个人中心
gpt4 book ai didi

Python实现的朴素贝叶斯算法经典示例【测试可用】

转载 作者：qq735679552 更新时间：2022-09-27 22:32:09 26 4
gpt4 key购买 nike

CFSDN坚持开源创造价值，我们致力于搭建一个资源共享平台，让每一个IT人在这里找到属于你的精彩世界.

这篇CFSDN的博客文章Python实现的朴素贝叶斯算法经典示例【测试可用】由作者收集整理，如果你对这篇文章有兴趣，记得点赞哟.

本文实例讲述了Python实现的朴素贝叶斯算法。分享给大家供大家参考，具体如下:

代码主要参考机器学习实战那本书，发现最近老外的书确实比中国人写的好，由浅入深，代码通俗易懂，不多说上代码:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
#encoding:utf-8
'''''
Created on 2015年9月6日
@author: ZHOUMEIXU204
朴素贝叶斯实现过程
'''
#在该算法中类标签为1和0，如果是多标签稍微改动代码既可
import numpy as np
path = u "D:\\Users\\zhoumeixu204\Desktop\\python语言机器学习\\机器学习实战代码  python\\机器学习实战代码\\machinelearninginaction\\Ch04\\"
def loadDataSet():
   postingList = [[ 'my' , 'dog' , 'has' , 'flea' , 'problems' , 'help' , 'please' ],\
          [ 'maybe' , 'not' , 'take' , 'him' , 'to' , 'dog' , 'park' , 'stupid' ],\
          [ 'my' , 'dalmation' , 'is' , 'so' , 'cute' , 'I' , 'love' , 'him' ],\
          [ 'stop' , 'posting' , 'stupid' , 'worthless' , 'garbage' ],\
          [ 'mr' , 'licks' , 'ate' , 'my' , 'steak' , 'how' , 'to' , 'stop' , 'him' ],\
          [ 'quit' , 'buying' , 'worthless' , 'dog' , 'food' , 'stupid' ]]
   classVec = [ 0 , 1 , 0 , 1 , 0 , 1 ]  #1 is abusive, 0 not
   return postingList,classVec
def createVocabList(dataset):
   vocabSet = set ([])
   for document in dataset:
     vocabSet = vocabSet| set (document)
   return list (vocabSet)
def setOfWordseVec(vocabList,inputSet):
   returnVec = [ 0 ] * len (vocabList)
   for word in inputSet:
     if word in vocabList:
       returnVec[vocabList.index(word)] = 1  #vocabList.index() 函数获取vocabList列表某个元素的位置，这段代码得到一个只包含0和1的列表
     else :
       print ( "the word :%s is not in my Vocabulary!" % word)
   return returnVec
listOPosts,listClasses = loadDataSet()
myVocabList = createVocabList(listOPosts)
print ( len (myVocabList))
print (myVocabList)
print (setOfWordseVec(myVocabList, listOPosts[ 0 ]))
print (setOfWordseVec(myVocabList, listOPosts[ 3 ]))
#上述代码是将文本转化为向量的形式，如果出现则在向量中为1，若不出现 ，则为0
def trainNB0(trainMatrix,trainCategory):  #创建朴素贝叶斯分类器函数
   numTrainDocs = len (trainMatrix)
   numWords = len (trainMatrix[ 0 ])
   pAbusive = sum (trainCategory) / float (numTrainDocs)
   p0Num = np.ones(numWords);p1Num = np.ones(numWords)
   p0Deom = 2.0 ;p1Deom = 2.0
   for i in range (numTrainDocs):
     if trainCategory[i] = = 1 :
       p1Num + = trainMatrix[i]
       p1Deom + = sum (trainMatrix[i])
     else :
       p0Num + = trainMatrix[i]
       p0Deom + = sum (trainMatrix[i])
   p1vect = np.log(p1Num / p1Deom)  #change to log
   p0vect = np.log(p0Num / p0Deom)  #change to log
   return p0vect,p1vect,pAbusive
listOPosts,listClasses = loadDataSet()
myVocabList = createVocabList(listOPosts)
trainMat = []
for postinDoc in listOPosts:
   trainMat.append(setOfWordseVec(myVocabList, postinDoc))
p0V,p1V,pAb = trainNB0(trainMat, listClasses)
if __name__! = '__main__' :
   print ( "p0的概况" )
   print (p0V)
   print ( "p1的概率" )
   print (p1V)
   print ( "pAb的概率" )
   print (pAb)

