.01% :     0.000004s :     22: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000006s :     32: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000003s :     15: predicate.value_based_eliminate
 0.40% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000913    13
44.79% :     0.000409s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 4.92% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.29% :     0.000459s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.051628    76
 0.14% :     0.000072s :      1: add_recomputation
 0.37% :     0.000193s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000358s :      1: bootstrap
 0.07% :     0.000034s :      1: cconv
 0.03% :     0.000017s :      1: convert_after_rewriter
 0.07% :     0.000037s :      1: cse_after_recomputation
 0.03% :     0.000015s :      1: environ_conv
 0.05% :     0.000027s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000006s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
14.20% :     0.007333s :      1: jit_opt_a
 0.42% :     0.000219s :      1: jit_opt_after_cconv
 0.16% :     0.000080s :      1: jit_opt_b
 1.13% :     0.000583s :      1: loop_unroll
 1.55% :     0.000801s :      1: mutable_eliminate
 2.46% :     0.001269s :     26: opt.transform.jit_opt_a
 0.18% :     0.000093s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000046s :      4: opt.transform.jit_opt_b
 0.04% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000036s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000592s :      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
 0.31% :     0.000161s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000022s :      1: remove_dup_value
 1.43% :     0.000740s :      1: renormalize.infer
 0.95% :     0.000489s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.62% :     0.000320s :      1: rewriter_after_opt_a
 0.21% :     0.000111s :      1: rewriter_before_opt_a
 0.23% :     0.000117s :      1: symbol_engine_optimizer
21.50% :     0.011101s :      1: task_emit
51.25% :     0.026457s :      1: type_inference
 0.17% :     0.000090s :      1: validate


op_name: mint.clamp, mode:pynative, test_op_reference...

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

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

op_name: mint.cat, mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="trilinear"), mode:pynative, test_op_reference...

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

op_name: mint.nn.BatchNorm2d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.l1_loss, mode:pynative, test_op_reference...

op_name: mint.nn.functional.mse_loss, mode:pynative, test_op_reference...

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

op_name: mint.nonzero, mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="bilinear"), mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="trilinear"), mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="bicubic"), mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="linear"), mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="nearest")-2d, mode:pynative, test_op_reference...

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

op_name: mint.select, mode:pynative, test_op_reference...

op_name: mint.matmul, mode:pynative, test_op_reference...

op_name: mint.nn.functional.log_softmax, mode:pynative, test_op_reference...

op_name: mint.nn.functional.softmax, mode:pynative, test_op_reference...

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

op_name: mint.chunk, mode:kbk, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="nearest")-1d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="nearest")-3d, mode:pynative, test_op_reference...

op_name: mint.repeat_interleave, mode:pynative, test_op_reference...

op_name: mint.arange, mode:pynative, test_op_reference...

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

op_name: mint.broadcast_to, mode:pynative, test_op_reference...

op_name: mint.nn.BatchNorm1d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.batch_norm, mode:pynative, test_op_reference...

op_name: mint.nn.functional.binary_cross_entropy, mode:pynative, test_op_reference...

op_name: mint.nn.functional.binary_cross_entropy_with_logits, mode:pynative, test_op_reference...

op_name: mint.nn.functional.l1_loss, mode:pynative, test_op_reference...

op_name: mint.nn.functional.mse_loss, mode:pynative, test_op_reference...

op_name: mint.unique, mode:pynative, test_op_reference...

op_name: mint.nn.functional.linear, mode:pynative, test_op_reference...

op_name: mint.gather, mode:kbk, test_op_reference...

op_name: mint.stack, mode:pynative, test_op_reference...

op_name: mint.scatter, mode:pynative, test_op_reference...

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

op_name: mint.nn.functional.interpolate(mode="nearest")-3d, mode:pynative, test_op_reference...

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

op_name: mint.nn.functional.batch_norm, mode:pynative, test_op_reference...

op_name: mint.nn.functional.log_softmax, mode:pynative, test_op_reference...

op_name: mint.unique, mode:pynative, test_op_reference...

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

op_name: mint.nn.BatchNorm3d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.conv1d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.conv3d, mode:pynative, test_op_reference...

op_name: mint.nn.Linear, mode:pynative, test_op_reference...

op_name: mint.nn.Conv1d, mode:pynative, test_op_reference...

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

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

op_name: mint.nn.functional.layer_norm, mode:pynative, test_op_reference...

op_name: mint.permute, mode:pynative, test_op_reference...

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

op_name: mint.nonzero, mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="nearest")-1d, mode:pynative, test_op_reference...

op_name: mint.select, mode:pynative, test_op_reference...

op_name: mint.nn.BatchNorm1d, mode:pynative, test_op_reference...

op_name: mint.nn.BatchNorm3d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.binary_cross_entropy, mode:pynative, test_op_reference...

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

op_name: mint.stack, mode:pynative, test_op_reference...

op_name: mint.nn.functional.dropout, mode:pynative, test_op_reference...

op_name: mint.nn.functional.pad(mode="constant"), mode:pynative, test_op_reference...

