def do_biKmeans(datMat, result, n=3):
from kMeans import biKmeans
from sklearn.metrics import adjusted_rand_score as ari
import numpy as np
import pandas as pd
datMat = np.mat(datMat)
i = 0
pred = []
ari_count = []
'''while i<6:
#做好多次以确认类标
try:
centList, clustAssing = biKmeans(datMat, n)
y_pred = clustAssing.A[:, 0].astype(pd.np.int)
pred.append(y_pred)
ari_count.append(ari(y_pred, result))
except:
y_pred = []
pred.append(y_pred)
ari_count.append(0)
i = i + 1
'''
centList, clustAssing = biKmeans(datMat, n)
y_pred = clustAssing.A[:, 0].astype(pd.np.int)
f = count_f(y_pred, result)
RI = ri(f)
J = j(f)
return y_pred, [RI, J]
def test1():
datMat = mat(kMeans.loadDataSet('testSet.txt'))
print(min(datMat[:, 0]))
print(kMeans.randCent(datMat, 2))
print(kMeans.distEclud(datMat[0], datMat[1]))
#myCentroids, clustAssing = kMeans.kMeans(datMat,3)
myCentroids, clustAssing = kMeans.biKmeans(datMat, 4)
print(myCentroids)
kMeans.plot1(datMat, myCentroids)
def main():
dataSet = kMeans.loadfromcsv('./data/8.csv')
dataMat = np.mat(dataSet)
# normalize dataMat
norMat = kMeans.normalize(dataMat)
# centroids is the center of clusters
# clusterAssment[cluster_index,deviation],in which deviation represents the dist
# from current point to centroids.
# 使用Bisceting Kmeans算法对游客进行聚类，预期聚类簇数目为4
centroids, clusterAssment = kMeans.biKmeans(norMat,4)
cluster_label = clusterAssment[:,0]
clusters = [[],[],[],[]]
for i in range(0,len(cluster_label)):
clusters[(int)(cluster_label[i])].append(np.asarray(norMat)[i])
clusters = np.asarray(clusters)
for i in range(0,len(clusters)):
clusters[i] = np.asarray(clusters[i])
clusters = np.asarray(clusters)
#找出含有元素最多的簇，以及最少的簇
minCount = 10000
maxCount = 0
max_cluster = 0
min_cluster = 0
for i in range(0,len(clusters)):
if minCount > len(clusters[i]):
minCount = len(clusters[i])
if maxCount < len(clusters[i]):
maxCount = len(clusters[i])
if len(clusters[max_cluster]) < len(clusters[i]):
max_cluster = i
if len(clusters[min_cluster]) > len(clusters[i]):
min_cluster = i
print "%d cluster has %d elements " % (i, len(clusters[i])),
print "the centroids is",
print centroids[i]
number_weight = float(len(clusters[max_cluster]))/(len(clusters[min_cluster]))
print centroids[max_cluster]
# 计算Dunn指标
di = base.dunn(clusters)
# 计算N-Dunn指标
NDunnIndex = di*(maxCount/minCount)
print di
print NDunnIndex
print "original dunn is %f" % di
print "weighted dunn is %f" % (number_weight*di)
def biKmeans_func(data_set, k, cent_file="", clus_file=""): print "kMeans : " + cent_file cent, clus = kMeans.biKmeans(data_set, k) # print cent # print clus kmean_res_cent_file = open(cent_file, 'w') for item in cent.A: item_str = "" for column in item: item_str = item_str + str(column) + " " item_str = item_str + '\n' kmean_res_cent_file.write(item_str) kmean_res_cent_file.close kmean_res_clus_file = open(clus_file, 'w') for item in clus.A: item_str = "" for column in item: item_str = item_str + str(column) + " " item_str = item_str + '\n' kmean_res_clus_file.write(item_str) kmean_res_clus_file.close
