Diabetes Follow-up Prediction API

Project Overview

A machine learning service as part of the MAIO course that helps predict diabetes progression to determind it the patients need follow-up care.

Getting Started

Build and Train Locally

python src/train.py

Run API locally

fastapi dev .\src\main.py

API Definition

Predict

POST /predict

curl --location --request POST 'http://127.0.0.1:8000/predict' \
--header 'Content-Type: application/json' \
--data-raw '{"age": 0.02, "sex": -0.044, "bmi": 0.06, "bp": -0.03, "s1": -0.02, "s2": 0.03, "s3": -0.02, "s4": 0.02, "s5": 0.02, "s6": -0.001}'

Returns:

{ "prediction": "123" }

Health check

GET /health

curl --location --request GET 'http://127.0.0.1:8000/health' \
--header 'Content-Type: application/json'

Returns:

{ "status": "ok", "model_version": "v0.1" }

Docker

Build:

docker build -t diabetes-triage-ml:v0.1 .

Run:

docker run -p 8000:8000 diabetes-triage-ml:v0.1