path ='iris.data'  # 數(shù)據(jù)文件路徑
    data = pd.read_csv(path, header=None)
    x = data[[0, 1]]
    # y = pd.Categorical(data[4]).codes
    y = LabelEncoder().fit_transform(data[4])
    # 為了可視化，僅使用前兩列特征
    x = x.iloc[:, :2]
    # x = x[[0,1]]
    x_train, x_test, y_train, y_test = train_test_split(x, y, train_size=0.7, random_state=1)

    # 決策樹參數(shù)估計(jì)
    # min_samples_split = 10：如果該結(jié)點(diǎn)包含的樣本數(shù)目大于10，則(有可能)對(duì)其分支
    # min_samples_leaf = 10：若將某結(jié)點(diǎn)分支后，得到的每個(gè)子結(jié)點(diǎn)樣本數(shù)目都大于10，則完成分支；否則，不進(jìn)行分支
    model = DecisionTreeClassifier(criterion='entropy', min_samples_leaf=3)
    model.fit(x_train, y_train)
    y_test_hat = model.predict(x_test)      # 測(cè)試數(shù)據(jù)
    print('accuracy_score:', accuracy_score(y_test, y_test_hat))