This is the author's officially unofficial TensorFlow implementation of CDL. Note that the code is a simplified version of CDL and is used for demonstration only.

To train the model

• pretrain model (CDL) run 'python train_sdae.py'
• train CDL run 'python train_cdl.py'

To evaluate performance

• run lib/mainEvaluate.m under Matlab

Other materials

• More details on the work and some direct extensions can be found at http://wanghao.in/CDL.htm.

• Matlab version for full CDL: https://github.com/js05212/CDL.

• MXNet version for simplified CDL: https://github.com/js05212/MXNet-for-CDL.

• Data: https://www.wanghao.in/data/ctrsr_datasets.rar.

• Slides: http://wanghao.in/slides/CDL_slides.pdf and http://wanghao.in/slides/CDL_slides_long.pdf.

Other implementations (third-party):

• Tensorflow code by gtshs2.

• Keras code by zoujun123.

• Python code by xiaoouzhang.

Acknowledgement

• This code is based on my original Matlab implementation of CDL, the MXNet version, as well as Xiaopeng Li's Tensorflow code.

Reference:

Collaborative Deep Learning for Recommender Systems

@inproceedings{DBLP:conf/kdd/WangWY15,
  author    = {Hao Wang and
               Naiyan Wang and
               Dit{-}Yan Yeung},
  title     = {Collaborative Deep Learning for Recommender Systems},
  booktitle = {SIGKDD},
  pages     = {1235--1244},
  year      = {2015}
}