运行结果:

32 ['him', 'garbage', 'problems', 'take', 'steak', 'quit', 'so', 'is', 'cute', 'posting', 'dog', 'to', 'love', 'licks', 'dalmation', 'flea', 'I', 'please', 'maybe', 'buying', 'my', 'stupid', 'park', 'food', 'stop', 'has', 'ate', 'help', 'how', 'mr', 'worthless', 'not'] [0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0] [0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0] 。

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
# -*- coding:utf-8 -*-
#!python2
#构建样本分类器testEntry=['love','my','dalmation'] testEntry=['stupid','garbage']到底属于哪个类别
import numpy as np
def loadDataSet():
   postingList = [[ 'my' , 'dog' , 'has' , 'flea' , 'problems' , 'help' , 'please' ],\
          [ 'maybe' , 'not' , 'take' , 'him' , 'to' , 'dog' , 'park' , 'stupid' ],\
          [ 'my' , 'dalmation' , 'is' , 'so' , 'cute' , 'I' , 'love' , 'him' ],\
          [ 'stop' , 'posting' , 'stupid' , 'worthless' , 'garbage' ],\
          [ 'mr' , 'licks' , 'ate' , 'my' , 'steak' , 'how' , 'to' , 'stop' , 'him' ],\
          [ 'quit' , 'buying' , 'worthless' , 'dog' , 'food' , 'stupid' ]]
   classVec = [ 0 , 1 , 0 , 1 , 0 , 1 ]  #1 is abusive, 0 not
   return postingList,classVec
def createVocabList(dataset):
   vocabSet = set ([])
   for document in dataset:
     vocabSet = vocabSet| set (document)
   return list (vocabSet)
def setOfWordseVec(vocabList,inputSet):
   returnVec = [ 0 ] * len (vocabList)
   for word in inputSet:
     if word in vocabList:
       returnVec[vocabList.index(word)] = 1  #vocabList.index() 函数获取vocabList列表某个元素的位置，这段代码得到一个只包含0和1的列表
     else :
       print ( "the word :%s is not in my Vocabulary!" % word)
   return returnVec
def trainNB0(trainMatrix,trainCategory):  #创建朴素贝叶斯分类器函数
   numTrainDocs = len (trainMatrix)
   numWords = len (trainMatrix[ 0 ])
   pAbusive = sum (trainCategory) / float (numTrainDocs)
   p0Num = np.ones(numWords);p1Num = np.ones(numWords)
   p0Deom = 2.0 ;p1Deom = 2.0
   for i in range (numTrainDocs):
     if trainCategory[i] = = 1 :
       p1Num + = trainMatrix[i]
       p1Deom + = sum (trainMatrix[i])
     else :
       p0Num + = trainMatrix[i]
       p0Deom + = sum (trainMatrix[i])
   p1vect = np.log(p1Num / p1Deom)  #change to log
   p0vect = np.log(p0Num / p0Deom)  #change to log
   return p0vect,p1vect,pAbusive
def  classifyNB(vec2Classify,p0Vec,p1Vec,pClass1):
   p1 = sum (vec2Classify * p1Vec) + np.log(pClass1)
   p0 = sum (vec2Classify * p0Vec) + np.log( 1.0 - pClass1)
   if p1>p0:
     return 1
   else :
     return 0
def testingNB():
   listOPosts,listClasses = loadDataSet()
   myVocabList = createVocabList(listOPosts)
   trainMat = []
   for postinDoc in listOPosts:
     trainMat.append(setOfWordseVec(myVocabList, postinDoc))
   p0V,p1V,pAb = trainNB0(np.array(trainMat),np.array(listClasses))
   print ( "p0V={0}" . format (p0V))
   print ( "p1V={0}" . format (p1V))
   print ( "pAb={0}" . format (pAb))
   testEntry = [ 'love' , 'my' , 'dalmation' ]
   thisDoc = np.array(setOfWordseVec(myVocabList, testEntry))
   print (thisDoc)
   print ( "vec2Classify*p0Vec={0}" . format (thisDoc * p0V))
   print (testEntry, 'classified as :' ,classifyNB(thisDoc, p0V, p1V, pAb))
   testEntry = [ 'stupid' , 'garbage' ]
   thisDoc = np.array(setOfWordseVec(myVocabList, testEntry))
   print (thisDoc)
   print (testEntry, 'classified as :' ,classifyNB(thisDoc, p0V, p1V, pAb))
if __name__ = = '__main__' :
   testingNB()

