Migrate to JAX SC Embedding PreprocessedInput

[manuadg-google]

The JAX SC library is evolving to support minibatching via the dataclass PreprocessedInput which wraps the previous input SparseDenseMatmulInput. The num_minibatches field in PreprocessedInput needs to be replicated hence the updated sharding.

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[gemini-code-assist]

Summary of Changes

Hello @manuadg-google, 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!

I've updated the JAX SC Embedding lookup mechanism to leverage the PreprocessedInput class. This change standardizes the handling and partitioning of sparse inputs, ensuring better integration and consistency within the JAX framework for embedding operations.

Highlights

• Configuration Update: I've introduced sparse_input_partition within the EmbeddingLookupConfiguration to explicitly manage the partitioning of preprocessed sparse inputs, ensuring proper sharding.
• Input Preprocessing Migration: I've updated both the forward (embedding_lookup) and backward (embedding_lookup_bwd) passes to wrap the sparse_dense_matmul_input with embedding.PreprocessedInput. This standardizes how sparse inputs are prepared for embedding operations.
• Sharding Specification Alignment: I've adjusted the in_specs for shard_map calls in both the forward and backward functions to correctly reference the new config.sparse_input_partition, aligning sharding with the PreprocessedInput structure.
• Type Hint Refinement: I've updated the type hint for sparse_input in the grad_func within embedding_lookup_bwd to embedding.PreprocessedInput, reflecting the migration to the new input class.

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 in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

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 issue 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 or fill out our survey to provide feedback.

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 migrates the JAX sparse-core embedding lookup to use PreprocessedInput. The changes are consistent and correctly update the configuration, input processing, and sharding specifications. I have one suggestion to improve code maintainability by refactoring duplicated code into a helper function. Otherwise, the changes look good.