kNN算法python代码学习2-手写识别系统

来源:互联网 发布:sql中的distinct 编辑:程序博客网 时间:2024/05/22 11:33

构造的系统能识别数字0到9,已经将图像转换为文本格式。

为了使用之前构造的分类器,要将图像格式化处理为一个向量。

def img2vector(filename):    returnVect = zeros((1,1024))    fr = open(filename)    for i in range(32): #图片大小为32*32        lineStr = fr.readline()        for j in range(32):            returnVect[0,32*i+j] = int(lineStr[j])    return returnVect

我们得到了分类器可以识别的数据格式,接下来就要将数据输入到分类器,对分类器的准确率进行测试:

def handwritingClassTest():    hwLabels = []    trainingFileList = listdir('trainingDigits')#使用此函数之前要将from os import listdir写在文件开头    m = len(trainingFileList)    trainingMat = zeros((m,1024))#每一行代表一个图片数据    for i in range(m):        fileNameStr = trainingFileList[i]        fileStr = fileNameStr.split('.')[0]        classNumStr = int(fileStr.split('_')[0]) #通过两次split获得分类数字(label)        hwLabels.append(classNumStr)        trainingMat[i,:] = img2vector('trainingDigits/%s'% fileNameStr)    testFileList = listdir('testDigits')    errorCount = 0.0    mTest = len(testFileList)    for i in range(mTest):        fileNameStr = testFileList[i]        fileStr = fileNameStr.split('.')[0]        classNumStr = int(fileStr.split('_')[0])        vectorUnderTest = img2vector('testDigits/%s' % fileNameStr)        classifierResult = classify0(vectorUnderTest, trainingMat, hwLabels, 3)#由于文件中的值都在0和1之间,所以不再需要归一化        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))

命令行中输入:

>>> reload(kNN)<module 'kNN' from 'C:\Users\mrzhang\Desktop\prac\python\kNN.py'>>>> kNN.handwritingClassTest()

测试结果为:

the classifier came back with: 0, the real answer is: 0the classifier came back with: 0, the real answer is: 0.....省略一些the classifier came back with: 9, the real answer is: 9the classifier came back with: 9, the real answer is: 9the total number of errors is: 11the total error rate is: 0.011628

得到的错误率为1.16%。

提高计算效率?KD树


0 0
原创粉丝点击