/
Inter-cluster distance measures.py
350 lines (304 loc) · 17.5 KB
/
Inter-cluster distance measures.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
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
import pandas as pd
import numpy as np
import matplotlib.pylab as plt
import time
import copy
import matplotlib.ticker as ticker
from sklearn import metrics
from sklearn.neighbors import KDTree
from sklearn.cluster import AgglomerativeClustering
from scipy.optimize import linear_sum_assignment as linear_assignment
from NaN import nan_2
import math
def findKNN1(inst, r, tree):
_, ind = tree.query([inst], r)
return _[0]
def acc(y_true, y_pred):
y_true = y_true.astype(np.int64)
assert y_pred.size == y_true.size
D = max(int(y_pred.max()), int(y_true.max())) + 1
w = np.zeros((D, D), dtype=np.int64)
for i in range(y_pred.size):
w[y_pred[i], y_true[i]] += 1
ind = linear_assignment(w.max() - w)
return sum([w[i, j] for i, j in zip(ind[0],ind[1])]) * 1.0 / y_pred.size
if __name__ == "__main__":
plt.figure(figsize=(48, 40))
font = {'family': 'Times New Roman', 'weight': 'bold', 'size': 57}
font1 = {'family': 'Times New Roman', 'weight': 'bold', 'size': 60}
hh = 56
dd = 400
index=1
flag=1
yflag=1
ccc=1
zz = 1
for jj in range(0, 5):
txtDF0 = pd.read_csv(r'E:\Datasets\Gaussian_4_original{s1}.csv'.format(s1=jj))
txtDF0.drop_duplicates(inplace=True)
txtDF0.to_csv(r'E:\Datasets\Gaussian_4_original{s1}.csv'.format(s1=jj),index=False)
txtDF1 = pd.read_csv(r'E:\Datasets\Gaussian_4_original{s1}.csv'.format(s1=jj))
col = txtDF1.shape[1]
final_true_data1 = txtDF1.iloc[:, 0:col]
df_data1 = txtDF1.iloc[:, 0:col]
ar_data1 = np.array(df_data1)
nat_neighbor1 = nan_2.Natural_Neighbor()
nat_neighbor1.load(r'E:\Datasets\Gaussian_4_original{s1}_1.csv'.format(s1=jj)) # 改2
nane1, temp_density1 = nat_neighbor1.algorithm()
final_density1 = []
for i in temp_density1:
final_density1.append(i / nane1)
df_density1 = pd.DataFrame(final_density1, columns=['Density'])
D_data_all1 = pd.concat([txtDF1, df_density1], axis=1)
D_data_all1_sort = D_data_all1.sort_values('Density', ascending=False)
D_data_all1_sort.to_csv(r'E:\Datasets\density+T_Gaussian_4_original{s1}_sorted.csv'.format(s1=jj),index=False)
para_list_noise_1 = [0.95]
para_list_center_1 = [0.05]
D_data_all1_sort1 = pd.read_csv(r'E:\Datasets\density+T_Gaussian_4_original{s1}_sorted.csv'.format(s1=jj))
NC1 = 4
for cpor1 in para_list_center_1:
x1 = "%.f%%" % (cpor1 * 100)
c_count1 = math.ceil(len(D_data_all1_sort1) * cpor1)
fc_count1 = np.max([c_count1, NC1])
c_sample1 = D_data_all1_sort1.head(fc_count1)
df_center1 = c_sample1.iloc[:, 0:(D_data_all1_sort1.shape[1]-1)]
df_carr1 = df_center1.values
df_cli1 = df_carr1.tolist()
remain_sample_count1 = len(D_data_all1_sort1) - fc_count1
r_sample1 = D_data_all1_sort1.loc[fc_count1:(len(D_data_all1_sort1) - 1)]
df_noise1 = r_sample1.iloc[:, 0:(D_data_all1_sort1.shape[1]-1)].values.tolist()
estimator1 = AgglomerativeClustering(n_clusters=NC1, linkage='single')
estimator1.fit(df_carr1)
Center_Clusters1 = [[] for i in range(NC1)]
for c_nc1 in range(NC1):
for cpoint1, cid1 in zip(df_cli1, estimator1.labels_):
if c_nc1 == cid1:
Center_Clusters1[c_nc1].append(cpoint1)
else:
continue
if len(df_noise1) > 0:
Singel_centers = copy.deepcopy(Center_Clusters1)
Singel_noise = copy.deepcopy(df_noise1)
start_1 = time.perf_counter()
