Fix endianness conversion in numpy_helper.to_array() #5904
Merged
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
Signed-off-by: Tabari Alexander <tehbone@gmail.com>
Codecov ReportAttention:
Additional details and impacted files@@ Coverage Diff @@
## main #5904 +/- ##
=======================================
Coverage 56.65% 56.65%
=======================================
Files 506 506
Lines 30192 30193 +1
Branches 4555 4555
=======================================
+ Hits 17104 17105 +1
Misses 12266 12266
Partials 822 822 ☔ View full report in Codecov by Sentry. |
justinchuby
approved these changes
Feb 2, 2024
|
Thank you! |
linshokaku
pushed a commit
to linshokaku/onnx
that referenced
this pull request
Oct 2, 2024
### Description This removes the invocation of `convert_endian()` from `numpy_helper.to_array()` in favor of another implementation, since convert_endian() modifies the TensorProto. ### Motivation and Context This change was needed due to observed issues when trying to load an ONNX model that was exported on a s390x system via tf2onnx. Debugging showed that the TeensorProto was being modified by endianness conversion operations, which led to model incorrectness under certain circumstances. Fixes onnx#5901. Signed-off-by: Tabari Alexander <tehbone@gmail.com> Signed-off-by: Linsho Kaku <linsho@preferred.jp>
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Outdated suggestions cannot be applied.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.
Description
This removes the invocation of
convert_endian()fromnumpy_helper.to_array()in favor of another implementation, since convert_endian() modifies the TensorProto.Motivation and Context
This change was needed due to observed issues when trying to load an ONNX model that was exported on a s390x system via tf2onnx. Debugging showed that the TeensorProto was being modified by endianness conversion operations, which led to model incorrectness under certain circumstances.
Fixes #5901.