机器学习实战笔记一：K-近邻算法在约会网站上的应用

#kNN分类器
def classify0(inX, dataSet, labels, k):
    dataSetSize = dataSet.shape[0]  #得到数据总量
    diffMat = tile(inX,(dataSetSize,1)) - dataSet #将输入数据扩充成与数据集同样大小的矩阵并作差
    sqDiffMat = diffMat**2
    sqDistances = sqDiffMat.sum(axis=1) #axis = 1 参数是维度参数等于1在此处表示将一个矩阵的每一行向量相加
    distances = sqDistances** 0.5
    sortedDistancesIndicies = distances .argsort() #将列表值进行对比返回一个按照数值升序的下标值
    classCount={}
    for i in range(k):
        voteIlabel = labels[sortedDistancesIndicies[i]]
        classCount[voteIlabel] = classCount.get(voteIlabel,0)+1
        #dict.get("key") 返回value  dict.get("key",default= None)如果能找到就返回对应的value找不到返回默认值
    sortedClassCount = sorted(classCount.items(),key = operator.itemgetter(1),reverse=True)
    #sorted 返回一个list  operator.itemgetter(x,y)表示根据x+1维度的第y+1维度
    return sortedClassCount[0][0]

# Author：Thomas Wang
from numpy import *
def file2matrix(filename):
    with open(filename,'r') as fr:
        arrayOLines = fr.readlines()
        numberOfLines = len(arrayOLines)#存储文件数据数目
        returnMat = zeros((numberOfLines,3))#准备接收数据的numpy数组
        classLabelVector = []#准备接收标签向量的数组python列表
        index = 0
        for line in arrayOLines:
            line = line.strip() #截去每行的回车
            listFromLine = line.split('\t')#以‘\t’为分隔符将字符串截取成字符串数组
            returnMat[index,:] = listFromLine[0:3]  #[index,:] = [index][:] 将字符串数组赋值给numpy数组自动转换为浮点型
            classLabelVector.append(int(listFromLine[-1]))
            index +=1
        return returnMat,classLabelVector

datingDataMat,datingLabels = file2matrix('datingTestSet2.txt')
print(datingDataMat)
print(datingLabels[0:20])

[[4.0920000e+04 8.3269760e+00 9.5395200e-01]
 [1.4488000e+04 7.1534690e+00 1.6739040e+00]
 [2.6052000e+04 1.4418710e+00 8.0512400e-01]
 ...
 [2.6575000e+04 1.0650102e+01 8.6662700e-01]
 [4.8111000e+04 9.1345280e+00 7.2804500e-01]
 [4.3757000e+04 7.8826010e+00 1.3324460e+00]]
[3, 2, 1, 1, 1, 1, 3, 3, 1, 3, 1, 1, 2, 1, 1, 1, 1, 1, 2, 3]

import matplotlib
import  matplotlib.pyplot as plt
fig = plt.figure()
ax = fig.add_subplot(111)
ax.scatter(datingDataMat[:,0],datingDataMat[:,2],15.0*array(datingLabels),15.0*array(datingLabels))
plt.show()

def autoNorm(dataSet):
    minVals = dataSet.min(0) #Return the minimum along a given axis.
    maxVals = dataSet.max(0) #Return the maximum along a given axis.
    ranges = maxVals - minVals #取到特征值最大最小值之间的范围
    normDataSet = zeros(shape(dataSet)) 
    m = dataSet.shape[0] #取到数据量
    normDataSet = dataSet - tile(minVals,(m,1))
    normDataSet = normDataSet/tile(ranges,(m,1))
    return normDataSet,ranges,minVals

#完整程序验证分类器
def datingClassTest():
    hoRatio  = 0.1 #测试数据比例
    datingDataMat, datingLabels = file2matrix('datingTestSet2.txt') #文本文件转换成可以处理的矩阵
    normDataSet, ranges, minVals = autoNorm(datingDataMat) #将数据进行归一化
    m = normDataSet.shape[0] #获取数据总量
    numTestVecs = int(m * hoRatio) #获取测试数据数量
    errorCount = 0.0 #错误数据个数计数器
    for i in range(numTestVecs):
        #对每个测试数据利用KNN算法进行分类
        classifierResult = classify0(normDataSet[i,:],normDataSet[numTestVecs:m,:],datingLabels[numTestVecs:m],4)
        #打印预测结果与实际结果
        print("分类器结果：%s,真实结果：%s"%(classifierResult,datingLabels[i]))
        #计算总错误个数并计算错误率
        if (classifierResult != datingLabels[i]): errorCount+=1.0
    print("the total error rate is: %f"%(errorCount/float(numTestVecs)))
datingClassTest()

分类器结果：3,真实结果：3
分类器结果：2,真实结果：2
分类器结果：1,真实结果：1
......
分类器结果：2,真实结果：2
分类器结果：2,真实结果：1
分类器结果：1,真实结果：1
the total error rate is: 0.040000

#完整可用系统
def classifyPerson():
    resultList = ['not at all','in small does','in large does']
    percetTats = float(input("玩视频游戏所耗时间百分比："))
    ffMiles = float(input("每年获取的飞行常客里程数："))
    iceCream = float(input("每年消耗冰淇淋的公升数："))
    datingDataMat,datingLabels = file2matrix("datingTestSet2.txt")
    normMat,ranges,minVals = autoNorm(datingDataMat)
    inArr = array([ffMiles,percetTats,iceCream])
    classifierResult = classify0((inArr-minVals)/ranges,normMat,datingLabels,4)
    print("You probably like this person:",resultList[classifierResult-1])