import kMeans
from numpy import *
# 导入txt数据
datMat = mat(kMeans.loadDataSet('data2.txt'))
# datMat矩阵的第2-4列分别对应半长轴、偏心率和轨道倾角
datMat[0, 2:5]
# 计算距离
delta_v = kMeans.distdeltaV(datMat[0, 2:5], datMat[1, 2:5])
# 随机生成k个质心
centroids = kMeans.randCent(datMat[:, 2:5], 4)
# k-均值聚类
myCentroids, clustAssing = kMeans.kMeans(datMat[:, 2:5], 5, kMeans.distdeltaV)
# 二分 k-均值聚类
centList, myNewAssments = kMeans.biKmeans(datMat[:, 2:5], 5, kMeans.distdeltaV)
# 画图
kMeans.showCluster_SRQ(datMat[:, 2:5], myNewAssments)
import kMeans
from numpy import*
import matplotlib
import matplotlib.pyplot as plt
k = 4
datmat = array(kMeans.loadDataSet('testSet.txt'))
centerList, clusterAssment = kMeans.biKmeans(datmat,k)
print 'The cendroids is:',centerList
fig = plt.figure()
fig.add_subplot(111)
colorList = ['b','c','g','k','r','y']
makerList = ['.','^','*','o','+']
for i in range(k):
ax = plt.scatter(datmat[nonzero(clusterAssment[:,0].A == i)[0],0],datmat[nonzero(clusterAssment[:,0].A == i)[0],1],
c = colorList[i],marker=makerList[i])
ax = plt.scatter(array(centerList[:,0]),array(centerList[:,1]),c = colorList[4],marker=makerList[3])
plt.title('Graph of k_Means ',)
plt.xlabel('x')
plt.ylabel('y')
plt.show()
# -*- coding:utf-8 -*-
import kMeans
from numpy import *
datMat = mat(kMeans.loadDataSet("testSet.txt"))
'''
myCentroids,clusterAssing = kMeans.kMeans(datMat, 4)
print("myCentroids is %s " % myCentroids)
print("clusterAssing is %s " % clusterAssing)
'''
#kMeans test example two
dataMat2 = mat(kMeans.loadDataSet('testSet2.txt'))
centList,myNewAssment = kMeans.biKmeans(dataMat2, 3)
print(centList)
#geoResult = kMeans.geoGrab('1 VA Center', 'Augusta,ME')
import kMeans
from numpy import *
dataMat = mat(kMeans.loadDataSet('testSet.txt'))
# print dataMat
randMat = kMeans.randCent(dataMat, 2)
# print dataMat[:, 0]
# print randMat
res = kMeans.kMeans(dataMat, 4)
# print res
dataMat3 = mat(kMeans.loadDataSet('testSet2.txt'))
kMeans.biKmeans(dataMat3, 3)
# centList, myNewAssments =
import kMeans
import numpy as np
import matplotlib.pyplot as plt
dataMat = np.mat(kMeans.loadDataSet('testSet.txt'))
#print(kMeans.randCent(dataMat,2))
myCentroids, clustAssing = kMeans.kMeans(dataMat,4)
#print(myCentroids,clustAssing)
datalist = dataMat.tolist()
#print([x[0] for x in datalist])
'''plt.figure()
plt.scatter([x[0] for x in datalist],[x[1] for x in datalist])
plt.scatter([x[0] for x in myCentroids.tolist()],[x[1] for x in myCentroids.tolist()])
plt.title('kmeans')
plt.show()'''
dataMat = kMeans.loadDataSet("testSet2.txt")
centList, clusteAssment = kMeans.biKmeans(dataMat, 3)
print(centList)
import kMeans
import ProbIN
from numpy import *
import subprocess
import numpy as np
datMat = mat(kMeans.loadDataSet('motionData_Training.txt'))
kMeans.biKmeans(datMat,12)
