RAG FASTAPI For Pdf,txt,doc files

Table of Contents

• API Documentation
• Setup Instructions
  • Prerequisites
  • Installation
  • Running the Application
  • Docker Setup
• Potential Improvements
• Contributing

API Documentation

The API provides several endpoints for interacting with the application. You can access the detailed API documentation via Postman. To get started:

1. API DOCS: check api-documentation.pdf file
2. API Endpoints:

• Process Document
  • Endpoint: POST /api/documents/process
  • Input: File path
  • Output: Asset ID
  • Description: This endpoint processes the document at the given file path and returns an Asset ID.
• Start Chat
  • Endpoint: POST /api/chat/start
  • Input: Asset ID
  • Output: Chat thread ID
  • Description: This endpoint initiates a chat session using the provided Asset ID and returns a Chat thread ID.
• Send Chat Message
  • Endpoint: POST /api/chat/message
  • Input: Chat thread ID, User message
  • Output: Agent response (streamed)
  • Description: This endpoint sends a user message to the chat thread and streams the agent's response in real-time.
• Get Chat History
  • Endpoint: GET /api/chat/history
  • Input: Chat thread ID
  • Output: Chat history
  • Description: This endpoint retrieves the full history of a chat session using the provided Chat thread ID.

Setup Instructions

Prerequisites

Ensure you have the following installed:

• Python 3.x
• pip (Python package manager)
• Docker (for containerized setup)

Clone

Clone the repository:

git clone https://github.com/Jayanth-MKV/RAG-fastapi-chroma-langchain-docker.git

cd RAG-fastapi-chroma-langchain-docker

Setting Up a Virtual Environment

It is recommended to use a virtual environment to manage dependencies for your project. Here’s how you can set one up:

1. Navigate to your project directory:

   cd RAG-fastapi-chroma-langchain-docker
   

2. Create a virtual environment:

   python -m venv venv
   

3. Activate the virtual environment:

   • On macOS and Linux:

     source venv/bin/activate
     

   • On Windows:

     ./venv/Scripts/activate
     

4. Once activated, your terminal should show the virtual environment name, indicating that you are working inside the virtual environment.

Installation

1. Install the required Python packages:

   pip install -r requirements.txt

2. ENV Setup create a .env file and replace the api key

   GROQ_API_KEY = "api-key-here"
   

   You can get this from https://console.groq.com/keys

Running the Application

To run the application locally, use the following command:

python main.py

This will start the server on http://localhost:8000. You can test the endpoints using at http://localhost:8000/docs

You can also run the sample streamlit app using - streamlit run streamlit.py

Docker Setup

To set up and run the application using Docker, follow these steps:

1. Build the Docker image:

   docker build -t myapp .

2. Run the Docker container:

   docker run -p 8000:8000 myapp

3. The application will be accessible at http://localhost:8000.

Potential Improvements

• User authentication and authorization to secure endpoints.
• Adding different formats of data + adding connectors to drive, notion, dropbox, etc..
• Improving file handling and storing.
• Add unit and integration tests to ensure code reliability. Also the stress test.
• Scaling Vector Database like quadrant/pinecone
• Implementing the rag evaluation pipeline
• Experimenting differnt embedding models and parameters
• Combining traditional rag with graph db - hybrid rag improves the results
• Agentic approach to encounter hallucinations

Contributing

Contributions are welcome! Please feel free to submit a Pull Request or open an Issue to discuss improvements or fixes.