K近邻算法总结

在这里插入代码片@TOC
机器学习实战中K近领域算法优化约会网站配对结果

1.特征值数据集

第一列表示每年的飞行里程数
第二列表示玩视频游戏所耗时间比
第三列表示每周消费的冰淇淋公升数
最后一列表示约会对象等级

2. 数据处理

def classify0(inX, dataSet, labels, k):
    dataSetSize = dataSet.shape[0] # 1
    diffMat = tile(inX, (dataSetSize, 1)) - dataSet # 2
    sqDiffMat = diffMat ** 2
    sqDistances = sqDiffMat.sum(axis=1) # 3
    distances = sqDistances ** 0.5
    sortedDistIndicies = distances.argsort() # 4
    classCount = {}
    for i in range(k):
        voteIlabel = labels[sortedDistIndicies[i]] # 5
        classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
    sortedClassCount = sorted(classCount.items(), key=operator.itemgetter(1), reverse=True)
    return sortedClassCount[0][0]

2.1 shape

注释1： dataSetSize = dataSet.shape[0]:

import numpy as np
x = np.array([[1,2,5],[2,3,5],[3,4,5],[2,3,6]])
#输出数组的行和列数
print x.shape  #结果： (4, 3)
#只输出行数
print x.shape[0] #结果： 4
#只输出列数
print x.shape[1] #结果： 3

2.2 title

注释2： diffMat = tile(inX, (dataSetSize, 1)) - dataSet

>>> tile(1,2)
array([1, 1])

>>> tile((1,2,3),3)
array([1, 2, 3, 1, 2, 3, 1, 2, 3])
 
>>> tile(a,2)
array([[1, 2, 3, 1, 2, 3],
       [4, 5, 5, 4, 5, 5]])
 
>>> b=[1,3,5]
>>> tile(b,[2,3])
array([[1, 3, 5, 1, 3, 5, 1, 3, 5],
       [1, 3, 5, 1, 3, 5, 1, 3, 5]])

2.3 sum

注释3：sqDistances = sqDiffMat.sum(axis=1)
一个多维数组精确到某个元素需要a[n0][n1][n2][…][n]，则axis的取值就是n

2.4 argsort

注释4 ： sortedDistIndicies = distances.argsort()

import numpy as np
x=np.array([1,4,3,-1,6,9])
x.argsort()

输出定义为y=array([3,0,2,1,4,5])。
argsort()函数是将x中的元素从小到大排列，提取其对应的index(索引)，然后输出到y。例如：x[3]=-1最小，所以y[0]=3,x[5]=9最大，所以y[5]=5。

3. 将数据集转换为特征值矩阵

def classify0(inX, dataSet, labels, k):
    dataSetSize = dataSet.shape[0] # 1
    diffMat = tile(inX, (dataSetSize, 1)) - dataSet # 2
    sqDiffMat = diffMat ** 2
    sqDistances = sqDiffMat.sum(axis=1) # 3
    distances = sqDistances ** 0.5
    sortedDistIndicies = distances.argsort() # 4
    classCount = {}
    for i in range(k):
        voteIlabel = labels[sortedDistIndicies[i]]
        classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
    sortedClassCount = sorted(classCount.items(), key=operator.itemgetter(1), reverse=True)
    return sortedClassCount[0][0]


# print(classify0([0,0],array([[1.0, 1.1], [1.0, 1.0], [0, 0], [0, 0.1]]), ['A', 'A', 'B', 'B'], 3 ))

def file2matrix(filename):
    fr = open(filename)
    arrayOlines = fr.readlines()
    numberOfLines = len(arrayOlines)  # 文件行数
    returnMat = zeros((numberOfLines, 3))  # 初始化数组
    classsLabelVector = []
    index = 0
    for line in arrayOlines:
        line = line.strip()  # 去掉回车字符
        listFromLine = line.split('\t')
        returnMat[index, :] = listFromLine[0:3]
        # classsLabelVector.append(listFromLine[-1])
        if listFromLine[-1] == "largeDoses":
            classsLabelVector.append(3)
        elif listFromLine[-1] == "smallDoses":
            classsLabelVector.append(2)
        elif listFromLine[-1] == "didntLike":
            classsLabelVector.append(1)
        index += 1
    return returnMat, classsLabelVector


dataMat, dataLabels = file2matrix('data\datingTestSet.txt')

import matplotlib
import matplotlib.pyplot as plt

fig = plt.figure()
ax = fig.add_subplot(111)


