deepctr.models.dcn module

Author:
Weichen Shen,wcshen1994@163.com
Reference:
[1] Wang R, Fu B, Fu G, et al. Deep & cross network for ad click predictions[C]//Proceedings of the ADKDD‘17. ACM, 2017: 12. (https://arxiv.org/abs/1708.05123)
deepctr.models.dcn.DCN(feature_dim_dict, embedding_size='auto', cross_num=2, hidden_size=(128, 128), l2_reg_embedding=1e-05, l2_reg_cross=1e-05, l2_reg_deep=0, init_std=0.0001, seed=1024, keep_prob=1, use_bn=False, activation='relu', final_activation='sigmoid')[source]

Instantiates the Deep&Cross Network architecture.

Parameters:
  • feature_dim_dict – dict,to indicate sparse field and dense field like {‘sparse’:{‘field_1’:4,’field_2’:3,’field_3’:2},’dense’:[‘field_4’,’field_5’]}
  • embedding_size – positive int or str,sparse feature embedding_size.If set to “auto”,it will be 6*pow(cardinality,025)
  • cross_num – positive integet,cross layer number
  • hidden_size – list,list of positive integer or empty list, the layer number and units in each layer of deep net
  • l2_reg_embedding – float. L2 regularizer strength applied to embedding vector
  • l2_reg_cross – float. L2 regularizer strength applied to cross net
  • l2_reg_deep – float. L2 regularizer strength applied to deep net
  • init_std – float,to use as the initialize std of embedding vector
  • seed – integer ,to use as random seed.
  • keep_prob – float in (0,1]. keep_prob used in deep net
  • use_bn – bool. Whether use BatchNormalization before activation or not.in deep net
  • activation – Activation function to use in deep net
  • final_activation – str,output activation,usually 'sigmoid' or 'linear'
Returns:

A Keras model instance.