pytorch implements of some popular deep network , mainly for ctr and other prediction, for example DCN, etc. What I want to mention is that this project is mainly used for learning the algorithm, not the higher result and lower loss.
the using example:
python main.py --model_type dcn --early_stopping 4
update on 01/03/2023
- paper: Deep & Cross Network for Ad Click Predictions[2017], from google
- website: https://arxiv.org/pdf/1708.05123.pdf
update on 07/03/2023
- paper: Wide & Deep Learning for Recommender Systems[2016], from google
- website: https://arxiv.org/pdf/1606.07792.pdf
continuing ... ...