Tech stack - Langchain, Fast API, ChromaDB, NodeJS, React, MongoDB, Automatic1111
- Engaging conversational chatbot with guardrails powered by Llama2 13B leveraging Langchain RAG pipelines
- Personalised outfit try-on experience using Stable diffusion in-painting over user’s images
- Collaborative & content-based recommendations using trending knowledge-base scraped from fashion magazines
# Install dependencies
npm run setup
# Rename .env.sample files to .env -
- tryon_backend/.env
- frontend/.env
# Run Backend server (port - 4000)
nodemon backend/server.js
# Run
# Run frontend (port - 3000)
cd frontend && npm start
# Tryon demo (API proxy for Huggingface spaces) (port - 5002)
cd tryon_backend
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
uvicorn app:app --reload --port=5002
# or
fastapi dev --port=5002
# Expose tryon via ngrok (for twilio bot)
ngrok config add-authtoken <your_token>
ngrok http 5002 --domain=merry-perfectly-jay.ngrok-free.app