# datMat2 = mat(kMeans.loadDataSet('GPS_1Hz_training.txt'))
# kMeans.biKmeans(datMat2,7)
# coding:utf-8
import kMeans
from numpy import *
datMat=mat(kMeans.loadDataSet('testSet.txt'))
print datMat[1:5,:]
myCentroids,clustAssing=kMeans.kMeans(datMat,4)
print myCentroids
print ' '
print clustAssing
datMat3=mat(kMeans.loadDataSet('testSet2.txt'))
centList,myNewAssments=kMeans.biKmeans(datMat3,3)
print centList,myNewAssments
def test4():
dataMat = np.mat(kMeans.loadDataSet('testSet2.txt'))
centList, myNewAssments = kMeans.biKmeans(dataMat, 3)
kMeans.plotScatter(dataMat, centList, myNewAssments)
print centList
#coding=utf-8
import kMeans
from numpy import *
# datMat=mat(kMeans.loadDataSet('testSet.txt'))
# print(datMat)
# myCentroids,clustAssing=kMeans.kMeans(datMat,4)
# print(clustAssing)
datMat = mat(kMeans.loadDataSet('testSet2.txt'))
myCentroids, clustAssing = kMeans.biKmeans(datMat, 3)
print(myCentroids)
# kMeans.clusterClubs(5)
import kMeans
from numpy import *
datMat = mat(kMeans.loadDataSet('testSet.txt'))
print min(datMat[:,0])
print min(datMat[:,1])
print max(datMat[:,0])
print max(datMat[:,1])
print kMeans.randCent(datMat, 2)
print kMeans.distEclud(datMat[0], datMat[1])
myCentroids, clustAssing = kMeans.kMeans(datMat, 4)
#print myCentroids, clustAssing
datMat3 = mat(kMeans.loadDataSet('testSet2.txt'))
centList, myNewAssments = kMeans.biKmeans(datMat3, 3)
print centList
#!/usr/bin/env python
__coding__ = "utf-8"
__author__ = "Ng WaiMing"
from kMeans import kMeans
from kMeans import loadDataSet
from kMeans import randCent
from kMeans import distEclud
from kMeans import biKmeans
from numpy import *
if __name__ == '__main__':
dataMat = mat(loadDataSet('testSet.txt'))
print('min(dataMat[:, 0])', min(dataMat[:, 0]), '\n')
print('min(dataMat[:, 1])', min(dataMat[:, 1]), '\n')
print('max(dataMat[:, 0])', max(dataMat[:, 0]), '\n')
print('max(dataMat[:, 1])', max(dataMat[:, 1]), '\n')
print(randCent(dataMat, 2), '\n')
print(distEclud(dataMat[0], dataMat[1]))
centroids, clusterAssment = kMeans(dataMat, 4)
print('centroids:\n', centroids, '\n')
print('clusterAssment:\n', clusterAssment, '\n')
dataMat3 = mat(loadDataSet('testSet2.txt'))
centList, myNewAssments = biKmeans(dataMat3, 3)
print('centList: \n', centList, '\n')
# fileName = '../../../../data/k-means/places.txt'
# imgName = '../../../../data/k-means/Portland.png'
# kMeans.clusterClubs(fileName=fileName, imgName=imgName, numClust=5)
#-*- coding:utf-8 -*-
import kMeans
from numpy import *
datamat = mat(kMeans.loadDataSet('testSet.txt'))
centList, myNewAssments = kMeans.biKmeans(datamat, 3) #每次给出四个质心，三次迭代后收敛
print centList
#!/usr/bin/python2.7
# _*_ coding: utf-8 _*_
"""
@Author: MarkLiu
"""
import numpy as np
import kMeans
import matplotlib.pyplot as plt
dataArr = kMeans.loadDataSet('datasets/testSet2.txt')
dataMat = np.matrix(dataArr)
k = 3
centroids, clusterAssment = kMeans.biKmeans(dataMat, k)