# print(dataMat[:,1])

# ax.scatter(dataMat[:, 1], dataMat[:, 2])
# ax.scatter(dataMat[:, 1], dataMat[:, 2], 15.0 * array(dataLabels), 15.0 * array(dataLabels))
# plt.show()

完整运行代码

from numpy import *
import operator


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


# print(createDataSet())

def classify0(inX, dataSet, labels, k):
    dataSetSize = dataSet.shape[0]
    diffMat = tile(inX, (dataSetSize, 1)) - dataSet
    sqDiffMat = diffMat ** 2
    sqDistances = sqDiffMat.sum(axis=1)
    distances = sqDistances ** 0.5
    sortedDistIndicies = distances.argsort()
    classCount = {}
    for i in range(k):
        voteIlabel = labels[sortedDistIndicies[i]]
        classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
    sortedClassCount = sorted(classCount.items(), key=operator.itemgetter(1), reverse=True)
    return sortedClassCount[0][0]


# print(classify0([0,0],array([[1.0, 1.1], [1.0, 1.0], [0, 0], [0, 0.1]]), ['A', 'A', 'B', 'B'], 3 ))

def file2matrix(filename):
    fr = open(filename)
    arrayOlines = fr.readlines()
    numberOfLines = len(arrayOlines)  # 文件行数
    returnMat = zeros((numberOfLines, 3))  # 初始化数组
    classsLabelVector = []
    index = 0
    for line in arrayOlines:
        line = line.strip()  # 去掉回车字符
        listFromLine = line.split('\t')
        returnMat[index, :] = listFromLine[0:3]
        # classsLabelVector.append(listFromLine[-1])
        if listFromLine[-1] == "largeDoses":
            classsLabelVector.append(3)
        elif listFromLine[-1] == "smallDoses":
            classsLabelVector.append(2)
        elif listFromLine[-1] == "didntLike":
            classsLabelVector.append(1)
        index += 1
    return returnMat, classsLabelVector


dataMat, dataLabels = file2matrix('data\datingTestSet.txt')

import matplotlib
import matplotlib.pyplot as plt

fig = plt.figure()
ax = fig.add_subplot(111)


# print(dataMat[:,1])

# ax.scatter(dataMat[:, 1], dataMat[:, 2])
# ax.scatter(dataMat[:, 1], dataMat[:, 2], 15.0 * array(dataLabels), 15.0 * array(dataLabels))
# plt.show()


# 归一化数值
def autoNorm(dataSet):
    minVals = dataSet.min(0)
    maxVals = dataSet.max(0)
    ranges = maxVals - minVals
    normDateSet = zeros(shape(dataSet))
    m = dataSet.shape[0]
    normDateSet = dataSet - tile(minVals, (m, 1))
    normDateSet = normDateSet / tile(ranges, (m, 1))
    return normDateSet, ranges, minVals


def datingClassTest():
    hoRatio = 0.10
    datingDataMat, datingLabels = file2matrix('data\datingTestSet.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, :], dataLabels[numTestVecs:m], 3)
        print("the classifier came back with: %d , the real answer is : %d" % (classifierResult, dataLabels[i]))
        if (classifierResult != dataLabels[i]):
            errorCount += 1.0
    print("the total error rate is :%f" % (errorCount / float(numTestVecs)))


datingClassTest()

示例文件

knn算法示例