Plutonian-coder/churn-prediction-mlops-pipeline
📈 Build a production-ready churn prediction system with an end-to-end MLOps pipeline using FastAPI, scikit-learn, and automated CI/CD practices.
What's novel
📈 Build a production-ready churn prediction system with an end-to-end MLOps pipeline using FastAPI, scikit-learn, and automated CI/CD practices.
Code Analysis
11 files read · 3 roundsA minimal Random Forest churn prediction script with 3 hardcoded features, a single FastAPI inference endpoint, and a basic HTML demo form.
Strengths
The code is simple and readable, and the FastAPI app includes sensible touches like Prometheus metrics, CORS configuration, and a token-gated model reload endpoint.
Weaknesses
The README is wildly misleading (describes a downloadable desktop app with installers that don't exist), the sample data is random with no signal, evaluate.py has data leakage, there is only one trivial test, and the 'MLOps pipeline' has no orchestration, validation, or deployment infrastructure.
Score Breakdown
Signal breakdown
Innovation
Craft
Traction
Scope
Evidence
Commits
9
Contributors
3
Files
16
Active weeks
4
Repository
Language
Python
Stars
2
Forks
0
License
—