Basic RAG System with FastAPI

This project demonstrates a simple Retrieval-Augmented Generation (RAG) system built using FastAPI, Pinecone, and Google Generative AI. It allows users to upload documents (PDF/DOCX), index their contents for efficient retrieval, and query the indexed documents to receive AI-generated responses.

---

Features

• Upload multiple PDF and DOCX files for indexing.
• Automatic chunking of documents for efficient retrieval.
• Query the system to receive AI-powered answers based on uploaded content.
• Powered by Pinecone for vector similarity search and Google Generative AI for language model-based responses.

---

Folder Structure

├── app.py               # Main FastAPI application
├── example.env          # Example environment variables file
├── requirements.txt     # Python dependencies for the project

---

Prerequisites

Before running the project, ensure you have the following:

1. Python 3.8+ installed on your machine.
2. Access to the following API keys:
   • Pinecone API Key
   • Google API Key
3. Docker (optional, for containerized deployment).

---

Setup Instructions

1. Clone the Repository

git clone <repository-url>
cd <repository-name>

2. Create and Activate a Virtual Environment

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install Dependencies

pip install -r requirements.txt

4. Configure Environment Variables

Rename example.env to .env and update the following variables with your credentials:

PINECONE_API_KEY=<your-pinecone-api-key>
GOOGLE_API_KEY=<your-google-api-key>

5. Run the FastAPI Server

uvicorn app:app --reload

The server will start at http://127.0.0.1:8000.

---

API Endpoints

1. Upload Documents

• URL: /upload_documents/
• Method: POST
• Description: Upload one or more documents (PDF/DOCX) to index their content.
• Request Body: Form data containing file uploads.
• Response: JSON message indicating the number of indexed document chunks.

2. Query System

• URL: /query/
• Method: GET
• Description: Query the system and get an AI-generated response based on the indexed documents.
• Query Parameter: query (string) - The question to ask.
• Response: JSON containing the query and the generated answer.

---

Example Usage

Uploading Documents

curl -X POST "http://127.0.0.1:8000/upload_documents/" \
-H "accept: application/json" \
-F "files=@example.pdf" \
-F "files=@example.docx"

Querying the System

curl -X GET "http://127.0.0.1:8000/query/?query=What is data preprocessing?" \
-H "accept: application/json"

---

Dependencies

This project uses the following Python libraries:

• fastapi
• uvicorn
• pinecone
• langchain_google_genai
• langchain_community
• langchain_core

Refer to requirements.txt for the full list.

---