deepctr 数据预处理部分,解读_deedctr-CSDN博客
文章浏览阅读424次。import pandas as pdfrom sklearn.preprocessing import LabelEncoder, MinMaxScalerdata_frame = pd.read_csv('./data/criteo_sample.txt')# C开头的列,值都是乱码,所以按枚举 转为idsparse_feature_names = ['C' + str(i) for i in range(1, 27)] # sparse的意思,......