Sample RAG Service

This project implements a Retrieval-Augmented Generation (RAG) service using FastAPI and Python. It processes PDF files containing medical codes and names, stores them in a vector database, and provides an API for querying the data.

Features

• Extracts medical codes and names from PDF files in the docs/ directory.
• Uses SentenceTransformers for embedding generation.
• Stores embeddings in a FAISS vector database in the rag/ directory.
• Provides a FastAPI endpoint for querying the data.
• Includes API key authentication for secure access.

Project Structure

• main.py: Entry point for the FastAPI application.
• routers/: Contains API route definitions.
• controllers/: Handles business logic for API endpoints.
• services/: Contains the RAG service implementation.
• middlewares/: Includes middleware for API key authentication.
• utils/: Utility functions for initialization and setup.

Environment Variables

• API_KEY: The API key for authenticating requests. Add this to a .env file in the root directory.

Setup

1. Clone the repository:

   git clone <repository-url>
   cd sample-rag

2. Install dependencies:

   pip install -r requirements.txt

3. Create the required directories:

   mkdir docs rag

4. Add a .env file with the following content:

   API_KEY=your_api_key_here

5. Add PDF files to the docs/ directory.

6. Run the application:

   uvicorn main:app --reload

API Endpoints

• GET /query: Query the RAG service with a question.
  • Headers: X-API-Key: <your_api_key>
  • Query Parameters: question=<your_question>

Notes

• The rag/ directory stores the FAISS index and is ignored by version control.
• Ensure the docs/ directory contains valid PDF files for processing.