-
Notifications
You must be signed in to change notification settings - Fork 19
/
imsat_hash.py
268 lines (198 loc) · 8.74 KB
/
imsat_hash.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
import argparse, sys
import numpy as np
import chainer
import chainer.functions as F
from chainer import FunctionSet, Variable, optimizers, cuda, serializers
from sklearn import metrics
parser = argparse.ArgumentParser()
parser.add_argument('--gpu', type=int, help='which gpu device to use', default=0)
parser.add_argument('--lam', type=float, help='trade-off parameter for mutual information and smooth regularization',
default=0.1)
parser.add_argument('--prop_eps', type=float, help='epsilon', default=0.25)
parser.add_argument('--n_bit', type=int, help='number of bits', default=16)
parser.add_argument('--hidden_list', type=str, help='hidden size list', default='400-400')
parser.add_argument('--seed', type=int, help='seed for random variable', default=0)
parser.add_argument('--dataset', type=str, default='mnist')
args = parser.parse_args()
lam = args.lam
n_bit = args.n_bit
N_query = 1000
args = parser.parse_args()
if args.dataset == 'mnist':
sys.path.append('mnist')
from load_mnist import *
whole = load_mnist_whole(PATH='mnist/', scale=1.0 / 128.0, shift=-1.0)
else:
print 'The dataset is not supported.'
raise NotImplementedError
n_class = np.max(whole.label) + 1
print n_class
dim = whole.data.shape[1]
data = whole.data
target = whole.label
np.random.seed(args.seed)
perm = np.random.permutation(len(target))
cnt_query = [0] * 10
ind_query = []
ind_gallary = []
for i in range(len(target)):
l = target[perm[i]]
if cnt_query[l] < 100:
ind_query.append(perm[i])
cnt_query[l] += 1
else:
ind_gallary.append(perm[i])
x_query = data[ind_query]
x_gallary = data[ind_gallary]
y_query = target[ind_query]
y_gallary = target[ind_gallary]
print x_query.shape
print x_gallary.shape
query = Data(x_query, y_query)
gallary = Data(x_gallary, y_gallary)
print 'use gpu'
chainer.cuda.get_device(args.gpu).use()
print 'query data: ' + str(N_query)
xp = cuda.cupy
hidden_list = map(int, args.hidden_list.split('-'))
def call_bn(bn, x, test=False, update_batch_stats=True):
if not update_batch_stats:
return F.batch_normalization(x, bn.gamma, bn.beta, use_cudnn=False)
if test:
return F.fixed_batch_normalization(x, bn.gamma, bn.beta, bn.avg_mean, bn.avg_var, use_cudnn=False)
else:
return bn(x)
def distance(y0, y1):
p0 = F.sigmoid(y0)
p1 = F.sigmoid(y1)
return F.sum(p0 * F.log((p0 + 1e-8) / (p1 + 1e-8)) + (1 - p0) * F.log((1 - p0 + 1e-8) / (1 - p1 + 1e-8))) / \
p0.data.shape[0]
def vat(forward, distance, x, eps_list, xi=10, Ip=1):
y = forward(Variable(x))
y.unchain_backward()
# calc adversarial direction
d = xp.random.normal(size=x.shape, dtype=np.float32)
d = d / xp.sqrt(xp.sum(d ** 2, axis=1)).reshape((x.shape[0], 1))
for ip in range(Ip):
d_var = Variable(d.astype(np.float32))
y2 = forward(x + xi * d_var)
kl_loss = distance(y, y2)
kl_loss.backward()
d = d_var.grad
d = d / xp.sqrt(xp.sum(d ** 2, axis=1)).reshape((x.shape[0], 1))
d_var = Variable(d.astype(np.float32))
eps = args.prop_eps * eps_list
y2 = forward(x + F.transpose(eps * F.transpose(d_var)))
return distance(y, y2)
class Encoder(chainer.Chain):
def __init__(self):
super(Encoder, self).__init__(
l1=F.Linear(dim, hidden_list[0], wscale=0.1),
l2=F.Linear(hidden_list[0], hidden_list[1], wscale=0.1),
l3=F.Linear(hidden_list[1], n_bit, wscale=0.0001),
bn1=F.BatchNormalization(hidden_list[0]),
bn2=F.BatchNormalization(hidden_list[1])
)
def __call__(self, x, test=False, update_batch_stats=True):
h = F.relu(call_bn(self.bn1, self.l1(x), test=test, update_batch_stats=update_batch_stats))
h = F.relu(call_bn(self.bn2, self.l2(h), test=test, update_batch_stats=update_batch_stats))
y = self.l3(h)
return y
def enc_aux_noubs(x):
return enc(x, test=False, update_batch_stats=False)
def enc_test(x):
return enc(x, test=True)
def loss_unlabeled(x, eps_list):
L = vat(enc_aux_noubs, distance, x.data, eps_list)
return L
def loss_test(x_query, t_query, x_gallary, t_gallary):
query_hash = F.sigmoid(enc(x_query, test=True)).data > 0.5
gallary_hash = F.sigmoid(enc(x_gallary, test=True)).data > 0.5
