search

maciejkula / spotlight

Deep recommender models using PyTorch.
MIT License
2.96k stars 421 forks source link

Session-based recommendation #125

Open Edogawa-Konan opened 5 years ago

Edogawa-Konan commented 5 years ago

Hello, Recently, I am trying to implement paper Recurrent Collaborative Filtering for Unifying General and Sequential Recommender with pytorch. By accident I found this project, which is very cool. However, I have some questions about ImplicitSequenceModel. Paper Session-based Recommendations with Recurrent Neural Networks didn't refer to any methods like user_representation. Besides, that paper used GRU, this project used LSTM instead.There are many technologies such as mini_batch, output_sample..etc in that paper. I think the implementation of this model is completely different from that paper. I am a newcomer in the field of Session-based recommendation systems. I wondered whether you can explain it to me? Thank you. @maciejkula

maciejkula commented 5 years ago

Yes, it is different! But the principles are very similar.