决策树

基础内容可以直接看这篇博客

下面的demo是使用决策树算法的一个例子，使用的数据如下

from sklearn.feature_extraction import DictVectorizer
from sklearn import tree
from sklearn import preprocessing
import csv
# 读入数据
Dtree = open(r'AllElectronics.csv', 'r')
reader = csv.reader(Dtree)

# 获取第一行数据
headers = reader.__next__()
print(headers)

# 定义两个列表
featureList = []
labelList = []

#
for row in reader:
    # 把label存入list
    labelList.append(row[-1])#类别标签
    rowDict = {}
    for i in range(1, len(row)-1):
        #建立一个数据字典
        rowDict[headers[i]] = row[i]
    # 把数据字典存入list
    featureList.append(rowDict)

print(featureList)
# 把数据转换成01表示
vec = DictVectorizer()
x_data = vec.fit_transform(featureList).toarray()
print("x_data: " + str(x_data))

# 打印属性名称
print(vec.get_feature_names())

# 打印标签
print("labelList: " + str(labelList))

# 把标签转换成01表示
lb = preprocessing.LabelBinarizer()
y_data = lb.fit_transform(labelList)
print("y_data: " + str(y_data))
# 创建决策树模型
model = tree.DecisionTreeClassifier(criterion='entropy')##建立决策树模型,基于信息熵
# 输入数据建立模型
model.fit(x_data, y_data)
# 测试
x_test = x_data[0]
print("x_test: " + str(x_test))

predict = model.predict(x_test.reshape(1,-1))
print("predict: " + str(predict))
# 导出决策树
# pip install graphviz
# http://www.graphviz.org/
import graphviz

dot_data = tree.export_graphviz(model,
                                out_file = None,
                                feature_names = vec.get_feature_names(),
                                class_names = lb.classes_,
                                filled = True,
                                rounded = True,
                                special_characters = True)
graph = graphviz.Source(dot_data)
graph.render('computer')#生成一个决策树图表文件

下面的demo是用决策树进行线性二分类

 1 import matplotlib.pyplot as plt
 2 import numpy as np
 3 from sklearn.metrics import classification_report
 4 from sklearn import tree
 5 # 载入数据
 6 data = np.genfromtxt("LR-testSet.csv", delimiter=",")
 7 x_data = data[:,:-1]
 8 y_data = data[:,-1]
 9
10 plt.scatter(x_data[:,0],x_data[:,1],c=y_data)
11 plt.show()
12 # 创建决策树模型
13 model = tree.DecisionTreeClassifier()
14 # 输入数据建立模型
15 model.fit(x_data, y_data)
16 # 导出决策树
17 import graphviz # http://www.graphviz.org/
18
19 dot_data = tree.export_graphviz(model,
20                                 out_file = None,
21                                 feature_names = ['x','y'],
22                                 class_names = ['label0','label1'],
23                                 filled = True,
24                                 rounded = True,
25                                 special_characters = True)
26 graph = graphviz.Source(dot_data)
27 # 获取数据值所在的范围
28 x_min, x_max = x_data[:, 0].min() - 1, x_data[:, 0].max() + 1
29 y_min, y_max = x_data[:, 1].min() - 1, x_data[:, 1].max() + 1
30
31 # 生成网格矩阵
32 xx, yy = np.meshgrid(np.arange(x_min, x_max, 0.02),
33                      np.arange(y_min, y_max, 0.02))
34
35 z = model.predict(np.c_[xx.ravel(), yy.ravel()])# ravel与flatten类似，多维数据转一维。flatten不会改变原始数据，ravel会改变原始数据
36 z = z.reshape(xx.shape)
37 # 等高线图
38 cs = plt.contourf(xx, yy, z)
39 # 样本散点图
40 plt.scatter(x_data[:, 0], x_data[:, 1], c=y_data)
41 plt.show()
42 predictions = model.predict(x_data)
43 print(classification_report(predictions,y_data))

下面这个demo是用决策树进行非线性二分类

 1 import matplotlib.pyplot as plt
 2 import numpy as np
 3 from sklearn.metrics import classification_report
 4 from sklearn import tree
 5 from sklearn.model_selection import train_test_split
 6
 7 # 载入数据
 8 data = np.genfromtxt("LR-testSet2.txt", delimiter=",")
 9 x_data = data[:, :-1]
10 y_data = data[:, -1]
11
12 plt.scatter(x_data[:, 0], x_data[:, 1], c=y_data)
13 plt.show()
14 #分割数据
15 x_train,x_test,y_train,y_test = train_test_split(x_data, y_data)
16
17 # 创建决策树模型
18 # max_depth，树的深度
19 # min_samples_split 内部节点再划分所需最小样本数
20 model = tree.DecisionTreeClassifier(max_depth=7,min_samples_split=4)
21 # 输入数据建立模型
22 model.fit(x_train, y_train)
23 # 导出决策树
24 import graphviz # http://www.graphviz.org/
25
26 dot_data = tree.export_graphviz(model,
27                                 out_file = None,
28                                 feature_names = ['x','y'],
29                                 class_names = ['label0','label1'],
30                                 filled = True,
31                                 rounded = True,
32                                 special_characters = True)
33 graph = graphviz.Source(dot_data)
34 # 获取数据值所在的范围
35 x_min, x_max = x_data[:, 0].min() - 1, x_data[:, 0].max() + 1
36 y_min, y_max = x_data[:, 1].min() - 1, x_data[:, 1].max() + 1
37
38 # 生成网格矩阵
39 xx, yy = np.meshgrid(np.arange(x_min, x_max, 0.02),
40                      np.arange(y_min, y_max, 0.02))
41
42 z = model.predict(np.c_[xx.ravel(), yy.ravel()])# ravel与flatten类似，多维数据转一维。flatten不会改变原始数据，ravel会改变原始数据
43 z = z.reshape(xx.shape)
44 # 等高线图
45 cs = plt.contourf(xx, yy, z)
46 # 样本散点图
47 plt.scatter(x_data[:, 0], x_data[:, 1], c=y_data)
48 plt.show()
49 predictions = model.predict(x_train)
50 print(classification_report(predictions,y_train))
51 predictions = model.predict(x_test)
52 print(classification_report(predictions,y_test))