sklearn 三种分类算法评估

产生一组随机的二分类数据，取训练集分别训练NaiveBayes，SVC，随机森林三种算法，对于SVC和随机森林算法，需要额外评估其参数的最佳取值。最后分别使用ACC，f1，rocauc方法评估三种算法。

from sklearn import cross_validation
from sklearn import datasets
from sklearn import naive_bayes
from sklearn import svm
from sklearn import ensemble
from sklearn import metrics
from numpy import argmax

acc_for_NB = []             #使用accuracy评估三个算法
acc_for_SVC = []
acc_for_RFC = []

f1_for_NB = []              # 使用F1-score评估三个算法
f1_for_SVC = []
f1_for_RFC = []

auc_for_NB = []             # 使用AUC ROC评估三个算法
auc_for_SVC = []
auc_for_RFC = []

X, Y = datasets.make_classification(n_samples = 1000, n_features = 10)
kf = cross_validation.KFold(1000, n_folds = 10, shuffle = True)
for train_index, test_index in kf:
    X_train, y_train = X[train_index], Y[train_index]
    X_test, y_test = X[test_index], Y[test_index]
    
    #使用NB算法
    NBclf = naive_bayes.GaussianNB()
    NBclf.fit(X_train, y_train)
    NBpred = NBclf.predict(X_test)
    
    #分别用三种方法评估NB算法结果
    acc_for_NB.append(metrics.accuracy_score(y_test, NBpred))
    f1_for_NB.append(metrics.f1_score(y_test, NBpred))
    auc_for_NB.append(metrics.roc_auc_score(y_test, NBpred))

    #使用SVC算法，先找出最佳的参数C
    nn = len(X_train)
    Cvalues = [1e-02, 1e-01, 1e00, 1e01, 1e02]
    Cscore = []       #记录每个C取值的评估分数
    for C in Cvalues:
        #将X_train分为5个fold
        ikf = cross_validation.KFold(nn, n_folds = 5, shuffle = True)
        innerscore = []
        #对一个C的取值进行评估
        for inner_train_index, inner_test_index in ikf:
            inner_X_train, inner_X_test = X_train[inner_train_index], X_train[inner_test_index]
            inner_y_train, in