代码注释：机器学习实战第10章 利用K-均值聚类算法对未标注数据分组

写在开头的话：在学习《机器学习实战》的过程中发现书中很多代码并没有注释，这对新入门的同学是一个挑战，特此贴出我对代码做出的注释，仅供参考，欢迎指正。

1、K-均值聚类算法

#coding:gbk
from numpy import *

#作用：从文件中导入数据集
#输入：文件名
#输出：数据集
def loadDataSet(fileName): #general function to parse tab -delimited floats
dataMat = [] #assume last column is target value
fr = open(fileName)
for line in fr.readlines():
curLine = line.strip().split('\t')
fltLine = map(float,curLine) #map all elements to float()
dataMat.append(fltLine)
return dataMat

#作用：计算两个向量的欧式距离
#输入：向量A，向量B
#输出：向量间的欧式距离
def distEclud(vecA, vecB):
return sqrt(sum(power(vecA - vecB, 2)))

#作用：为给定数据集构建一个包含k个随机质心的集合
#输入：数据集，随机质心数
#输出：包含k个随机质心的集合
def randCent(dataSet, k):
n = shape(dataSet)[1]
centroids = mat(zeros((k, n)))
for j in range(n):
minJ = min(dataSet[:, j])
rangeJ = float(max(dataSet[:, j]) - minJ)
centroids[:, j] = minJ + rangeJ * random.rand(k, 1)
return centroids

#作用：k-均值算法
#输入：数据集，簇数目，距离计算方法，质心集合创造方法
#输出：簇质心集合，簇分配结果矩阵
def kMeans(dataSet, k, distMeas = distEclud, createCent = randCent):
m = shape(dataSet)[0]
#簇分配结果矩阵，包含两列，一列记录簇索引值，第二列存储误差
clusterAssment = mat(zeros((m, 2)))
centroids = createCent(dataSet, k)
clusterChanged = True
while clusterChanged:
clusterChanged = False
#对每个样本，寻找最近的质心
for i in range(m):
minDist = inf
minIndex = -1#从属簇的索引值
for j in range(k):
distJI = distMeas(centroids[j, :], dataSet[i, :])
if distJI < minDist:
minDist = distJI
minIndex = j
#只要有数据点的簇分配结果发生改变，clusterChanged = True
if clusterAssment[i, 0] != minIndex:
clusterChanged = True
clusterAssment[i, :] = minIndex, minDist ** 2
#print centroids
#遍历所有质心并更新它们的取值
for cent in range(k):
ptsInClust = dataSet[nonzero(clusterAssment[:, 0].A == cent)[0]]
centroids[cent, :] = mean(ptsInClust, axis = 0)#axis = 0表示沿矩阵的列方向进行均值计算
return centroids, clusterAssment

2、二分K-均值聚类算法

#作用：二分k-均值聚类算法
#输入：数据集，簇数目，距离计算方法
#输出：簇质心集合，簇分配结果矩阵
def biKmeans(dataSet, k, distMeas = distEclud):
m = shape(dataSet)[0]#数据点个数
clusterAssment = mat(zeros((m, 2)))#簇分配结果矩阵，包含两列，一列记录簇索引值，第二列存储误差
centroid0 = mean(dataSet, axis = 0).tolist()[0]#计算整个数据集的质心
centList = [centroid0]#使用列表保留所有簇的质心，将初始簇的质心压入
#遍历数据集中所有的点来计算每个点到质心的误差值
for j in range(m):
clusterAssment[j, 1] = distMeas(mat(centroid0), dataSet[j, :]) ** 2
#不停对簇进行划分，直到得到想要的簇数目为止
while (len(centList) < k):
lowestSSE = inf
#遍历已有的簇来决定最佳的簇进行划分
for i in range(len(centList)):
#只有第i个簇的数据集
ptsInCurrCluster = dataSet[nonzero(clusterAssment[:, 0].A == i)[0], :]
#对第i个簇一分为二
centroidMat, splitClustAss = kMeans(ptsInCurrCluster, 2, distMeas)
#对第i个簇划分后得到的误差平方和
sseSplit = sum(splitClustAss[:, 1])
#除了第i个簇的数据集的误差平方和
sseNotSplit = sum(clusterAssment[nonzero(clusterAssment[:, 0].A != i)[0], 1])
print "sseSplit, and notSplit: ", sseSplit, sseNotSplit
#如果划分后的簇有最小的总误差
if (sseSplit + sseNotSplit) < lowestSSE:
bestCentToSplit = i
bestNewCents = centroidMat
bestClustAss = splitClustAss.copy()
lowestSSE = sseSplit +sseNotSplit
#将新的二分好的簇的第二个的索引值设为cenList + 1，即现有的centList后的一个
bestClustAss[nonzero(bestClustAss[:, 0].A == 1)[0], 0] = len(centList)
#将新的二分好的簇的第二个的索引值设为bestCentToSplit，即要二分的簇
bestClustAss[nonzero(bestClustAss[:, 0].A == 0)[0], 0] = bestCentToSplit
print 'the bestCentToSplit is: ', bestCentToSplit
print 'the len of bestClustAss is: ', len(bestClustAss)
#均需要加tolist()[0]，否则后面会出错误
centList[bestCentToSplit] = bestNewCents[0, :].tolist()[0]#将i个簇换成新的二分好的簇的第一个
centList.append(bestNewCents[1, :].tolist()[0])#将新的二分好的簇的第二个压入列表
clusterAssment[nonzero(clusterAssment[:, 0].A == bestCentToSplit)[0], :] = bestClustAss#更新簇的分配结果
#print centList
return mat(centList), clusterAssment

