-
Notifications
You must be signed in to change notification settings - Fork 5
/
Copy pathDenPeakcode.py
199 lines (179 loc) · 6.57 KB
/
DenPeakcode.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
# -*- coding: utf-8 -*-‘
from time import time
import math
import numpy as np
from sklearn.metrics.pairwise import euclidean_distances
import metrics
from time import time
def load_data(InputFileName):
#-------X and Y in 2 Files----------
x = np.loadtxt(InputFileName+".txt")
y = np.loadtxt(InputFileName+"_y.txt")
# xy = np.concatenate((x,y),axis=1)
#-------X and Y in 1 File-----------
# xy = np.loadtxt(InputFileName + ".txt")
# x = xy[:, 0:xy.shape[1]-1]# get just x without y
# y = xy[:,-1]
# print("x.shape:", x.shape, "y.shape:", y.shape)
return x,y
def ChooseCenter_N_Cluster(dc_percent,X):
#calculate points distance
dis_matrix = euclidean_distances(X, X)
avg_dis = np.average(dis_matrix)
#choose dc
dc = dc_percent * avg_dis
#calculate rho
dis_matrix1 = dis_matrix / dc
dis_matrix1 = np.multiply(dis_matrix1, dis_matrix1)
dis_matrix1 = np.exp(-dis_matrix1)
rho_ = dis_matrix1.sum(axis=1)
rho = rho_ - 1
avg_rho = np.average(rho)
rho_list = [(rho[i], i) for i in range(rho.shape[0])]
rho_sorted = sorted(rho_list, reverse=1)
#calculate delta
delta = [rho_sorted[0][0] for i in range(X.shape[0])]# avg_dis 存疑
nneigh = [-1 for i in range(X.shape[0])]
#如果rho不排序的话，就只能遍历整个矩阵找topK
for ii in range(1, X.shape[0]):
for jj in range(0, ii):
id_p1 = rho_sorted[ii][1] # get point1's id
id_p2 = rho_sorted[jj][1] # get point2's id
if (dis_matrix[id_p1, id_p2] < delta[id_p1]):
delta[id_p1] = dis_matrix[id_p1, id_p2]
nneigh[id_p1] = id_p2
# assignment
cl = [-1 for i in range(X.shape[0])]
center_num = 0
each_rho_set=[]# every cluster's points rho
each_cluster_num=[]# every cluster's points num
centers_idx=[]
centers_label=[]
cluster_set=[]# every cluster state
dis_m=dis_matrix
for i in range(X.shape[0]):
idx = rho_sorted[i][1]
each_cluster = []
if rho[idx] > avg_rho and delta[idx] > dc:
cl[idx] = center_num
each_cluster.append(idx)
cluster_set.append(each_cluster)
centers_idx.append(idx)
centers_label.append(center_num)
each_rho_set.append(rho[idx])
each_cluster_num.append(1)
center_num += 1
else:
if (cl[idx] == -1 and cl[nneigh[idx]] != -1):
cl[idx] = cl[nneigh[idx]]
each_rho_set[cl[idx]] +=rho[idx]
each_cluster_num[cl[idx]] +=1
# for k in range(len(cluster_set[cl[idx]])):
# id=cluster_set[cl[idx]][k]
# dis_m[id][idx]=0
# dis_m[idx][id]=0
cluster_set[cl[idx]].append(idx)
each_avg_rho=[]
for j in range(len(each_rho_set)):
each_avg_rho.append(each_rho_set[j]/each_cluster_num[j]) #every cluster's average rho
print("inital cluster num:", len(cluster_set))
return dc,cl,rho,delta,cluster_set,each_avg_rho,dis_m
def Corepoints_N_Merge(dc,cl,rho,cluster_set,each_avg_rho,dis_m):
core_point=[]
cluster_num=len(cluster_set)
cl_borader=cl.copy()
merged_cluster_num=cluster_num
