Online STEM and Robotics Creative Club in Samut Prakan

1. What Flask Is

Notebooks and local scripts cannot serve clients by themselves. Flask is a lightweight Python web framework that turns models and pipelines into websites or APIs that browsers and apps can call.

2. Core Capabilities

• Minimalist core — flexible, low boilerplate
• Built-in server & debugger — fast local testing
• Jinja2 — dynamic HTML templates
• URL routing — map paths to Python functions
• REST endpoints — expose predictions over HTTP

3. Serving an ML Model

Typical flow: client sends features → Flask parses the request → loaded model runs inference → Flask returns JSON or HTML.

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

app = Flask(__name__)

class DummyModel:
    def predict(self, features):
        return [int(np.sum(features) > 50)]

model = DummyModel()

@app.route("/predict", methods=["POST"])
def predict():
    data = request.get_json()
    features = data.get("features", [])
    prediction = model.predict(features)
    return jsonify({"status": "success", "prediction": prediction})

if __name__ == "__main__":
    app.run(debug=True)

4. Flask vs Django vs FastAPI

Factor	Flask	Django	FastAPI
Complexity	Minimal	Full-featured	Minimal
Learning curve	Very easy	Moderate	Easy
Speed	Moderate	Moderate	High (async)
Flexibility	Very high	Lower (opinionated)	Moderate
Best for	APIs, light apps, ML demos	Full web apps, ORM, auth	Fast APIs, async inference

For data science teams shipping models quickly, Flask’s low overhead and clean integration with Scikit-Learn, PyTorch, and TensorFlow make it a practical first step.