pro-grammer-SD/food-rec
🥗 ML-powered food recommendations app using scikit-learn
What's novel
🥗 ML-powered food recommendations app using scikit-learn
Code Analysis
2 files read · 2 roundsA content-based food recommendation system that constructs a synthetic user profile from CLI or Streamlit inputs and finds the top-10 matching foods using TF-IDF text features, one-hot encoded categories, and scaled nutrients via K-Nearest Neighbors.
Strengths
The project provides a functional, offline-first solution with a clean UX in both CLI and Streamlit. The synthetic profile generation (quantile-based target setting) is a clever way to handle cold-start without complex training loops.
Weaknesses
Significant code duplication between main.py and app.py violates DRY principles. There is no error handling for missing data beyond basic fillna, no input validation, and zero test coverage.
Score Breakdown
Signal breakdown
Innovation
Craft
Traction
Scope
Evidence
Commits
5
Contributors
1
Files
5
Active weeks
1
Repository
Language
Python
Stars
6
Forks
0
License
—