If you have any questions, feel free to ask! :)

CETN: Contrast-enhanced Through Network for CTR Prediction

Model Overview

Requirements

python>=3.6
pytorch>=1.0
fuxictr
PyYAML
pandas
scikit-learn
numpy
h5py
tqdm

Experiment results

Datasets

Get the datasets from https://github.com/reczoo/Datasets

Hyperparameter settings and logs

Get the result from https://github.com/salmon1802/CETN/tree/main/CETN_torch/checkpoints

Acknowledgement

This implementation is based on FuxiCTR and BARS. Thanks for their sharing and contribution.
BARS: https://github.com/openbenchmark
FuxiCTR: https://github.com/xue-pai/FuxiCTR

Citation

If you find our code helpful for your research, please cite the following paper:

@article{li2023cetn,
  title={CETN: Contrast-enhanced Through Network for CTR Prediction},
  author={Li, Honghao and Sang, Lei and Zhang, Yi and Zhang, Xuyun and Zhang, Yiwen},
  journal={arXiv preprint arXiv:2312.09715},
  year={2023}
}