运行结果:

p0V=[-3.25809654 -2.56494936 -3.25809654 -3.25809654 -2.56494936 -2.56494936  -3.25809654 -2.56494936 -2.56494936 -3.25809654 -2.56494936 -2.56494936  -2.56494936 -2.56494936 -1.87180218 -2.56494936 -2.56494936 -2.56494936  -2.56494936 -2.56494936 -2.56494936 -3.25809654 -3.25809654 -2.56494936  -2.56494936 -3.25809654 -2.15948425 -2.56494936 -3.25809654 -2.56494936  -3.25809654 -3.25809654] p1V=[-2.35137526 -3.04452244 -1.94591015 -2.35137526 -1.94591015 -3.04452244  -2.35137526 -3.04452244 -3.04452244 -1.65822808 -3.04452244 -3.04452244  -2.35137526 -3.04452244 -3.04452244 -3.04452244 -3.04452244 -3.04452244  -3.04452244 -3.04452244 -3.04452244 -2.35137526 -2.35137526 -3.04452244  -3.04452244 -2.35137526 -2.35137526 -3.04452244 -2.35137526 -2.35137526  -2.35137526 -2.35137526] pAb=0.5 [0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0] vec2Classify*p0Vec=[-0.         -0.         -0.         -0.         -0.         -0.         -0.  -0.         -0.         -0.         -0.         -0.         -0.         -0.  -1.87180218 -0.         -0.         -2.56494936 -0.         -0.         -0.  -0.         -0.         -0.         -0.         -0.         -0.  -2.56494936 -0.         -0.         -0.         -0.        ] ['love', 'my', 'dalmation'] classified as : 0 [0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1] ['stupid', 'garbage'] classified as : 1 。