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

op_name: mint.flatten, mode:pynative, test_op_reference...

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

op_name: mint.nn.functional.dropout, mode:pynative, test_op_reference...

op_name: mint.nn.functional.pad(mode="constant"), mode:pynative, test_op_reference...

op_name: mint.transpose, mode:pynative, test_op_reference...

op_name: mint.topk, mode:pynative, test_op_reference...

op_name: mint.normal, mode:pynative, test_op_reference...

op_name: mint.nn.functional.adaptive_avg_pool1d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.adaptive_avg_pool2d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.adaptive_avg_pool3d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.avg_pool2d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.max_pool2d, mode:pynative, test_op_reference...

op_name: mint.nn.AdaptiveAvgPool2d, mode:pynative, test_op_reference...

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

op_name: mint.chunk, mode:pynative, test_op_reference...

op_name: mint.cumsum, mode:pynative, test_op_reference...

op_name: mint.index_select, mode:pynative, test_op_reference...

op_name: mint.masked_select, mode:pynative, test_op_reference...

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

op_name: mint.nn.functional.interpolate(mode="bilinear"), mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="linear"), mode:pynative, test_op_reference...

op_name: mint.repeat_interleave, mode:pynative, test_op_reference...

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

op_name: mint.broadcast_to, mode:pynative, test_op_reference...

op_name: mint.matmul, mode:pynative, test_op_reference...

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

op_name: mint.nn.GroupNorm, mode:pynative, test_op_reference...

op_name: mint.permute, mode:pynative, test_op_reference...

op_name: mint.topk, mode:pynative, test_op_reference...

op_name: mint.normal, mode:pynative, test_op_reference...

op_name: mint.nn.functional.adaptive_avg_pool2d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.max_pool2d, mode:pynative, test_op_reference...

op_name: mint.nn.AdaptiveAvgPool2d, mode:pynative, test_op_reference...

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

op_name: mint.gather, mode:kbk, test_op_reference...

op_name: mint.nn.BatchNorm2d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.conv2d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.linear, mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="bicubic"), mode:pynative, test_op_reference...

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

op_name: mint.nn.functional.pad(mode="reflect"), mode:pynative, test_op_reference...

op_name: mint.nn.functional.pad(mode="replicate"), mode:pynative, test_op_reference...

op_name: mint.nn.functional.pad(mode="circular"), mode:pynative, test_op_reference...

op_name: mint.nn.functional.conv1d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.conv3d, mode:pynative, test_op_reference...

op_name: mint.nn.Linear, mode:pynative, test_op_reference...

op_name: mint.nn.Conv1d, mode:pynative, test_op_reference...

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

op_name: mint.nn.functional.layer_norm, mode:pynative, test_op_reference...

op_name: mint.transpose, mode:pynative, test_op_reference...

op_name: mint.nn.functional.adaptive_avg_pool1d, mode:pynative, test_op_reference...

op_name: mint.nn.functional.adaptive_avg_pool3d, mode:pynative, test_op_reference...

op_name: mint.nn.AdaptiveAvgPool3d, mode:pynative, test_op_reference...

op_name: mint.chunk, mode:kbk, test_op_reference...

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

op_name: mint.nn.functional.one_hot, mode:pynative, test_op_reference...

op_name: mint.reshape, mode:pynative, test_op_reference...

op_name: mint.nn.functional.pad(mode="reflect"), mode:pynative, test_op_reference...

op_name: mint.nn.functional.pad(mode="replicate"), mode:pynative, test_op_reference...

op_name: mint.nn.functional.pad(mode="circular"), mode:pynative, test_op_reference...

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

op_name: mint.nn.GroupNorm, mode:pynative, test_op_reference...

op_name: mint.nn.LayerNorm, mode:pynative, test_op_reference...

op_name: mint.unsqueeze, mode:pynative, test_op_reference...

op_name: mint.nn.AdaptiveAvgPool3d, mode:pynative, test_op_reference...

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

op_name: mint.gather, mode:pynative, test_op_reference...

op_name: mint.split, mode:pynative, test_op_reference...

op_name: mint.flatten, mode:pynative, test_op_reference...

op_name: mint.nn.functional.interpolate(mode="nearest")-2d, mode:pynative, test_op_reference...

op_name: mint.arange, mode:pynative, test_op_reference...

op_name: mint.nn.functional.one_hot, mode:pynative, test_op_reference...

op_name: mint.reshape, mode:pynative, test_op_reference...

op_name: mint.nn.functional.binary_cross_entropy_with_logits, mode:pynative, test_op_reference...

op_name: mint.nn.functional.softmax, mode:pynative, test_op_reference...

op_name: mint.clamp, mode:pynative, test_op_reference...

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

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

op_name: mint.scatter, mode:pynative, test_op_reference...

op_name: mint.nn.functional.conv2d, mode:pynative, test_op_reference...

