Machine Learning（Dtree）决策树分类学习中遇到的问题

# -*- coding: utf-8 -*-
"""
Created on Thu Feb 28 15:57:14 2019

@author: 
"""

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

#文件读取
alldata=open(r'E:\Python exc\data.csv','rt')
reader = csv.reader(alldata)
headers=next(reader)
print(headers)

featureList=[]
labelList=[]

for row in reader:
    labelList.append(row[len(row)-1])
    rowDict={}
    for i in range(1,len(row)-1):
        rowDict[headers[i]]=row[i]
    featureList.append(rowDict)
    
print(featureList)

#vectorize file
#vec = DictVectorizer
#dummyX = vec.fit_transform(featureList).toarray()

#print("dummyX:"+ str(dummyX))
#print(vec.get_feature_names())

vec = DictVectorizer()
dummyX = vec.fit_transform(featureList).toarray()
print("dummyX : " + str(dummyX))
print ("feature mapping : " + str(vec.get_feature_names()))

print("labelList:"+str(labelList))

#vectorize classlabels
lb = preprocessing.LabelBinarizer()
dummyY = lb.fit_transform(labelList)
print("dummyY :"+ str(dummyY))

#tree分类器，ID3算法
clf = tree.DecisionTreeClassifier(criterion = "entropy")
clf = clf.fit(dummyX,dummyY)
print("clf:" + str(clf))

#创建dot文件并输出树数据
with open('treeclass.dot','w') as f:
     f = tree.export_graphviz(clf, feature_names = vec.get_feature_names() , out_file = f)

#dot -Tpdf E:\Python exc\treeclass.dot -o E:\Python exc\treeclass.pdf

testSet = open(r'E:\Python exc\test.csv','rt')
rea = csv.reader(testSet)

 

testList = []

for row in rea:
    #将每一行的结果放入labelList
    labelList.append(row[len(row)-1])
    #对每一行数据创建一个字典(将每行特征数据转为JSON格式),将headers中的字段与实际值对应 如age:youth
    rowDict = {}
    #i从1开始，取消RID的影响
    for i in range(1,len(row)-1):
        rowDict[headers[i]] = row[i]
    testList.append(rowDict)
# print testList

#把testList向量化
vec = DictVectorizer()
testX = vec.fit_transform(testList).toarray()
print("testX : " + str(testX))
predictSet = clf.predict(testX)
print (predictSet)

注：将Dtree输出为dot文件，并将其转化为PDF格式：安装Graphviz：http：//Graphviz.org/ 并配置环境变量。