Update CI dependencies and fix tests; generate opset 20 #1335
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Contributor
thiagocrepaldi
commented
Apr 2, 2024
•
thiagocrepaldi
commented
- Update ONNX, ORT version in CI
- Pin IR version in run_converter_test to 9 because ONNX runtime does not support IRv10 yet. Need a better mechanism to handle the IR version change in tests.
- Generate opset 20 for onnxscript
- Skip opset 21 tests
- Fixes [Broken CI] torch-nightly and onnx-weekly are failing #1318
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #1335 +/- ##
==========================================
- Coverage 75.61% 75.58% -0.04%
==========================================
Files 132 134 +2
Lines 17666 17773 +107
Branches 2928 2927 -1
==========================================
+ Hits 13358 13433 +75
- Misses 3865 3898 +33
+ Partials 443 442 -1 ☔ View full report in Codecov by Sentry. |
thiagocrepaldi
force-pushed
the
thiagofc/update-onnx-weekly
branch
from
e0c15eb to
3172fe8
Compare
Test Results 24 files + 6 24 suites +6 2h 17m 50s ⏱️ + 31m 28s For more details on these failures, see this check. Results for commit c951bf3. ± Comparison against base commit d96dcf2. ♻️ This comment has been updated with latest results. |
|
Unexpected error: 2024-04-02T16:33:59.5097958Z �[31m�[1m_ TestModelSaving.test_input_output_and_initializer_are_not_stored_in_value_info _�[0m
2024-04-02T16:33:59.5099409Z [gw1] linux -- Python 3.10.14 /home/runner/work/onnxscript/onnxscript/.nox/test/bin/python
2024-04-02T16:33:59.5101242Z �[1m�[31m.nox/test/lib/python3.10/site-packages/torch/onnx/_internal/exporter.py�[0m:1195: in dynamo_export
2024-04-02T16:33:59.5102596Z ).export()
2024-04-02T16:33:59.5103515Z �[1m�[31m.nox/test/lib/python3.10/site-packages/torch/onnx/_internal/exporter.py�[0m:950: in export
2024-04-02T16:33:59.5104359Z graph_module = pre_export_passes(
2024-04-02T16:33:59.5105339Z �[1m�[31m.nox/test/lib/python3.10/site-packages/torch/onnx/_internal/exporter.py�[0m:1261: in pre_export_passes
2024-04-02T16:33:59.5106358Z module = passes.Modularize(diagnostic_context, module).run()
2024-04-02T16:33:59.5107521Z �[1m�[31m.nox/test/lib/python3.10/site-packages/torch/onnx/_internal/diagnostics/infra/decorator.py�[0m:151: in wrapper
2024-04-02T16:33:59.5108450Z ctx.log_and_raise_if_error(diag)
2024-04-02T16:33:59.5109567Z �[1m�[31m.nox/test/lib/python3.10/site-packages/torch/onnx/_internal/diagnostics/infra/context.py�[0m:366: in log_and_raise_if_error
2024-04-02T16:33:59.5110608Z raise diagnostic.source_exception
2024-04-02T16:33:59.5111642Z �[1m�[31m.nox/test/lib/python3.10/site-packages/torch/onnx/_internal/diagnostics/infra/decorator.py�[0m:135: in wrapper
2024-04-02T16:33:59.5112589Z return_values = fn(*args, **kwargs)
2024-04-02T16:33:59.5113319Z �[1m�[31m.nox/test/lib/python3.10/site-packages/torch/onnx/_internal/fx/_pass.py�[0m:267: in run
2024-04-02T16:33:59.5113915Z module = self._run(*args, **kwargs)
2024-04-02T16:33:59.5114655Z �[1m�[31m.nox/test/lib/python3.10/site-packages/torch/onnx/_internal/fx/passes/modularization.py�[0m:831: in _run
2024-04-02T16:33:59.5115379Z root_module_node.add_leaf_node(_LeafNode(fx_node))
2024-04-02T16:33:59.5116218Z �[1m�[31m.nox/test/lib/python3.10/site-packages/torch/onnx/_internal/fx/passes/modularization.py�[0m:701: in __init__
2024-04-02T16:33:59.5116975Z self._stack_meta = _module_stack_meta_from_node(node)
2024-04-02T16:33:59.5117924Z �[1m�[31m.nox/test/lib/python3.10/site-packages/torch/onnx/_internal/fx/passes/modularization.py�[0m:330: in _module_stack_meta_from_node
2024-04-02T16:33:59.5118806Z return _ModuleStackMeta(node.meta.get("nn_module_stack"))
2024-04-02T16:33:59.5121524Z �[1m�[31mE beartype.roar.BeartypeCallHintParamViolation: Method torch.onnx._internal.fx.passes.modularization._ModuleStackMeta.__init__() parameter nn_module_stack_meta={'L__self___fc1': ("L['self'].fc1", <class 'torch.nn.modules.linear.Linear'>)} violates type hint typing.Union[collections.OrderedDict, typing.Dict[str, typing.Tuple[str, typing.Tuple[str, type]]], NoneType], as dict {'L__self___fc1': ("L['self'].fc1", <class 'torch.nn.modules.linear.Linear'>)}:�[0m
2024-04-02T16:33:59.5123841Z �[1m�[31mE * Not <class "builtins.NoneType"> or <class "collections.OrderedDict">.�[0m
2024-04-02T16:33:59.5124832Z �[1m�[31mE * dict key str 'L__self___fc1' value tuple index 1 item type <class 'torch.nn.modules.linear.Linear'> not instance of tuple.�[0m |
|
Updated failing tests |
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
thiagocrepaldi
commented
Apr 2, 2024
| @@ -138,6 +140,7 @@ def test_add_initializer_allows_adding_the_same_tensor_twice_using_same_name(sel | |||
| graph.add_initializer("x", x_tensor) | |||
|
|
|||
|
|
|||
| @unittest.skipIf(IS_WINDOWS, "dynamo_export not supported in Windows") | |||
There was a problem hiding this comment.
dynamo is supposed to be supported on windows on pytorch nightly. if something is failing, we need to track and fix
There was a problem hiding this comment.
We are not pulling the latest pytorch. I will add additional pytorch version check for this.
There was a problem hiding this comment.
Do you know since which version?
There was a problem hiding this comment.
|
I will work on this pr and get the rest of the stuff green |
justinchuby
approved these changes
Apr 2, 2024
justinchuby
changed the title
justinchuby
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.