机器学习-决策树实战应用

# -*- coding:utf-8 -*-

from sklearn.feature_extraction import DictVectorizer
import csv
from sklearn import preprocessing
from sklearn import tree
from sklearn.externals.six import StringIO

#read the csv file
allElectronicsDate = open(r‘D:\Python\date\play.csv‘,‘rt‘)
reader = csv.reader(allElectronicsDate)
headers = next(reader)

# print(headers)#打印输出第一行标题
#[‘RID‘, ‘age‘, ‘income‘, ‘student‘, ‘credit_rating‘, ‘Class_buys_computer‘]

featureList = [] #用来存储特征值
labelList = [] #用来存储类标签

#获取特征值并打印输出
for row in reader:
    labelList.append(row[len(row) - 1])#每一行最后的值，类标签
    rowDict = {}
    for i in range(1,len(row) - 1):#每一行 遍历除第一列和最后一列的值
        rowDict[headers[i]] = row[i]

    featureList.append(rowDict)

# print(featureList)

#vectorize feature 使用sklearn自带的方法将特征值离散化为数字标记
vec = DictVectorizer()
dumpyX = vec.fit_transform(featureList).toarray()

# print("dunmpyX" + str(dumpyX))
# print("feature_name" + str(vec.get_feature_names()))

# print("labelList" + str(labelList))

#vectorize class labels #数字化类标签
lb = preprocessing.LabelBinarizer()
dummyY = lb.fit_transform(labelList)
# print("dummyY:" + str(dummyY))

#use the decision tree for classification
clf = tree.DecisionTreeClassifier(criterion=‘entropy‘)
clf = clf.fit(dumpyX,dummyY) #构建决策树

#打印构建决策树采用的参数
print("clf:" + str(clf))

#visilize the model
with open(‘allElectronicInformationGainOri.dot‘,‘w‘) as f:
  f=tree.export_graphviz(clf,feature_names=vec.get_feature_names,out_file=f)
#这时就生成了allElectronicInformationGainOri.dot文件

# dot -Tpdf in.dot -o out.pdf dot文件输出为pdf文件

#验证数据，取出一行数据，修改几个属性预测结果
oneRowX = dummyY[0,:]
print("oneRowX:" + str(oneRowX))

newRowX = oneRowX
newRowX[0] = 1
newRowX[2] = 0
print("newRowX:" + str(newRowX))

predictedY = clf.predict(newRowX)
print("predictedY:"+str(predictedY))