tests(transformers): increase the RoCBert/Speech2Text/UniSpeech models test error passing threshold for BF16

[alien-0119]

What does this PR do?

Fixes # (issue)

1. global_rng, a random number generator, when used as a global variable, will become inactive after being called a second time within a single process. To solve this, function get_rng creates a new random number generator on every call, meaning that different calls to functions like ids_numpy and floats_numpy will receive a random number generator that is reset to the same initial state.

2. Increase the RoCBert/Speech2Text/UniSpeech models test error passing threshold for BF16 to avoid unnecessary test failures for valid results.

refer to: #1357

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline?
• Did you make sure to update the documentation with your changes? E.g. record bug fixes or new features in What's New. Here are the
  documentation guidelines
• Did you build and run the code without any errors?
• Did you report the running environment (NPU type/MS version) and performance in the doc? (better record it for data loading, model inference, or training tasks)
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

@xxx

[gemini-code-assist]

Summary of Changes

Hello @alien-0119, 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 adjusts the acceptable error margins for tests involving Speech2Text and UniSpeech models when operating with BFloat16 precision. This change is crucial for preventing test failures that arise from the inherent numerical characteristics of BF16, ensuring that tests pass within a reasonable tolerance. Additionally, a minor refactoring was implemented to improve the management of random number generation within the testing utilities.

Highlights

• Test Threshold Adjustment: The error passing thresholds for BFloat16 (BF16) precision tests have been increased for both Speech2Text and UniSpeech models to accommodate numerical variations.
• Random Number Generator Refactoring: The global random number generator in modeling_common.py has been refactored into a get_rng() function to ensure better isolation and reproducibility in tests.

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

1. 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. ↩

[gemini-code-assist]

Code Review

This pull request aims to fix failing tests for Speech2Text and UniSpeech models on BF16 by increasing the error thresholds. However, the root cause of the increased error appears to be a subtle but significant change in tests/transformers_tests/models/modeling_common.py concerning random number generation. The new get_rng() function resets the random seed on every call, leading to repetitive random sequences, which is different from the previous global RNG behavior. This likely alters the test data in a way that increases the discrepancy between framework implementations. My review focuses on fixing this underlying issue, which should make the threshold increases unnecessary. I've suggested a fix for the RNG logic and recommended reverting the threshold changes pending re-evaluation after the fix.