for single_noise_point in Singel_noise:
single_distance = []
for already_clu in Singel_centers:
single_tree = KDTree(np.array(already_clu))
single_distance.append(findKNN1(np.array(single_noise_point), 1, single_tree)[0])
Singel_centers[single_distance.index(min(single_distance))].append(single_noise_point)
end_1 = time.perf_counter()
print("Runtime：",end_1 - start_1)
final_single_results=[]
for label_sin in range(4):
for ele_sin in Singel_centers[label_sin]:
ele_sin.append(label_sin)
final_single_results.append(ele_sin)
Complete_centers = copy.deepcopy(Center_Clusters1)
Complete_noise = copy.deepcopy(df_noise1)
start_2 = time.perf_counter()
for complete_noise_point in Complete_noise:
complete_distance = []
for already_clu_c in Complete_centers:
complete_tree = KDTree(np.array(already_clu_c))
complete_distance.append(findKNN1(np.array(complete_noise_point), len(already_clu_c), complete_tree)[len(already_clu_c)-1])
Complete_centers[complete_distance.index(min(complete_distance))].append(complete_noise_point)
end_2 = time.perf_counter()
print("Runtime：", end_2 - start_2)
final_complete_results = []
for label_com in range(4):
for ele_com in Complete_centers[label_com]:
ele_com.append(label_com)
final_complete_results.append(ele_com)
Average_centers = copy.deepcopy(Center_Clusters1)
Average_noise = copy.deepcopy(df_noise1)
start_3 = time.perf_counter()
for Average_noise_point in Average_noise:
Average_distance = []
for already_clu_a in Average_centers:
Average_tree = KDTree(np.array(already_clu_a))
Average_distance.append(np.mean(findKNN1(np.array(Average_noise_point), len(already_clu_a), Average_tree)))
Average_centers[Average_distance.index(min(Average_distance))].append(Average_noise_point)
end_3 = time.perf_counter()
print("Runtime：", end_3 - start_3)
final_average_results = []
for label_avg in range(4):
for ele_avg in Average_centers[label_avg]:
ele_avg.append(label_avg)
final_average_results.append(ele_avg)
Centroid_centers = copy.deepcopy(Center_Clusters1)
Centroid_noise = copy.deepcopy(df_noise1)
start_4 = time.perf_counter()
for Centroid_noise_point in Centroid_noise:
Centroid_distance = []
for already_clu_Cen in Centroid_centers:
centro_sum = np.array([0, 0])
for centro in range(len(already_clu_Cen)):
centro_sum = centro_sum + np.array(already_clu_Cen[centro])
cluster_cen=centro_sum / len(already_clu_Cen)
Centroid_distance.append(np.linalg.norm(cluster_cen-np.array(Centroid_noise_point)))
Centroid_centers[Centroid_distance.index(min(Centroid_distance))].append(Centroid_noise_point)
end_4 = time.perf_counter()
print("Runtime：", end_4 - start_4)
final_Centroid_results = []
for label_Centroid in range(4):
for ele_Centroid in Centroid_centers[label_Centroid]:
ele_Centroid.append(label_Centroid)
final_Centroid_results.append(ele_Centroid)
# -------------------------------------Ward linkage------------------------
Ward_centers = copy.deepcopy(Center_Clusters1)
Ward_noise = copy.deepcopy(df_noise1)
start_5 = time.perf_counter()
for Ward_noise_point in Ward_noise:
Ward_distance = []
for already_clu_war in Ward_centers:
war_sum = np.array([0, 0])
for wartro in range(len(already_clu_war)):
war_sum= war_sum + np.array(already_clu_war[wartro])
cluster_war = war_sum / len(already_clu_war)
Ward_distance.append((np.linalg.norm(cluster_war - np.array(Ward_noise_point))*len(already_clu_war))/(len(already_clu_war)+1))
