0925Stacking集成学习+StratifiedKFold

短命女 2023-10-09 12:33 20阅读 0赞

#导入包
    import pandas as pd
    import numpy as np
    import math
    
    from sklearn.linear_model import LogisticRegressionCV
    from sklearn.ensemble import RandomForestClassifier
    from xgboost.sklearn import XGBClassifier
    from sklearn.linear_model import LogisticRegression
    from sklearn.model_selection import train_test_split
    
    from sklearn import metrics
    from sklearn.metrics import confusion_matrix
    from sklearn.metrics import accuracy_score
    from sklearn.metrics import precision_score
    from sklearn.metrics import recall_score
    from sklearn.metrics import roc_auc_score
    from sklearn.metrics import roc_curve
    from sklearn.model_selection import GridSearchCV
    from sklearn.model_selection import StratifiedKFold
    
    xtrain0,xtest0,ytrain0,ytest0 = train_test_split(X,y,test_size=0.3,random_state=42)
    
    ytrain0 = np.array(ytrain0)
    ytrain0 = np.reshape(ytrain0,[36803,])
    
    skf = StratifiedKFold(n_splits=3,random_state=2019,shuffle=True)
    clfs = [
        LogisticRegressionCV()
        ,RandomForestClassifier(criterion='gini',max_depth=10,max_features='sqrt',min_samples_leaf=3,n_estimators=50)
        ,XGBClassifier(n_estimators=110,max_depth=6,learning_rate=0.5,gamma=0.05)
    ]
    ############################################################
    ############################################################
    #stacking第一层模型训练
    blend_train = np.zeros((xtrain0.shape[0],3)) #训练集行数，3列（3个基分类器）
    blend_test = np.zeros((xtest0.shape[0],3))#测试集行数，3列（3个基分类器）
    
    for j,clf in enumerate(clfs):
        print('Training classifier[%s]'%(j))
        blend_train_j = np.zeros((xtrain0.shape[0],3)) #每个基分类器对应的训练集
        blend_test_j = np.zeros((xtest0.shape[0],3)) #每个基分类器对应的测试集
        for i,(train_index,cv_index) in enumerate(skf.split(xtrain0,ytrain0)):
            print('Fold[%s]'%(i))
            xtrain00 = xtrain0.iloc[train_index,:] #新的训练集X
            ytrain00 = ytrain0[train_index] #新的训练集Y
            xtrain0_cv = xtrain0.iloc[cv_index,:]  #验证集X
            ytrain0_cv = ytrain0[cv_index]  #验证集Y
            #训练模型
            clf.fit(xtrain00,ytrain00)
            #用训练的模型进行预测
            blend_test_j[:,i] = clf.predict(xtest0)  #每个基分类器对测试集进行预测
        #对每个分类器下测试数据中三个预测值求平均作为该分类器的预测值
        blend_test[:,j] = blend_test_j.mean(1)
    #stacking第二层模型训练
    clf = LogisticRegression()
    result = clf.fit(blend_train,ytrain0) #训练第二层学习器
    y_test_predict = clf.predict(blend_test)  #预测
    #输出模型效果
    print('AUC:%.4f'%metrics.roc_auc_score(ytest0,y_test_predict))
    print('ACC:%.4f'%metrics.accuracy_score(ytest0,y_test_predict))
    print('Recall:%.4f'%metrics.recall_score(ytest0,y_test_predict))
    print('F1-score:%.4f'%metrics.f1_score(ytest0,y_test_predict))
    print('Precision:%.4f'%metrics.precision_score(ytest0,y_test_predict))

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L0d1aWFiYmV5_size_16_color_FFFFFF_t_70]  
参考资料：https://www.jianshu.com/p/5905f19c4df6

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L0d1aWFiYmV5_size_16_color_FFFFFF_t_70]: https://img-blog.csdnimg.cn/20190925200933898.jpg?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L0d1aWFiYmV5,size_16,color_FFFFFF,t_70