使用Python进行文本分类(一)
准备数据:从文本中构建词向量
def loadDataSet():#创建实验样本
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]
return postingList,classVec
def createVocabList(dataSet):#将输入文档的不重复词创建为一个列表
vocabSet = set ([])
for document in dataSet:
vocabSet = vocabSet | set(document)
return list(vocabSet)
def setOfWords2Vec(vocabList, inputSet):#输入为词汇表及文档,判断词汇表中每个词是否在文档中出现过
returnVec = [0]*len(vocabList)
for word in inputSet:
if word in vocabList:
returnVec[vocabList.index(word)] = 1
else: print("the word: %s is not in my Vocabulary!" % word)
return returnVec
>>> listOPosts,listClasses = bayes.loadDataSet()
>>> myVocabList = bayes.createVocabList(listOPosts)
>>> myVocabList
['is', 'flea', 'help', 'cute', 'food', 'stop', 'ate', 'stupid', 'how', 'not', 'maybe', 'take', 'steak', 'has', 'dalmation', 'worthless', 'quit', 'problems', 'garbage', 'love', 'dog', 'my', 'to', 'him', 'posting', 'buying', 'mr', 'licks', 'I', 'so', 'park', 'please']
>>> bayes.setOfWords2Vec(myVocabList,listOPosts[0])
[0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1]
>>> bayes.setOfWords2Vec(myVocabList,listOPosts[1])
[0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0]
训练算法:从词向量计算概率
用法:zeros(shape, dtype=float, order=’C’)
返回:返回来一个给定形状和类型的用0填充的数组;
参数:shape:形状
dtype:数据类型,可选参数,默认numpy.float64
def trainNB0(trainMatrix,trainCategory):
numTrainDocs = len(trainMatrix)
numWords = len(trainMatrix[0])
pAbusive = sum(trainCategory)/float(numTrainDocs)
p0Num = zeros(numWords); p1Num = zeros(numWords)
p0Denom = 0.0; p1Denom = 0.0
for i in range(numTrainDocs):
if trainCategory[i] == 1:
p1Num += trainMatrix[i]
p1Denom += sum(trainMatrix[i])
else:
p0Num += trainMatrix[i]
p0Denom += sum(trainMatrix[i])
p1Vect = p1Num/p1Denom
p0Vect = p0Num/p0Denom
return p0Vect,p1Vect,pAbusive
根据之后的调用,传入的矩阵应该是[[0,0,0,1,1,…], [0,1,…… ] , [ ]…… ]
一个6行32列的矩阵,因为之前统计的不重复的词汇表应该是有32个,文档总共有6个,传入的由标签构成的向量意味着这每一行是否是侮辱性文档。
pAbusive计算的是侮辱性文档的概率6个里面有3个 0.5比较好理解
在判断好 if trainCategory[i] == 1:条件的情况下,
p1Num += trainMatrix[i]
p1Denom += sum(trainMatrix[i])
p1Num本来时一个具有32个0元素的数组,trainMatrix[i]是i行中是否含有32个不重复词汇的判断,就是[0,1,0,0,0,1……](这只是个例子)
p1Num += trainMatrix[i]按照循环 最后p1Num还是一个32数组,数组里的每个元素都是这5个文档在这个位置有没有i位置不重复词汇的和,就是每个词汇出现的次数
p1Denom += sum(trainMatrix[i]) p1Denom是数值,应该是这5行中侮辱性文档每一行相加的值,就是这32个词汇在侮辱性文档中出现的次数
p1Vect = p1Num/p1Denom 就是在已知文档是否为侮辱性文档的情况下,32个词汇中 每个词汇出现的次数 /32个词汇出现的总次数
起初,一直提醒我这个问题,我查了一下,说是解释器没法区分这个到底是局部变量,还是全局变量,加一个global,这我就很迷了,因为我的变量全部在trainNB0里面啊,我还加了global试了试,结果是,当然不行了,不是这个问题
UnboundLocalError: local variable 'p1Denom' referenced before assignment
没办法,我就找了之前我下的一个机器学习实战的一个代码集和数据集来,是吧,代码这种东西,看起来一样,可能跑起来就不一样了,然后,我就跑出了一个新的不一样的错误,相当无语
>>> import bayes
>>> from imp import reload
>>> reload(bayes)
<module 'bayes' from 'E:\\Python\\bayes.py'>
>>> listOPosts,listClasses = bayes.loadDataSet()
>>> myVocabList = bayes.createVocabList(listOPosts)
>>> trainMat=[]
>>> for postinDoc in listOPosts:
trainMat.append(bayes.setOfWords2Vec(myVocabList,postinDoc))
>>> p0V,p1V,pAb=bayes.trainNBO(trainMat,listClasses)
Traceback (most recent call last):
File "<pyshell#9>", line 1, in <module>
p0V,p1V,pAb=bayes.trainNBO(trainMat,listClasses)
AttributeError: module 'bayes' has no attribute 'trainNBO'
然后,我就又迷了,说我的函数trainNB0没有训练,然后找朋友看,说出了让我吐血三升的一句话,你这个trainNB0是0啊还是O啊
啊 然后我就检查了一下
p0V,p1V,pAb=bayes.trainNB0(trainMat,listClasses)
我当时确实是搞错了,可能之前那个错误也是,我也没法确定,没有保留之前的代码,真是无语了都,学习的过程总是无限踩坑
千呼万唤始出来
>>> import bayes
>>> from imp import reload
>>> reload(bayes)
<module 'bayes' from 'E:\\Python\\bayes.py'>
>>> listOPosts,listClasses = bayes.loadDataSet()
>>> myVocabList = bayes.createVocabList(listOPosts)
>>> trainMat=[]
>>> for postinDoc in listOPosts:
trainMat.append(bayes.setOfWords2Vec(myVocabList,postinDoc))
>>> p0V,p1V,pAb=bayes.trainNB0(trainMat,listClasses)
>>> pAb
0.5
>>> p0V
array([0.04166667, 0. , 0.04166667, 0.04166667, 0. ,
0.04166667, 0.04166667, 0.04166667, 0.04166667, 0.04166667,
0. , 0. , 0.04166667, 0. , 0.04166667,
0.04166667, 0. , 0.04166667, 0. , 0.04166667,
0.04166667, 0.04166667, 0.04166667, 0.04166667, 0. ,
0. , 0. , 0.04166667, 0.04166667, 0.08333333,
0. , 0.125 ])
>>> p1V
array([0. , 0.05263158, 0. , 0. , 0.15789474,
0. , 0. , 0. , 0. , 0.05263158,
0.05263158, 0.05263158, 0. , 0.05263158, 0. ,
0. , 0.05263158, 0. , 0.05263158, 0. ,
0. , 0.10526316, 0. , 0. , 0.05263158,
0.10526316, 0.05263158, 0. , 0.05263158, 0.05263158,
0.05263158, 0. ])