分类算法-----KNN思想和python实现学习笔记

1.1K近邻算法的简介

K近邻算法用于测量不同特征值之间的距离，根据最近的前N个样本的标签给数据进行分类

优点：精度高、对异常值不敏感、无数据输入假设

缺点：计算复杂度高、空间复杂度高

1.2K近邻算计算流程

1.3 K近邻算法例子

下图个K近邻算法的计算步骤流程，由于K=4，Top4有3个为flag1则A的类别为flag1，下图选用最常用的计算距离的公式两点之间距离（欧式距离）

1.5Python代码实现

一下代码参加机器学习实战

 

from numpy import *
import  matplotlib.pyplot  as plt
#读数据集
def file2matrix(filename):
    fr = open(filename)
    arror=fr.readlines()
    numberOfLines = len(arror)         #get the number of lines in the file
    returnMat = zeros((numberOfLines,3))        #prepare matrix to return
    classLabelVector = []                       #prepare labels return   
    index = 0
    for line in arror:
        line = line.strip()
        listFromLine = line.split('\t')
        returnMat[index,:] = listFromLine[0:3]
        classLabelVector.append(int(listFromLine[-1]))
        index += 1
    return returnMat,classLabelVector
 #数据集归一化   
def autoNorm(dataSet):
    minVals = dataSet.min(0)
    maxVals = dataSet.max(0)
    ranges = maxVals - minVals
    normDataSet = zeros(shape(dataSet))
    m = dataSet.shape[0]
    normDataSet = dataSet - tile(minVals, (m,1))
    normDataSet = normDataSet/tile(ranges, (m,1))   #element wise divide
    return normDataSet, ranges, minVals

#分类器
def classify0(inX, dataSet, labels, k):
    dataSetSize = dataSet.shape[0]#数据集记录数
    diffMat = tile(inX, (dataSetSize,1)) - dataSet#生成一个和dataSet相同维度的数据集
    sqDiffMat = diffMat**2#平方项
    sqDistances = sqDiffMat.sum(axis=1)#按行求和
    distances = sqDistances**0.5#开方
    sortedDistIndicies = distances.argsort()     #对结果进行排序
    classCount={}          
    for i in range(k):
        voteIlabel = labels[sortedDistIndicies[i]]
        classCount[voteIlabel] = classCount.get(voteIlabel,0) + 1
    sortedClassCount = sorted(classCount.iteritems(), key=operator.itemgetter(1), reverse=True)
    return sortedClassCount[0][0]

#对钱hoRatio*项进行分类，
def datingClassTest():
    hoRatio = 0.50      #训练集的数据量
    datingDataMat,datingLabels = file2matrix('datingTestSet2.txt')       #load data setfrom file
    normMat, ranges, minVals = autoNorm(datingDataMat)#数据归一化，normMat归一化的结果，ranges最大值和最小值范围，minVals最小值
    m = normMat.shape[0]#行数
#    print shape(normMat)
    numTestVecs = int(m*hoRatio)#
    print numTestVecs
    errorCount = 0.0
    for i in range(numTestVecs):
        classifierResult = classify0(normMat[i,:],normMat[numTestVecs:m,:],datingLabels[numTestVecs:m],5)
#        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 classifyPerson():
    resultList=['not at all','in some doses','in large doses']
    ffmiles=float(raw_input('frequent filer miles earned'))
    precents=float(raw_input('precents filer miles earned'))
    iceCream=float(raw_input('liters of ice'))
    datingDataMat,datingLabels = file2matrix('datingTestSet2.txt') 
    normMat, ranges, minVals = autoNorm(datingDataMat)
    inArray=[ffmiles,precents,iceCream]
    classifierResult = classify0((inArray-minVals)*1.0/ranges,normMat,datingLabels,3)  
    print resultList[classifierResult-1]
    print classifierResult




if __name__=='__main__':
    datingClassTest()
    classifyPerson()

下图为数据的分布趋势

 

下面为用户手动输入数据观测数据所属类别

至此K近邻算法的实现基本完成