Create Gemma3n finetuning script for local, offline inference of emot…

[linhkid]

…ional first aid assistance using Unsloth.

Features:

• Prepare different mental health conversation datasets for finetuning
• Unsloth with LoRA 4bit finetuning
• Merge back and load full model for inference and ready for deployment
• Optional: Upload model artifact to GCP Bucket.

(this is my first contribution so please let me know what I lack, I would be happy to improve right away)

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[gemini-code-assist]

Summary of Changes

Hello @linhkid, 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 introduces a new finetuning script for the Gemma-3N model, focusing on developing an emotional first aid assistant capable of local and offline inference. The script leverages Unsloth with LoRA 4-bit finetuning on mental health conversation datasets, streamlining the process of adapting the model for specialized applications. It also prepares the model for deployment and offers an optional integration for uploading the model artifact to a GCP Bucket.

Highlights

• New Finetuning Script: Introduced a new finetuning script for the Gemma-3N model, specifically designed for local, offline inference.
• Emotional First Aid Assistant: The script aims to create an emotional first aid assistant by finetuning Gemma-3N on mental health conversation datasets.
• Unsloth with LoRA: The finetuning process utilizes Unsloth with LoRA 4-bit finetuning for efficient model adaptation.
• Deployment Readiness: The script includes steps to merge the finetuned model back and load the full model, making it ready for deployment.
• GCP Bucket Upload: An optional feature has been added to upload the finetuned model artifact to a GCP Bucket.

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 by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

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 pull request 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. ↩

[gemini-code-assist]

Code Review

This pull request introduces an excellent and well-structured Jupyter notebook for fine-tuning a Gemma-3N model on a mental health dataset using Unsloth. The notebook covers the entire process from data preparation and training to inference and optional deployment to Google Cloud. It's a valuable contribution. My review focuses on improving clarity, usability, and fixing a critical inconsistency between training and inference data formats that could impact the model's performance. Addressing these points will make the notebook more robust and user-friendly, especially for those new to the process. Congratulations on your first contribution!

[bebechien]

LGTM!
will fix the nbfmt later