任务5-CSDN博客

本文链接：https://blog.csdn.net/qq_43585786/article/details/86244213

首先调用所需要的库

    from sklearn import metrics
    from sklearn.metrics import confusion_matrix
    from sklearn.metrics import precision_score        
    from sklearn.metrics import recall_score           
    from sklearn.metrics import f1_score                
    from sklearn.metrics import accuracy_score

1、逻辑回归

#训练集
Y_pred_lr = logreg.predict(x_train)
precision=metrics.accuracy_score(y_train,Y_pred_lr)
cnf_matrix=metrics.confusion_matrix(y_train,Y_pred_lr)
print('精准率：','%.3f' %metrics.precision_score(y_train,Y_pred_lr) ,
      '召回率：','%.3f' %metrics.recall_score(y_train,Y_pred_lr) ,
      'f1_score：','%.3f' %metrics.f1_score(y_train,Y_pred_lr),
      '准确率：','%.3f' %metrics.accuracy_score(y_train,Y_pred_lr))
      y_train_prob=logreg.predict_proba(x_train)
y_pred_prob=y_train_prob[:,1]                             #正例率 
fpr,tpr,thresholds=metrics.roc_curve(y_train,y_pred_prob) #计算ROC曲线
auc=metrics.auc(fpr,tpr)
auc   
  #测试集
  Y_pred_lr = logreg.predict(x_test)
  precision=metrics.accuracy_score(y_test,Y_pred_lr)
  cnf_matrix=metrics.confusion_matrix(y_test,Y_pred_lr)
  print('精准率：','%.3f' %metrics.precision_score(y_test,Y_pred_lr) ,
         '召回率：','%.3f' %metrics.recall_score(y_test,Y_pred_lr) ,
         'f1_score：','%.3f' %metrics.f1_score(y_test,Y_pred_lr),
         '准确率：','%.3f' %metrics.accuracy_score(y_test,Y_pred_lr))

  
#绘制ROC曲线
plt.plot(fpr1,tpr1,lw=2,label='train ROC curve (area={:.2f})'.format(auc1))
plt.plot(fpr2,tpr2,lw=2,label='test ROC curve (area={:.2f})'.format(auc2))
plt.plot([0,1],[0,1],'r--')
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('train ROC')
plt.legend(loc='lower right')
plt.show()

ROC
在这里插入图片描述

2.svm

#训练集
from sklearn.svm import SVC
svc = SVC(probability=True)
svc.fit(x_train, y_train)
Y_pred_svc = svc.predict(x_train)
precision=metrics.accuracy_score(y_train,Y_pred_svc)
print('精准率：','%.3f' %metrics.precision_score(y_train,Y_pred_svc) ,
          '召回率：','%.3f' %metrics.recall_score(y_train,Y_pred_svc) ,
          'f1_score：','%.3f' %metrics.f1_score(y_train,Y_pred_svc),
          '准确率：','%.3f' %metrics.accuracy_score(y_train,Y_pred_svc))
#测试集
Y_pred_svc = svc.predict(x_test)
precision=metrics.accuracy_score(y_test,Y_pred_svc)
print('精准率：','%.3f' %metrics.precision_score(y_test,Y_pred_svc) ,
          '召回率：','%.3f' %metrics.recall_score(y_test,Y_pred_svc) ,
          'f1_score：','%.3f' %metrics.f1_score(y_test,Y_pred_svc),
          '准确率：','%.3f' %metrics.accuracy_score(y_test,Y_pred_svc))
y_train_prob=svc.predict_proba(x_train)
y_test_prob=svc.predict_proba(x_test)
y_pred_prob1=y_train_prob[:,1]#正例率 
y_pred_prob2=y_test_prob[:,1]#正例率 
fpr1,tpr1,thresholds1=metrics.roc_curve(y_train,y_pred_prob1) #计算ROC曲线
fpr2,tpr2,thresholds2=metrics.roc_curve(y_test,y_pred_prob2)
auc1=metrics.auc(fpr1,tpr1)
auc2=metrics.auc(fpr2,tpr2)

#绘制ROC曲线
plt.plot(fpr1,tpr1,lw=2,label='train ROC curve (area={:.2f})'.format(auc1))
plt.plot(fpr2,tpr2,lw=2,label='test ROC curve (area={:.2f})'.format(auc2))
plt.plot([0,1],[0,1],'r--')
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('ROC curve')
plt.legend(loc='lower right')
plt.show()

