deepctr.models.mlr module

Weichen Shen,
[1] Gai K, Zhu X, Li H, et al. Learning Piece-wise Linear Models from Large Scale Data for Ad Click Prediction[J]. arXiv preprint arXiv:1704.05194, 2017.(
deepctr.models.mlr.MLR(region_feature_dim_dict, base_feature_dim_dict={'dense': [], 'sparse': []}, region_num=4, l2_reg_linear=1e-05, init_std=0.0001, seed=1024, final_activation='sigmoid', bias_feature_dim_dict={'dense': [], 'sparse': []})[source]

Instantiates the Mixed Logistic Regression/Piece-wise Linear Model.

  • region_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’]}
  • base_feature_dim_dict – dict or None,to indicate sparse field and dense field of base learner.if None, it is same as region_feature_dim_dict
  • region_num – integer > 1,indicate the piece number
  • l2_reg_linear – float. L2 regularizer strength applied to weight
  • init_std – float,to use as the initialize std of embedding vector
  • seed – integer ,to use as random seed.
  • final_activation – str,output activation,usually 'sigmoid' or 'linear'
  • bias_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’]}

A Keras model instance.