[OpenVINO backend] fix randomuniform issue #21670
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Summary of Changes
Hello @Mohamed-Ashraf273, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!
This pull request aims to enhance the reliability of random number generation within the Keras OpenVINO backend. It addresses a known issue where the randomuniform operation yielded inconsistent results, ensuring that random operations are now deterministic and reproducible for improved model stability and debugging.
Highlights
- Addressing
randomuniformindeterminism: This PR resolves an issue where therandomuniformoperation in the OpenVINO backend produced undeterministic outputs, as reported in OpenVINO issue #32045 and discovered via keras-team/keras-hub/pull/2389. - Switching to NumPy for random number generation: The implementation of
uniformin the OpenVINO backend has been updated to usenumpy.random.default_rngfor generating random values, replacing the direct use ofov_opset.random_uniform. - Simplified seed handling: The seed handling logic within the
uniformfunction has been streamlined to directly passseed_datato the NumPy random number generator, improving consistency.
Using Gemini Code Assist
The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.
Invoking Gemini
You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.
| Feature | Command | Description |
|---|---|---|
| Code Review | /gemini review |
Performs a code review for the current pull request in its current state. |
| Pull Request Summary | /gemini summary |
Provides a summary of the current pull request in its current state. |
| Comment | @gemini-code-assist | Responds in comments when explicitly tagged, both in pull request comments and review comments. |
| Help | /gemini help |
Displays a list of available commands. |
Customization
To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.
Limitations & Feedback
Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.
You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.
Footnotes
-
Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩
There was a problem hiding this comment.
Code Review
This pull request aims to fix a non-determinism issue with random.uniform in the OpenVINO backend by switching from ov_opset.random_uniform to a NumPy-based implementation.
While this correctly addresses the determinism for concrete shapes, it introduces a critical regression by breaking support for symbolic shapes, which is a common use case in Keras. I've left a detailed comment with a suggested fix that uses a hybrid approach to support both concrete and symbolic shapes.
Additionally, the random.categorical function also uses ov_opset.random_uniform and is likely affected by the same non-determinism bug. This should also be investigated and fixed to fully resolve the issue.
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #21670 +/- ##
==========================================
+ Coverage 82.55% 82.56% +0.01%
==========================================
Files 571 571
Lines 57626 57713 +87
Branches 9001 9016 +15
==========================================
+ Hits 47572 47650 +78
- Misses 7759 7765 +6
- Partials 2295 2298 +3
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Mohamed-Ashraf273
changed the title
@rkazants
@fchollet
Solve openvino issue for the undeterministic output of
randomuniform, by implementing it likeops.normalabove.Issue: openvinotoolkit/openvino#32045
Discovred from: keras-team/keras-hub#2389.
running
pytest keras_hub/src/models/gemma/gemma_causal_lm_test.py::GemmaCausalLMTest::test_causal_lm_basics -v -swithout fix gives:
with fix: