首页
所有分类
TAGS
文字工具 EMOJI BIBLE
编程 Python

Python实现的朴素贝叶斯算法经典示例【测试可用】

Posted in Python onJune 13, 2018

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：

#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]

# -*- 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

# -*- 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\\3waterPyDemo\\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。

注：本文算法源自《机器学习实战》一书。

希望本文所述对大家Python程序设计有所帮助。

Python实现的朴素贝叶斯算法经典示例【测试可用】

- Author -

旭旭_哥

声明：登载此文出于传递更多信息之目的，并不意味着赞同其观点或证实其描述。

Python 相关文章推荐
Python-基础-入门 简介
Aug 09 Python
Python中用Decorator来简化元编程的教程
Apr 13 Python
详解Python进程间通信之命名管道
Aug 28 Python
Django中的CBV和FBV示例介绍
Feb 25 Python
对pandas中apply函数的用法详解
Apr 10 Python
基于Python开发chrome插件的方法分析
Jul 07 Python
学生信息管理系统python版
Oct 17 Python
pandas 根据列的值选取所有行的示例
Nov 07 Python
Django连接数据库并实现读写分离过程解析
Nov 13 Python
Python 复平面绘图实例
Nov 21 Python
pytorch  网络参数 weight bias 初始化详解
Jun 24 Python
python 如何使用find和find_all爬虫、找文本的实现
Oct 16 Python
Python使用matplotlib和pandas实现的画图操作【经典示例】
Jun 13 #Python
使用python爬虫获取黄金价格的核心代码
Jun 13 #Python
Python实现爬虫从网络上下载文档的实例代码
Jun 13 #Python
Pycharm导入Python包,模块的图文教程
Jun 13 #Python
mac下pycharm设置python版本的图文教程
Jun 13 #Python
使用Python来开发微信功能
Jun 13 #Python
python爬取足球直播吧五大联赛积分榜
Jun 13 #Python
netty(1) 变量提升(1) JOIN(1) 缓冲区(1) NoSQL(1) 弹窗(1) Launch Mode(1) json_extract(1) form表单(1) 类加载器(1)
You might like
如何在PHP中使用Oracle数据库(5)
2006/10/09 PHP
cache_lite试用
2007/02/14 PHP
php从字符串创建函数的方法
2015/03/16 PHP
简单介绍PHP的责任链编程模式
2015/08/11 PHP
Javascript document.referrer判断访客来源网址
2020/05/15 Javascript
jQuery 表格插件整理
2010/04/27 Javascript
扩展easyui.datagrid，添加数据loading遮罩效果代码
2010/11/02 Javascript
使用Firebug对js进行断点调试的图文方法
2011/04/02 Javascript
JavaScript栏目列表隐藏/显示简单实现
2013/04/03 Javascript
jquery用ajax方式从后台获取json数据后如何将内容填充到下拉列表
2015/08/26 Javascript
JS非Alert实现网页右下角“未读信息”效果弹窗
2015/09/26 Javascript
JavaScript encodeURI 和encodeURIComponent
2015/12/04 Javascript
AngularJS入门教程之Scope(作用域)
2016/07/27 Javascript
概述javascript在Google IE中的调试技巧
2016/11/24 Javascript
基于jQuery实现简单人工智能聊天室
2017/02/10 Javascript
canvas实现刮刮卡效果
2017/03/14 Javascript
深入理解vue.js中$watch的oldvalue与newValue
2017/08/07 Javascript
vue权限路由实现的方法示例总结
2018/07/29 Javascript
利用Angular7开发一个Radio组件的全过程
2019/07/11 Javascript
基于ts的动态接口数据配置的详解
2019/12/18 Javascript
JS实现图片切换特效
2019/12/23 Javascript
electron踩坑之dialog中的callback解决
2020/10/06 Javascript
[01:59]游戏“zheng”当时试玩会
2019/08/21 DOTA
Python时间模块datetime、time、calendar的使用方法
2016/01/13 Python
python中的字典使用分享
2016/07/31 Python
在Python中使用AOP实现Redis缓存示例
2017/07/11 Python
python select.select模块通信全过程解析
2017/09/20 Python
GAZMAN官网：澳大利亚领先的男装品牌
2019/12/19 全球购物
升职自荐信
2013/11/28 职场文书
中专毕业生自荐信范文
2013/11/28 职场文书
电子信息工程专业推荐信
2014/02/14 职场文书
岁月神偷观后感
2015/06/11 职场文书
Python基础详解之描述符
2021/04/28 Python
python使用PySimpleGUI设置进度条及控件使用
2021/06/10 Python
vue自定义右键菜单之全局实现
2022/04/09 Vue.js
vue route新窗口跳转页面并且携带与接收参数
2022/04/10 Vue.js