Fix evaluation trigger and enhance benchmark metadata

[uygnef]

Motivation

Enable evaluation when only eval_hidden_states_path (offline mode) is provided and prevent configuration conflicts between online/offline modes.

Modifications

• Evaluation Logic: Updated the trigger condition in scripts/train_eagle3.py to support both online and offline evaluation paths.
• Sanity Check: Added a check to ensure eval_data_path and eval_hidden_states_path are mutually exclusive.
• Benchmark Metadata: Added speculative_draft_model_path to benchmark results in benchmarks/bench_eagle3.py for better experiment tracking.

Related Issues

Accuracy Test

Benchmark & Profiling

Checklist

• Format your code according to the Code Formatting with Pre-Commit.
• Add unit tests as outlined in the Running Unit Tests.
• Update documentation / docstrings / example tutorials as needed, according to Writing Documentation.
• Provide throughput / latency benchmark results and accuracy evaluation results as needed, according to Benchmark and Profiling and Accuracy Results.
• For reviewers: If you haven't made any contributions to this PR and are only assisting with merging the main branch, please remove yourself as a co-author when merging the PR.
• Please feel free to join our Slack channel at https://sgl-fru7574.slack.com/archives/C09784E3EN6 to discuss your PR.

[gemini-code-assist]

Summary of Changes

Hello @uygnef, 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 enhances the evaluation system for the Eagle3 model by enabling offline evaluation capabilities and preventing configuration conflicts between online and offline modes. It also improves the traceability of benchmark results by incorporating specific model paths into the metadata. These changes aim to provide greater flexibility in evaluating model performance and ensure more robust experiment tracking.

Highlights

• Evaluation Trigger Logic: The evaluation trigger in scripts/train_eagle3.py has been updated to enable evaluation when either eval_data_path (online mode) or eval_hidden_states_path (offline mode) is provided, allowing for more flexible evaluation setups.
• Configuration Conflict Prevention: A new sanity check was introduced in scripts/train_eagle3.py to prevent simultaneous configuration of eval_data_path and eval_hidden_states_path, ensuring clear distinction between online and offline evaluation modes.
• Benchmark Metadata Enhancement: The benchmark results in benchmarks/bench_eagle3.py now include speculative_draft_model_path, improving experiment tracking and reproducibility for speculative decoding benchmarks.

🧠 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 introduces a crucial sanity check to prevent conflicting evaluation paths (eval_data_path and eval_hidden_states_path) from being set simultaneously, ensuring a clearer distinction between online and offline evaluation modes. It also enhances the benchmark metadata by including the speculative_draft_model_path, which is valuable for tracking experiment configurations. Additionally, the evaluation trigger logic has been updated to correctly initiate evaluation when either an online or offline evaluation path is provided, aligning with the new flexibility in evaluation modes.