PyTorch based implementation of Graph Convolutional Matrix Completion for recommender systems, based on Kipf and Welling (2017) paper. We also implemented them based on their source code.
This code only covers the Movielens 100K, 1M Dataset.
After downloading ml_1m to the ./data directory, you need to preprocess it by Preprocess.ipynb.
- Python 3.5
- PyTorch (0.4)
python main.py| Kipf and Welling | Ours | |
|---|---|---|
| ML-100K | 0.910 | 0.935 |
| ML-100K (+Feat) | 0.905 | 0.932 |
| ML-1M | 0.832 | - |