在这里插入图片描述
3.决策树

    #训练集
    Y_pred_decision_tree = decision_tree.predict(x_train)
    precision=metrics.accuracy_score(y_train,Y_pred_decision_tree)
    print('精准率：','%.3f' %metrics.precision_score(y_train,Y_pred_decision_tree) ,
              '召回率：','%.3f' %metrics.recall_score(y_train,Y_pred_decision_tree) ,
              'f1_score：','%.3f' %metrics.f1_score(y_train,Y_pred_decision_tree),
              '准确率：','%.3f' %metrics.accuracy_score(y_train,Y_pred_decision_tree))
    #测试集
    Y_pred_decision_tree = decision_tree.predict(x_test)
    precision=metrics.accuracy_score(y_test,Y_pred_decision_tree)
    print('精准率：','%.3f' %metrics.precision_score(y_test,Y_pred_decision_tree) ,
              '召回率：','%.3f' %metrics.recall_score(y_test,Y_pred_decision_tree) ,
              'f1_score：','%.3f' %metrics.f1_score(y_test,Y_pred_decision_tree),
              '准确率：','%.3f' %metrics.accuracy_score(y_test,Y_pred_decision_tree))
y_train_prob=decision_tree.predict_proba(x_train)
y_test_prob=decision_tree.predict_proba(x_test)
y_pred_prob1=y_train_prob[:,1]#正例率 
y_pred_prob2=y_test_prob[:,1]#正例率 
fpr1,tpr1,thresholds1=metrics.roc_curve(y_train,y_pred_prob1) #计算ROC曲线
fpr2,tpr2,thresholds2=metrics.roc_curve(y_test,y_pred_prob2)
auc1=metrics.auc(fpr1,tpr1)
auc2=metrics.auc(fpr2,tpr2)

#绘制ROC曲线
plt.plot(fpr1,tpr1,lw=2,label='train ROC curve (area={:.2f})'.format(auc1))
plt.plot(fpr2,tpr2,lw=2,label='test ROC curve (area={:.2f})'.format(auc2))
plt.plot([0,1],[0,1],'r--')
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('ROC curve')
plt.legend(loc='lower right')
plt.show()

在这里插入图片描述
4随机森林

#训练集
Y_pred_random_forest = random_forest.predict(x_train)
precision=metrics.accuracy_score(y_train,Y_pred_random_forest)
print('精准率：','%.3f' %metrics.precision_score(y_train,Y_pred_random_forest) ,
          '召回率：','%.3f' %metrics.recall_score(y_train,Y_pred_random_forest) ,
          'f1_score：','%.3f' %metrics.f1_score(y_train,Y_pred_random_forest),
          '准确率：','%.3f' %metrics.accuracy_score(y_train,Y_pred_random_forest))
#测试集
Y_pred_random_forest = random_forest.predict(x_test)
precision=metrics.accuracy_score(y_test,Y_pred_random_forest)
print('精准率：','%.3f' %metrics.precision_score(y_test,Y_pred_random_forest) ,
          '召回率：','%.3f' %metrics.recall_score(y_test,Y_pred_random_forest) ,
          'f1_score：','%.3f' %metrics.f1_score(y_test,Y_pred_random_forest),
          '准确率：','%.3f' %metrics.accuracy_score(y_test,Y_pred_random_forest))
y_train_prob=random_forest.predict_proba(x_train)
y_test_prob=random_forest.predict_proba(x_test)
y_pred_prob1=y_train_prob[:,1]#正例率 
y_pred_prob2=y_test_prob[:,1]#正例率 
fpr1,tpr1,thresholds1=metrics.roc_curve(y_train,y_pred_prob1) #计算ROC曲线
fpr2,tpr2,thresholds2=metrics.roc_curve(y_test,y_pred_prob2)
auc1=metrics.auc(fpr1,tpr1)
auc2=metrics.auc(fpr2,tpr2)

#绘制ROC曲线
plt.plot(fpr1,tpr1,lw=2,label='train ROC curve (area={:.2f})'.format(auc1))
plt.plot(fpr2,tpr2,lw=2,label='test ROC curve (area={:.2f})'.format(auc2))
plt.plot([0,1],[0,1],'r--')
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('ROC curve')
plt.legend(loc='lower right')
plt.show()

在这里插入图片描述