Add strict_shape_type_inference config option #11081
Merged
Conversation
This file contains 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
Prior to this, certain shape and type errors were surfaced only when the model was using the latest known op set version. Providing users an explicit option allows for better testing of code that produces models, which includes unit tests within this repo and other repos such as the TF-ONNX and PT-ONNX converters. The option defaults to false to avoid causing errors for users that rely on the previous permissive behavior. Fixes microsoft#9506.
… C++20 which we don't use
|
@pranavsharma could you please review or find someone else who can review? I'm not very familiar with who would be a good reviewer. |
jcwchen
reviewed
Apr 5, 2022
skottmckay
reviewed
Apr 6, 2022
garymm
changed the title
|
@skottmckay I'd appreciate a review before the weekend if possible. Thanks! |
skottmckay
reviewed
Apr 19, 2022
skottmckay
reviewed
Apr 19, 2022
pranavsharma
approved these changes
Apr 19, 2022
|
@pranavsharma the web CI pipelines seem to be timing out consistently. Any way to skip them? |
@fs-eire - can you please take a look? |
seddonm1
pushed a commit
to seddonm1/onnxruntime
that referenced
this pull request
May 15, 2022
Prior to this, certain shape and type errors were surfaced only when the model was using the latest known op set version. Providing users an explicit option allows for better testing of code that produces models, which includes unit tests within this repo and other repos such as the TF-ONNX and PT-ONNX converters. Remove the previous behavior which seems quite counter-intuitive: an otherwise identical model with a later op set version should be treated identically in this regard. The option defaults to false to avoid causing errors for users that rely on the previous permissive behavior. Turned on the strict enforcement by default in OpTester, which revealed a few disagreements between ORT and ONNX on what the correct output shape should be. Fix shape inference bug in ReduceSumTraining with noop_with_empty_axes=1 which was revealed. Fix TensorOpTest.Unsqueeze_scalar, which was testing negative axes on an op set version where the op did not actually support negative axes. Fixes microsoft#9506.
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.
Prior to this, certain shape and type errors were surfaced only when
the model was using the latest known op set version.
Providing users an explicit option allows for better testing of code
that produces models, which includes unit tests within this repo and
other repos such as the TF-ONNX and PT-ONNX converters.
Remove the previous behavior which seems quite counter-intuitive:
an otherwise identical model with a later op set version should be treated
identically in this regard.
The option defaults to false to avoid causing errors for users that
rely on the previous permissive behavior.
Turned on the strict enforcement by default in
OpTester, which revealed a fewdisagreements between ORT and ONNX on what the correct output shape should
be.
Fix shape inference bug in
ReduceSumTrainingwithnoop_with_empty_axes=1which was revealed.
Fix
TensorOpTest.Unsqueeze_scalar, which was testing negative axes on anop set version where the op did not actually support negative axes.
Fixes #9506.