sklearn实现多种机器学习中的集成算法。包括Adaboost，随机森林，梯度提升回归等算法

àì夳堔傛蜴生んèń 2022-04-05 14:41 200阅读 0赞

求各位进来的老铁帮个忙。。帮我把最后自己写的那个提升算法完善一下。。测试集该怎么测试准确率？？？ 求大佬补充

from sklearn.datasets import load_iris
    # 用决策树作为基础模型
    from sklearn.tree import DecisionTreeClassifier
    from sklearn.model_selection import train_test_split
    from sklearn.ensemble import GradientBoostingClassifier
    from sklearn.ensemble import AdaBoostClassifier
    from sklearn.ensemble import RandomForestClassifier
    import pandas as pd
    
    # bagging模型
    def RandomForestBagging(X, y):
        '''
        随机森林
        :param X:
        :param y:
        :return:
        '''
        x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.3)
        rd = RandomForestClassifier(
            n_estimators=100,
            criterion="gini",
            max_depth=4,
        )
        rd.fit(x_train, y_train)
        print("随机森林的测试集上的准确率:", rd.score(x_test, y_test))
        print("随机森林的训练集上的准确率:", rd.score(x_train, y_train))
    
    
    # boosting模型汇总
    def GardientBoosting(X, y):
        '''
        梯度提升算法
        :return:
        '''
        x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.3)
        gbrt = GradientBoostingClassifier(max_depth=2, n_estimators=3, learning_rate=1.0)
        gbrt.fit(x_train, y_train)
        print("梯度提升回归树的测试集上准确率:", gbrt.score(x_test, y_test))
        print("梯度提升回归树的训练集上准确率:", gbrt.score(x_train, y_train))
    
    def AdaBoosting(X, y):
        '''
        自适应提升算法
        :param X:
        :param y:
        :return:
        '''
        x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0)
        ada = AdaBoostClassifier(
            base_estimator=DecisionTreeClassifier(max_depth=1),
            n_estimators=100,
            learning_rate=0.5,
            algorithm='SAMME.R',
            random_state=0
        )
        ada.fit(x_train, y_train)
        print("自适应提升算法的测试集上准确率:", ada.score(x_test, y_test))
        print("自适应提升算法的训练集上准确率:", ada.score(x_train, y_train))
    
    
    # stacking模型汇总
    def selfsuanfa(x_train, y_train):
    
        from sklearn.svm import SVC
        from sklearn.naive_bayes import GaussianNB
        from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier
        from sklearn.linear_model import LogisticRegression
        from sklearn.neighbors import KNeighborsClassifier
        from sklearn.neural_network import MLPClassifier
    
        SEED = 1
        nb = GaussianNB()
        svc = SVC(C=100, probability=True)
        knn = KNeighborsClassifier(n_neighbors=3)
        lr = LogisticRegression(C=100, random_state=SEED)
        nn = MLPClassifier((80, 10), early_stopping=False, random_state=SEED)
        gb = GradientBoostingClassifier(n_estimators=100, random_state=SEED)
        rf = RandomForestClassifier(n_estimators=10, max_features=3, random_state=SEED)
    
    
        # 第一层定义好了
        nb.fit(x_train, y_train)
        data1 = nb.predict(x_train).reshape(-1, 1)
        svc.fit(x_train, y_train)
        data2 = svc.predict(x_train).reshape(-1, 1)
        knn.fit(x_train, y_train)
        data3 = knn.predict(x_train).reshape(-1, 1)
        data_level1 = pd.DataFrame(data1, columns=['nb'])
        data_level1['svc'] = data2
        data_level1['knn'] = data3
        data_level1['real'] = y_train
        # print(data_level1)
    
        # 第二层
        X1 = data_level1[['nb', 'svc', 'knn']]
        y1 = data_level1['real']
        x_train1 = X1
        y_train1 = y1
        nn.fit(x_train1, y_train1)
        data11 = nn.predict(x_train1).reshape(-1, 1)
        data_level2 = pd.DataFrame(data11, columns=['nn'])
    
        gb.fit(x_train1, y_train1)
        data22 = gb.predict(x_train1).reshape(-1, 1)
        data_level2['gb'] = data22
    
        lr.fit(x_train1, y_train1)
        data33 = lr.predict(x_train1).reshape(-1, 1)
        data_level2['lr'] = data33
        data_level2['real2'] = y_train1
        # print(data_level2)
    
        # 第三层
        X2 = data_level2[['nn', 'gb', 'lr']]
        y2 = data_level2[['real2']]
        x_train2 = X2
        y_train2 = y2
        rf.fit(x_train2, y_train2)
        print("最强集成算法的测试集上准确率:", rf.score(x_train2, y_train2))
    
    
    if __name__ == '__main__':
        iris = load_iris()
        X = iris.data
        y = iris.target
        # model1(X, y)
        RandomForestBagging(X, y)  # 随机森林算法
        GardientBoosting(X, y)  # 梯度提升算法
        AdaBoosting(X, y)   # 自适应提升算法
    
        x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0)
        # 训练集上的准确率
        selfsuanfa(x_train, y_train)   # 就这个算法，求老铁补充一下