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
# -*- coding:utf-8 -*-
#! python2
#使用朴素贝叶斯过滤垃圾邮件
# 1.收集数据：提供文本文件
# 2.准备数据：讲文本文件见习成词条向量
# 3.分析数据：检查词条确保解析的正确性
# 4.训练算法：使用我们之前简历的trainNB0()函数
# 5.测试算法：使用classifyNB()，并且对建一个新的测试函数来计算文档集的错误率
# 6.使用算法，构建一个完整的程序对一组文档进行分类，将错分的文档输出到屏幕上
# import re
# mySent='this book is the best book on python or M.L. I hvae ever laid eyes upon.'
# print(mySent.split())
# regEx=re.compile('\\W*')
# print(regEx.split(mySent))
# emailText=open(path+"email\\ham\\6.txt").read()
import numpy as np
path = u "C:\\py\\jb51PyDemo\\src\\Demo\\Ch04\\"
def loadDataSet():
   postingList = [[ 'my' , 'dog' , 'has' , 'flea' , 'problems' , 'help' , 'please' ],\
          [ 'maybe' , 'not' , 'take' , 'him' , 'to' , 'dog' , 'park' , 'stupid' ],\
          [ 'my' , 'dalmation' , 'is' , 'so' , 'cute' , 'I' , 'love' , 'him' ],\
          [ 'stop' , 'posting' , 'stupid' , 'worthless' , 'garbage' ],\
          [ 'mr' , 'licks' , 'ate' , 'my' , 'steak' , 'how' , 'to' , 'stop' , 'him' ],\
          [ 'quit' , 'buying' , 'worthless' , 'dog' , 'food' , 'stupid' ]]
   classVec = [ 0 , 1 , 0 , 1 , 0 , 1 ]  #1 is abusive, 0 not
   return postingList,classVec
def createVocabList(dataset):
   vocabSet = set ([])
   for document in dataset:
     vocabSet = vocabSet| set (document)
   return list (vocabSet)
def setOfWordseVec(vocabList,inputSet):
   returnVec = [ 0 ] * len (vocabList)
   for word in inputSet:
     if word in vocabList:
       returnVec[vocabList.index(word)] = 1  #vocabList.index() 函数获取vocabList列表某个元素的位置，这段代码得到一个只包含0和1的列表
     else :
       print ( "the word :%s is not in my Vocabulary!" % word)
   return returnVec
def trainNB0(trainMatrix,trainCategory):  #创建朴素贝叶斯分类器函数
   numTrainDocs = len (trainMatrix)
   numWords = len (trainMatrix[ 0 ])
   pAbusive = sum (trainCategory) / float (numTrainDocs)
   p0Num = np.ones(numWords);p1Num = np.ones(numWords)
   p0Deom = 2.0 ;p1Deom = 2.0
   for i in range (numTrainDocs):
     if trainCategory[i] = = 1 :
       p1Num + = trainMatrix[i]
       p1Deom + = sum (trainMatrix[i])
     else :
       p0Num + = trainMatrix[i]
       p0Deom + = sum (trainMatrix[i])
   p1vect = np.log(p1Num / p1Deom)  #change to log
   p0vect = np.log(p0Num / p0Deom)  #change to log
   return p0vect,p1vect,pAbusive
def  classifyNB(vec2Classify,p0Vec,p1Vec,pClass1):
   p1 = sum (vec2Classify * p1Vec) + np.log(pClass1)
   p0 = sum (vec2Classify * p0Vec) + np.log( 1.0 - pClass1)
   if p1>p0:
     return 1
   else :
     return 0
def textParse(bigString):
   import re
   listOfTokens = re.split(r '\W*' ,bigString)
   return [tok.lower() for tok in listOfTokens if len (tok)> 2 ]
def spamTest():
   docList = [];classList = [];fullText = []
   for i in range ( 1 , 26 ):
     wordList = textParse( open (path + "email\\spam\\%d.txt" % i).read())
     docList.append(wordList)
     fullText.extend(wordList)
     classList.append( 1 )
     wordList = textParse( open (path + "email\\ham\\%d.txt" % i).read())
     docList.append(wordList)
     fullText.extend(wordList)
     classList.append( 0 )
   vocabList = createVocabList(docList)
   trainingSet = range ( 50 );testSet = []
   for i in range ( 10 ):
     randIndex = int (np.random.uniform( 0 , len (trainingSet)))
     testSet.append(trainingSet[randIndex])
     del (trainingSet[randIndex])
   trainMat = [];trainClasses = []
   for  docIndex in trainingSet:
     trainMat.append(setOfWordseVec(vocabList, docList[docIndex]))
     trainClasses.append(classList[docIndex])
   p0V,p1V,pSpam = trainNB0(np.array(trainMat),np.array(trainClasses))
   errorCount = 0
   for  docIndex in testSet:
     wordVector = setOfWordseVec(vocabList, docList[docIndex])
     if classifyNB(np.array(wordVector), p0V, p1V, pSpam)! = classList[docIndex]:
       errorCount + = 1
   print 'the error rate is :' , float (errorCount) / len (testSet)
if __name__ = = '__main__' :
   spamTest()

运行结果:

the error rate is : 0.0 。

其中，path路径所使用到的Ch04文件点击此处本站下载.

希望本文所述对大家Python程序设计有所帮助.

原文链接：https://blog.csdn.net/luoyexuge/article/details/49104837 。

最后此篇关于Python实现的朴素贝叶斯算法经典示例【测试可用】的文章就讲到这里了,如果你想了解更多关于Python实现的朴素贝叶斯算法经典示例【测试可用】的内容请搜索CFSDN的文章或继续浏览相关文章，希望大家以后支持我的博客！ 。

26 4 0
文章推荐： C中qsort快速排序使用实例
文章推荐： linux使用gcc编译c语言共享库步骤
文章推荐： 插入排序的顺序表实现代码
文章推荐： C语言堆栈入门指南
qq735679552
个人简介

我是一名优秀的程序员,十分优秀！

滴滴打车优惠券免费领取
滴滴打车优惠券
全站热门文章
Copyright 2021 - 2024 cfsdn All Rights Reserved 蜀ICP备2022000587号
广告合作:1813099741@qq.com 6ren.com