Skip to content

[Stable APIs] Create torchlib_opset for torch 2.6 #1963

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 3 commits into from
Nov 23, 2024
Merged

[Stable APIs] Create torchlib_opset for torch 2.6 #1963

merged 3 commits into from
Nov 23, 2024

Conversation

justinchuby
Copy link
Collaborator

Create torchlib_opset for torch 2.6. This will be used for creating the model opset import as well as in _building for creating constant/concat nodes etc.

github-advanced-security[bot]
github-advanced-security bot found potential problems Nov 22, 2024
View reviewed changes
@codecov Codecov
Copy link

codecov bot commented Nov 22, 2024 •
edited
Loading

❌ 16 Tests Failed:

Tests completed Failed Passed Skipped
14307 16 14291 1623
View the full list of 3 ❄️ flaky tests 
tests.eager_mode_test.TestEagerModeArguments_0_reference_runtime::test_function_all_input_by_kwargs

Flake rate in main: 39.73% (Passed 9442 times, Failed 6223 times)

Stack Traces | 0.002s run time 
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:91: in run
    res = self._run(x, y)
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:139: in _run
    res = (convert_from_ml_dtypes(res[0]),)
..../test_torch_nightly/lib/python3.12.../onnx/reference/custom_element_types.py:50: in convert_from_ml_dtypes
    return array.view(dtype=dtype)
E   ValueError: Changing the dtype of a 0d array is only supported if the itemsize is unchanged

The above exception was the direct cause of the following exception:
tests/eager_mode_test.py:109: in test_function_all_input_by_kwargs
    self.assertEqual(add_with_alpha(this=1.0, other=2.0), 3.0)
onnxscript/values.py:576: in __call__
    return evaluator.default().eval_function(self, args, kwargs)
onnxscript/evaluator.py:307: in eval_function
    result = function.function(*adapted_args, **adapted_kwargs)
tests/eager_mode_test.py:59: in add_with_alpha
    other = op.Mul(other, alpha)
.../onnx_opset/_impl/opset14.py:696: in Mul
    return op(*self._prepare_inputs(schema, A, B))
onnxscript/values.py:304: in __call__
    return evaluator.default().eval(schema, args, kwargs)
onnxscript/evaluator.py:194: in eval
    outputs = self._eval(schema, inputs, attributes, closure)
onnxscript/evaluator.py:524: in _eval
    result = session.run(None, session_run_input)
..../test_torch_nightly/lib/python3.12.../onnx/reference/reference_evaluator.py:599: in run
    outputs = node.run(*inputs, **linked_attributes)
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:114: in run
    res = OpRunBinary.run(self, x, y)
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:93: in run
    raise TypeError(
E   TypeError: Issues with types <class 'numpy.ndarray'>, <class 'numpy.ndarray'> (binary operator 'Mul').
tests.eager_mode_test.TestEagerModeArguments_0_reference_runtime::test_function_some_input_by_kwargs

Flake rate in main: 39.73% (Passed 9442 times, Failed 6223 times)

Stack Traces | 0.003s run time 
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:91: in run
    res = self._run(x, y)
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:139: in _run
    res = (convert_from_ml_dtypes(res[0]),)
..../test_torch_nightly/lib/python3.12.../onnx/reference/custom_element_types.py:50: in convert_from_ml_dtypes
    return array.view(dtype=dtype)
E   ValueError: Changing the dtype of a 0d array is only supported if the itemsize is unchanged

The above exception was the direct cause of the following exception:
tests/eager_mode_test.py:106: in test_function_some_input_by_kwargs
    self.assertEqual(add_with_alpha(1.0, other=2.0), 3.0)
onnxscript/values.py:576: in __call__
    return evaluator.default().eval_function(self, args, kwargs)
onnxscript/evaluator.py:307: in eval_function
    result = function.function(*adapted_args, **adapted_kwargs)
tests/eager_mode_test.py:59: in add_with_alpha
    other = op.Mul(other, alpha)
.../onnx_opset/_impl/opset14.py:696: in Mul
    return op(*self._prepare_inputs(schema, A, B))
onnxscript/values.py:304: in __call__
    return evaluator.default().eval(schema, args, kwargs)
onnxscript/evaluator.py:194: in eval
    outputs = self._eval(schema, inputs, attributes, closure)
onnxscript/evaluator.py:524: in _eval
    result = session.run(None, session_run_input)
..../test_torch_nightly/lib/python3.12.../onnx/reference/reference_evaluator.py:599: in run
    outputs = node.run(*inputs, **linked_attributes)
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:114: in run
    res = OpRunBinary.run(self, x, y)
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:93: in run
    raise TypeError(
E   TypeError: Issues with types <class 'numpy.ndarray'>, <class 'numpy.ndarray'> (binary operator 'Mul').
tests.eager_mode_test.TestEagerModeArguments_0_reference_runtime::test_function_attribute_by_positional_args

