[Transforms] Update examples for R4 and transform_block_size option

[brian-dellabetta]

SUMMARY:
Prerequisites:

• [Transform] Revert deprecation of TransformScheme.head_dim for compatibility with vllm compressed-tensors#472

This PR updates the SpinQuant and Quip examples to include transform_block_size and the latest R4 feature in SpinQuant. It also reverts the TransformScheme.block_size changes previously introduced into CT, and updated in Pr linked above. While block_size is a more appropriate name, head_dim has already landed in vllm, and it would be too much of a pain to change. Users will rarely create their own TransformScheme anyway.

TEST PLAN:

• Both examples run and the saved model can be run in vllm, output is meaningful.
• with prints, confirmed hadacore is used for QuIPModifier(rotations=["v", "u"], transform_block_size=64, transform_type="hadamard")
• and dense gemm is used for QuIPModifier(rotations=["v", "u"], transform_block_size=64, transform_type="random-hadamard")

[github-actions]

👋 Hi! Thank you for contributing to llm-compressor. Please add the ready label when the PR is ready for review.

Note: This is required to complete the testing suite, please only add the label once the PR is code complete and local testing has been performed.

[gemini-code-assist]

Summary of Changes

Hello @brian-dellabetta, 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 focuses on enhancing the usability and consistency of the transformation examples and underlying API. It updates the QuIP and SpinQuant examples to demonstrate new features like transform_block_size and the R4 rotation, while also ensuring that the internal API aligns with external dependencies like vLLM by standardizing parameter names.

Highlights

• Example Updates: The QuIP and SpinQuant examples have been updated to incorporate the transform_block_size parameter and the latest R4 rotation feature in SpinQuant.
• API Consistency: The TransformScheme.block_size parameter has been reverted to head_dim across relevant modules to maintain consistency with vLLM's existing terminology, despite block_size being a more semantically appropriate name.

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 updates the SpinQuant and QuIP examples to use the transform_block_size option and adds support for R4 rotation in the SpinQuant example. The core change is the renaming of the block_size parameter to head_dim in the internal TransformScheme class to maintain compatibility with vllm, as explained in the description. The user-facing API in the modifiers correctly retains the transform_block_size parameter. The changes are consistent, and the examples are updated accordingly. I've identified a couple of areas where validation could be improved to prevent potential runtime errors.

[shanjiaz]

Looks good to me!

[fynnsu]

Looks good! Added a question about comment below.