k-Nearest Neighbors

# -*- coding: UTF-8 -*-    //显示中文，中文编码
'''
Created on Sep 16, 2010
kNN: k Nearest Neighbors

Input:      inX: vector to compare to existing dataset (1xN)
            dataSet: size m data set of known vectors (NxM)
            labels: data set labels (1xM vector)
            k: number of neighbors to use for comparison (should be an odd number)
            
Output:     the most popular class label

@author: pbharrin
'''
from numpy import *         #scientific computing
import operator             #for sorting
from os import listdir

'''
just for convenience to create dataset and labels
'''
def createDataSet():
    #numpy中 array创建时参数必须为list
    group = array([[1.0,1.1],[1.0,1.0],[0,0],[0,0.1]])
    labels = ['A','A','B','B']
    return group, labels


def classify0(inX, dataSet, labels, k):
    '''
    inX:the input vector to classfy
    dataSet:training examples
    labels:a vector of labels 
    k:the number of nearest neighbor to use In the voting  
    '''
    #distance calculation
    dataSetSize = dataSet.shape[0]  #get the length of fist dimension 
    diffMat = tile(inX, (dataSetSize,1)) - dataSet  #将inX向量复制dataSetSize次得到一个矩阵，再将去dataset矩阵得到差值矩阵
    sqDiffMat = diffMat**2              #对差值矩阵中的值进行平方（求点之间距离的公式）
    sqDistances = sqDiffMat.sum(axis=1) #axis=1将矩阵的每一行相加,得到一维向量，默认是anis=0，即普通的相加
    distances = sqDistances**0.5    #开方，得到inX向量点到其他点的距离
    sortedDistIndicies = distances.argsort()    # argsort对distances中数据进行由小到大排序，返回排序下标数组
    
    #dictionary 该变量在后面多次用到，所以此处提前进行声明
    classCount={}        
    for i in range(k):
        voteIlabel = labels[sortedDistIndicies[i]]#sortedDistIndicies[i]为下标值
        #get():Return the value for key if key is in the dictionary, else default（here default is 0）.
        classCount[voteIlabel] = classCount.get(voteIlabel,0) + 1
    #以classCount的第二个域进行排序,即以value进行排序，倒序
    sortedClassCount = sorted(classCount.iteritems(), key=operator.itemgetter(1), reverse=True)
	'''
	注意Python中array 与list的区别。 二维list的访问方式为a[0][1]
	二维array的访问方式为a[1,:] 即访问a的第二行中所有成员，具体区别查看CSDN
	'''
    return sortedClassCount[0][0]
	
'''
将文本文件内容存储到array矩阵中
'''
def file2matrix(filename):
    fr = open(filename)
    numberOfLines = len(fr.readlines())         #get the number of lines in the file
    returnMat = zeros((numberOfLines,3))        #prepare matrix to return(numberOfLines*3大小的矩阵array)
    classLabelVector = []                       #prepare labels return   
    fr = open(filename)							#fr.readlines()会将游标放到末尾，此处使用open()将游标放到开头
    index = 0
    for line in fr.readlines():
        line = line.strip()
        listFromLine = line.split('\t')     #\t是制表符，相当于一个很大的空格。返回一个list
        returnMat[index,:] = listFromLine[0:3]  #将每行的数据存入矩阵returnMat
        classLabelVector.append((listFromLine[-1]))  #数据的最后一个label
        index += 1
    return returnMat,classLabelVector
'''
将不同的属性值进行变为0-1，使得影响力变为相同
'''
def autoNorm(dataSet):
    minVals = dataSet.min(0)						#返回的是一个list，每个项都是各列中最小值
    maxVals = dataSet.max(0)
    ranges = maxVals - minVals
    normDataSet = zeros(shape(dataSet))             #shape(dataSet)返回一个dataSet大小的array
    #shape[0] 只读取第一维的长度
    m = dataSet.shape[0]
    normDataSet = dataSet - tile(minVals, (m,1))    #得到所有数据集合与集合中最小值的差
    normDataSet = normDataSet/tile(ranges, (m,1))   #element wise divide
    return normDataSet, ranges, minVals

'''
error rate
'''
def datingClassTest():
    hoRatio = 0.10      #hold out 10%
    datingDataMat,datingLabels = file2matrix('datingTestSet2.txt')       #load data setfrom file
    normMat, ranges, minVals = autoNorm(datingDataMat)
    m = normMat.shape[0]
    numTestVecs = int(m*hoRatio)
    errorCount = 0.0
    for i in range(numTestVecs):
        classifierResult = classify0(normMat[i,:],normMat[numTestVecs:m,:],datingLabels[numTestVecs:m],3)
        print "the classifier came back with: %d, the real answer is: %d" % (classifierResult, datingLabels[i])
        if (classifierResult != datingLabels[i]): errorCount += 1.0
    print "the total error rate is: %f" % (errorCount/float(numTestVecs))
    print errorCount
    
def img2vector(filename):
    returnVect = zeros((1,1024))
    fr = open(filename)
    for i in range(32):
        lineStr = fr.readline()
        for j in range(32):
            returnVect[0,32*i+j] = int(lineStr[j])
    return returnVect

def handwritingClassTest():
    hwLabels = []
    trainingFileList = listdir('trainingDigits')           #load the training set
    m = len(trainingFileList)
    trainingMat = zeros((m,1024))
    for i in range(m):
        fileNameStr = trainingFileList[i]
        fileStr = fileNameStr.split('.')[0]     #take off .txt
        classNumStr = int(fileStr.split('_')[0])
        hwLabels.append(classNumStr)
        trainingMat[i,:] = img2vector('trainingDigits/%s' % fileNameStr)
    testFileList = listdir('testDigits')        #iterate through the test set
    errorCount = 0.0
    mTest = len(testFileList)
    for i in range(mTest):
        fileNameStr = testFileList[i]
        fileStr = fileNameStr.split('.')[0]     #take off .txt
        classNumStr = int(fileStr.split('_')[0])
        vectorUnderTest = img2vector('testDigits/%s' % fileNameStr)
        classifierResult = classify0(vectorUnderTest, trainingMat, hwLabels, 3)
        print "the classifier came back with: %d, the real answer is: %d" % (classifierResult, classNumStr)
        if (classifierResult != classNumStr): errorCount += 1.0
    print "\nthe total number of errors is: %d" % errorCount
    print "\nthe total error rate is: %f" % (errorCount/float(mTest))