ci: make find_local_hf_snapshot_dir more robust

[mickqian]

Motivation

Modifications

several modifications to find_local_hf_snapshot_dir:

1. if any incomplete file exists, return None
2. check if all safetensors with name model-{i}-of-{n}.safetensors exists

Accuracy Tests

Benchmarking and Profiling

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.

[gemini-code-assist]

Summary of Changes

Hello @mickqian, 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 significantly enhances the reliability of loading local Hugging Face model snapshots by introducing comprehensive validation steps. It ensures that only fully downloaded and intact models are utilized, preventing issues that could arise from incomplete files, broken references, or missing sharded components. This makes the model loading process more resilient to network interruptions and corrupted cache states.

Highlights

• Enhanced Local Snapshot Validation: The find_local_hf_snapshot_dir function now includes more robust checks to ensure the integrity and completeness of locally cached Hugging Face model snapshots.
• Incomplete Download Detection: Added a mechanism to detect .incomplete files within the Hugging Face cache's blobs directory, which indicates a corrupted or unfinished download, prompting a re-download.
• Broken Symlink Handling: Improved the file validation loop to explicitly check os.path.exists for each matched weight file, correctly identifying and excluding broken symlinks from valid local files.
• Sharded Model Completeness Check: Implemented logic to verify that all shards of a sharded model (e.g., model-00001-of-00009.safetensors) are present and valid, preventing the use of partially downloaded sharded models.

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 enhances the robustness of find_local_hf_snapshot_dir by adding several checks for local Hugging Face model snapshots. It now verifies the absence of incomplete download files, ensures that weight files are not broken symbolic links, and checks for the completeness of sharded models. While these are valuable improvements, I've found a critical issue in the implementation of the sharded model check that needs to be addressed.