基于决策树实现葡萄酒分类
将葡萄酒数据集拆分成训练集和测试集，搭建tree_1和tree_2两个决策树模型，tree_1使用信息增益作为特征选择指标，B树使用基尼指数作为特征选择指标，各自对训练集进行训练，然后分别对训练集和测试集进行预测。输出以下结果：
（1）tree_1（信息增益）在训练集上的准确率，在测试集上的准确率。
（2）tree_2（基尼指数）在训练集上的准确率，在测试集上的准确率。
源码
python
from sklearn.datasets import load_wine
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
if __name__ == "__main__":
print("2 基于决策树实现葡萄酒分类")
print("李思强 20201107148")
wine = load_wine()
x_train,x_test,y_train,y_test = train_test_split(wine.data,wine.target)
print("tree_1(信息增益)")
tree_1 = DecisionTreeClassifier(criterion="entropy")
tree_1.fit(x_train,y_train)
train_score = tree_1.score(x_train,y_train)
test_score = tree_1.score(x_test,y_test)
print("训练集")
print("准确率:", train_score)
print("测试集")
print("准确率:", test_score)
print("tree_2(基尼指数)")
tree_2 = DecisionTreeClassifier(criterion="gini")
tree_2.fit(x_train,y_train)
train_score = tree_2.score(x_train,y_train)
test_score = tree_2.score(x_test,y_test)
print("训练集:")
print("准确率:", train_score)
print("测试集")
print("准确率:", test_score)运行结果
