/
cluster_kl.py
81 lines (64 loc) · 1.73 KB
/
cluster_kl.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
import numpy as np
import json
import torch
import matplotlib.pyplot as plt
fc_mask = np.load("/data/mxy/Clustering/LanguageSpecify/rebuttal_measure/KL_v2_nonorm.npy")
lan = [
'bg',
'sr',
'hr',
'uk',
'sk',
# 'mk',
# 'sl',
# 'bs',
'be',
'fa',
'hi',
# 'mr',
'bn',
'id',
'ms',
'kk',
'tr',
'de',
'sv',
'ja',
'ko'
]
for k in range(len(lan)):
print("============================")
sort_dict = {}
l = lan[k]
for mask, key in zip(fc_mask[k],lan):
if key == l :
continue
sort_dict[key] = np.abs(round(mask,5))
# sort_dict[key] = mask
sort_dict = sorted(sort_dict.items(),key=lambda d:d[1])
# sort_dict = sorted(sort_dict.items(), key=lambda d:d[1])
waitinglist = []
name = []
for key,v in sort_dict:
waitinglist.append(v)
name.append(key)
best = 1
gap = waitinglist[1] - waitinglist[0]
for i in range(2,10):
if waitinglist[i] - waitinglist[i - 1] == gap:
best = i
gap = gap /2
elif waitinglist[i] - waitinglist[i - 1] < gap:
best = i
gap = waitinglist[i] - waitinglist[i - 1]
else:
break
idx = 0
coorp = []
for k,v in sort_dict:
coorp.append(k)
idx +=1
if idx > best:
break
print(f"target langauge {l}: auxiliary languages {coorp}")
print("============================")