/CTR

CTR模型代码和学习笔记总结

Primary LanguagePython

CTR学习笔记

  • Run: python main.py --model DeepFM --step train --dataset census --clear_model 1
  • Requirement: tensorflow 1.15
  1. 已完成模型列表[支持数据集]
  • FM [census]
  • FFM [census]
  • Embedding+MLP [census]
  • wide & Deep [census]
  • FNN [census]
  • PNN [census]
  • DeepFM [census & frappe]
  • AFM [census & frappe]
  • NFM [census & frappe]
  • Deep Crossing [census]
  • Deep & Cross [census & frappe]
  • xDeepFM [census & frappe]
  • FiBiNET [census & frappe]
  • DIN [amazon]
  1. 数据集 当前支持census, frappe数据集,详情见data目录,training parameter和preprocess与数据集绑定

  2. 参考论文列表

  • [GBDT+LR] Practical Lessons from Predicting Clicks on Ads at Facebook
  • [FM] S. Rendle, Factorization machines
  • [FM Model] Fast Context-aware Recommendations with Factorization Machines
  • [FFM] Yuchin Juan,Yong Zhuang,Wei-Sheng Chin,Field-aware Factorization Machines for CTR Prediction
  • [NCF] Neural Collaborative Filtering
  • [Wide&Deep] Cheng H T, Koc L, Harmsen J, et al. Wide & deep learning for recommender systems
  • [FNN] Weinan Zhang, Tianming Du, and Jun Wang. Deep learning over multi-field categorical data - - A case study on user response
  • [PNN] Qu Y, Cai H, Ren K, et al. Product-based neural networks for user response prediction
  • [DeepFM] Huifeng Guo et all. DeepFM: A Factorization-Machine based Neural Network for CTR Prediction
  • [AFM] Attentional Factorization Machines - Learning the Weight of Feature Interactions via Attention Networks
  • [NFM] Neural Factorization Machines for Sparse Predictive Analytics
  • [DCN] Deep & Cross Network for Ad Click Predictions
  • [Deep Crossing] Deep Crossing - Web-Scale Modeling without Manually Crafted Combinatorial Features
  • [xDeepFM] xDeepFM- Combining Explicit and Implicit Feature Interactions for Recommender Systems
  • [FiBiNET]- Combining Feature Importance and Bilinear feature Interaction for Click-Through Rate Prediction
  • [AutoInt]- Automatic Feature Interaction Learning via Self-Attentive Neural Networks
  • [DIN] Deep Interest Network for Click-Through Rate Prediction.
  • [DIEN] Deep Interest Evolution Network for Click-Through Rate Prediction
  1. 总结博客