fix: "Save as LoRA adapter" option added to prompt selection#254
fix: "Save as LoRA adapter" option added to prompt selection#254alejandroCartas wants to merge 1 commit intop-e-w:masterfrom
Conversation
Summary of ChangesHello, 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 enhances the model merging workflow by offering users the ability to save only the LoRA adapter, rather than always merging it into the full model. It also refines the prompt display logic within the 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. 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. Footnotes
|
There was a problem hiding this comment.
Code Review
This pull request modifies the obtain_merge_strategy function in src/heretic/main.py. The changes ensure that the user is always prompted to select a strategy for handling LoRA models, removing the previous conditional auto-merge behavior. Additionally, a new option to 'Save LoRA adapter' has been introduced, providing users with more flexibility. There is no feedback to provide from the code review.
|
Did you actually test this? It does NOT work when uploading, which is exactly why it has been disabled. See #152. While I'm aware that it appears to work when exporting locally, as discussed in #152, I will not re-enable this option without understanding why it fails on upload, because there might be situations where the same failure mode happens for local export. If you can actually find the problem it would be great, but unfortunately, re-enabling a menu is not enough. |
The option of saving only the LoRA adapter was added to the obtain_merge_strategy function. Additionally, the prompt selection now shows even if the settings.quantization is not equal to QuantizationMethod.BNB_4BIT, as this might be a bug.