# centroids, clusterAssment = kMeans.kMeans(dataMat, k)
# 计算原始数据加上中心数据，将数据分离
m = np.shape(dataMat)[0]
# 分离出不同簇的x，y坐标
xPoint_0 = []
yPoint_0 = []
xPoint_1 = []
yPoint_1 = []
xPoint_2 = []
yPoint_2 = []
xPoint_3 = []
yPoint_3 = []
for i in range(m):
if int(clusterAssment[i, 0]) == 0:
def plotCluster():
dataSetShangHai = kMeans.loadfromcsv('./data/8.csv')
dataSetXiAn = kMeans.loadfromcsv('./data/10195.csv')
dataSetQingDao = kMeans.loadfromcsv('./data/10444.csv')
dataSetSanYa = kMeans.loadfromcsv('./data/10030.csv')
dataSetJiuZhaigou = kMeans.loadfromcsv('./data/10136.csv')
dataSetTaiShan = kMeans.loadfromcsv('./data/10284.csv')
dataMatXiAn = np.mat(dataSetXiAn)
dataMatQingDao = np.mat(dataSetQingDao)
dataMatShangHai = np.mat(dataSetShangHai)
dataMatSanYa = np.mat(dataSetSanYa)
dataMatTaiShan = np.mat(dataSetTaiShan)
dataMatJiuZhaigou = np.mat(dataSetJiuZhaigou)
norMatXiAn = kMeans.normalize(dataMatXiAn)
norMatQingDao = kMeans.normalize(dataMatQingDao)
norMatShangHai = kMeans.normalize(dataMatShangHai)
norMatSanYa = kMeans.normalize(dataMatSanYa)
norMatTaiShan = kMeans.normalize(dataMatTaiShan)
norMatJiuZhaigou = kMeans.normalize(dataMatJiuZhaigou)
centroidsShangHai, clusterAssmentShangHai = kMeans.biKmeans(norMatShangHai,4)
centroidsXiAn, clusterAssmentXiAn = kMeans.biKmeans(norMatXiAn,4)
centroidsQingDao, clusterAssmentQingDao = kMeans.biKmeans(norMatQingDao,4)
centroidsSanYa, clusterAssmentSanYa = kMeans.biKmeans(norMatSanYa,4)
centroidsJiuZhaigou, clusterAssmentJiuZhaigou = kMeans.biKmeans(norMatJiuZhaigou,4)
centroidsTaiShan, clusterAssmentTaiShan = kMeans.biKmeans(norMatTaiShan,4)
print "==================上海聚类结果=========="
printBasicInfo(centroidsShangHai,clusterAssmentShangHai,norMatShangHai)
print "==================西安聚类结果=========="
printBasicInfo(centroidsXiAn, clusterAssmentXiAn, norMatXiAn)
print "==================青岛聚类结果=========="
printBasicInfo(centroidsQingDao, clusterAssmentQingDao, norMatQingDao)
print "==================三亚聚类结果=========="
printBasicInfo(centroidsSanYa, clusterAssmentSanYa, norMatSanYa)
print "==================九寨沟聚类结果========"
printBasicInfo(centroidsJiuZhaigou, clusterAssmentJiuZhaigou, norMatJiuZhaigou)
print "==================泰山聚类结果=========="
printBasicInfo(centroidsTaiShan, clusterAssmentTaiShan, norMatTaiShan)
# shanghai
plt.subplot(321)
pointClusNumShangHai = clusterAssmentShangHai[:,0].A.T
n = np.shape(pointClusNumShangHai)[1]
plt.title(u'上海')
for i in range(n):
if 0.0 == pointClusNumShangHai.item(i):
plt.plot(norMatShangHai[i,0],norMatShangHai[i,1],'g^')
elif 1.0 == pointClusNumShangHai.item(i):
plt.plot(norMatShangHai[i,0],norMatShangHai[i,1],'b*')
elif 2.0 == pointClusNumShangHai.item(i):
plt.plot(norMatShangHai[i,0],norMatShangHai[i,1],'k<')
elif 3.0 == pointClusNumShangHai.item(i):
plt.plot(norMatShangHai[i,0],norMatShangHai[i,1],'ms')
plt.plot(centroidsShangHai[:,0],centroidsShangHai[:,1],'ro')