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

op_name: mint.nn.LayerNorm, mode:pynative, test_op_reference...

op_name: mint.unsqueeze, mode:pynative, test_op_reference...

op_name: mint.nn.functional.avg_pool2d, mode:pynative, test_op_reference...

op_name: Tensor.transpose, mode:pynative, test_op_reference...
group_cases_0 have all been run, results of sub cases are below:
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.cat') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.split') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.index_select') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nonzero') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.interpolate(mode="bilinear")') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.interpolate(mode="trilinear")') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.interpolate(mode="bicubic")') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.chunk') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.interpolate(mode="linear")') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.interpolate(mode="nearest")-1d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.interpolate(mode="nearest")-3d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.interpolate(mode="nearest")-2d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.index_select') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.repeat_interleave') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.repeat_interleave') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.arange') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.repeat') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.one_hot') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.masked_select') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.select') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.select') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.broadcast_to') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.cumsum') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.matmul') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.BatchNorm1d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.batch_norm') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.binary_cross_entropy') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.binary_cross_entropy_with_logits') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.matmul') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.l1_loss') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.log_softmax') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.mse_loss') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.softmax') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.split') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.clamp') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.gather') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.clamp') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.flatten') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.stack') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.expand_as') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.dropout') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.reshape') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.pad(mode="reflect")') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.pad(mode="replicate")') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.BatchNorm3d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.BatchNorm2d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.conv1d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.conv2d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.conv3d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.Linear') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.Conv1d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.pad(mode="circular")') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.masked_scatter') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.add_') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.masked_scatter_') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.GroupNorm') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.layer_norm') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.pad(mode="constant")') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.LayerNorm') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.transpose') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.topk') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.normal') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.adaptive_avg_pool1d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.adaptive_avg_pool2d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.adaptive_avg_pool3d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.avg_pool2d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.max_pool2d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.unsqueeze') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.AdaptiveAvgPool2d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.scatter') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.AdaptiveAvgPool3d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.transpose') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.topk') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.gather') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.chunk') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.cumsum') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.permute') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.index_select') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.gather') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.masked_select') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.index_select') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.split') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.split') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.cat') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.nn.functional.linear') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.interpolate(mode="bilinear")') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nonzero') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.flatten') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.unsqueeze') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.interpolate(mode="linear")') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.interpolate(mode="nearest")-3d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.interpolate(mode="trilinear")') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.interpolate(mode="nearest")-2d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.interpolate(mode="nearest")-1d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.repeat_interleave') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.arange') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.one_hot') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.interpolate(mode="bicubic")') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.select') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.broadcast_to') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.select') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.repeat') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.BatchNorm1d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.repeat_interleave') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.BatchNorm3d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.reshape') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.BatchNorm2d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.batch_norm') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.l1_loss') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.binary_cross_entropy_with_logits') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.log_softmax') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.mse_loss') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.softmax') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.binary_cross_entropy') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.clamp') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.clamp') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.gather') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.expand_as') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.stack') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.dropout') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.matmul') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.pad(mode="reflect")') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.pad(mode="replicate")') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.pad(mode="circular")') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.scatter') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.conv1d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'mint.unique') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.conv3d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.Linear') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.conv2d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.Conv1d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.matmul') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.masked_scatter_') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.add_') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.masked_scatter') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.GroupNorm') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.layer_norm') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.LayerNorm') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.permute') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.topk') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.transpose') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.normal') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.adaptive_avg_pool1d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.unsqueeze') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.adaptive_avg_pool2d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.max_pool2d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.adaptive_avg_pool3d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.avg_pool2d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.pad(mode="constant")') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.AdaptiveAvgPool2d') {} pass.
case: <function test_other_op_reference_forward at 0xffff979c13a0> ('pynative', 'Tensor.unique') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.topk') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.AdaptiveAvgPool3d') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.nn.functional.linear') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.transpose') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.unsqueeze') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'mint.unique') {} pass.
case: <function test_other_op_reference_backward_kbk at 0xffff979c1550> ('kbk', 'mint.gather') {} pass.
case: <function test_other_op_reference_backward_kbk at 0xffff979c1550> ('kbk', 'mint.chunk') {} pass.
case: <function test_other_op_reference_forward_kbk at 0xffff979c14c0> ('kbk', 'mint.gather') {} pass.
case: <function test_other_op_reference_backward at 0xffff979c1430> ('pynative', 'Tensor.unique') {} pass.
case: <function test_other_op_reference_forward_kbk at 0xffff979c14c0> ('kbk', 'mint.chunk') {} pass.

.
[hook] pytest_runtest_teardown:test_other_op_group_case_level0

tests/st/ops/allcases_onecard/test_ops_group_cases.py::test_other_op_group_case_level0,max_mem:2.0M


=============================== 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
================== 1 passed, 25 warnings in 571.37s (0:09:31) ==================
[LOG_WARNING] can not open file, file: /home/jenkins/ascend/log/debug/plog/plog-169938_20260129173655645.log, possible reason: Permission denied.