Skip to content

[bugfix] fix megatron convert #5416

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

Merged
merged 1 commit into from
Aug 17, 2025

Conversation

Jintao-Huang
Copy link
Collaborator

No description provided.

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Summary of Changes

Hello @Jintao-Huang, 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 a bug by relocating a critical validation check related to model loading and initialization. Previously, this check was incorrectly placed within the Megatron-specific arguments. The change moves this validation to the general training arguments, ensuring that the system correctly enforces the requirement to set --no_initialization false when no model is being loaded, thereby preventing potential issues during model initialization, especially in conversion or training workflows.

Highlights

  • Relocation of Argument Validation: The validation logic for ensuring that --no_initialization is false when --load is not provided has been relocated from swift/megatron/argument/megatron_args.py to swift/megatron/argument/train_args.py.
  • Bug Fix for Megatron Conversion: This change fixes a bug, likely related to the Megatron conversion process, by correctly applying the necessary validation for model loading and initialization arguments within the training argument parsing.
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

  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.

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a 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 moves a validation check from the general MegatronArguments class to the more specific MegatronTrainArguments class. The goal is to fix an issue encountered during model conversion. While this change correctly addresses the conversion scenario, I have raised one high-severity concern. Removing the safety check from the base class could lead to silent failures in other non-training scripts (e.g., standalone evaluation) that might use MegatronArguments directly and inadvertently run on an uninitialized model. I've recommended ensuring that all execution paths requiring a fully initialized model have this or a similar validation in place.

@Jintao-Huang Jintao-Huang merged commit 91ce23f into modelscope:main Aug 17, 2025
2 checks passed
Jintao-Huang added a commit that referenced this pull request Aug 17, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant