Librecommender

🚀 Features
+ Support train & save & retrain PyTorch models (141)
+ Add GraphSage, GraphSageDGL (161)
+ Add PinSage (146), PinSageDGL (150)
+ Add serving models with [Rust](https://www.rust-lang.org/) (#157)
+ Add serving using Docker Compose (156)

🐛 Bug Fixes
+ Fix retraining `item2vec` and `deepwalk` models (163)

📚 Documentation
+ Add [Rust Serving Guide](https://github.com/massquantity/LibRecommender/blob/master/doc/rust_serving_guide.md) (#157)
+ Refactor docs (155)

Bug Fixes
+ Fix mistakenly preserving feature state after running (132)

API Changes
+ remove `reset_state` parameter in `DatasetFeat.build_trainset` and `DatasetPure.build_trainset` (132)

+ Add [focal loss](https://github.com/massquantity/LibRecommender/tree/master/examples#loss).
+ Add [embedding APIs](https://github.com/massquantity/LibRecommender/tree/master/examples#embedding).
+ Refactor the serving part. The original `serving` module is renamed to [`libserving`](https://github.com/massquantity/LibRecommender/tree/master/libserving).

Add graph algorithms: DeepWalk, NGCF, LightGCN