XGBoost与决策树对Iris数据集的预测

傷城~ 2022-11-22 00:26 164阅读 0赞

XGBoost模型预测

from xgboost import XGBClassifier
    import pandas as pd
    import numpy as np
    
    def load_data():
        data = pd.read_csv('D:/Studyfile/machine/ensemble/Iris.csv')#设置数据路径
        #前4/5作为训练集，后1/5作为测试集
        data_training = data[0:int(len(data)*4/5)]
        data_test = data[int(len(data)*4/5):len(data)]
        #分割
        train_x = np.array(data_training.iloc[:, [i for i in range(data_training.shape[1]-1)]])
        train_y = np.array(data_training['Species'])
        test_x = np.array(data_test.iloc[:, [i for i in range(data_test.shape[1]-1)]])
        test_y = np.array(data_test['Species'])
    
        return train_x, train_y, test_x, test_y
    
    
    def XGBoost():
        train_x, train_y, test_x, test_y = load_data()
        #训练
        clf=XGBClassifier(base_score=0.5, booster='gbtree', learning_rate=0.05, max_depth=8, n_estimators=50)
        clf.fit(train_x, train_y)
        #测试
        print(clf.score(test_x, test_y))
    
    
    if __name__ == '__main__':
        XGBoost()

决策树模型预测：

import pydot
    from sklearn.tree import export_graphviz
    from sklearn.tree import DecisionTreeClassifier
    #3：1拆分数据集
    from sklearn.model_selection import train_test_split
    from sklearn import datasets
     
    if __name__ == '__main__':
        cancer = datasets.load_iris()
        print('数据集规格大小',cancer.data.shape)
        print('标签规格大小',cancer.target.shape)
        print('标签有',cancer.target_names)
        print('数据属性有',cancer.feature_names)
        # 参数random_state是指随机生成器，0表示函数输出是固定不变的
        X_train, X_test, y_train, y_test = train_test_split(cancer['data'], cancer['target'], random_state=42)
        tree = DecisionTreeClassifier(random_state=0)
        tree.fit(X_train, y_train)
        print('Train score:{:.3f}'.format(tree.score(X_train, y_train)))
        print('Test score:{:.3f}'.format(tree.score(X_test, y_test)))
        # 生成可视化图
        export_graphviz(tree, out_file="tree.dot", class_names=cancer.target_names, feature_names=cancer.feature_names,
                        impurity=False, filled=True)
        # 展示可视化图
        (graph,) = pydot.graph_from_dot_file('tree.dot')
        graph.write_png('tree.png')

树可视化：

import pandas as pd
    from sklearn import preprocessing
    from sklearn import tree
    from sklearn.datasets import load_iris
    #加载数据集
    iris=load_iris()
    #dir(iris)  #查看数据集的列标签
    iris_feature_name=iris.feature_names
    iris_feature=iris.data
    iris_target_name=iris.target_names
    iris_target=iris.target
    #构建模型
    clf=tree.DecisionTreeClassifier(max_depth=4)
    clf=clf.fit(iris_feature,iris_target)
    print(clf)
    #可视化
    import pydotplus
    from IPython.display import Image,display
    dot_data=tree.export_graphviz(clf,
                   out_file=None,
                   feature_names=iris_feature_name,
                   class_names=iris_target_name,
                   filled=True,
                   rounded=True
                  )
    graph=pydotplus.graph_from_dot_data(dot_data)
    display(Image(graph.create_png()))