Fixed mask sharding if inputs is sharded

[vfdev-5]

Fixes #5209

[gemini-code-assist]

Summary of Changes

Hello @vfdev-5, 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 addresses a critical issue in the Dropout module where the generated mask was not correctly sharded when applied to sharded input tensors. By ensuring the Bernoulli mask generation respects the input's sharding, the PR guarantees consistent behavior in distributed computations. The inclusion of a new dedicated test case validates this fix, confirming that dropout operations now correctly maintain data sharding for distributed inputs.

Highlights

• Dropout Mask Sharding Fix: The random.bernoulli function call within the Dropout module now explicitly passes the input's sharding information (out_sharding=inputs.sharding) when generating the dropout mask. This ensures that the mask is sharded consistently with the input tensor, resolving an issue where masks might not align correctly with sharded inputs.
• New Dropout Sharding Test: A new test case, test_out_sharding_dropout, has been added to tests/nnx/spmd_test.py. This test verifies that the nnx.Dropout module correctly handles sharded inputs, asserting that the output sharding matches the expected distributed layout after applying dropout.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

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 addresses an issue in nnx.Dropout where the dropout mask was not created with the same sharding as the input tensor, potentially causing performance issues due to data movement. The fix involves passing the input's sharding to random.bernoulli to ensure the mask is correctly sharded. A new test, test_out_sharding_dropout, has been added to verify that the output of the Dropout layer maintains the correct sharding when given a sharded input. The change is correct and the test effectively validates the fix.