→ Watch it directly on YouTube

✅ Logistic Regression churn prediction
✅ FastAPI REST endpoint
✅ OpenAI summarisation for human-readable explanations

1. Install requirements:

pip install -r requirements.txt

2. Run locally:

uvicorn app:app --reload

• app:app loads the FastAPI app from app.py.
• --reload enables auto-reload for development (useful for code changes, not for production).

3. Test with:

curl -X POST http://localhost:8000/predict -H "Content-Type: application/json" -d '{
  "age": 45.0,
  "tenure": 24.0,
  "monthly_charges": 79.85,
  "total_charges": 1800.0,
  "contract_type": "Month-to-month",
  "payment_method": "Electronic check"
}'

Set your OpenAI API key:

export OPENAI_API_KEY="your_actual_openai_api_key"

Python_GML_ML_Pipeline/
├── app.py                  # FastAPI application
├── requirements.txt        # Python dependencies
├── logistic_model.pkl     # Trained ML model (placeholder)
├── scaler.pkl            # Feature scaler (placeholder)
├── Dockerfile            # Container configuration
├── .gitignore           # Git ignore rules
└── README.md           # This file

1. Build the Docker image:

docker build -t python-gml-ml-pipeline .

2. Run the Docker container:

docker run -p 8000:8000 -e OPENAI_API_KEY="your_actual_openai_api_key" python-gml-ml-pipeline

3. Access the FastAPI app:

• API: http://localhost:8000
• Health Check: http://localhost:8000/health
• Interactive Docs: http://localhost:8000/docs
• OpenAPI Schema: http://localhost:8000/openapi.json

Execution Flow:

1. Load trained model and scaler (using joblib).
2. API endpoint receives JSON data (new user or input data).
3. Dataframe creation & scaling for consistency with training.
4. Model predicts churn probability (or other target).
5. Returns JSON response with prediction for integration into apps or dashboards.

✅ Run locally:

uvicorn app:app --reload

✅ Run in Docker:

docker build -t python-gml-ml-pipeline .
docker run -p 8000:8000 -e OPENAI_API_KEY="your_actual_openai_api_key" python-gml-ml-pipeline

Key reasons to use FastAPI:

1. Modern async Python framework
2. Automatic OpenAPI schema & Swagger docs
3. Production-grade performance
4. Can be integrated into microservices/SaaS

Made with ❤️ by Pierre-Henry Soria. A super passionate & enthusiastic Problem-Solver / Senior Software Engineer. Also a true cheese 🧀, ristretto ☕️, and dark chocolate lover! 😋

• logistic_model.pkl and scaler.pkl are placeholders. Train and export your own models using joblib.dump.
• This project is a modern, production-ready ML pipeline, showcasing deployment and explainability best practices for 2025 and beyond.

“AI models become valuable when they’re deployable, explainable, and integrated into real products that create business value.”