Semantic Recipe Finder Documentation
Welcome!
This application uses natural language semantic search to help you discover recipes intelligently.
📚 Documentation Sections
Learn about the system architecture, data flow, and core technologies powering the semantic search.
Complete API documentation with request/response examples and endpoint specifications.
🚀 Quick Links
Live Resources
- 🌐 Live Demo: HuggingFace Spaces
- 💻 GitHub Repository: semantic-recipe-finder
- 🐛 Issue Tracker: GitHub Issues
✨ Key Features
Query recipes using everyday language
Powered by ChromaDB and sentence-transformers
FastAPI backend + Streamlit frontend
Docker support with comprehensive testing
61 tests with 100% passing rate
🎯 Use Cases
Recipe Discovery
Find recipes based on ingredients, cuisine, or cooking style
Dietary Filtering
Search for recipes matching specific dietary requirements
Time-Based Search
Discover quick meals or elaborate dishes
Semantic Understanding
Natural queries like "healthy vegetarian dinner under 30 minutes"
🛠️ Technology Overview
| Component | Technology |
|---|---|
| Backend | FastAPI 0.115.6 |
| Frontend | Streamlit 1.41.1 |
| ML Model | sentence-transformers (all-MiniLM-L6-v2) |
| Vector DB | ChromaDB 0.5.23 |
| Testing | pytest 9.0.2 (61 tests) |
| Deployment | Docker, docker-compose |
📊 Dataset
Dataset Details
- 📚 10000 recipes from Food.com dataset
- 🔢 384-dimensional embeddings for semantic search
- 📝 Rich metadata: ingredients, instructions, nutrition, ratings
- 💾 Persistent storage with ChromaDB
🤝 Getting Help
Need Support?
- 🐛 Found a bug? Open an issue
- 📧 Contact: hanifekaptan.dev@gmail.com
📄 License
License
This project is open source and available under the Apache Licence.
Navigation: Home | Architecture | API