Ward_centers[Ward_distance.index(min(Ward_distance))].append(Ward_noise_point)
final_Ward_results = []
end_5 = time.perf_counter()
print("Runtime：", end_5 - start_5)
for label_Ward in range(4):
for ele_Ward in Ward_centers[label_Ward]:
ele_Ward.append(label_Ward)
final_Ward_results.append(ele_Ward)
noise_class1 = []
NN_centers = copy.deepcopy(Center_Clusters1)
NN_noise = copy.deepcopy(df_noise1)
start_6 = time.perf_counter()
for noise_point1 in NN_noise:
dist_naverage1 = []
for n_cluster1 in NN_centers:
n_tree1 = KDTree(np.array(n_cluster1))
min_nnane1 = np.min([nane1, len(n_cluster1)])
dist_naverage1.append(np.mean(findKNN1(np.array(noise_point1), min_nnane1, n_tree1)))
temp_nid1 = dist_naverage1.index(min(dist_naverage1))
noise_class1.append(temp_nid1)
NN_centers[temp_nid1].append(noise_point1)
end_6 = time.perf_counter()
print("Runtime：", end_6 - start_6)
final_NN_results = []
for label_NN in range(4):
for ele_NN in NN_centers[label_NN]:
ele_NN.append(label_NN)
final_NN_results.append(ele_NN)
# ----------------------------------------------------------------------------------------------
Single_results = pd.DataFrame(np.array(final_single_results))
Complete_results = pd.DataFrame(np.array(final_complete_results))
Average_results = pd.DataFrame(np.array(final_average_results))
Centroid_results = pd.DataFrame(np.array(final_Centroid_results))
Ward_results = pd.DataFrame(np.array(final_Ward_results))
NN_results = pd.DataFrame(np.array(final_NN_results))
Single_data = np.array(Single_results.iloc[:, 0:2]).tolist()
Single_label_1 = np.array(Single_results.iloc[:, 2:3])
Single_label=[]
for single_1 in range(len(Single_data)):
Single_label.append(int(Single_label_1[single_1][0]))
Complete_data = np.array(Complete_results.iloc[:, 0:2]).tolist()
Complete_label_1 = np.array(Complete_results.iloc[:, 2:3])
Complete_label = []
for Complete_1 in range(len(Complete_data)):
Complete_label.append(int(Complete_label_1[Complete_1][0]))
Average_data = np.array(Average_results.iloc[:, 0:2]).tolist()
Average_label_1 = np.array(Average_results.iloc[:, 2:3])
Average_label = []
for Average_1 in range(len(Average_data)):
Average_label.append(int(Average_label_1[Average_1][0]))
Centroid_data = np.array(Centroid_results.iloc[:, 0:2]).tolist()
Centroid_label_1 = np.array(Centroid_results.iloc[:, 2:3])
Centroid_label = []
for Centroid_1 in range(len(Centroid_data)):
Centroid_label.append(int(Centroid_label_1[Centroid_1][0]))
Ward_data = np.array(Ward_results.iloc[:, 0:2]).tolist()
Ward_label_1 = np.array(Ward_results.iloc[:, 2:3])
Ward_label = []
for Ward_1 in range(len(Ward_data)):
Ward_label.append(int(Ward_label_1[Ward_1][0]))
NN_data = np.array(NN_results.iloc[:, 0:2]).tolist()
NN_label_1 = np.array(NN_results.iloc[:, 2:3])
NN_label = []
for NN_1 in range(len(NN_data)):
NN_label.append(int(NN_label_1[NN_1][0]))
plt.subplot(5, 6, index)
center_cnames1 = ['blue', 'red', 'green', "purple"]
for Single_point1, Single_id1 in zip(Single_data, np.array(Single_label)):
plt.scatter(Single_point1[0], Single_point1[1], c=center_cnames1[Single_id1], s=dd, alpha=1)
if flag == 1:
plt.title("NNLD-SL-SL", fontdict=font, bbox=dict(ec='pink', fc='orange'), verticalalignment='bottom')
flag=flag+1
if yflag == 1:
plt.ylabel('Illustrative dataset 1', font1, bbox=dict(ec='orange', fc='w'), verticalalignment='bottom')
yflag = 0
