br_encoding_dict[col] = br_encoding['bad_rate']
num_features.append(col + '_br_encoding')
file2 = open(folderOfData + 'br_encoding_dict.pkl', 'wb+')
pickle.dump(br_encoding_dict, file2)
file2.close()
# （iii）对连续型变量进行分箱，包括（ii）中的变量
continous_merged_dict = {}
for col in num_features:
print("{} is in processing".format(col))
if -1 not in set(trainData[col]): # －1会当成特殊值处理。如果没有－1，则所有取值都参与分箱
max_interval = 5 # 分箱后的最多的箱数
cutOff = scorecard_function.ChiMerge(trainData,
col,
'y',
max_interval=max_interval,
special_attribute=[],
minBinPcnt=0)
trainData[col + '_Bin'] = trainData[col].map(
lambda x: scorecard_function.AssignBin(
x, cutOff, special_attribute=[]))
monotone = scorecard_function.BadRateMonotone(trainData, col + '_Bin',
'y') # 检验分箱后的单调性是否满足
while (not monotone):
# 检验分箱后的单调性是否满足。如果不满足，则缩减分箱的个数。
max_interval -= 1
cutOff = scorecard_function.ChiMerge(trainData,
col,
'y',
max_interval=max_interval,
special_attribute=[],
x2_bin = range(25,105,5)
x3_bin = range(1,11)
x7_bin = range(1,11)
x9_bin = range(1,11)
x4_bin = list(np.linspace(0.1,0.9,num=9))
x5_bin = range(1000,14000,1000)+range(20000,50000,10000)
x6_bin = range(1,20)
x8_bin = range(1,10)
x10_bin = range(1,5)
## 卡方分箱
print '卡方分箱结果为：'
for x in X.columns:
locals()[x+'_chi'] = scorecard_function.ChiMerge(train, x, 'y', max_interval=10, special_attribute=[], minBinPcnt=0)
print x , locals()[x+'_chi']
##决策树分箱
from sklearn import tree
def decession_tree_bin(X,y,bin):
clf = tree.DecisionTreeClassifier(criterion = 'entropy',max_leaf_nodes= bin,min_samples_leaf = 0.05).fit(X,y)
n_nodes = clf.tree_.node_count
children_left = clf.tree_.children_left
children_right = clf.tree_.children_right
threshold = clf.tree_.threshold
boundary = []
for i in range(n_nodes):
if children_left[i]!=children_right[i]:
boundary.append(threshold[i])
sort_boundary = sorted(boundary)