-
Notifications
You must be signed in to change notification settings - Fork 0
/
utils.py
141 lines (115 loc) · 4.28 KB
/
utils.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
import numpy as np
from sklearn.mixture import GaussianMixture
from sklearn.neighbors import KernelDensity
def JSV_Gaussian_u(Fea, len_s, n_components):
"""compute one v-div with GMM value"""
X = Fea[0:len_s, :] # fetch the sample 1
Y = Fea[len_s:, :] # fetch the sample 2
ind = np.random.choice(Fea.shape[0], len_s, replace=False)
XY = Fea[ind]
jsv = L_Gaussian(0,Fea,X,Y,n_components)
return jsv
def JSV_Gaussian(Fea, N_per, N1, alpha, n_components):
"""run permutation test with v-div with GMM"""
jsv_vector = np.zeros(N_per)
jsv_value = JSV_Gaussian_u(Fea, N1, n_components)
count = 0
nxy = Fea.shape[0]
nx = N1
for r in range(N_per):
# print r
ind = np.random.choice(nxy, nxy, replace=False)
# divide into new X, Y
indx = ind[:nx]
# print(indx)
indy = ind[nx:]
Kx = Fea[indx]
# print(Kx)
Ky = Fea[indy]
indxy = np.concatenate((indx[:int(nx/2)], indy[:int(nx/2)]))
Kxy = Fea[indxy]
jsv_r = L_Gaussian(r + 1,Fea,Kx,Ky,n_components)
jsv_vector[r] = jsv_r
if jsv_vector[r] > jsv_value:
count = count + 1
if count > np.ceil(N_per * alpha):
h = 0
threshold = "NaN"
break
else:
h = 1
if h == 1:
S_jsv_vector = np.sort(jsv_vector)
threshold = S_jsv_vector[int(np.ceil(N_per * (1 - alpha)))]
return h, threshold, jsv_value
def JSV_KDE_u(Fea, len_s, bandwidth):
"""compute value of deep-kernel MMD and std of deep-kernel MMD using merged data."""
X = Fea[0:len_s] # fetch the sample 1
Y = Fea[len_s:] # fetch the sample 2
ind = np.random.choice(Fea.shape[0], len_s, replace=False)
XY = Fea[ind]
model_x = KernelDensity(bandwidth=bandwidth[0])
model_y = KernelDensity(bandwidth=bandwidth[1])
model_xy = KernelDensity(bandwidth=bandwidth[2])
model_x.fit(X)
model_y.fit(Y)
model_xy.fit(XY)
jsv = L_KDE(0,Fea,X,Y,bandwidth)
return jsv
def JSV_KDE(Fea, N_per, N1, alpha, bandwidth):
"""run two-sample test (TST) using ordinary Gaussian kernel."""
jsv_vector = np.zeros(N_per)
jsv_value = JSV_KDE_u(Fea, N1, bandwidth)
count = 0
nxy = Fea.shape[0]
nx = N1
for r in range(N_per):
# print r
ind = np.random.choice(nxy, nxy, replace=False)
# divide into new X, Y
indx = ind[:nx]
# print(indx)
indy = ind[nx:]
Kx = Fea[indx]
# print(Kx)
Ky = Fea[indy]
indxy = np.concatenate((indx[:int(nx/2)], indy[:int(nx/2)]))
Kxy = Fea[indxy]
model_x = KernelDensity(bandwidth=bandwidth[0])
model_y = KernelDensity(bandwidth=bandwidth[1])
model_xy = KernelDensity(bandwidth=bandwidth[2])
model_x.fit(Kx)
model_y.fit(Ky)
model_xy.fit(Kxy)
jsv_r = L_KDE(r + 1,Fea,Kx,Ky,bandwidth)
jsv_vector[r] = jsv_r
S_jsv_vector = np.sort(jsv_vector)
threshold = S_jsv_vector[int(np.ceil(N_per * (1 - alpha)))]
h = 0
if jsv_value > threshold:
h = 1
return h, threshold, jsv_value
def L_KDE(r,Fea,x,y,bandwidth):
model_x = KernelDensity(bandwidth=bandwidth[0])
model_y = KernelDensity(bandwidth=bandwidth[1])
model_xy = KernelDensity(bandwidth=bandwidth[2])
model_x.fit(x)
model_y.fit(y)
model_xy.fit(Fea)
mixed = 1/2*np.mean(-model_xy.score_samples(x))+1/2*np.mean(-model_xy.score_samples(y))
x_prob = np.mean(-model_y.score_samples(x))
y_prob = np.mean(-model_x.score_samples(y))
gap = abs(x_prob - mixed) + abs(y_prob - mixed)
return abs(gap)
def L_Gaussian(r,Fea,x,y, n_components):
model_x = GaussianMixture(n_components=n_components)
model_y = GaussianMixture(n_components=n_components)
model_xy = GaussianMixture(n_components=n_components)
model_x.fit(x)
model_y.fit(y)
model_xy.fit(Fea)
mixed = 1/2*np.mean(-model_xy.score_samples(x))+1/2*np.mean(-model_xy.score_samples(y))
x_prob = np.mean(-model_y.score_samples(x))
y_prob = np.mean(-model_x.score_samples(y))
gap = abs(x_prob - mixed) + abs(y_prob - mixed)
return abs(gap)