传智播客旗下技术交流社区北京校区

 找回密码
 加入黑马

QQ登录

只需一步,快速开始


K折交叉验证: sklearn.model_selection.KFold(n_splits=3,shuffle=False,random_state=None)
思路:将训练/测试数据划分n_splits个互斥子集,每次用其中一个子集当作验证集,剩下的n_splits-1个作为训练集,进行n_splits次训练和测试,得到n_splits个结果

注意:对于不能均等分的数据集,前n_samples%n_splits子集拥有n_samples//n_splits+1个样本,其余子集只有n_samples//n_splits个样本

参数:

n_splits:表示划分几等份

shuffle:在每次划分时,是否进行洗牌

1)若为False,其效果等同于random_state等于整数,每次划分的结果相同

2)若为True时,每次划分的结果都不一样,表示经过洗牌,随机取样的

random_state:随机种子数

属性:

①get_n_splits(X=None, y=None, groups=None):获取参数n_splits的值

②split(X, y=None, groups=None):将数据集划分成训练集和测试集,返回索引生成器



A:设置shuffle=False,运行两次,发现两次结果相同

from sklearn.model_selection import KFold
import numpy as np

#设置shuffle=False,运行两次,发现两次结果相同
X=np.arange(24).reshape(12,2)
y=np.random.choice([1,2],12,p=[0.4,0.6])#1,2 总共出现12次,其中1出现的概率为0.4,2出现的概率为0.6
kf=KFold(n_splits=5,shuffle=False)
for train_index,test_index in kf.split(X):
    print('train_index %s, test_index %s'%(train_index,test_index))

X=np.arange(24).reshape(12,2)
y=np.random.choice([1,2],12,p=[0.4,0.6])#1,2 总共出现12次,其中1出现的概率为0.4,2出现的概率为0.6
kf=KFold(n_splits=5,shuffle=False)
for train_index,test_index in kf.split(X):
    print('train_index %s, test_index %s'%(train_index,test_index))

'''
train_index [ 3  4  5  6  7  8  9 10 11], test_index [0 1 2]
train_index [ 0  1  2  6  7  8  9 10 11], test_index [3 4 5]
train_index [ 0  1  2  3  4  5  8  9 10 11], test_index [6 7]
train_index [ 0  1  2  3  4  5  6  7 10 11], test_index [8 9]
train_index [0 1 2 3 4 5 6 7 8 9], test_index [10 11]
train_index [ 3  4  5  6  7  8  9 10 11], test_index [0 1 2]
train_index [ 0  1  2  6  7  8  9 10 11], test_index [3 4 5]
train_index [ 0  1  2  3  4  5  8  9 10 11], test_index [6 7]
train_index [ 0  1  2  3  4  5  6  7 10 11], test_index [8 9]
train_index [0 1 2 3 4 5 6 7 8 9], test_index [10 11]
'''
B. 设置shuffle=True时,运行两次,发现两次运行的结果不同

from sklearn.model_selection import KFold
import numpy as np

#设置shuffle=False,运行两次,发现两次结果相同
X=np.arange(24).reshape(12,2)
y=np.random.choice([1,2],12,p=[0.4,0.6])#1,2 总共出现12次,其中1出现的概率为0.4,2出现的概率为0.6
kf=KFold(n_splits=5,shuffle=True)
for train_index,test_index in kf.split(X):
    print('train_index %s, test_index %s'%(train_index,test_index))
print('-----------------------------------------------------------')
X=np.arange(24).reshape(12,2)
y=np.random.choice([1,2],12,p=[0.4,0.6])#1,2 总共出现12次,其中1出现的概率为0.4,2出现的概率为0.6
kf=KFold(n_splits=5,shuffle=True)
for train_index,test_index in kf.split(X):
    print('train_index %s, test_index %s'%(train_index,test_index))
'''
train_index [ 0  2  3  4  5  6  8 10 11], test_index [1 7 9]
train_index [ 1  2  3  4  5  7  9 10 11], test_index [0 6 8]
train_index [ 0  1  2  3  5  6  7  8  9 10], test_index [ 4 11]
train_index [ 0  1  2  3  4  6  7  8  9 11], test_index [ 5 10]
train_index [ 0  1  4  5  6  7  8  9 10 11], test_index [2 3]
-----------------------------------------------------------
train_index [ 0  1  2  3  4  6  7  8 11], test_index [ 5  9 10]
train_index [ 0  3  4  5  6  8  9 10 11], test_index [1 2 7]
train_index [ 0  1  2  4  5  6  7  8  9 10], test_index [ 3 11]
train_index [ 0  1  2  3  5  7  8  9 10 11], test_index [4 6]
train_index [ 1  2  3  4  5  6  7  9 10 11], test_index [0 8]
'''
C: 设置shuffle=True和random_state=整数,发现每次运行的结果都相同

from sklearn.model_selection import KFold
import numpy as np

#设置shuffle=False,运行两次,发现两次结果相同
X=np.arange(24).reshape(12,2)
y=np.random.choice([1,2],12,p=[0.4,0.6])#1,2 总共出现12次,其中1出现的概率为0.4,2出现的概率为0.6
kf=KFold(n_splits=5,shuffle=True,random_state=10)
for train_index,test_index in kf.split(X):
    print('train_index %s, test_index %s'%(train_index,test_index))
print('-----------------------------------------------------------')
X=np.arange(24).reshape(12,2)
y=np.random.choice([1,2],12,p=[0.4,0.6])#1,2 总共出现12次,其中1出现的概率为0.4,2出现的概率为0.6
kf=KFold(n_splits=5,shuffle=True,random_state=10)
for train_index,test_index in kf.split(X):
    print('train_index %s, test_index %s'%(train_index,test_index))
'''
0  1  3  4  5  8  9 10 11], test_index [2 6 7]
train_index [ 0  1  2  3  4  6  7  9 10], test_index [ 5  8 11]
train_index [ 0  1  2  4  5  6  7  8  9 11], test_index [ 3 10]
train_index [ 2  3  4  5  6  7  8  9 10 11], test_index [0 1]
train_index [ 0  1  2  3  5  6  7  8 10 11], test_index [4 9]
-----------------------------------------------------------
train_index [ 0  1  3  4  5  8  9 10 11], test_index [2 6 7]
train_index [ 0  1  2  3  4  6  7  9 10], test_index [ 5  8 11]
train_index [ 0  1  2  4  5  6  7  8  9 11], test_index [ 3 10]
train_index [ 2  3  4  5  6  7  8  9 10 11], test_index [0 1]
train_index [ 0  1  2  3  5  6  7  8 10 11], test_index [4 9]
'''
分享至 : QQ空间
收藏

1 个回复

倒序浏览
奈斯,加油
回复 使用道具 举报
您需要登录后才可以回帖 登录 | 加入黑马