kNN算法实现和基础应用

ml的学习笔记。

之前学的知识都较为零散，现在需要系统的学习各种算法和思想，目前主要练习内容是基于《机器学习实战》。

在学习过程中，除了学习算法的本身之外，对python库的运用和矩阵运算的代码风格是另外一个学习的重心，c风格的编程习惯如果放到python里远不如矩阵运算和内置函数的效率高，学习过程 的一些记录在代码里以注释方式表示。

kNN（k近邻）为无监督算法，原理即对每个测试数据向量，计算它与训练数据中每个点的距离，选择距离最小的前k个点，统计这k个点的标签，给该测试数据打上统计标签中出现频率最高的标签。

kNN.py

from numpy import *
import operator
from os import listdir
#创建一个简单的测试数据矩阵和标签向量
def createDataSet():
    group = array([[1.0,1.1],[1.0,1.0],[0,0],[0,0.1]])
    labels = ['A','A','B','B']
    return group, labels

#分类器
def classify0(inX, dataSet, labels, k):
    dataSetSize = dataSet.shape[0]
    diffMat = tile(inX, (dataSetSize,1))-dataSet
    sqDiffMat = diffMat**2
    sqDistances = sqDiffMat.sum(axis=1) #axis=1表示矩阵的每行相加
    distances = sqDistances**0.5
    sortedDistIndicies = distances.argsort() #从小到大排序，返回的矩阵存储的是原索引号
    classCount = {}
    for i in range(k):
        voteIlabel = labels[sortedDistIndicies[i]]
        classCount[voteIlabel] = classCount.get(voteIlabel,0) + 1
    sortedClassCount = sorted(classCount.items(), key=lambda d:d[1], reverse = True)
    #sortedClassCount = sorted(classCount.iteritems(), key=operator.itemgetter(1), reversed=True)
    return sortedClassCount[0][0]

#数据归一化：newValue = (oldValue - min)/(max - min)
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))
    return normDataSet,ranges,minVals

#将文件中存储好的数据转换为内存中的矩阵
def file2matrix(filename):
    fr = open(filename)
    arrayOLines = fr.readlines()
    numberOfLines = len(arrayOLines)
    returnMat = zeros((numberOfLines,3))
    classLabelVector = []
    index = 0
    for line in arrayOLines:
        line = line.strip() #除去回车符号
        listFromLine = line.split('\t')
        returnMat[index,:] = listFromLine[0:3]
        classLabelVector.append(int(listFromLine[-1])) # -1取最后一列元素
        index = index + 1
    return returnMat,classLabelVector

def datingClassTest():
    hoRatio = 0.10
    datingDataMat,datingLabels = file2matrix('datingTestSet2.txt')
    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 =errorCount + 1.0
    print("the total error rate is : %f" %(errorCount/float(numTestVecs)))


datingClassTest()
'''
#测试kNN分类器
group,labels = createDataSet()
print(classify0([0,0], group, labels, 3))
'''

'''
data=array([[1,1,3],[1,2,3],[1,0,3]])
data1=array([[3,2,1],[3,2,1],[1,0,3]])
diffmat = data - data1
print(diffmat)
print(diffmat**2)
sumdiffmat = diffmat**2
sumvec = sumdiffmat.sum(axis=1)#矩阵每行相加
print(sumvec)
print(sumvec.argsort())
'''

'''
#字典的运用
classCount= {}
a='A'
classCount[a] = 1
classCount[a] = classCount[a] + 1
classCount[a] = classCount.get(a,0) + 1
classCount['b']=2
classCount['c']=1
#python2
#sortedDscendCount = sorted(classCount.iteritems(), key=operator.itemgetter(1), reversed=True)
#python3
sortedDscendCount = sorted(classCount.items(), key=lambda d:d[1], reverse = True)
print(sortedDscendCount)
sortedAscendCount = sorted(classCount.items(), key=lambda d:d[1], reverse = False)
print(sortedAscendCount)
'''

'''
group,labels = createDataSet()
print(group)
print(labels)
'''

plot.py 测试Matplotlib创建散点图

import matplotlib
import matplotlib.pyplot as plt
import ch1.kNN as knn
returnMat,classLabelVector = knn.file2matrix("datingTestSet2.txt")
fig = plt.figure()
ax = fig.add_subplot(111)
ax.scatter(returnMat[:,1],returnMat[:,2])
plt.show()

test.py 学习的一些库函数的知识

#在python 3.2.3中  input和raw_input 整合了，没有了raw_input,input返回字符串型
'''
整数字符串转换为对应的整数
int('12')
小数字符串转换为对应小数
float('12.34')
数字转换为字符串
str(123.45)
ASCII码转换为相应字符
chr(97)
字符转换为响应ASCII码
ord('a')
'''

'''
#测试listdir,可得到文件夹下一层所有文件和文件夹的名字
from os import listdir
filelist = listdir("D:/testdata")
m = len(filelist)
print(m)
for i in filelist:
    print(i)
    fileStr = i.split('.')
    print(fileStr[0])
    #print(fileStr[1])
'''

'''
a = input("hello :")
print(a)
c=2+int(a)
print(c)
'''