Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Summary of Changes
Hello @garyzhang99, 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 resolves schema inconsistencies encountered when loading Supervised Fine-Tuning (SFT) data, particularly for dynamic 'messages' and 'tools' columns, using Huggingface's load_dataset. The changes involve adopting the JSON Lines format for data storage and enhancing the data formatter to intelligently parse these fields, ensuring consistent schema interpretation and preventing unintended null value auto-filling, thereby improving data loading reliability.
Highlights
- SFT Data Schema Handling: Implemented robust parsing for 'messages' and 'tools' fields within the SFT data formatter, allowing them to be loaded as JSON strings to prevent schema misalignment issues when using Huggingface Datasets.
- Data Format Transition: Replaced the traditional JSON test data file (
.json) with a JSON Lines (.jsonl) file for SFT tools, providing a more consistent and reliable format forload_datasetwhen dealing with dynamic or varying schemas per record. - Documentation Update: Updated an example in the FAQ documentation to demonstrate the correct method for loading FSDP trainer checkpoints.
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 effectively addresses the schema misalignment issue when loading SFT data with Huggingface by serializing the messages and tools fields into JSON strings. The changes to the test data and documentation are consistent with this approach. My main feedback is on improving the error handling in trinity/buffer/schema/formatter.py. Currently, a JSON decoding error is only logged, which allows the program to continue with invalid data, likely causing a less informative crash later. I've suggested raising an exception to ensure the process fails fast and clearly on malformed data.
|
/unittest-module-trainer |
Summary
Skipped
Tests
Github Test Reporter by CTRF 💚 |
|
/unittest-module-buffer |
Summary
Tests
Github Test Reporter by CTRF 💚 |
|
/unittest-module-trainer |
Summary
Skipped
Tests
Github Test Reporter by CTRF 💚 |
2 participants
Description
This PR addresses schema misalignment issues when reading SFT data in JSON format with Huggingface, particularly with columns like messages and tools containing dynamic content. When using Huggingface load_dataset to read JSON files, columns with different structures are auto-filled with null values, resulting in inconsistent schemas.
Checklist
Please check the following items before code is ready to be reviewed.