def decision_iris():
"""
用决策树对鸢尾花进行分类
:return:
"""
# 1）获取数据集
iris = load_iris()
# 2）划分数据集
x_train, x_test, y_train, y_test = train_test_split(iris.data, iris.target, random_state=22)
# 3）决策树预估器
estimator = DecisionTreeClassifier(criterion="entropy")
estimator.fit(x_train, y_train)
# 4）模型评估
# 方法1：直接比对真实值和预测值
y_predict = estimator.predict(x_test)
print("y_predict:\n", y_predict)
print("直接比对真实值和预测值:\n", y_test == y_predict)
# 方法2：计算准确率
score = estimator.score(x_test, y_test)
print("准确率为：\n", score)
# 可视化决策树
export_graphviz(estimator, out_file="iris_tree.dot", feature_names=iris.feature_names)
return None
