Inplace mul check_graph=True unittest failed

MARD1NO commented 2 years ago

@flow.unittest.skip_unless_1n1d()
    @autotest(check_graph=True)
    def test_mul_inplace_tensor(test_case):
        device = random_device()
        rand_tensor = random_pytorch_tensor(low=-2, high=2, ndim=4, dim0=constant(16), dim1=constant(9), dim2=constant(4), dim3=constant(7)).to(device)
        y = rand_tensor + 1
        x = random_pytorch_tensor(low=-2, high=2, ndim=4, dim0=constant(16), dim1=constant(9), dim2=constant(4), dim3=constant(7)).to(device)
        y.mul_(
            x
        )
        return y

Tensor([16, 9, 4, 7]).to(cuda)
Tensor([16, 9, 4, 7]).__add__(1)
Tensor([16, 9, 4, 7]).to(cuda)
Tensor([16, 9, 4, 7]).mul_(Tensor([16, 9, 4, 7]))
-----------------------------------------------------------
This program has 0 input tensor: 
E..............................s..................s..............sss....
======================================================================
ERROR: test_mul_inplace_tensor (__main__.TestTensor)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/zzk/oneflow/python/oneflow/test_utils/automated_test_util/torch_flow_dual_object.py", line 384, in dual_method
    test_g_res = test_g()
  File "/home/zzk/oneflow/python/oneflow/nn/graph/graph.py", line 258, in __call__
    self._compile(*args)
  File "/home/zzk/oneflow/python/oneflow/nn/graph/graph.py", line 506, in _compile
    eager_outputs = self._build_graph(*args)
  File "/home/zzk/oneflow/python/oneflow/nn/graph/graph.py", line 614, in _build_graph
    ) = self._build_io("output", graph_build_util.build_graph_output, *outputs)
  File "/home/zzk/oneflow/python/oneflow/nn/graph/graph.py", line 818, in _build_io
    build_args.append(build_tensor_or_none(arg, name, repr_str))
  File "/home/zzk/oneflow/python/oneflow/nn/graph/graph.py", line 798, in build_tensor_or_none
    build_arg = build_func(name, tensor)
  File "/home/zzk/oneflow/python/oneflow/framework/graph_build_util.py", line 187, in build_graph_output
    assert out.is_lazy
AssertionError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/zzk/oneflow/python/oneflow/test_utils/automated_test_util/torch_flow_dual_object.py", line 703, in new_f
    res = f(test_case)
  File "test_tensor.py", line 482, in test_mul_inplace_tensor
    x
  File "/home/zzk/oneflow/python/oneflow/test_utils/automated_test_util/torch_flow_dual_object.py", line 387, in dual_method
    raise OneFlowGraphBuildOrRunError(e)
oneflow.test_utils.automated_test_util.torch_flow_dual_object.OneFlowGraphBuildOrRunError: OneFlow nn.Graph Build Or Run Error: 

----------------------------------------------------------------------
Ran 146 tests in 18.178s

FAILED (errors=1, skipped=8)

strint commented 2 years ago

你的测试分支有合并这个pr不：https://github.com/Oneflow-Inc/oneflow/pull/7254

strint commented 2 years ago

Inplace的测试比较特殊，它把输入改掉了，还依赖autotest的一个改动的PR合并：https://github.com/Oneflow-Inc/oneflow/pull/7276

Oneflow-Inc / oneflow

Inplace mul check_graph=True unittest failed #7299