Skip to content

BPR, Bayesian Personalized Ranking (BPR), extremely convenient BPR & Multiple Pairwise Ranking

License

Notifications You must be signed in to change notification settings

RunlongYu/BPR_MPR

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

75 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

BPR_Bayesian-Personalized-Ranking_MPR_Multiple-Pairwise-Ranking

  • Implement Steffen Rendle, et al. Bayesian personalized ranking from implicit feedback (run BPR.py);
  • Implement Runlong Yu, et al. Multiple Pairwise Ranking with Implicit Feedback (run MPR.py) in Python3.

Dataset describe: MovieLens100K; 943 Users; 1682 Items

  • More details about MovieLens datasets at https://grouplens.org/datasets/movielens .
  • Our code does not depend on the datasets. Just split the dataset you are interested in into the training set and test set, replace the train.txt and test.txt files, you can run BPR and MPR easily.

About MPR Framework:

avatar

avatar


Note:

In the MPR paper, the dataset is divided into different sets according to popularity, and the method is too dependent on the dataset. In the open source MPR code, we have used the negative sampling method (refer to AoBPR, DNS methods) instead of the method of data division, and achieved better results.


About Runlong Yu:

About

BPR, Bayesian Personalized Ranking (BPR), extremely convenient BPR & Multiple Pairwise Ranking

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages