dayOneAbu/amharic-rag-supabase

Retrieval-Augmented Generation (RAG) system for Amharic language processing. # Amharic RAG Bun ## Overview Amharic RAG Bun is a project designed to implement a Retrieval-Augmented Generation (RAG) system for Amharic language processing. This project leverages Supabase for backend services and includes functionalities for querying, data ingestion, and more. ## Project Structure ``` supabase/ index.ts # Main entry point for the Supabase integration package.json # Project dependencies and scripts README.md # Project documentation tsconfig.json # TypeScript configuration functions/ rag-query/ # Contains the RAG query function index.ts migrations/ # Database migration files 001_initial_schema.sql scripts/ # Utility scripts for the project ingest-pdf.ts # Script for ingesting PDF data supabase/ # Supabase configuration and related files ``` ## Features - **RAG Query Functionality**: Implements a retrieval-augmented generation query system for Amharic language. - **PDF Ingestion**: A script to ingest PDF documents into the system. - **Supabase Integration**: Utilizes Supabase for database and backend services. ## Prerequisites - Node.js (v16 or later) - Supabase CLI - PostgreSQL ## Setup 1. Clone the repository: ```bash git clone cd amharic-rag-bun ``` 2. Install dependencies: ```bash cd supabase npm install ``` 3. Set up Supabase: - Install the Supabase CLI if you haven't already: supabase.com - Run the following command to start Supabase: ```bash supabase start ``` 4. Run database migrations: ```bash supabase db reset ``` 5. Ingest PDF data: ```bash node scripts/ingest-pdf.js ``` ## Usage - To run the RAG query function, deploy the function in the `functions/rag-query` directory to Supabase. - Use the provided scripts to ingest data and query the system. ## Contributing Contributions are welcome! Please fork the repository and submit a pull request for any features or bug fixes. ## License This project is licensed under the MIT License. See the LICENSE …