dingjingtao / negativesamplerbpr Goto Github PK

Hi dingjingtao,

Thank you for sharing the code and datasets.

I notice that the statics of Beibei is not consistent with the one presented in "Sampler Design for Bayesian Personalized Ranking by Leveraging View Data”. For example, the number of users in the provided Beibei dataset is 10,000 while the paper stated it is 158,907. It seems that the Beibei dataset is a partial one.

Could you please share the full Beibei dataset?

Thank you very much!