Recommender System in PyTorch

Implementations of various top-N recommender systems in PyTorch for practice.

Movielens 100k & 1M are used as datasets.

Available models

Model	Paper
BPRMF	Steffen Rendle et al., BPR: Bayesian Personalized Ranking from Implicit Feedback. UAI 2009. Link
ItemKNN	Jun Wang et al., Unifying user-based and item-based collaborative filtering approaches by similarity fusion. SIGIR 2006. Link
PureSVD	Paolo Cremonesi et al., Performance of Recommender Algorithms on Top-N Recommendation Tasks. RecSys 2010. Link
SLIM	Xia Ning et al., SLIM: Sparse Linear Methods for Top-N Recommender Systems. ICDM 2011. Link
P3a	Colin Cooper et al., Random Walks in Recommender Systems: Exact Computation and Simulations. WWW 2014. Link
RP3b	Bibek Paudel et al., Updatable, accurate, diverse, and scalablerecommendations for interactive applications. TiiS 2017. Link
DAE, CDAE	Yao Wu et al., Collaborative denoising auto-encoders for top-n recommender systems. WSDM 2016.Link
MultVAE	Dawen Liang et al., Variational Autoencoders for Collaborative Filtering. WWW 2018. Link
EASE	Harald Steck, Embarrassingly Shallow Autoencoders for Sparse Data. WWW 2019. Link
NGCF	Xiang Wang, et al., Neural Graph Collaborative Filtering. SIGIR 2019. Link
LightGCN	Xiangnan He, et al., LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation. SIGIR 2020. Link

Enable C++ evaluation

To evaluate with C++ backend, you have to compile C++ and cython with the following script:

python setup.py build_ext --inplace

If compiled NOT successfully, "evaluation with python backend.." will be printed in the beginning.

How to run

1. Edit experiment configurations in config.py
2. Edit model hyperparameters you choose in conf/[MODEL_NAME]
3. run main.py

Implement your own model

You can add your own model into the framework if:

1. Your model inherits BaseModel class in models/BaseModel.py
2. Implement necessary methods and add additional methods if you want.
3. Make YourModel.conf file in conf
4. Add your model in models.__init__

Reference

Some model implementations and util functions refers to these nice repositories.

• NeuRec: An open source neural recommender library. Repository
• RecSys 2019 - DeepLearning RS Evaluation. Paper Repository