Deployment Production
Real-World APIs

Model Deployment Basics

Learn how to move a trained model from your notebook into a simple web API that real applications can call.

Deployment Concepts

  • Batch predictions: run model on large data periodically.
  • Online predictions: real-time API for single requests.
  • Monitoring: track latency, errors, data drift.

Simple Flask API Example

Serve a Scikit-Learn Model with Flask
# pip install flask joblib scikit-learn

from flask import Flask, request, jsonify
import joblib
import numpy as np

app = Flask(__name__)

# Load trained model (previously saved with joblib.dump)
model = joblib.load("model.pkl")

@app.route("/predict", methods=["POST"])
def predict():
    data = request.get_json()
    # Expecting JSON like: {"features": [5.1, 3.5, 1.4, 0.2]}
    features = np.array(data["features"]).reshape(1, -1)
    pred = model.predict(features)[0]
    return jsonify({"prediction": int(pred)})

if __name__ == "__main__":
    app.run(host="0.0.0.0", port=5000)
Next: MLOps Concepts