The is the code for"Bayesian Deep Collaborative Matrix Factorization(BDCMF)" (AAAI 2019). It consists of two parts: a Matlab component and a Python component.
Requirements:
Python 3.4
Tensorflow 1.7
Matlab
To run "train.py", it generates two '.mat' files: U.mat and V.mat which are latent factors of users and items, respectively. Then you can use the two files to evaluate recommendation performance on test data.