3、对地图上的点进行聚类

import urllib
import json
#作用：对地址进行地理编码
#输入：地址，城市
#输出：地理编码
def geoGrab(stAddress, city):
apiStem = 'http://where.yahooapis.com/geocode?'
params = {}
params['flags'] = 'J'#将返回类型设为JSON
params['appid'] = 'ppp68N8t'
params['location'] = '%s %s' % (stAddress, city)
url_params = urllib.urlencode(params)#将创建的字典转换为可以通过URL进行传递的字符串格式
yahooApi = apiStem + url_params
print yahooApi#打印输出的URL
c = urllib.urlopen(yahooApi)#打开URL
return json.loads(c.read())#读取返回值

from time import sleep
#作用：服务器不存在，失败
#输入：
#输出：
def massPlaceFind(fileName):
fw = open('places.txt', 'w')
for line in open(fileName).readlines():
line = line.strip()
lineArr = line.split('\t')
retDict = geoGrab(lineArr[1], lineArr[2])
if retDict['ResultSet']['Error'] == 0:
lat = float(retDict['ResultSet']['Result'][0]['latitude'])#维度
lng = float(retDict['ResultSet']['Result'][0]['longitude'])#经度
print "%s\t%f\t%f" % (lineArr[0], lat, lng)
fw.write('%s\t%f\t%f\n' % (line, lat, lng))
else:
print "error fetching"
sleep(1)
fw.close()

def distSLC(vecA, vecB):
a = sin(vecA[0, 1] * pi / 180) * sin(vecB[0, 1] * pi / 180)
b = cos(vecA[0, 1] * pi / 180) * cos(vecB[0, 1] * pi / 180) * cos(pi * (vecB[0, 0] -vecA[0, 0]) / 180)
return arccos(a + b) * 6371.0

import matplotlib
import matplotlib.pyplot as plt
def clusterClubs(numClust = 5):
datList = []#表示每个地点的经度、维度
for line in open('places.txt').readlines():
lineArr = line.split('\t')
datList.append([float(lineArr[4]), float(lineArr[3])])
datMat = mat(datList)
#二分k-均值聚类算法
myCentroids, clustAssing = biKmeans(datMat, numClust, distMeas = distSLC)
fig = plt.figure()
rect = [0.1, 0.1, 0.8, 0.8]
scatterMarkers = ['s', 'o', '^', '8', 'p', 'd', 'v', 'h', '>', '<']
axprops = dict(xticks = [], yticks = [])
ax0 = fig.add_axes(rect, label = 'ax0', **axprops)
imgP = plt.imread('Portland.png')
ax0.imshow(imgP)
ax1 = fig.add_axes(rect, label = 'ax1', frameon = False)
#ax1.scatter(myCentroids[:, 0].flatten().A[0], myCentroids[:, 1].flatten().A[0], marker='+', s=300)
#绘制坐标点
for i in range(numClust):
ptsInCurrCluster = datMat[nonzero(clustAssing[:, 0].A == i)[0], :]
markerStyle = scatterMarkers[i % len(scatterMarkers)]
ax1.scatter(ptsInCurrCluster[:, 0].flatten().A[0], ptsInCurrCluster[:, 1].flatten().A[0], \
marker = markerStyle, s = 90)
#myCentroids = mat(myCentroids)
#print myCentroids
#绘制簇中心
ax1.scatter(myCentroids[:, 0].flatten().A[0], myCentroids[:, 1].flatten().A[0], marker = '+', s = 300)
plt.show()