Go to file

SHREY c62bf15d4f Update README.md		2024-12-10 04:31:00 +00:00
app.py	Add app.py	2024-12-09 20:22:08 +00:00
medquad.csv	Upload files to "/"	2024-12-09 20:25:45 +00:00
medquad_index.faiss	Upload files to "/"	2024-12-09 20:26:23 +00:00
preprocess.py	Upload files to "/"	2024-12-09 20:28:34 +00:00
README.md	Update README.md	2024-12-10 04:31:00 +00:00
requirements.txt	Upload files to "/"	2024-12-09 20:29:33 +00:00

README.md

DigiDoctor: AI-Powered Medical Assistant

DigiDoctor is a medical chatbot powered by Retrieval-Augmented Generation (RAG). It uses the MedQuAD dataset for retrieval and Llama3.2 (via Ollama) for generating real-time AI responses.

Features

Intelligent medical assistant using a RAG pipeline.
Provides reliable responses based on the MedQuAD dataset.
Real-time streaming of AI-generated answers via Ollama.

Prerequisites

Python 3.8 or later.
Ollama installed and running on http://localhost:11434.

Installation and Setup

1. Clone this repository:

git clone https://git.digimantra.com/SHREY/DigiDoctor.git
cd DigiDoctor

2. Install dependencies in virtual environment:

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

3. Start the Ollama server:

curl -fsSL https://ollama.com/install.sh | sh
ollama serve

4. (Optional) Generate FAISS index if not already present:

If the medquad_index.faiss file is missing or needs to be rebuilt:
python3 preprocess.py

5. Running the Application

Start the Streamlit app:
streamlit run app.py