scorecardpy

scorecardpy copied to clipboard

两个地方算出来的iv不同

1

谢老师，您好！我在使用评分卡模型包的时候，发现var_filter中计算出来的iv值和使用woebin中计算出来的iv是不一样，前一个是没有分箱直接计算出来的iv，后一个是通过分箱计算出来的iv。那么问题来了，为什么变量筛选这一步不放在分箱后进行，而是一开始就进行变量筛选？

RAO-sy

error on pip install git+git://github.com/shichenxie/scorecardpy.git

1

Sukhman-symend

Can only pass argument "on" OR "left_index" and "right_index", not a combination of both.

3

sc.woebin的时候报这个错，是什么意思呀，有没有大佬帮忙解答一下

gc277000

MergeError when executing "woebin" function

9

Hi,  few days ago after updating the PANDAS library to version 1.2.0, the "woebin" function of scorerapy version '0.1.9.2' stopped working. When trying to execute it, the error is...

kendalvictor

woebin_plot graphs are not rescalable

2

Currently I am using scorecardpy version 0.1.9.2 and I am trying to plot plot using `woebin_plot() ` and it is generating plots. but when there is lengthy string in "bin"...

divatemangesh

代码报错

4

import scorecardpy as sc dat = sc.germancredit() bins = sc.woebin(dt_s, y="creditability") 运行这一步的时候报错，错误提示为： TypeError: unhashable type: 'numpy.ndarray' 是不是源码有异常？ 使用的版本是今天安装的最新版

xiaofengfengye

About the monotonicity of bins.

2

Normally，it is essential for the bins besides missing to be monotonically increasing or decreasing on the badrate for better explainability. For exmple， the bigger a certain feature X1，the better chance...

UTimeStrange

perf.py The calculation of KS

6

当我使用`sc.perf_eva(y_train, train_pred , plot_type = ["ks"])`计算KS时，KS曲线的定义貌似不是正确的。 正确的KS曲线，应该是以阈值点为x轴，TPR和FPR为y轴，显然源码中计算是错误。这里附上我，参考网上的部分代码，并自己更改后的代码，希望谢老师能参考一下： ``` def model_score_pro_ks(df_fact_tag, df_expected_score_or_pro, buckets, type_input='score'): # 初始等间隔分段区间列表 breakpoints = np.arange(0, buckets + 1) / (buckets) # 将预期得分，按照等距分箱，分成buckets个，返回的是每段的上下界限,array # input指的是，初始分箱间隔，组成的列表 # min和max指的是，df_expected_score_or_pro的最大最小值 def...

sunrisehang

Hi! Sometimes I need to keep different versions of scorecardpy with some modifications at the same place, e.g. `scorecardpy_019/`, `scorecardpy_0193ab/` etc. To make this possible the sub-modules must be imported...

evgenykurbatov

Woe won't work with NaNs

2

Thanks for putting this package together - it's great to have a scorecard package in Python. However, it appears the weight of evidence binning algorithm only works with complete data,...

rcg-uab