-
Notifications
You must be signed in to change notification settings - Fork 47
Bug fix in benchmark ckpt loading and megatron hf save #392
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
Summary of ChangesHello @chenyushuo, 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 addresses two critical bug fixes related to model management. It refines the checkpoint loading mechanism to prevent erroneous behavior in benchmark mode and enhances the stability of Hugging Face model saving by introducing comprehensive error handling. These changes contribute to more reliable and robust model operations within the system. Highlights
Using Gemini Code AssistThe 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
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 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
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Code Review
This pull request introduces two bug fixes. The first prevents the synchronizer from automatically loading the latest checkpoint when running in 'bench' mode, which is a correct change to allow benchmarking specific checkpoints. The second change adds error handling when saving Hugging Face models with the Megatron engine, preventing crashes from NotImplementedError and providing a helpful message to the user. I've suggested a small improvement to the error logging to include a full traceback for easier debugging. Overall, the changes improve the robustness and correctness of the codebase.
2. check `micro_batch_size` when not using `use_dynamic_bsz` in fsdp_workers.
|
/unittest-all |
|
/unittest-all |
Summary
Skipped
Tests
Github Test Reporter by CTRF 💚 |
Description
kl_loss != 0in first step.kl_loss_coefandkl_loss_typein verl_config.micro_batch_sizewhen not usinguse_dynamic_bszin fsdp_workers.modecheck.bots_reward.pyChecklist
Please check the following items before code is ready to be reviewed.