高级编程技术作业_20 sklearn

最新推荐文章于 2018-07-08 22:10:43 发布

akago9

最新推荐文章于 2018-07-08 22:10:43 发布

阅读量198

点赞数

本文链接：https://blog.csdn.net/akago9/article/details/80740714

版权

这篇博客探讨了使用sklearn库进行高级编程的技术，包括特定的代码实现和实验结果。通过实例展示了如何有效地运用sklearn进行机器学习模型训练和评估。

摘要由CSDN通过智能技术生成

题目：

这里写图片描述

代码展示：

from sklearn import datasets, cross_validation, metrics
from sklearn.naive_bayes import GaussianNB
from sklearn.svm import SVC
from sklearn.ensemble import RandomForestClassifier


dataset = datasets.make_classification(n_samples = 1500, n_features = 10, n_informative=2, n_redundant=2, n_classes = 2)
X, y = dataset[0], dataset[1]

kf = cross_validation.KFold(len(dataset[0]), 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]

clf = GaussianNB()
clf.fit(X_train, y_train)
pred = clf.predict(X_test)

print("GaussianNB:")
acc = metrics.accuracy_score(y_test, pred)
print('ACU score:', acc)
f1 = metrics.f1_score(y_test, pred)
print('f1 score:', f1)
auc = metrics.roc_auc_score(y_test, pred)
print('ROC ACU score:', auc)
print('\n')


for C in  [1e-02, 1e-01, 1e00, 1e01, 1e02]:
    clf = SVC(C)
    clf.fit(X_train, y_train)
    pred = clf.predict(X_test)

    print("SVM for C="+str(C))
    acc = metrics.accuracy_score(y_test, pred)
    print('ACU score:', acc)
    f1 = metrics.f1_score(y_test, pred)
    print('f1 score:', f1)
    auc = metrics.roc_auc_score(y_test, pred)
    print('ROC ACU score:', auc)
    print('\n')


for n_estimators in [10, 100, 1000]:
    clf = RandomForestClassifier(n_estimators)
    clf.fit(X_train, y_train)
    pred = clf.predict(X_test)

    print("SVM for n_estimators="+str(n_estimators))
    acc = metrics.accuracy_score(y_test, pred)
    print('ACU score:', acc)
    f1 = metrics.f1_score(y_test, pred)
    print('f1 score:', f1)
    auc = metrics.roc_auc_score(y_test, pred)
    print('ROC ACU score:', auc)
    print('\n')

结果：

GaussianNB:
ACU score: 0.9333333333333333
f1 score: 0.9367088607594937
ROC ACU score: 0.9326923076923076

SVM for C=0.01
ACU score: 0.9333333333333333
f1 score: 0.9390243902439025
ROC ACU score: 0.9310897435897436

SVM for C=0.1
ACU score: 0.9466666666666667
f1 score: 0.9506172839506173
ROC ACU score: 0.9449786324786326

SVM for C=1.0
ACU score: 0.9533333333333334
f1 score: 0.9565217391304347
ROC ACU score: 0.9519230769230769

SVM for C=10.0
ACU score: 0.94
f1 score: 0.9447852760736197
ROC ACU score: 0.9380341880341879

SVM for C=100.0
ACU score: 0.9266666666666666
f1 score: 0.9325153374233129
ROC ACU score: 0.9246794871794872

SVM for n_estimators=10
ACU score: 0.9466666666666667
f1 score: 0.9493670886075949
ROC ACU score: 0.9460470085470086

SVM for n_estimators=100
ACU score: 0.9466666666666667
f1 score: 0.9500000000000001
ROC ACU score: 0.9455128205128205

SVM for n_estimators=1000
ACU score: 0.9466666666666667
f1 score: 0.9500000000000001
ROC ACU score: 0.9455128205128205