RAG-based QA System

Description

The RAG-based QA System is a question-answering application that leverages Retrieval-Augmented Generation (RAG) to provide accurate and contextually relevant answers. It combines the power of retrieval-based models and generative models to enhance the quality of responses. The system supports file uploads, asking question, and logs management to deliver a seamless user experience.

Installation

Prerequisites

• Python 3.8 or higher
• Docker (optional, for containerized deployment)
• MongoDB

Steps

1. Clone the repository:

   git clone https://github.com/vahidaskari/RAG-QA-system

2. Navigate to the project directory:

   cd RAG-QA-system

3. Create and activate a virtual environment:

   python3 -m venv venv
   source venv/bin/activate

4. Install the required dependencies:

   pip install -r requirements.lock

Configuration

Before running the application, copy the .env.sample file to .env and update the environment variables based on your setup:

cp .env.sample .env

Usage

To start the QA system, use the following command:

fastapi run

Docker Deployment

Build and run the application using Docker:

docker build -t rag_app .
docker run --rm --name rag-app -p 127.0.0.1:8000:8000 --network host --env-file ./.env -v ./../data:/app/data rag_app

Docker compose Deployment (Recommended)

1. change the MONGODB_CONNECTION_STRING value to "mongodb://mongodb:27017/" in .env file\
2. run:

docker compose up

API Routes

1. Upload File

Upload one or more files for processing. Files are saved, and text is extracted to create a document stored in ChromaDB.

• Endpoint: POST /file
• Description: Uploads multiple files and returns a unique document_id to reference the stored documents.
• Request:
  • Form-data: documents (list of files)
• Response:
  • {"document_id": "<unique_document_id>"}
• Example:

  curl -X POST "http://127.0.0.1:8000/file" -F "[email protected]"

---

2. Chat

Send a query and receive a response based on the provided document context.

• Endpoint: POST /chat
• Description: Retrieves context from the document and generates a response using an LLM.
• Request Body:

  {
    "query": "What is the content of the uploaded file?",
    "document_id": "<unique_document_id>"
  }

• Response:

  {
    "response": "<generated_response>"
  }

• Example:

  curl -X POST "http://127.0.0.1:8000/chat" -H "Content-Type: application/json" -d '{"query":"What is in the file?","document_id":"12345"}'

---

3. Logs

Retrieve logs or records of interactions with the system.

• Endpoint: POST /logs
• Description: Fetches logs for a specific document or all logs if no document_id is provided (to get all logs, send an empty JSON as body). here, "document_id" refers to mongodb document id (_id) .
• Request Body:

  {
    "document_id": "<optional_document_id>"
  }
  

• Response:
  • Single log: { ...log_data... }
  • Multiple logs: [ ...list_of_log_data... ]
• Example:

  curl -X POST "http://127.0.0.1:8000/logs" -H "Content-Type: application/json" -d '{"document_id":"12345"}'