for i in range(cluster_num):
temp_set=cluster_set[i]
for j in range(len(cluster_set[i])):
id = temp_set[j]
if rho[id] <= each_avg_rho[i]:#boarder point
cl_borader[id]= -1
for k in range(len(cl)):
dis_m[id][k] = 0
dis_m[k][id] = 0
else:
core_point.append(id)
for i in range(len(core_point)):
idx=core_point[i]
for j in range(len(cl)):
if 0.0 <dis_m[idx][j] < dc and cl[idx] != cl[j]:
cl = merge(idx,j,cluster_set,cl)
merged_cluster_num -=1
print("merged cluster num:",merged_cluster_num)
return cl,cl_borader,cluster_num,merged_cluster_num
def merge(i,j,cluster_set,cl):
label1 = cl[i]
label2 = cl[j]
if(cl[i]>cl[j]):
label1=cl[j]
label2=cl[i]
print("merge:",label1,label2)
for k in range(len(cluster_set[label2])):
id = cluster_set[label2][k]
cl[id]=label1
cluster_set[label1].extend(cluster_set[label2])
return cl
def match(y,cl):
cl=np.array(cl)
y=np.array(y)
acc = np.round(metrics.acc(y, cl), 5)
nmi = np.round(metrics.nmi(y, cl), 5)
ari = np.round(metrics.ari(y, cl), 5)
return acc,nmi,ari
def drawOriginGraph(pl, points, cl,cl_boarder, colorNum):
x = points[:,0]
y = points[:,1]
cm = pl.get_cmap("RdYlGn")
for i in range(len(points)):
if (cl_boarder[i] != -1):
pl.plot(x[i], y[i], '.', color=cm(cl[i] * 1.0 / colorNum))
else:
pl.plot(x[i], y[i], '.', color='k')
def drawDecisionGraph(pl, rho, delta, cl, cl_boarder, colorNum):
cm = pl.get_cmap("RdYlGn")
for i in range(len(rho)):
if(cl_boarder[i]!=-1):
pl.plot(rho[i], delta[i], '.', color=cm(cl[i] * 1.0 / colorNum))
else:
pl.plot(rho[i], delta[i], '.', color='k')
pl.xlabel(r'$\rho$')
pl.ylabel(r'$\delta$')
def DenPeakCluster(x):
# InputFileName="flame"
# InputFileName="Compund"
# InputFileName = "D31"
# InputFileName = "Spiral"
# InputFileName = "Jain"
# InputFileName = "R15"
# InputFileName = "usps"
# x,y=load_data(InputFileName)
#--------------------------coefficient：dc---------------------------------(avg_delta*dc_percent)
dc_percent = 0.1
# print("dc_percent:",dc_percent)
t1=time()
dc, cl, rho, delta, cluster_set, each_avg_rho, dis_m=ChooseCenter_N_Cluster(dc_percent,x)
t2=time()
print("ChooseCenter_N_Cluster time:",t2-t1)
# import pylab as pl
# fig1 = pl.figure(1)
# pl.subplot(321)
# drawOriginGraph(pl, x, cl, cl,len(cluster_set))
# pl.subplot(322)
# drawDecisionGraph(pl, rho, delta, cl, cl,len(cluster_set))
cl,cl_boarder,cluster_num,merged_cluster_num = Corepoints_N_Merge(dc,cl,rho,cluster_set,each_avg_rho,dis_m)
t3=time()
print("Corepoints_N_Merge time:",t3-t2)
# pl.subplot(323)
# drawOriginGraph(pl, x, cl, cl, cluster_num)
# pl.subplot(324)
# drawDecisionGraph(pl, rho, delta, cl, cl, cluster_num)
#
# pl.subplot(325)
# drawOriginGraph(pl, x, cl, cl_boarder, cluster_num)
# pl.subplot(326)
# drawDecisionGraph(pl, rho, delta, cl, cl_boarder, cluster_num)
# pl.savefig('./'+InputFileName+'.png')
# pl.show()
cl=np.array(cl)
return cl,cl_boarder,merged_cluster_num,dc_percent,dc