Fine-Tuning Repository for Causal Language Models

Overview 🌟

This repository offers tools for fine-tuning causal language models, leveraging the power of the PEFT (PyTorch Efficient Fine-Tuning) library. It's designed for enhancing code generation tailored to specific repositories.

Features 🛠️

• Model Flexibility 📈: Load any causal language model.
• Configuration-Based Setup ⚙️: Easily define model selection and tokenization through a conf/config.yaml file.
• Efficient Fine-Tuning 🏃‍♂️: Currently utilizing LoRA (Low-Rank Adaptation) via the PEFT library. More methods coming soon!
• Future-Ready 🔮: Stay tuned for training methods for instruct and QA models.

Getting Started 🚦

Prerequisites 📋

• Python 3.x 🐍
• PyTorch 🔥
• PEFT Library 📚
• Other dependencies listed in requirements.txt

Installation 📥

1. Clone the repository: git clone [repository URL] :git:
2. Make a place for storing data mkdir data 📂 make sure to update config.yaml accordingly.
3. Install required packages: pip install -r requirements.txt

Configuration 🛠️

Tweak the conf\config.yaml file to set your model and tokenization parameters.

Usage 🖥️

Kick off the fine-tuning process by running the main script (python main.py). Fine tuned model will be saved in directory specified in config.yaml -> "optimization" -> result_dir.

Contributing 🤝

We welcome your contributions to make this tool even more efficient and feature-rich. Please adhere to standard open-source contribution guidelines.

License 📄

Apache License 2.0