Conversation
Contributor
Summary of ChangesHello @kevssim, 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 represents a major overhaul and rebranding of the project, transitioning from 'SWIFT' to 'Twinkle'. It establishes a comprehensive, modular, and distributed training framework for large language models, with a strong focus on multi-adapter LoRA, deep integration with Ray for distributed execution, and broad support for diverse model architectures like Transformers and Megatron-Core, including MoE and VLM. The changes encompass a complete restructuring of the project's core components, new data handling mechanisms, a client-server architecture for remote interaction, and extensive documentation to guide users and developers. Highlights
🧠 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
Activity
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
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request refactors the expert parallelism (EP) cookbook example for Qwen3 MoE models. The changes simplify the device mesh setup by introducing ep_size support into the DeviceMesh.from_sizes factory method, which is a nice API improvement. Consequently, the example script is updated to use this new, more declarative approach, and dependencies on DeviceGroup and remote_group are removed, streamlining the code. The script's execution mode is also switched from 'ray' to 'local', enhancing its accessibility for users without a Ray cluster. Additionally, a minor bug is fixed to prevent saving the model checkpoint at the very beginning of the training loop. My review includes a suggestion to add input validation to the new ep_size parameter for better robustness. Overall, these are solid improvements to the example.
Comment on lines
+89
to
+93
| if ep_size is not None: | ||
| mesh_dim_sizes.append(ep_size) | ||
| mesh_dim_names.append("ep") | ||
| if origin_world_size == 1: | ||
| world_size *= ep_size |
Contributor
There was a problem hiding this comment.
To improve robustness, it's a good practice to validate input parameters. The ep_size should be a positive integer. Consider adding a check to ensure ep_size > 0 to prevent potential runtime errors during mesh creation if an invalid value is passed.
Suggested change
| if ep_size is not None: | |
| mesh_dim_sizes.append(ep_size) | |
| mesh_dim_names.append("ep") | |
| if origin_world_size == 1: | |
| world_size *= ep_size | |
| if ep_size is not None: | |
| if ep_size <= 0: | |
| raise ValueError(f'ep_size must be positive, but got {ep_size}') | |
| mesh_dim_sizes.append(ep_size) | |
| mesh_dim_names.append("ep") | |
| if origin_world_size == 1: | |
| world_size *= ep_size |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.