请教一下交叉验证相关问题

Question

请问一下10×10交叉验证是重复分层交叉验证吗，10×10交叉验证的python代码怎么实现啊

Answer 1

10×10交叉验证是将数据集分成十份，轮流将其中9份做训练1份做验证，10次的结果的均值作为对算法精度的估计，以此进行10次10折交叉验证求均值
这是一个例子

# -*- coding: utf-8 -*-
import numpy as np
from sklearn import cross_validation
from sklearn import datasets
from sklearn import svm
iris = datasets.load_iris()
X=iris.data
Y=iris.target

def tenFolds(X,Y):
    from sklearn.model_selection import StratifiedKFold
    skf= StratifiedKFold(n_splits=10)
    from sklearn.cross_validation import cross_val_score
    clf = svm.SVC(kernel='linear', C=5)
    zhou=[]
    #shoushou=list(skf.split(X,Y)) #发现每一个都是tuple类型的数据
    for train_index,test_index in skf.split(X,Y):
        print('Train: ',train_index,'Test: ',test_index)
        X_train,X_test=X[train_index],X[test_index]
        Y_train,Y_test=Y[train_index],Y[test_index]
        v10=cross_val_score(clf,X_train,Y_train,cv=5).mean() 
        zhou.append(v10)
    print('均值:',np.mean(zhou)) 
    print('方差:',np.std(zhou)) 
    return zhou,np.mean(zhou)+np.std(zhou)

# X:features  Y:targets  cv:k
import pandas as pd
all_valid=[]
mean_var=[]
#做10次10折交叉验证结果分析
for i in range(1,11):
    df1=pd.DataFrame(X)
    df2=pd.DataFrame(Y)
    df3=pd.concat([df1,df2],axis=1)
    df3.columns=['f1','f2','f3','f4','label']
    df3=df3.sample(frac = 1) #随机打乱样本
    df3.index=range(150)
    X1=np.array(df3[['f1','f2','f3','f4']])
    Y1=np.array(df3['label'])
    zhou,vsd= tenFolds(X1,Y1)
    all_valid.append(zhou)
    mean_var.append(vsd)
np.mean(mean_var)