Flake rate in main: 39.73% (Passed 9442 times, Failed 6223 times)

Stack Traces | 0.003s run time 
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:91: in run
    res = self._run(x, y)
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:139: in _run
    res = (convert_from_ml_dtypes(res[0]),)
..../test_torch_nightly/lib/python3.12.../onnx/reference/custom_element_types.py:50: in convert_from_ml_dtypes
    return array.view(dtype=dtype)
E   ValueError: Changing the dtype of a 0d array is only supported if the itemsize is unchanged

The above exception was the direct cause of the following exception:
tests/eager_mode_test.py:112: in test_function_attribute_by_positional_args
    self.assertEqual(add_with_alpha(1.0, 2.0, 3.0), 7.0)
onnxscript/values.py:576: in __call__
    return evaluator.default().eval_function(self, args, kwargs)
onnxscript/evaluator.py:307: in eval_function
    result = function.function(*adapted_args, **adapted_kwargs)
tests/eager_mode_test.py:59: in add_with_alpha
    other = op.Mul(other, alpha)
.../onnx_opset/_impl/opset14.py:696: in Mul
    return op(*self._prepare_inputs(schema, A, B))
onnxscript/values.py:304: in __call__
    return evaluator.default().eval(schema, args, kwargs)
onnxscript/evaluator.py:194: in eval
    outputs = self._eval(schema, inputs, attributes, closure)
onnxscript/evaluator.py:524: in _eval
    result = session.run(None, session_run_input)
..../test_torch_nightly/lib/python3.12.../onnx/reference/reference_evaluator.py:599: in run
    outputs = node.run(*inputs, **linked_attributes)
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:114: in run
    res = OpRunBinary.run(self, x, y)
..../test_torch_nightly/lib/python3.12.../reference/ops/_op.py:93: in run
    raise TypeError(
E   TypeError: Issues with types <class 'numpy.ndarray'>, <class 'numpy.ndarray'> (binary operator 'Mul').

To view more test analytics, go to the Test Analytics Dashboard
Got feedback? Let us know on Github

@justinchuby justinchuby mentioned this pull request Nov 23, 2024
Closed
@justinchuby justinchuby merged commit 9592227 into main Nov 23, 2024
22 of 39 checks passed
@justinchuby justinchuby deleted the justinchu/api-opset branch November 23, 2024 03:03
pytorchmergebot pushed a commit to pytorch/pytorch that referenced this pull request Nov 25, 2024
@justinchuby @pytorchmergebot
[ONNX] Use the torchlib opset number and fix opset import logic (#141413
419b566 
)

- Update the ONNX IR `add_opset_imports` pass to remove the heuristics of taking the `max` of the seen opsets. Instead, it uses the torchlib default opset version for the model's opset_import. The version converter is able to take the true opset versions in the nodes and convert the model to the correct version.
- Update all hard coding of opset 18 to instead query the default torchlib opset from onnxscript, introduced in microsoft/onnxscript#1963

Fixes #141260
Pull Request resolved: #141413
Approved by: https://github.com/titaiwangms
pobin6 pushed a commit to pobin6/pytorch that referenced this pull request Dec 5, 2024
@justinchuby
[ONNX] Use the torchlib opset number and fix opset import logic (pyto…
efbbced 
…rch#141413)

- Update the ONNX IR `add_opset_imports` pass to remove the heuristics of taking the `max` of the seen opsets. Instead, it uses the torchlib default opset version for the model's opset_import. The version converter is able to take the true opset versions in the nodes and convert the model to the correct version.
- Update all hard coding of opset 18 to instead query the default torchlib opset from onnxscript, introduced in microsoft/onnxscript#1963

Fixes pytorch#141260
Pull Request resolved: pytorch#141413
Approved by: https://github.com/titaiwangms
Esquains pushed a commit to Esquains/study1 that referenced this pull request Dec 15, 2024
@justinchuby
Update on "[ONNX] Use the torchlib opset number and fix opset import …
f8ac3f0 
…logic"


- Update the ONNX IR `add_opset_imports` pass to remove the heuristics of taking the `max` of the seen opsets. Instead, it uses the torchlib default opset version for the model's opset_import. The version converter is able to take the true opset versions in the nodes and convert the model to the correct version.
- Update all hard coding of opset 18 to instead query the default torchlib opset from onnxscript, introduced in microsoft/onnxscript#1963

Fixes pytorch/pytorch#141260

[ghstack-poisoned]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@titaiwangms titaiwangms titaiwangms approved these changes

Assignees

No one assigned

Labels

topic: api

Projects

ONNX Script Review Board
Status: Done

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@justinchuby @titaiwangms