pytorch - 💡(How to fix) Fix DISABLED test_opcheck_opinfo_NumpyViewCopyCustomOp_cpu_float32 (__main__.TestCustomOpTestingCPU) [2 comments, 1 participants]

Traceback (most recent call last): File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/testing/_internal/optests/generate_tests.py", line 695, in opcheck tester(op, args, kwargs, rtol=rtol, atol=atol) ~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/testing/_internal/optests/generate_tests.py", line 61, in safe_schema_check result = op(*args, **kwargs) File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/_ops.py", line 871, in call return self._op(*args, **kwargs) ~~~~~~~~^^^^^^^^^^^^^^^^^ File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/_compile.py", line 54, in inner return disable_fn(*args, **kwargs) File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/_dynamo/eval_frame.py", line 1281, in _fn return fn(*args, **kwargs) File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/_subclasses/schema_check_mode.py", line 184, in torch_dispatch out = func(*args, **kwargs) File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/_ops.py", line 871, in call return self._op(*args, **kwargs) ~~~~~~~~^^^^^^^^^^^^^^^^^ File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/_library/custom_ops.py", line 375, in backend_impl result = self._backend_fns[device_type](*args, **kwargs) File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/_compile.py", line 54, in inner return disable_fn(*args, **kwargs) File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/_dynamo/eval_frame.py", line 1281, in _fn return fn(*args, **kwargs) File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/_library/custom_ops.py", line 410, in wrapped_fn return fn(*args, **kwargs) File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/testing/_internal/custom_op_db.py", line 238, in numpy_view_copy return torch.tensor(np.copy(to_numpy(x).reshape(shape)), device=x.device) ~~~~~~~~^^^ File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/testing/_internal/custom_op_db.py", line 29, in to_numpy return tensor.cpu().numpy() ~~~~~~~~~~~~~~~~~~^^ RuntimeError: Can't call numpy() on Tensor that requires grad. Use tensor.detach().numpy() instead.

The above exception was the direct cause of the following exception:

Traceback (most recent call last): File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/testing/_internal/common_device_type.py", line 1175, in test_wrapper return test(*args, **kwargs) File "/__w/pytorch/pytorch/test/test_custom_ops.py", line 444, in test_opcheck_opinfo torch.library.opcheck(op.op, args, kwargs) ~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^ File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/library.py", line 1749, in opcheck return optests.opcheck( ~~~~~~~~~~~~~~~^ op, ^^^ ...<5 lines>... atol=atol, ^^^^^^^^^^ ) ^ File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/testing/_internal/optests/generate_tests.py", line 699, in opcheck raise OpCheckError( ...<2 lines>... ) from ex torch.testing._internal.optests.generate_tests.OpCheckError: opcheck(op, ...): test_schema failed with Can't call numpy() on Tensor that requires grad. Use tensor.detach().numpy() instead. (scroll up for stack trace)

The above exception was the direct cause of the following exception:

Traceback (most recent call last): File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/testing/_internal/common_utils.py", line 3444, in wrapper method(*args, **kwargs) ~~~~~~^^^^^^^^^^^^^^^^^ File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/testing/_internal/common_device_type.py", line 439, in instantiated_test result = test(self, **param_kwargs) File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/testing/_internal/common_utils.py", line 1821, in wrapper fn(*args, **kwargs) ~~^^^^^^^^^^^^^^^^^ File "/opt/conda/envs/py_3.14/lib/python3.14t/site-packages/torch/testing/_internal/common_device_type.py", line 1187, in test_wrapper raise e_tracked from e Exception: opcheck(op, ...): test_schema failed with Can't call numpy() on Tensor that requires grad. Use tensor.detach().numpy() instead. (scroll up for stack trace)

Caused by sample input at index 0: SampleInput(input=Tensor[size=(2, 3, 4), device="cpu", dtype=torch.float32], args=((2,12)), kwargs={}, broadcasts_input=False, name='')

To execute this test, run the following from the base repo dir: PYTORCH_OPINFO_SAMPLE_INPUT_INDEX=0 python test/test_custom_ops.py TestCustomOpTestingCPU.test_opcheck_opinfo_NumpyViewCopyCustomOp_cpu_float32

This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0

TL;DR

The most likely fix is to use tensor.detach().numpy() instead of tensor.numpy() when calling numpy() on a Tensor that requires grad.

Guidance

• Identify the line of code causing the error: return tensor.cpu().numpy() in to_numpy function.
• Replace tensor.cpu().numpy() with tensor.detach().cpu().numpy() to detach the tensor before converting it to a numpy array.
• Verify the fix by re-running the test with the modified code.
• If the issue persists, investigate other potential causes of the error, such as incorrect tensor creation or modification.

Example

def to_numpy(tensor):
    return tensor.detach().cpu().numpy()

Notes

The error message explicitly states that tensor.detach().numpy() should be used instead of tensor.numpy() when the tensor requires grad. This suggests that the fix is to modify the to_numpy function to detach the tensor before converting it to a numpy array.

Recommendation

Apply the workaround by modifying the to_numpy function to use tensor.detach().cpu().numpy() instead of tensor.cpu().numpy(). This should fix the error and allow the test to pass.