KNN学习笔记1

#学习自机器学习实战，部分更改来自自己
from numpy import*
import operator
import matplotlib
import matplotlib.pyplot as plt
def createDataset():
    group=array([[1.0,1.1],[1.0,1.0],[0,0],[0,0.1]])
    labels=['A' ,'A' ,'B' ,'B']
    return group,labels
def classify(inX,dataSet,labels,k):
   # print(inX)
    dataSetSize=dataSet.shape[0]
   # print(tile(inX,(dataSetSize)))
    #print((dataSet.shape)) #4 1
    diffMat=tile(inX,(dataSetSize,1))-dataSet#tile like copy   it is [inx inx inx inx]

    print("diffMat" ,diffMat)
    sqDiffMat=diffMat**2
   # print("sqDiffMat\n" ,sqDiffMat)
    #print(sqDiffMat.sum(axis=0))
    sqDistances=sqDiffMat.sum(axis=1)# sum of each list
    distances=sqDistances**0.5
    #print(sqDistances)
    sortedDistIndicies=distances.argsort()#return the index
   # print(sortedDistIndicies)
    classCount={}
    for i in range(k):
        voteIlabel=labels[sortedDistIndicies[i]]
        #print(voteIlabel)
        classCount[voteIlabel]=classCount.get(voteIlabel,0)+1
        #print(classCount)
    sortedClassCount=sorted(classCount.items(),key=operator.itemgetter(1),reverse=True)#sort from big to small
    print(sortedClassCount)
    return sortedClassCount[0][0]
def file2matrix(filename,n):
    fr=open(filename)#open the file
    arrayOLines=fr.readlines()# an iterator
    numberOfLines=len(arrayOLines)#num of line
    returnMat=zeros((numberOfLines,n-1))
   # print("arrayOLines:" , arrayOLines,"\n numberOfLines:" ,numberOfLines,"\nreturnMat:" ,returnMat)
    classLabelVector=[]
    index=0
    for line in arrayOLines:
        line=line.strip()#end with the end of line
        listFromLine=line.split('\t')#s
        #print(listFromLine)
        returnMat[index,:]=listFromLine[0:n-1]
        #print(returnMat)
        classLabelVector.append(float(listFromLine[-1]))
        index+=1
    return returnMat,classLabelVector
def autoNorm(dataSet):
    minVals=dataSet.min(0)
    maxVals=dataSet.max(0)
    ranges=maxVals-minVals
    normDataSet=zeros(shape(dataSet))
   # print(normDataSet)
    m=dataSet.shape[0]
    normDataSet=dataSet-tile(minVals,(m,1))# val-min)/ranges
    normDataSet=normDataSet/tile(ranges,(m,1))
    #print(minVals,maxVals,ranges,m,normDataSet)
    return normDataSet,ranges,minVals
def datingClassTest():
    hoRatio=0.10
    datingDataMat,datingLabels=file2matrix('dataset.txt' ,3)
    normMat,ranges,minVals=autoNorm(datingDataMat)
    m=normMat.shape[0]
    n=normMat.shape[1]
    print("n:" ,n)
    numTestVecs=int(m*hoRatio)
    print(numTestVecs)
    print("normMat" ,normMat)
    errorCount=0.0
    for i in range(numTestVecs):
        #print("test" ,normMat[i:2,:2])
        classifierResult=classify(normMat[i,:n],normMat[numTestVecs:m,:n],\
                                  datingLabels[numTestVecs:m],3)
        if classifierResult!=datingLabels[i] :
            errorCount+=1
        print(classifierResult)
    print(errorCount/numTestVecs)
if __name__=='__main__':
    datingClassTest()
    #inx,labels=createDataset()
    #y=classify([0.5,1],inx,labels,3)
    #print(y)
    # x,y=file2matrix("dataset.txt")
    # print(autoNorm(x))
    # fig=plt.figure()#build
    # ax=fig.add_subplot(111)#the rate of heigh width and depth
    # ax.scatter(x[:,1],x[:,2])#the x and y axis
    # plt.show()
    # print(x)