training/README.md

ORA Training Environment

This directory contains the scripts and configuration for Fine-Tuning a custom LLM (e.g., Llama 3) for the ORA Spiritual Assistant.

Prerequisites

• Python 3.10+
• CUDA-capable GPU (NVIDIA) or Mac with M-series chip
• Ollama (for local testing)

Setup

1. Create a virtual environment:

   python -m venv venv
   source venv/bin/activate # or venv\Scripts\activate on Windows
   

2. Install dependencies:

   pip install -r requirements.txt
   

Workflow

1. Prepare Data: Place your data (JSONL format) in data/.
2. Fine-Tune: Run the training script (coming soon).
3. Export: Convert LoRA adapters to GGUF format.
4. Serve: Load into Ollama.