plt.axis([0,4.0,0,12])
plt.xticks([0,1,2,3])
# plt.xlabel(u'Distance Index')
# plt.ylabel(u'Activity Degree Index')
# XiAn
plt.subplot(322)
pointClusNumXiAn = clusterAssmentXiAn[:,0].A.T
n = np.shape(pointClusNumXiAn)[1]
plt.title(u'西安')
for i in range(n):
if 0.0 == pointClusNumXiAn.item(i):
plt.plot(norMatXiAn[i,0],norMatXiAn[i,1],'g^')
elif 1.0 == pointClusNumXiAn.item(i):
plt.plot(norMatXiAn[i,0],norMatXiAn[i,1],'b*')
elif 2.0 == pointClusNumXiAn.item(i):
plt.plot(norMatXiAn[i,0],norMatXiAn[i,1],'k<')
elif 3.0 == pointClusNumXiAn.item(i):
plt.plot(norMatXiAn[i,0],norMatXiAn[i,1],'ms')
plt.plot(centroidsXiAn[:,0],centroidsXiAn[:,1],'ro')
plt.axis([0,4.0,0,12])
plt.xticks([0,1,2,3])
# plt.xlabel(u'Distance Index')
# plt.ylabel(u'Activity Degree Index')
# QingDao
plt.subplot(323)
pointClusNumQingDao = clusterAssmentQingDao[:,0].A.T
n = np.shape(pointClusNumQingDao)[1]
plt.title(u'青岛')
for i in range(n):
if 0.0 == pointClusNumQingDao.item(i):
plt.plot(norMatQingDao[i,0],norMatQingDao[i,1],'g^')
elif 1.0 == pointClusNumQingDao.item(i):
plt.plot(norMatQingDao[i,0],norMatQingDao[i,1],'b*')
elif 2.0 == pointClusNumQingDao.item(i):
plt.plot(norMatQingDao[i,0],norMatQingDao[i,1],'k<')
elif 3.0 == pointClusNumQingDao.item(i):
plt.plot(norMatQingDao[i,0],norMatQingDao[i,1],'ms')
plt.plot(centroidsQingDao[:,0],centroidsQingDao[:,1],'ro')
plt.axis([0,4.0,0,12])
plt.xticks([0,1,2,3])
# plt.xlabel(u'Distance Index')
# plt.ylabel(u'Activity Degree Index')
# SanYa
plt.subplot(324)
pointClusNumSanYa = clusterAssmentSanYa[:,0].A.T
n = np.shape(pointClusNumSanYa)[1]
plt.title(u'三亚')
for i in range(n):
if 0.0 == pointClusNumSanYa.item(i):
plt.plot(norMatSanYa[i,0],norMatSanYa[i,1],'g^')
elif 1.0 == pointClusNumSanYa.item(i):
plt.plot(norMatSanYa[i,0],norMatSanYa[i,1],'b*')
elif 2.0 == pointClusNumSanYa.item(i):
plt.plot(norMatSanYa[i,0],norMatSanYa[i,1],'k<')
elif 3.0 == pointClusNumSanYa.item(i):
plt.plot(norMatSanYa[i,0],norMatSanYa[i,1],'ms')
plt.plot(centroidsSanYa[:,0],centroidsSanYa[:,1],'ro')
plt.axis([0,4.0,0,12])
plt.xticks([0,1,2,3])
# plt.xlabel(u'Distance Index')
# plt.ylabel(u'Activity Index')
# JiuZhaigou
plt.subplot(325)
pointClusNumJiuZhaigou = clusterAssmentJiuZhaigou[:,0].A.T
n = np.shape(pointClusNumJiuZhaigou)[1]
plt.title(u'九寨沟')
for i in range(n):
if 0.0 == pointClusNumJiuZhaigou.item(i):
plt.plot(norMatJiuZhaigou[i,0],norMatJiuZhaigou[i,1],'g^')
elif 1.0 == pointClusNumJiuZhaigou.item(i):
plt.plot(norMatJiuZhaigou[i,0],norMatJiuZhaigou[i,1],'b*')
elif 2.0 == pointClusNumJiuZhaigou.item(i):
plt.plot(norMatJiuZhaigou[i,0],norMatJiuZhaigou[i,1],'k<')
elif 3.0 == pointClusNumJiuZhaigou.item(i):
plt.plot(norMatJiuZhaigou[i,0],norMatJiuZhaigou[i,1],'ms')
plt.plot(centroidsJiuZhaigou[:,0],centroidsJiuZhaigou[:,1],'ro')
plt.axis([0,4.0,0,12])
plt.xticks([0,1,2,3])
plt.xlabel(u'Distance Index')
plt.ylabel(u'Activity Index')
