Traceback (most recent call last): File "/opt/conda/envs/py_3.10/lib/python3.10/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_3.10/lib/python3.10/site-packages/torch/testing/_internal/optests/generate_tests.py", line 61, in safe_schema_check result = op(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_ops.py", line 871, in call return self._op(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_compile.py", line 54, in inner return disable_fn(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/eval_frame.py", line 1281, in _fn return fn(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_subclasses/schema_check_mode.py", line 184, in torch_dispatch out = func(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_ops.py", line 871, in call return self._op(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/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_3.10/lib/python3.10/site-packages/torch/_compile.py", line 54, in inner return disable_fn(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/eval_frame.py", line 1281, in _fn return fn(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_library/custom_ops.py", line 410, in wrapped_fn return fn(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/testing/_internal/custom_op_db.py", line 320, in numpy_split_copy x_np = to_numpy(x) File "/opt/conda/envs/py_3.10/lib/python3.10/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_3.10/lib/python3.10/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_3.10/lib/python3.10/site-packages/torch/library.py", line 1749, in opcheck return optests.opcheck( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/testing/_internal/optests/generate_tests.py", line 699, in opcheck raise OpCheckError( 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_3.10/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 3444, in wrapper method(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/testing/_internal/common_device_type.py", line 439, in instantiated_test result = test(self, **param_kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 1821, in wrapper fn(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/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, 9), device="cpu", dtype=torch.float32], args=((1,3,6),1), 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_NumpySplitCopyCustomOp_cpu_float32

This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0

TL;DR

The most likely fix is to modify the to_numpy function to use tensor.detach().numpy() instead of tensor.cpu().numpy() when the tensor requires grad.

Guidance

• Identify the to_numpy function in torch/testing/_internal/custom_op_db.py and modify it to check if the tensor requires grad before calling numpy().
• Use tensor.detach().numpy() when the tensor requires grad to avoid the RuntimeError.
• Verify the fix by running the test with the modified to_numpy function and checking if the error is resolved.
• Review the test case and ensure that the tensor is not expected to require grad in the test scenario.

Example

def to_numpy(x):
    if x.requires_grad:
        return x.detach().numpy()
    else:
        return x.cpu().numpy()

Notes

The provided fix assumes that the tensor requiring grad is not expected to be modified in the test scenario. If the tensor is expected to require grad, additional modifications may be necessary to handle this case.

Recommendation

Apply the workaround by modifying the to_numpy function to use tensor.detach().numpy() when the tensor requires grad, as this should resolve the immediate error and allow the test to run successfully.