[torchlib] Fix reflection pad

[justinchuby]

Fixes pytorch/pytorch#144382

[codecov]

❌ 27 Tests Failed:

Tests completed	Failed	Passed	Skipped
10993	27	10966	1862

View the top 3 failed test(s) by shortest run time

onnxscript.backend.onnx_export_test.TestOnnxBackEnd::test_export2python_produces_correct_onnx_script_model_0907_test_reduce_sum_empty_axes_input_noop_example

Stack Traces | 0.003s run time

onnxscript\backend\onnx_export_test.py:137: in extract_functions
    mod = importlib.import_module(import_name)
C:\hostedtoolcache\windows\Python\3.10.11\x64\lib\importlib\__init__.py:126: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
E   ModuleNotFoundError: No module named 'tests.onnx_backend_test_code.test_reduce_sum_empty_axes_input_noop_example'

The above exception was the direct cause of the following exception:
.nox\test\lib\site-packages\parameterized\parameterized.py:620: in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
onnxscript\backend\onnx_export_test.py:271: in test_export2python_produces_correct_onnx_script_model
    functions = extract_functions(backend_test.name, code, self.test_folder)
onnxscript\backend\onnx_export_test.py:139: in extract_functions
    raise AssertionError(
E   AssertionError: Unable to import 'tests.onnx_backend_test_code.test_reduce_sum_empty_axes_input_noop_example' (e=No module named 'tests.onnx_backend_test_code.test_reduce_sum_empty_axes_input_noop_example') (file: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_reduce_sum_empty_axes_input_noop_example.py', absolute path: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_reduce_sum_empty_axes_input_noop_example.py', current folder: D:\a\onnxscript\onnxscript
E   ---- CONTENT --
E   import numpy
E   from onnx import TensorProto
E   from onnx.helper import make_tensor
E   from onnxscript import script, external_tensor
E   from onnxscript.values import Opset
E   from onnxscript.onnx_types import FLOAT, INT64
E   from onnxscript.onnx_opset import opset13
E   
E   @script()
E   def bck_test_reduce_sum_empty_axes_input_noop_example(data: FLOAT[3,2,2], axes: INT64[0]) -> (FLOAT[3,2,2]):
E       reduced = opset13.ReduceSum(data, axes, keepdims=1, noop_with_empty_axes=1)
E       return reduced

onnxscript.backend.onnx_export_test.TestOnnxBackEnd::test_export2python_produces_correct_onnx_script_model_0507_test_layer_normalization_3d_axis0_epsilon

Stack Traces | 0.004s run time

onnxscript\backend\onnx_export_test.py:137: in extract_functions
    mod = importlib.import_module(import_name)
C:\hostedtoolcache\windows\Python\3.10.11\x64\lib\importlib\__init__.py:126: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
E   ModuleNotFoundError: No module named 'tests.onnx_backend_test_code.test_layer_normalization_3d_axis0_epsilon'

The above exception was the direct cause of the following exception:
.nox\test\lib\site-packages\parameterized\parameterized.py:620: in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
onnxscript\backend\onnx_export_test.py:271: in test_export2python_produces_correct_onnx_script_model
    functions = extract_functions(backend_test.name, code, self.test_folder)
onnxscript\backend\onnx_export_test.py:139: in extract_functions
    raise AssertionError(
E   AssertionError: Unable to import 'tests.onnx_backend_test_code.test_layer_normalization_3d_axis0_epsilon' (e=No module named 'tests.onnx_backend_test_code.test_layer_normalization_3d_axis0_epsilon') (file: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_layer_normalization_3d_axis0_epsilon.py', absolute path: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_layer_normalization_3d_axis0_epsilon.py', current folder: D:\a\onnxscript\onnxscript
E   ---- CONTENT --
E   import numpy
E   from onnx import TensorProto
E   from onnx.helper import make_tensor
E   from onnxscript import script, external_tensor
E   from onnxscript.values import Opset
E   from onnxscript.onnx_types import FLOAT
E   from onnxscript.onnx_opset import opset17
E   
E   @script()
E   def bck_test_layer_normalization_3d_axis0_epsilon(X: FLOAT[2,3,5], W: FLOAT[2,3,5], B: FLOAT[2,3,5]) -> (FLOAT[2,3,5], FLOAT[1,1,1], FLOAT[1,1,1]):
E       Y, Mean, InvStdDev = opset17.LayerNormalization(X, W, B, axis=0, epsilon=0.10000000149011612)
E       return Y, Mean, InvStdDev

onnxscript.backend.onnx_export_test.TestOnnxBackEnd::test_export2python_produces_correct_onnx_script_model_0541_test_layer_normalization_4d_axis3

Stack Traces | 0.004s run time

onnxscript\backend\onnx_export_test.py:137: in extract_functions
    mod = importlib.import_module(import_name)
C:\hostedtoolcache\windows\Python\3.11.9\x64\Lib\importlib\__init__.py:126: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
E   ModuleNotFoundError: No module named 'tests.onnx_backend_test_code.test_layer_normalization_4d_axis3'

The above exception was the direct cause of the following exception:
.nox\test_torch_nightly\Lib\site-packages\parameterized\parameterized.py:620: in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
onnxscript\backend\onnx_export_test.py:271: in test_export2python_produces_correct_onnx_script_model
    functions = extract_functions(backend_test.name, code, self.test_folder)
onnxscript\backend\onnx_export_test.py:139: in extract_functions
    raise AssertionError(
E   AssertionError: Unable to import 'tests.onnx_backend_test_code.test_layer_normalization_4d_axis3' (e=No module named 'tests.onnx_backend_test_code.test_layer_normalization_4d_axis3') (file: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_layer_normalization_4d_axis3.py', absolute path: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_layer_normalization_4d_axis3.py', current folder: D:\a\onnxscript\onnxscript
E   ---- CONTENT --
E   import numpy
E   from onnx import TensorProto
E   from onnx.helper import make_tensor
E   from onnxscript import script, external_tensor
E   from onnxscript.values import Opset
E   from onnxscript.onnx_types import FLOAT
E   from onnxscript.onnx_opset import opset17
E   
E   @script()
E   def bck_test_layer_normalization_4d_axis3(X: FLOAT[2,3,4,5], W: FLOAT[5], B: FLOAT[5]) -> (FLOAT[2,3,4,5], FLOAT[2,3,4,1], FLOAT[2,3,4,1]):
E       Y, Mean, InvStdDev = opset17.LayerNormalization(X, W, B, axis=3)
E       return Y, Mean, InvStdDev

To view more test analytics, go to the Test Analytics Dashboard
_{📋 Got 3 mins? Take this short survey to help us improve Test Analytics.}

[Copilot]

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

Comments suppressed due to low confidence (2)

onnxscript/function_libs/torch_lib/ops/nn.py:1511

• The new padding logic introduced by _process_padding should be covered by test cases to ensure it works as expected.

pad: Sequence[INT64],

onnxscript/function_libs/torch_lib/ops/nn.py:1515

• The parameter pad is described as SymInt[], but it should be Sequence[INT64].

"""pad(Tensor self, SymInt[] pad, str mode=\"constant\", float? value=None) -> Tensor"""

[titaiwangms]

Do we need a new reflection_pad test to guard the missing corner case?

[justinchuby]

Do we need a new reflection_pad test to guard the missing corner case?

Could you remind me which was the missing case? (will follow up if needed)

[titaiwangms]

pytorch/pytorch#144382

Aren't we refactoring because it has this corner case?

[justinchuby]

Let me test

[justinchuby]

Thanks for the reminder. It's fixed. Would be helpful to create some cases. I will do that