评分卡应用 - 利用Toad进行有监督分箱（卡方分箱/决策树分箱）

toad.detect(train)[:10]

to_drop = ['APP_ID_C','month'] # 去掉ID列和month列
toad.quality(data.drop(to_drop,axis=1),'target',iv_only=True)[:15]

toad.selection.select(dataframe, target=’target’, empty=0.9, iv=0.02, corr=0.7, return_drop=False, exclude=None):

train_selected, dropped = toad.selection.select(train,target = 'target', empty = 0.5, iv = 0.05, corr = 0.7, return_drop=True, exclude=['APP_ID_C','month'])
print(dropped)
print(train_selected.shape)

c.fit(dataframe, y = ‘target’, method = ‘chi’, min_samples = None, n_bins = None, empty_separate = False)

*查看分箱节点*：c.export()
*手动调整分箱*: c.load(dict)
*apply分箱结果*: c.transform(dataframe, labels=False):
	- labels: 是否将分箱结果转化成箱标签。False时输出0,1,2…（离散变量根据占比高低排序），True输出(-inf, 0], (0,10], (10, inf)。

# initialise
c = toad.transform.Combiner()

# 使用特征筛选后的数据进行训练：使用稳定的卡方分箱，规定每箱至少有5%数据, 空值将自动被归到最佳箱。
c.fit(train_selected.drop(to_drop, axis=1), y = 'target', method = 'chi', min_samples = 0.05) #empty_separate = False

# 为了演示，仅展示部分分箱
print('var_d2:',c.export()['var_d2'])
print('var_d5:',c.export()['var_d5'])
print('var_d6:',c.export()['var_d6'])

var_d2: [747.0, 782.0, 820.0]
var_d5: [['O', 'nan', 'F'], ['M']]
var_d6: [['PUBLIC LTD COMPANIES', 'NON-RESIDENT INDIAN', 'PRIVATE LTD COMPANIES', 'PARTNERSHIP FIRM', 'nan'], ['RESIDENT INDIAN', 'TRUST', 'TRUST-CLUBS/ASSN/SOC/SEC-25 CO.', 'HINDU UNDIVIDED FAMILY', 'CO-OPERATIVE SOCIETIES', 'LIMITED LIABILITY PARTNERSHIP', 'ASSOCIATION', 'OVERSEAS CITIZEN OF INDIA', 'TRUST-NGO']]

 toad.plot.bin_plot(dataframe, x = None, target = ‘target)

from toad.plot import bin_plot
col = 'sepal length (cm)'
bin_plot(c.transform(iris[[col,'target']], labels=True), x=col, target='target')

# 初始化
transer = toad.transform.WOETransformer()

# combiner.transform() & transer.fit_transform() 转化训练数据，并去掉target列
train_woe = transer.fit_transform(c.transform(train_selected), train_selected['target'], exclude=to_drop+['target'])
OOT_woe = transer.transform(c.transform(OOT))

print(train_woe.head(3))

gbdt_transer = toad.transform.GBDTTransformer()
gbdt_transer.fit(final_data[col+['target']], 'target', n_estimators = 10, max_depth = 2)

gbdt_vars = gbdt_transer.transform(final_data[col])

gbdt_vars.shape
>>> (43576, 40)

import pandas as pd
import numpy as np
import toad
from sklearn.datasets import load_iris

# 数据载入
iris = load_iris()
target = iris['target']
iris = pd.DataFrame(iris['data'],columns = iris['feature_names'])
iris['target'] = target

# 数据洞察
eda = toad.detect(iris)
qualitys = toad.quality(iris,'target',iv_only=True)


# 分箱
c = toad.transform.Combiner()
    # 使用特征筛选后的数据进行训练：使用稳定的卡方分箱，规定每箱至少有5%数据, 空值将自动被归到最佳箱。
c.fit(iris, y = 'target', method = 'chi', min_samples = 0.05) #empty_separate = False
    # 为了演示，仅展示部分分箱
c.export()
print('var_d2:',c.export()['var_d2'])
print('var_d5:',c.export()['var_d5'])
print('var_d6:',c.export()['var_d6'])


# 画分箱图
from toad.plot import bin_plot
# 看'var_d2'在时间内的分箱
col = 'sepal length (cm)'
bin_plot(c.transform(iris[[col,'target']], labels=True), x=col, target='target')
    # bar代表了样本量占比，红线代表了正样本占比（e.g. 坏账率）


# GBDT + LR的树模型特征输出
col = 'sepal length (cm)'
gbdt_transer = toad.transform.GBDTTransformer()
gbdt_transer.fit(iris, 'target', n_estimators = 10, max_depth = 2)
gbdt_vars = gbdt_transer.transform(iris[['sepal length (cm)', 'sepal width (cm)', 'petal length (cm)',
       'petal width (cm)']])
iris.shape,gbdt_vars.shape

# initialise
c = toad.transform.Combiner()

# 使用特征筛选后的数据进行训练：使用稳定的卡方分箱，规定每箱至少有5%数据, 空值将自动被归到最佳箱。
c.fit(data, method = 'quantile',n_bins  = 3) 

c.transform(data, labels=False)