sPPO #232
sPPO #232
Conversation
There was a problem hiding this comment.
Summary of Changes
Hello @yaochaorui, 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 introduces the sPPO (surrogate Proximal Policy Optimization) algorithm, integrating it into the system. It includes the core algorithm implementation, a dedicated policy loss function, and comprehensive example configurations for applying sPPO to the GSM8k dataset, facilitating its immediate use and testing.
Highlights
- sPPO Algorithm Integration: Implemented the sPPO algorithm, including its registration and default configuration within the trinity.algorithm module.
- Dedicated sPPO Policy Loss Function: Added sPPOPolicyLossFn to calculate the sPPO loss, which involves clipping the probability ratio based on a specified epsilon value.
- New sPPO GSM8k Example: Provided a complete example setup for running sPPO on the GSM8k dataset, including a README, main configuration (gsm8k.yaml), and training configuration (train_gsm8k.yaml).
- Configuration Refinement: Updated an existing configuration file (examples/asymre_gsm8k/gsm8k.yaml) with clearer project naming and relevant research paper references.
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. 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
-
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. ↩
There was a problem hiding this comment.
Code Review
This pull request introduces the sPPO algorithm, along with an example configuration for the GSM8k dataset. The implementation of the new algorithm in trinity/algorithm/algorithm.py and the associated example files are well-structured. However, I've found a critical issue in the sPPO loss function implementation which does not seem to match the formulation in the referenced paper. I've also noted a couple of minor issues in the configuration files, such as an incomplete comment and an invalid link. Please see my detailed comments for suggestions.
Co-authored-by: Yanxi Chen <153061753+yanxi-chen@users.noreply.github.com>
Description
Implemented sPPO. Added example config. Tested example config.
Checklist
Please check the following items before code is ready to be reviewed.