single_filter online - shuiwanghuohuo/scorecard_wiki GitHub Wiki
from online import feature
feature.single_filter(data_rdd, not_in_list=["None", "NaN", "NA", "nan", None, "-999", "-999.0", -999, "-1111", "-1111.0", -1111])
计算指标的单一阈值
Parameter Description
---------------------
data_rdd : pyspark.rdd.PipelinedRDD
spark dataframe经过trans_rdd转换过的数据集
Return
------
result : 一个pandas dataframe 包括了每个特征的变量名,单一阈值