add simple_math reward_fn

[andytwigg]

adds simple binary format and answer reward functions
move logging of example and its rewards into reward_manager.py

• I have added all the necessary unit tests for my change.
• I have verified that my change does not break existing code and all unit tests pass.
• I have added all appropriate doc-strings/documentation.
• My PR is based on the latest changes of the main branch (if unsure, rebase the code).
• I have signed the Contributor License Agreement.
• I have followed Contribution Guidelines.

[gemini-code-assist]

Summary of Changes

Hello, 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 expands the reward function capabilities by introducing new functions tailored for simple mathematical evaluations, focusing on both response formatting and accurate numerical answers. It also enhances the clarity of existing reward functions by documenting their scoring ranges and streamlines debugging by adding a focused logging feature within the reward management system.

Highlights

• New Reward Functions: Added a new simple_math module containing format and answer reward functions for evaluating responses based on binary format adherence and numerical correctness.
• Reward Function Documentation: Annotated existing reward functions in gsm8k.py with comments indicating their expected score ranges.
• Logging Improvements: Removed verbose debug logging from gsm8k.py and introduced a new mechanism in reward_manager.py to log a single example from the reward batch for easier debugging and monitoring.

🧠 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.

Changelog

• tunix/cli/reward_fn/gsm8k.py
  • Added range comments to match_format_exactly, match_format_approximately, check_answer, and check_numbers functions.
  • Removed debug logging statements from the check_numbers function.
• tunix/cli/reward_fn/simple_math.py
  • Added a new file simple_math.py containing format and answer reward functions.
• tunix/rl/reward_manager.py
  • Implemented a logging function _log_one_example to display a single example from the reward batch.

Activity

• No specific activity (comments, reviews, progress) has been recorded for this pull request yet.

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 new reward functions for simple math problems in tunix/cli/reward_fn/simple_math.py and refactors existing reward functions and logging. The changes in gsm8k.py are good improvements, removing unused code. The new simple_math.py file is a good addition, but I've suggested some improvements regarding function naming to avoid shadowing built-ins and to improve clarity. I've also pointed out an unused import. In reward_manager.py, I've suggested changing the log level of the new debugging output to avoid excessive noise in the logs and removing a stray blank line.