Titanic inference example

A FastAPI inference server.

Pre-requisites

• pip install -r requirements.txt

Usage

Train

python src/train.py

Serve

uvicorn src.serve:app --reload

Serve mounted

git xet mount https://xethub.com/xdssio/titanic-server-example.git server
uvicorn server.src.serve:app --reload
# Checkout http://127.0.0.1:8000/docs or run: 
curl -X 'POST' \
  'http://127.0.0.1:8000/predict' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "Sex": "male",
  "Age": 22,
  "Fare": 7.25,
  "Embarked": "S"
}'
umount server