sklearn构建svm分类模型及其模型评价 | 您所在的位置:网站首页 › svm调用 › sklearn构建svm分类模型及其模型评价 |
sklearn中常用的分类算法(模块名--函数名--算法名): (1) linear_model LogisticRegression 逻辑斯蒂回归 (2)svm SVC 支持向量机 (3)neighbors KNeighborsClassifier k近邻分类 (4)naive_bayes GaussianNB 高斯朴素贝叶斯 (5)tree Decision TreeClassifier 分类决策树 (6)ensemble RandomForestClassifier 随机森林分类 (7)ensemble GradientBoostingClassifier梯度提升分类树 使用sklearn估计器构建SVM模型 ##导入各模块和所需函数 import numpy as np from sklearn.datasets import load_breast_cancer from sklearn.svm import SVC from sklearn.model_selection import train_test_split from sklearn.preprocessing import StandardScaler ##cancer数据集特征 cancer = load_breast_cancer() cancer_data = cancer['data'] cancer_target = cancer['target'] cancer_names = cancer['feature_names'] ## 将数据划分为训练集测试集 cancer_data_train,cancer_data_test, cancer_target_train,cancer_target_test = \ train_test_split(cancer_data,cancer_target,test_size = 0.2,random_state = 22) ## 数据标准化 stdScaler = StandardScaler().fit(cancer_data_train) cancer_trainStd = stdScaler.transform(cancer_data_train) cancer_testStd = stdScaler.transform(cancer_data_test) ## 建立SVM模型 svm = SVC().fit(cancer_trainStd,cancer_target_train) print('建立的SVM模型为:\n',svm) ## 预测训练集结果 cancer_target_pred = svm.predict(cancer_testStd) print('预测前20个结果为:\n',cancer_target_pred[:20])将预测结果和真实结果做对比,求出准确率,代码如下: ## 求出预测和真实一样的数目 true = np.sum(cancer_target_pred == cancer_target_test ) print('预测对的结果数目为:', true) print('预测错的的结果数目为:', cancer_target_test.shape[0]-true) print('预测结果准确率为:', true/cancer_target_test.shape[0])单单准确率并不能很好的反映模型的性能,为了有效的判断一个预测模型的效能表现,需要结合真实值计算出精确率,召回率,F1值,Cohen's Kappa系数等指标。详情见下: 方法名称——最佳值——sklearn函数 Precision(精确率) 1.0 metrics.precision_score Recall(召回率) 1.0 metrics.recall_score F1值 1.0 metrics.f1_score Cohen's Kappa系数1.0 metrics.cohen_kappa_score ROC曲线 最靠近y轴 metrics.roc_curve 代码如下: from sklearn.metrics import accuracy_score,precision_score, \ recall_score,f1_score,cohen_kappa_score print('使用SVM预测breast_cancer数据的准确率为:', accuracy_score(cancer_target_test,cancer_target_pred)) print('使用SVM预测breast_cancer数据的精确率为:', precision_score(cancer_target_test,cancer_target_pred)) print('使用SVM预测breast_cancer数据的召回率为:', recall_score(cancer_target_test,cancer_target_pred)) print('使用SVM预测breast_cancer数据的F1值为:', f1_score(cancer_target_test,cancer_target_pred)) print('使用SVM预测breast_cancer数据的Cohen’s Kappa系数为:', cohen_kappa_score(cancer_target_test,cancer_target_pred))另外,sklearn的metrics模块除了提供precision等单一评价指标的函数外,还提供了一个能输出分类模型评价报告的函数classification_report,代码如下: from sklearn.metrics import classification_report print('使用SVM预测iris数据的分类报告为:','\n', classification_report(cancer_target_test, cancer_target_pred))除此之外,还可以用ROC曲线的方式来评价分类模型,代码如下: from sklearn.metrics import roc_curve import matplotlib.pyplot as plt ## 求出ROC曲线的x轴和y轴 fpr, tpr, thresholds = roc_curve(cancer_target_test,cancer_target_pred) plt.figure(figsize=(10,6)) plt.xlim(0,1) ##设定x轴的范围 plt.ylim(0.0,1.1) ## 设定y轴的范围 plt.xlabel('False Postive Rate') plt.ylabel('True Postive Rate') plt.plot(fpr,tpr,linewidth=2, linestyle="-",color='red') plt.show()
|
CopyRight 2018-2019 实验室设备网 版权所有 |