cheon-research / J-NCF-pytorch

PyTorch Implementation for J-NCF (ACM Transactions on Information Systems, 2019)

J-NCF pytorch implementation

Non-official Implementation for "Joint Neural Collaborative Filtering for Recommender Systems"
ACM Transactions on Information Systems, Vol. 37, No. 4, Article 39. (August 2019)

https://dl.acm.org/doi/10.1145/3343117
https://arxiv.org/pdf/1907.03459.pdf

It is not sure that this implementation has same performance with authors' experiments.
(Followed the authors' experimental setups as much as possible.)

Implemented on following environments

* python==3.7.6
* pytorch==1.4.0
* numpy==1.18.1
* pandas==1.0.0
* scipy==1.4.1

Notes

If you have any idea to make faster in sparse matrix indexing(input interaction history), please let me know.

cheon.research (at) gmail.com

(some figures will be updated...)