【人工智能】机器学习之用Python使用ID3算法实例及使用sklearn的决策树算法对葡萄酒数据集进行分类

# 导入相关库
from sklearn.model_selection import train_test_split
from sklearn import tree
from sklearn.datasets import load_wine
import pandas as pd
# 导入数据集
wine = load_wine()
X = wine.data
Y = wine.target
features_name = wine.feature_names
print(features_name)
print(pd.concat([pd.DataFrame(X), pd.DataFrame(Y)], axis=1))
# 打印数据
# 划分数据集，数据集划分为测试集占20%；
x_train, x_test, y_train, y_test = train_test_split(
        X, Y, test_size=0.2, random_state=0)

# 采用ID3算法进行计算
clf = tree.DecisionTreeClassifier(criterion="entropy")
# 采用CART算法进行计算
# clf = tree.DecisionTreeClassifier(criterion="gini")
# 获取模型
clf.fit(x_train, y_train)
score = clf.score(x_test, y_test)
y_predict = clf.predict(x_test)
print('准确率为:', score)
# 对测试集的预测类别标签和真实标签进行对比
print(pd.concat([pd.DataFrame(x_test), pd.DataFrame(y_test), pd.DataFrame(y_predict)], axis=1))

if __name__ == '__main__':
    def createDataSet():
        dataSet = [[1, 1, 1, 1, 'No'],
                   [1, 1, 1, 2,  'No'],
                   [2, 1, 1, 1,  'Yes'],
                   [3, 2, 1, 1,  'Yes'],
                   [3, 3, 2, 1,  'Yes'],
                   [3, 3, 2, 2,  'No'],
                   [2, 3, 2, 2,  'Yes'],
                   [1, 2, 1, 1,  'No'],
                   [1, 3, 2, 1,  'Yes'],
                   [3, 2, 2, 1,  'Yes'],
                   [1, 2, 2, 2,  'Yes'],
                   [2, 2, 1, 2,  'Yes'],
                   [2, 1, 2, 1,  'Yes'],
                   [3, 2, 1, 2,  'No'], ]
        features = ['outlook', 'temp', 'humidity', 'windy']

        return dataSet, features


    id3 = ID3Tree()  # 创建一个ID3决策树
    ds, labels = createDataSet()
    id3.getDataSet(ds, labels)
    id3.train()  # 训练ID3决策树
    print(id3.tree)  # 输出ID3决策树
    print(id3.predict(id3.tree, {'outlook': 2, 'temp': 2, 'humidity': 1, 'windy': 1}))
    treePlotter.createPlot(id3.tree)