Medical-Chatbot-Llama2

Medical Chatbot using LLama 2 and Pinecone

A specialized medical chatbot built with Llama 2, Pinecone vector database, and LangChain, designed to provide medical information and answers based on uploaded medical documents.

Features

• Medical Knowledge Base: Processes and stores medical documents for accurate responses
• Vector Search: Utilizes Pinecone for efficient similarity search
• LLM Integration: Powered by Llama 2 for natural language understanding
• Web Interface: Flask-based chat interface for easy interaction
• Document Processing: Handles PDF documents for knowledge base creation

Tech Stack

• LLM: Llama 2 (7B-chat quantized model)
• Vector Database: Pinecone
• Embeddings: HuggingFace Sentence Transformers (all-MiniLM-L6-v2)
• Framework: LangChain
• Web Framework: Flask
• PDF Processing: PyPDF

Prerequisites

• Python 3.8 or higher
• Pinecone API Key
• Llama 2 7B chat model (quantized version)
• Sufficient RAM for running the LLM (minimum 8GB recommended)

Project Structure

medical-chatbot/
├── app.py
├── requirements.txt
├── model/
│   └── llama-2-7b-chat.ggmlv3.q4_1.bin
├── src/
│   ├── __init__.py
│   ├── helper.py
│   └── prompt.py
├── Data/
│   └── [your PDF files]
└── templates/
    └── chat.html

Installation

1. Clone the repository:

git clone https://github.com/PrathamAhuja1/Medical-Chatbot-Llama2.git
cd Medical-Chatbot-Llama2
pip install -r requirements.txt
python app.py