To run server, run docker file:

# /model-api
$ docker-compose build
$ docker-compose up -d

Check service running at localhost:8008

Use different docker files depending on your env or need:

• Dockerfile (basic uvicorn)
• Dockerfile.comb (gunicorn + uvicorn) - currently for production
• Dockerfile.transformer (huggingface transformer + pytorch + GPU) - NOTE: comment out transformer & pytorch from requirements.txt