sklearn中KFold,StratifiedKFold的区别

最新推荐文章于 2022-10-14 19:59:56 发布

joker-G

最新推荐文章于 2022-10-14 19:59:56 发布

阅读量178

点赞数

分类专栏：机器学习

本文链接：https://blog.csdn.net/qq_40415753/article/details/105151849

版权

机器学习专栏收录该内容

14 篇文章 0 订阅

订阅专栏

StratifiedKFold不会出错

import numpy as np
from sklearn import model_selection
from sklearn.model_selection import KFold,StratifiedKFold
x = np.array(list(range(10)))
y = np.array(list([0]*5+[1]*5))

# kfold = KFold(n_splits=5,shuffle=False)
# for train, test in kfold.split(x,y):
#     print("%s %s %s %s" % (x[train], y[train],x[test],y[test]))

kfold = StratifiedKFold(n_splits=5, shuffle=False)
for train, test in kfold.split(x,y):
    print("%s %s %s %s" % (x[train], y[train],x[test],y[test]))

StratifiedKFold会出错

import numpy as np
from sklearn import model_selection
from sklearn.model_selection import KFold,StratifiedKFold
x = np.array(list(range(10)))
y = np.array(list(range(10)))
# y = np.array(list([0]*5+[1]*5))

# kfold = KFold(n_splits=5,shuffle=False)
# for train, test in kfold.split(x,y):
#     print("%s %s %s %s" % (x[train], y[train],x[test],y[test]))

kfold = StratifiedKFold(n_splits=5, shuffle=False)
for train, test in kfold.split(x,y):
    print("%s %s %s %s" % (x[train], y[train],x[test],y[test]))