How shafwansafi06/campus-entity-resolver works
Project Type and Purpose
• This is a full-stack, AI-powered security & intelligence platform (“CampusFlow”) built for smart campuses.
• Frontend: React + TypeScript SPA; Backend: FastAPI Python server; ML Engine: PyTorch + XGBoost + LightGBM; Database/Auth: Supabase (PostgreSQL + Realtime + Auth).
File Structure
• Top level:
• package.json, vite.config.ts, tailwind.config.ts → frontend build & config
• src/ → React application (components/, pages/, contexts/, hooks/, lib/)
• public/ → static assets
• backend/ → Python FastAPI service
– main.py → API entrypoint
– database.py → Supabase integration
– entity_resolution.py → fuzzy-matching logic (FuzzyWuzzy)
– ml_predictor.py → loads ML artifacts (.pt, .json, .txt, .pkl)
– predictive_analytics.py → forecasts & analytics endpoints
– models/ → serialized ML models
• Config scripts: .env, creds.env, create-test-user.sh, bun.lockb, package-lock.json.
README Insights
• Tech stack and ML ensemble breakdown.