# TaiShan
plt.subplot(326)
pointClusNumTaiShan = clusterAssmentTaiShan[:,0].A.T
n = np.shape(pointClusNumTaiShan)[1]
plt.title(u'泰山')
for i in range(n):
if 0.0 == pointClusNumTaiShan.item(i):
plt.plot(norMatTaiShan[i,0],norMatTaiShan[i,1],'g^')
elif 1.0 == pointClusNumTaiShan.item(i):
plt.plot(norMatTaiShan[i,0],norMatTaiShan[i,1],'b*')
elif 2.0 == pointClusNumTaiShan.item(i):
plt.plot(norMatTaiShan[i,0],norMatTaiShan[i,1],'k<')
elif 3.0 == pointClusNumTaiShan.item(i):
plt.plot(norMatTaiShan[i,0],norMatTaiShan[i,1],'ms')
plt.plot(centroidsTaiShan[:,0],centroidsTaiShan[:,1],'ro')
plt.axis([0,4.0,0,12])
plt.xticks([0,1,2,3])
plt.xlabel(u'Distance Index')
plt.ylabel(u'Activity Index')
plt.show()
st = time.time()
my_ima = imread('city.jpg')
fig0 = plt.figure()
ax0 = fig0.add_subplot(111)
imshow(my_ima)
lab = color.rgb2lab(my_ima)
ab = double(lab[:, :, 1:3])
nrows = ab.shape[0]
ncols = ab.shape[1]
X = ab.reshape(nrows * ncols, 2)
fig1 = plt.figure()
ax1 = fig1.add_subplot(111)
ax1.scatter(X[:, 0], X[:, 1])
k = 10
centList, clusterAssment = kMeans.biKmeans(X, k)
fig2 = plt.figure()
ax2 = fig2.add_subplot(111)
centroids = array(centList)
datMat = array(X)
colorList = ['b', 'c', 'g', 'k', 'r', 'y', 'm', 'w']
makerList = ['.', '^', '*', '+', 'o']
for i in range(k):
ax2.scatter(datMat[kMeans.find_all_index(clusterAssment[:, 0], i), 0],
datMat[kMeans.find_all_index(clusterAssment[:, 0], i), 1],
c=colorList[i % 8],
marker=makerList[3])
ax2.scatter(centroids[:, 0],
centroids[:, 1],
marker=makerList[4],
c=colorList[4])
# kmeansTest.py
import kMeans
from numpy import *
# dataMat = mat(kMeans.loadDataSet('testSet.txt'))
'''
print('min(dataMat[:, 0]) = ', min(dataMat[:, 0]))
print('min(dataMat[:, 1]) = ', min(dataMat[:, 1]))
print('max(dataMat[:, 0]) = ', max(dataMat[:, 0]))
print('max(dataMat[:, 1]) = ', max(dataMat[:, 1]))
print('randCent of dataset : ', kMeans.randCent(dataMat, 2))
print('distance of eclud : ', kMeans.distEclud(dataMat[0], dataMat[1]))
'''
# myCentroids, clusterAssing = kMeans.kMeans(dataMat, 4)
# print('myCentroids : ', myCentroids)
# print('clusterAssing : ', clusterAssing)
dataMat3 = mat(kMeans.loadDataSet('testSet2.txt'))
centList, newAssments = kMeans.biKmeans(dataMat3, 3)
print('centList = ', centList)
min(datMat[:,0])
max(datMat[:,0])
min(datMat[:,1])
max(datMat[:,1])
kMeans.randCent(datMat,2) # 看一下初始化的质心是否在取值范围内
kMeans.distEclud(datMat[0],datMat[1])
# 在实际数据上看下K-means
reload(kMeans)
datMat = np.mat(kMeans.loadDataSet('testSet.txt'))
myCentroids,clustAssing = kMeans.kMeans(datMat,4) # 不一定是全局最优解
# 二分k-means
reload(kMeans)
datMat3 = np.mat(kMeans.loadDataSet('testSet2.txt'))
centList,myNewAssments = kMeans.biKmeans(datMat3,3) # 其实依然无法保证全局最优解，只能是局部最优解
centList
myNewAssments
# 利用二分k-means在图上画出簇
reload(kMeans)
kMeans.clusterClubs(4)
def test3():
dataMat = np.mat(kMeans.loadDataSet('testSet.txt'))
kMeans.biKmeans(dataMat, 4)