README.md

metadata

title: NCF Movie Recommender
emoji: 🎬
colorFrom: purple
colorTo: blue
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: false
license: mit

🎬 NCF Movie Recommender

A Neural Collaborative Filtering (NCF) movie recommendation demo trained on the MovieLens 1M dataset.

How It Works

1. Rate some movies you've already watched (1-5 stars)
2. Click "Get Recommendations" to find your next favorite film
3. The model finds users with similar tastes and predicts what you'll enjoy

Model Details

• Architecture: Neural Collaborative Filtering (NCF)
• Training Data: MovieLens 1M (1 million ratings from 6,040 users on 3,900 movies)
• User Embeddings: 6,041 × 128
• Item Embeddings: 3,953 × 128
• MLP Layers: 256 → 128 → 64 → 1

Performance Metrics

• HR@10: 0.132 (Hit Rate at 10)
• NDCG@10: 0.091 (Normalized Discounted Cumulative Gain)

Citation

If you use the MovieLens dataset, please cite:

F. Maxwell Harper and Joseph A. Konstan. 2015. The MovieLens Datasets: History and Context. ACM Transactions on Interactive Intelligent Systems (TiiS) 5, 4, Article 19 (December 2015), 19 pages. DOI: http://dx.doi.org/10.1145/2827872

Files

├── app.py              # Main Gradio application
├── requirements.txt    # Python dependencies
├── data/
│   ├── model.pt        # Trained NCF model weights
│   ├── movies.dat      # Movie metadata
│   ├── ratings.dat     # User-movie ratings
│   └── users.dat       # User demographics