海伦约会-KNN1

# -*- coding: utf-8 -*-
"""
Created on Sun Oct 21 22:44:59 2018

@author: 国涛
"""
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
import operator
def file2matrix(filename):
    '''
        从文件中读取数据
        filename:是文件所在的路径
    '''
    #打开文件
    fr = open(filename)
    #读取文件的内容
    arrayOLines = fr.readlines()
    #计算文件的个数
    numberOfLines = len(arrayOLines)
    #创建出需要返回的数组
    returnMat = np.zeros((numberOfLines,3))
    #创建需要返回的类别字典
    classLabelVector = []
    index = 0
    #开始读数据进行处理
    for line in arrayOLines:
        line  = line.strip()
        listFromLine = line.split('\t')
        returnMat[index,:] = listFromLine[0:3]
        classLabelVector.append(listFromLine[-1])
        index += 1
    return returnMat,classLabelVector
def createMap(returnMat,classLabelVector):
    '''
        根据数据创建散点图
        returnMat:数据集
        classLabelVector:标签集
    '''
    fig = plt.figure()
    ax = fig.add_subplot(111, projection='3d')
    row,cols = returnMat.shape
    for i in range(row):
        if(classLabelVector[i] == 'largeDoses'):
            ax.scatter(returnMat[i,0],returnMat[i,1],returnMat[i,2],c='r',marker='o') 
        elif(classLabelVector[i] == 'smallDoses'):
            ax.scatter(returnMat[i,0],returnMat[i,1],returnMat[i,2],c='b',marker='*')
        else:
            ax.scatter(returnMat[i,0],returnMat[i,1],returnMat[i,2],c='g',marker='^')

def autoNorm(returnMat):
    '''
        对数据集进行归一化
        returnMat:数据集
        normData:归一化后的数据集
        ranges:maxVals - minVals
        minVals:数据集中每列的最小值
    '''
    minVals = returnMat.min(0)
    maxVals  = returnMat.max(0)
    ranges = (maxVals  - minVals )
    normDataSet = np.zeros((returnMat.shape))
    m = returnMat.shape[0]
    normDataSet = returnMat - np.tile(minVals ,(m,1))
    normDataSet = normDataSet/np.tile(ranges,(m,1))
    return normDataSet,ranges,minVals

def datingClassTest(dataingDataMat,dataingLabels):
    '''
        对网站分类器进行测试
    '''
    hoRate = 0.10
    errorCount = 0.0
    normData,ranges,minVals = autoNorm(dataingDataMat)
    m = dataingDataMat.shape[0]
    numTestVecs = int(hoRate * m)
    for i in range(numTestVecs):
        classifierResult = classify0(normData[i,:],normData[numTestVecs:m,:],
                                     dataingLabels[numTestVecs:m],3)
        print('the classifier came back with: %s,the real answer is : %s'
              %(classifierResult,dataingLabels[i]))
        if(classifierResult != dataingLabels):
            errorCount += 1.0
    print("the total error rate is %f%%" %(errorCount/float(numTestVecs)))

def classifyPerson():
    '''
        对实际的数据进行预测
    '''   
    percentTats = float(input("percent of time spend playing video games："))
    ffMiles = float(input("frequent flier miles earned per years："))
    iceCream = float(input("liters of icr creams per years："))
    filePath = 'E:\\机器学习实战\\datingTestSet.txt'
    returnMat,classLabelVector = file2matrix(filePath)
    normData,ranges,minVals = autoNorm(returnMat)
    inArr = np.array([ffMiles,percentTats,iceCream])
    classifierResult = classify0((nArr - minVals)/ranges,normData,
                                     classLabelVector,3)
    print('the classifier came back with: %s'
              %(classifierResult))  
def classify0(inX,dataSet,labels,k):
    '''
        inX:输入的测试集
        dataSet：数据集
        labels:标签
        k:选择距离最小的k个点
    '''
    #计算数据集的个数
    dataSetSize = dataSet.shape[0]
    #计算inX与数据集直接的距离差
    diffMat = np.tile(inX,(dataSetSize,1)) - dataSet
    #计算平方和
    sqDiffMat = diffMat**2
    sqDistances = sqDiffMat.sum(axis=1)
    #计算距离
    distance = sqDistances**0.5
    #根据距离进行排序
    sortedDistIndices = distance.argsort()
    classCount={}
    for i in range(k):
        votelabel = labels[sortedDistIndices[i]]
        classCount[votelabel] = classCount.get(votelabel,0) + 1
    sortedClassCount = sorted(classCount.items(),key=operator.itemgetter(1),
                              reverse=True)
    return sortedClassCount[0][0]

if __name__ == '__main__':
    #根据数据集创建散点图
    createMap(returnMat,classLabelVector)
    #由于数据之间相差的太多，对数据进行归一化
    #autoNorm(returnMat)
    #d对分类器进行测试
    datingClassTest(returnMat,classLabelVector)
    classifyPerson()
    
    
        
        
    

---

结果显示

散点图：

模型测试误差：

对数据进行预测：