1 代碼結構圖

2? 源碼下載地址

下載源碼

3? 代碼實現

from sklearn.feature_extraction import DictVectorizer import csv from sklearn import tree from sklearn import preprocessing from sklearn.externals.six import StringIO import numpy as np# csv文件里面的數據 csvData = open(r'D:\workspace\MachineLearning\07stage\1-fundamental\01\DecisionTree\AllElectronics.csv','r') # 解析csv文件數據 readerData = csv.reader(csvData) # 將csv數據轉為list數據 sourceData = [] for row in readerData:sourceData.append(row) headers = sourceData[0] # 得到表頭數據 labelList = [] # 分類結果 featureList = [] # 特征值數據 coreData = sourceData[1:] for data in coreData:rowDict = {}labelList.append( data[ len(data)-1 ] )for i in range(1,len(data)-1):rowDict[ headers[i] ] = data[i]featureList.append(rowDict) """ 將特征值數據進行向量化 比如： featureList = [{'age': 'youth'},{'age': 'old'},{'age': 'mid'}] 進行向量化后得到結果: [[0. 0. 1.][0. 1. 0.][1. 0. 0.]] 各維度的解釋為: ['age=mid', 'age=old', 'age=youth'] """ vec = DictVectorizer() # 特征值向量化工具類 featureArray = vec.fit_transform(featureList).toarray() #進行向量化 featureNames = vec.get_feature_names();#得到各個維度的含義""" 將[[no],[no],[yes],[yes],[yes] ] 轉為[ [0],[0],[1],[1],[1] ] """ lb = preprocessing.LabelBinarizer() labels = lb.fit_transform(labelList) """ 采用信息熵來做決策樹 """ classifier = tree.DecisionTreeClassifier(criterion='entropy') classifier = classifier.fit(featureArray,labels)#將算好的信息熵保存到文件中 with open('AllElectronicsDot.dot','w') as f:f = tree.export_graphviz(classifier,feature_names=featureNames,out_file=f)""" 使用graphviz繪圖工具將AllElectronicsDot.dot轉為pdf格式 使用命令: dot -Tpdf AllElectronicsDot.dot -o AllElectronicsPdf.pdf """""" 預測數據 [ youth,high,no,fair ] """ newData = featureArray[0,:]; # 拷貝csv文件中第一行數據 newData[0] = 1 newData[1] = 0 newData[2] = 0 predictParam = [] predictParam.append(newData)predictValue = classifier.predict(predictParam) print("predictValue " + str(predictValue))

?

總結

以上是生活随笔為你收集整理的1.2 决策树代码实现的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。