This is the complete implementation of the Campus Lost-and-Found Intelligence System (CLFIS).
- Docker & Docker Compose
- Python 3.11+
- Node.js 18+
- PostgreSQL 16 (via Docker)
docker-compose up -dThis will start:
- PostgreSQL database (port 5432)
- Backend FastAPI server (port 8000)
- Frontend Next.js application (port 3000)
docker run --name clfis-pg -e POSTGRES_USER=postgres -e POSTGRES_PASSWORD=postgres \
-e POSTGRES_DB=clfis_db -p 5432:5432 -d postgis/postgis:16-3.4cd backend
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
pip install -r requirements.txt
uvicorn app.main:app --reloadcd frontend
npm install
npm run dev- database/: PostgreSQL schema with pgvector & PostGIS
- backend/: FastAPI application with ML services
- frontend/: Next.js 14 app with React & Tailwind
- ml/: ML evaluation suite and model integration
- scripts/: Setup and utility scripts
✅ Multimodal item matching (SigLIP embeddings) ✅ Spatiotemporal scoring with decay functions ✅ Zero-Knowledge claim verification ✅ Cryptographic QR handshake system ✅ Campus zone awareness (PostGIS) ✅ OCR token extraction & matching ✅ Karma score system ✅ Admin vault management ✅ ML metrics evaluation (MRR, NDCG, Recall@K)
Visit http://localhost:8000/docs for interactive API docs (Swagger UI)
Create .env file:
DATABASE_URL=postgresql://postgres:postgres@localhost:5432/clfis_db
SECRET_KEY=your-secret-key
CAMPUS_EMAIL_DOMAIN=college.edu
NEXT_PUBLIC_API_URL=http://localhost:8000/api
- Next.js 14 (App Router)
- TypeScript
- Tailwind CSS
- Zustand for state management
- Axios for API calls
- FastAPI
- SQLAlchemy ORM
- PostgreSQL with pgvector & PostGIS
- Pydantic validation
- JWT authentication
- Hugging Face SigLIP model
- Tesseract OCR
- Vector embeddings (768-d)
- Scikit-learn metrics
cd backend
pytest tests/python ml/src/evaluation/run_eval.pySee CONTRIBUTING.md for guidelines
MIT