t_query = cuda.to_cpu(t_query.data)
t_gallary = cuda.to_cpu(t_gallary.data)
withinN_precision_label = 0
withinR_precision_label = 0
mAP = 0
for i in range(N_query):
hamming_distance = cuda.to_cpu(xp.sum((1 - query_hash[i]) == gallary_hash, axis=1))
mAP += metrics.average_precision_score(t_gallary == t_query[i], 1. / (1 + hamming_distance))
nearestN_index = np.argsort(hamming_distance)[:500]
withinN_precision_label += float(np.sum(t_gallary[nearestN_index] == t_query[i])) / 500
withinR_label = t_gallary[hamming_distance < 3]
num_withinR = len(withinR_label)
if not num_withinR == 0:
withinR_precision_label += np.sum(withinR_label == t_query[i]) / float(num_withinR)
return mAP / N_query, withinN_precision_label / N_query, withinR_precision_label / N_query
def loss_information(enc, x):
p_logit = enc(x)
p = F.sigmoid(p_logit)
p_ave = F.sum(p, axis=0) / x.data.shape[0]
cond_ent = F.sum(- p * F.log(p + 1e-8) - (1 - p) * F.log(1 - p + 1e-8)) / p.data.shape[0]
marg_ent = F.sum(- p_ave * F.log(p_ave + 1e-8) - (1 - p_ave) * F.log(1 - p_ave + 1e-8))
p_ave = F.reshape(p_ave, (1, len(p_ave.data)))
p_ave_separated = F.separate(p_ave, axis=1)
p_separated = F.separate(F.expand_dims(p, axis=2), axis=1)
p_ave_list_i = []
p_ave_list_j = []
p_list_i = []
p_list_j = []
for i in range(n_bit - 1):
p_ave_list_i.extend(list(p_ave_separated[i + 1:]))
p_list_i.extend(list(p_separated[i + 1:]))
p_ave_list_j.extend([p_ave_separated[i] for n in range(n_bit - i - 1)])
p_list_j.extend([p_separated[i] for n in range(n_bit - i - 1)])
p_ave_pair_i = F.expand_dims(F.concat(tuple(p_ave_list_i), axis=0), axis=1)
p_ave_pair_j = F.expand_dims(F.concat(tuple(p_ave_list_j), axis=0), axis=1)
p_pair_i = F.expand_dims(F.concat(tuple(p_list_i), axis=1), axis=2)
p_pair_j = F.expand_dims(F.concat(tuple(p_list_j), axis=1), axis=2)
p_pair_stacked_i = F.concat((p_pair_i, 1 - p_pair_i, p_pair_i, 1 - p_pair_i), axis=2)
p_pair_stacked_j = F.concat((p_pair_j, p_pair_j, 1 - p_pair_j, 1 - p_pair_j), axis=2)
p_ave_pair_stacked_i = F.concat((p_ave_pair_i, 1 - p_ave_pair_i, p_ave_pair_i, 1 - p_ave_pair_i), axis=1)
p_ave_pair_stacked_j = F.concat((p_ave_pair_j, p_ave_pair_j, 1 - p_ave_pair_j, 1 - p_ave_pair_j), axis=1)
p_product = F.sum(p_pair_stacked_i * p_pair_stacked_j, axis=0) / len(p.data)
p_ave_product = p_ave_pair_stacked_i * p_ave_pair_stacked_j
pairwise_mi = 2 * F.sum(p_product * F.log((p_product + 1e-8) / (p_ave_product + 1e-8)))
return cond_ent, marg_ent, pairwise_mi
enc = Encoder()
enc.to_gpu()
o_enc = optimizers.Adam(alpha=0.002, beta1=0.9)
o_enc.setup(enc)
batchsize = 250
N_gallary = len(gallary.data)
nearest_dist = np.loadtxt(args.dataset + '/10th_neighbor.txt').astype(np.float32)
x_query, t_query = cuda.to_gpu(query.data), cuda.to_gpu(query.label)
x_gallary, t_gallary = cuda.to_gpu(gallary.data), cuda.to_gpu(gallary.label)
n_epoch = 50
for epoch in range(n_epoch):
print epoch
sum_cond_ent = 0
sum_marg_ent = 0
sum_pairwise_mi = 0
sum_vat = 0
for it in range(N_gallary / batchsize):
x, _, ind = whole.get(batchsize, need_index=True)
cond_ent, marg_ent, pairwise_mi = loss_information(enc, Variable(x))
sum_cond_ent += cond_ent.data
sum_marg_ent += marg_ent.data
sum_pairwise_mi += pairwise_mi.data
loss_info = cond_ent - marg_ent + pairwise_mi
loss_ul = loss_unlabeled(Variable(x), cuda.to_gpu(nearest_dist[ind]))
sum_vat += loss_ul.data
o_enc.zero_grads()
(loss_ul + lam * loss_info).backward()
o_enc.update()
loss_ul.unchain_backward()
loss_info.unchain_backward()
condent = sum_cond_ent / (N_gallary / batchsize)
margent = sum_marg_ent / (N_gallary / batchsize)
pairwise = sum_pairwise_mi / (N_gallary / batchsize)
print 'conditional entropy: ' + str(condent)
print 'marginal entropy: ' + str(margent)
print 'pairwise mi: ' + str(pairwise)
print 'vat loss: ' + str(sum_vat / (N_gallary / batchsize))
sys.stdout.flush()
mAP, withNpreclabel, withRpreclabel = loss_test(Variable(x_query, volatile=True), Variable(t_query, volatile=True),
Variable(x_gallary, volatile=True), Variable(t_gallary, volatile=True))
print 'mAP: ', mAP
print 'withNpreclabel: ', withNpreclabel
print 'withRpreclabel: ', withRpreclabel