Meditron Chatbot [Medical RAG App]

Description

Meditron Chatbot is a Medical Retrieval-Augmented Generation (RAG) application designed to provide relevant answers along with page numbers and context. The chatbot is built to avoid hallucinations, ensuring accurate and reliable responses. It leverages an open-source instruct Large Language Model (LLM), embeddings, and a vector store that operates locally.

Tech Stack

Front End

• HTML
• Bootstrap

Back End

• Flask
• FastAPI

Model

• Meditron

Embeddings

• Hugging Face Instruct Embeddings

Vector Store

• Qdrant (running using Docker)

Framework

• Langchain
• CTransformers (as orchestration frameworks)

Installation

Prerequisites

• Docker
• Python 3.8+
• Flask
• FastAPI

Steps

1. Clone the Repository

   git clone https://github.com/HR-04/Medical_App_RAG.git
   cd meditron-chatbot

2. Set Up the Environment

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

3. Install Dependencies

   pip install -r requirements.txt

4. Run Qdrant using Docker

   docker run -p 6333:6333 -d qdrant/qdrant

5. Start the Back End

   # In one terminal
   flask run
   
   # In another terminal
   uvicorn app:rag.app --reload or python rag.py

6. Open the Front End

   • Open index.html in your browser.

Usage

1. Open the web application in your browser.
2. Enter your medical query into the chat interface.
3. Receive responses with relevant information, including page numbers and context.

Contributing

We welcome contributions! Please fork the repository and submit a pull request.