if yflag == 2:
plt.ylabel('Illustrative dataset 2', font1, bbox=dict(ec='orange', fc='w'), verticalalignment='bottom')
yflag = 0
if yflag == 3:
plt.ylabel('Illustrative dataset 3', font1, bbox=dict(ec='orange', fc='w'), verticalalignment='bottom')
yflag = 0
if yflag == 4:
plt.ylabel('Illustrative dataset 4', font1, bbox=dict(ec='orange', fc='w'), verticalalignment='bottom')
yflag = 0
if yflag == 5:
plt.ylabel('Illustrative dataset 5', font1, bbox=dict(ec='orange', fc='w'), verticalalignment='bottom')
yflag = 0
plt.xticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.yticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.tight_layout()
plt.subplot(5, 6, index+1)
center_cnames1 = ['blue', 'red', 'green', "purple"]
for Complete_point1, Complete_id1 in zip(Complete_data, np.array(Complete_label)):
plt.scatter(Complete_point1[0], Complete_point1[1], c=center_cnames1[Complete_id1], s=dd, alpha=1)
if flag == 2:
plt.title("NNLD-SL-CL", fontdict=font, bbox=dict(ec='pink', fc='orange'), verticalalignment='bottom')
flag = flag + 1
plt.xticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.yticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.tight_layout()
plt.subplot(5, 6, index + 2)
center_cnames1 = ['blue', 'red', 'green', "purple"]
for Average_point1, Average_id1 in zip(Average_data, np.array(Average_label)):
plt.scatter(Average_point1[0], Average_point1[1], c=center_cnames1[Average_id1], s=dd, alpha=1)
if flag == 3:
plt.title("NNLD-SL-AL", fontdict=font, bbox=dict(ec='pink', fc='orange'), verticalalignment='bottom')
flag = flag + 1
plt.xticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.yticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.tight_layout()
plt.subplot(5, 6, index + 3)
center_cnames1 = ['blue', 'red', 'green', "purple"]
for Centroid_point1, Centroid_id1 in zip(Centroid_data, np.array(Centroid_label)):
plt.scatter(Centroid_point1[0], Centroid_point1[1], c=center_cnames1[Centroid_id1], s=dd, alpha=1)
if flag == 4:
plt.title("NNLD-SL-CeL", fontdict=font, bbox=dict(ec='pink', fc='orange'), verticalalignment='bottom')
flag = flag + 1
plt.xticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.yticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.tight_layout()
plt.subplot(5, 6, index + 4)
center_cnames1 = ['blue', 'red', 'green', "purple"]
for Ward_point1, Ward_id1 in zip(Ward_data, np.array(Ward_label)):
plt.scatter(Ward_point1[0], Ward_point1[1], c=center_cnames1[Ward_id1], s=dd, alpha=1)
if flag == 5:
plt.title("NNLD-SL-WL", fontdict=font, bbox=dict(ec='pink', fc='orange'), verticalalignment='bottom')
flag = flag + 1
plt.xticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.yticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.tight_layout()
plt.subplot(5, 6, index + 5)
center_cnames1 = ['blue', 'red', 'green', "purple"]
for NN_point1, NN_id1 in zip(NN_data, np.array(NN_label)):
plt.scatter(NN_point1[0], NN_point1[1], c=center_cnames1[NN_id1], s=dd, alpha=1)
if flag == 6:
plt.title("NNLD-SL-NNLICSM", fontdict=font, bbox=dict(ec='pink', fc='orange'), verticalalignment='bottom')
flag = flag + 1
plt.xticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.yticks(fontsize=hh, family='Times New Roman', weight="bold")
plt.tight_layout()
index = index+6
zz = zz + 1
ccc=ccc+1
yflag = ccc
plt.savefig(r'E:\Datasets\ICDM(0.05).jpeg', dpi=600)
plt.show()