NST - application based on the Perceptual Losses for Real-Time Style Transfer and Super-Resolution paper and Justin Johnson's pre-trained models.
Application implements:
- Develop an asynchronous API with Python and FastAPI
- Serve up a machine learning model with FastAPI
- Develop a UI with Streamlit
- Containerize FastAPI and Streamlit with Docker
- Leverage asyncio to execute code in the background outside the request/response flow
- FastAPI: for the API
- streamlit : for the interface
- Docker: to containerize the app
./download_models.sh
docker-compose up -d