Relevant Priors API

HTTP API for the New Lantern relevant-priors-v1 challenge.

Endpoint

The app exposes both:

• POST /
• POST /predict

Both accept the challenge JSON payload and return:

{
  "predictions": [
    {
      "case_id": "1001016",
      "study_id": "2453245",
      "predicted_is_relevant": true
    }
  ]
}

Local run

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

uvicorn app.main:app --host 0.0.0.0 --port 8000

Then submit:

http://localhost:8000/predict

Train model

python scripts/train.py --data data/relevant_priors_public.json --out models/relevant_priors_model.joblib

The submitted API uses the saved models/relevant_priors_model.joblib.

Deploy on Render

Use these settings:

• Environment: Python
• Build command: pip install -r requirements.txt
• Start command: uvicorn app.main:app --host 0.0.0.0 --port $PORT
• Health check path: /health

Submit the deployed /predict URL.