pytorch - 💡(How to fix) Fix DISABLED test_ifft_norm_backward (__main__.TestFFT1D) [2 comments, 1 participants]

Traceback (most recent call last): File "/opt/conda/envs/py_3.10/lib/python3.10/unittest/case.py", line 59, in testPartExecutor yield File "/opt/conda/envs/py_3.10/lib/python3.10/unittest/case.py", line 591, in run self._callTestMethod(testMethod) File "/opt/conda/envs/py_3.10/lib/python3.10/unittest/case.py", line 549, in _callTestMethod method() File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 3528, in wrapper method(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 603, in instantiated_test test(self, **param_kwargs) File "/__w/pytorch/pytorch/test/torch_np/numpy_tests/fft/test_pocketfft.py", line 73, in test_ifft assert_allclose(x, np.fft.ifft(np.fft.fft(x, norm=norm), norm=norm), atol=1e-6) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 2506, in call result = self._torchdynamo_orig_backend( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 2217, in call result = self._inner_convert( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 772, in call result = _compile( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 2001, in _compile guarded_code, tracer_output = compile_inner(code, one_graph, hooks) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_utils_internal.py", line 96, in wrapper_function return function(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 1611, in compile_inner result = _compile_inner(code, one_graph, hooks) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 1670, in _compile_inner dynamo_output = compile_frame( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 1516, in compile_frame bytecode, tracer_output = transform_code_object(code, transform) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/bytecode_transformation.py", line 1626, in transform_code_object tracer_output = transformations(instructions, code_options) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 1488, in transform tracer_output = trace_frame( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 367, in _fn return fn(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 949, in trace_frame run_tracer() File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/convert_frame.py", line 930, in run_tracer tracer.run() File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/symbolic_convert.py", line 1809, in run while self.step(): File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/symbolic_convert.py", line 1476, in step self.dispatch_table[inst.opcode](self, inst) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/symbolic_convert.py", line 1013, in wrapper return inner_fn(self, inst) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/symbolic_convert.py", line 2882, in CALL_FUNCTION_KW self.call_function(fn, args, kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/symbolic_convert.py", line 1377, in call_function self.push(fn.call_function(self, args, kwargs)) # type: ignore[arg-type] File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/variables/lazy.py", line 476, in realize_and_forward return getattr(self.realize(), name)(args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/variables/misc.py", line 1831, in call_function return NumpyNdarrayVariable.create(tx, proxy) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/variables/tensor.py", line 2353, in create return wrap_fx_proxy_cls( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/variables/builder.py", line 3195, in wrap_fx_proxy_cls out: VTTypeAlias = _wrap_fx_proxy( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/variables/builder.py", line 3319, in _wrap_fx_proxy example_value = get_fake_value(proxy.node, tx, allow_non_graph_fake=True) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/utils.py", line 3761, in get_fake_value return _get_fake_value_impl(node, tx, allow_non_graph_fake) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/utils.py", line 3952, in _get_fake_value_impl _wrap_graph_break_with_torch_runtime_err( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/utils.py", line 3750, in _wrap_graph_break_with_torch_runtime_err raise exc.with_traceback(e.traceback) from None File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/utils.py", line 3747, in _wrap_graph_break_with_torch_runtime_err gb_fn() File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/utils.py", line 3953, in <lambda> lambda: unimplemented( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_dynamo/exc.py", line 658, in unimplemented raise Unsupported( torch._dynamo.exc.TorchRuntimeError: RuntimeError when making fake tensor call Explanation: Dynamo failed to run FX node with fake tensors: call_function <Wrapped function <original ifft>>(([],), **{'norm': 'backward'}): got RuntimeError('Invalid number of data points (0) specified') Hint: Your code may result in an error when running in eager. Please double check that your code doesn't contain a similar error when actually running eager/uncompiled. You can do this by removing the torch.compile call, or by using torch.compiler.set_stance("force_eager").

Developer debug context:

For more details about this graph break, please visit: https://meta-pytorch.github.io/compile-graph-break-site/gb/gb4315.html

from user code: File "/__w/pytorch/pytorch/test/torch_np/numpy_tests/fft/test_pocketfft.py", line 78, in torch_dynamo_resume_in_test_ifft_at_73 np.fft.ifft([], norm=norm)

Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"

To execute this test, run the following from the base repo dir: PYTORCH_TEST_WITH_DYNAMO=1 python test/torch_np/numpy_tests/fft/test_pocketfft.py TestFFT1D.test_ifft_norm_backward

This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0

TL;DR

The most likely fix is to modify the test case to handle the edge case where the input array is empty, causing the np.fft.ifft function to fail.

Guidance

• Review the test case in test_pocketfft.py to ensure it handles empty input arrays correctly.
• Consider adding a check for empty input arrays before calling np.fft.ifft to prevent the RuntimeError.
• Use the provided debugging instructions to investigate the issue further and gather more information about the error.
• Set TORCHDYNAMO_VERBOSE=1 and TORCH_LOGS="+dynamo" to get more detailed internal stack traces and logs.

Example

No specific code example is provided, but the test case in question is test_ifft_norm_backward in test_pocketfft.py, which can be modified to handle empty input arrays.

Notes

The error message suggests that the issue is related to the np.fft.ifft function being called with an empty input array, which is not a valid input. The test case should be modified to handle this edge case correctly.

Recommendation

Apply a workaround by modifying the test case to handle empty input arrays correctly, as the root cause of the issue is related to the input data rather than a bug in the PyTorch library.