-
Notifications
You must be signed in to change notification settings - Fork 16
/
DistributionShifting.py
205 lines (179 loc) · 6.99 KB
/
DistributionShifting.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
from tensorboardX import SummaryWriter
import numpy as np
import matplotlib.pyplot as plt
from datetime import datetime, timedelta
import os
from torch.utils.data import Dataset, DataLoader
from skimage.transform import resize
import torch
import torch.nn as nn
import torch.optim as optim
import torch.utils.data
import torch.nn.functional as F
import torchvision.utils as vutils
from torch.autograd import Variable
from torch import LongTensor, FloatTensor
from torchvision.utils import save_image
import sys
gpu_id = 0
class NWSDataset(Dataset):
"""
NWS Dataset
"""
def __init__(
self, fake='data/fake.pt', c=0.75, i=1, n=2557
):
val = int(n * (c ** i))
self.real = torch.load('data/real.pt').cuda(gpu_id)
self.real.requires_grad = False
self.fake = torch.load(fake).cuda(gpu_id)
self.fake.requires_grad = False
self.realdata = torch.cat([self.real[:val], self.fake[:-1 * val]], 0)
def __len__(self):
return self.realdata.shape[0]
def __getitem__(self, item):
return self.realdata[item]
def weights_init_normal(m):
classname = m.__class__.__name__
if classname.find("Conv") != -1:
torch.nn.init.normal_(m.weight.data, 0.0, 0.02)
def convTBNReLU(in_channels, out_channels, kernel_size=4, stride=2, padding=1):
return nn.Sequential(
nn.ConvTranspose2d(
in_channels,
out_channels,
kernel_size=kernel_size,
stride=stride,
padding=padding,
),
nn.InstanceNorm2d(out_channels),
nn.LeakyReLU(0.2, True),
)
def convBNReLU(in_channels, out_channels, kernel_size=4, stride=2, padding=1):
return nn.Sequential(
nn.Conv2d(
in_channels,
out_channels,
kernel_size=kernel_size,
stride=stride,
padding=padding,
),
nn.InstanceNorm2d(out_channels),
nn.LeakyReLU(0.2, True),
)
class Generator(nn.Module):
def __init__(self, in_channels, out_channels):
super(Generator, self).__init__()
self.in_channels = in_channels
self.out_channels = out_channels
self.block1 = convTBNReLU(in_channels, 512, 4, 1, 0)
self.block2 = convTBNReLU(512, 256)
self.block3 = convTBNReLU(256, 128)
self.block4 = convTBNReLU(128, 64)
self.block5 = nn.ConvTranspose2d(64, out_channels, 4, 2, 1)
def forward(self, noise):
out = self.block1(noise)
out = self.block2(out)
out = self.block3(out)
out = self.block4(out)
return torch.tanh(self.block5(out))
class Discriminator(nn.Module):
def __init__(self, in_channels):
super(Discriminator, self).__init__()
self.in_channels = in_channels
self.block1 = convBNReLU(self.in_channels, 64)
self.block2 = convBNReLU(64, 128)
self.block3 = convBNReLU(128, 256)
self.block4 = convBNReLU(256, 512)
self.block5 = nn.Conv2d(512, 64, 4, 1, 0)
self.source = nn.Linear(64, 1)
def forward(self, input):
out = self.block1(input)
out = self.block2(out)
out = self.block3(out)
out = self.block4(out)
out = self.block5(out)
size = out.shape[0]
out = out.view(size, -1)
source = torch.sigmoid(self.source(out))
return source
latentdim = 20
criterionSource = nn.BCELoss()
G = Generator(in_channels=latentdim, out_channels=1).cuda(gpu_id)
D = Discriminator(in_channels=1).cuda(gpu_id)
G.apply(weights_init_normal)
D.apply(weights_init_normal)
optimizerG = optim.Adam(G.parameters(), lr=0.00002, betas=(0.5, 0.999))
optimizerD = optim.Adam(D.parameters(), lr=0.00001, betas=(0.5, 0.999))
static_z = Variable(FloatTensor(torch.randn((81, latentdim, 1, 1)))).cuda(gpu_id)
def sample_image(stage, epoch):
static_sample = G(static_z).detach().cpu()
static_sample = (static_sample + 1) / 2.0
save_image(static_sample, DIRNAME + "stage%depoch%d.png" % (stage, epoch), nrow=9)
c = 0.75
k = 10
DIRNAME = 'DistShift/'
os.makedirs(DIRNAME, exist_ok=True)
board = SummaryWriter(log_dir=DIRNAME)
G.load_state_dict(torch.load('DCGAN/G999.pt'))
D.load_state_dict(torch.load('DCGAN/D999.pt'))
step = 0
fake_name = 'data/fake.pt'
n = 2557
for i in range(1, k):
dataloader = DataLoader(NWSDataset(fake=fake_name, c=c, i=i, n=n), batch_size=256, shuffle=True)
for epoch in range(0, 100):
print(epoch)
for realdata in dataloader:
noise = 1e-5 * max(1 - (epoch / 100.0), 0)
step += 1
batch_size = realdata[0].shape[0]
trueTensor = 0.7 + 0.5 * torch.rand(batch_size)
falseTensor = 0.3 * torch.rand(batch_size)
probFlip = torch.rand(batch_size) < 0.05
probFlip = probFlip.float()
trueTensor, falseTensor = (
probFlip * falseTensor + (1 - probFlip) * trueTensor,
probFlip * trueTensor + (1 - probFlip) * falseTensor,
)
trueTensor = trueTensor.view(-1, 1).cuda(gpu_id)
falseTensor = falseTensor.view(-1, 1).cuda(gpu_id)
realdata = realdata.cuda(gpu_id)
realSource = D(realdata)
realLoss = criterionSource(realSource, trueTensor.expand_as(realSource))
latent = Variable(torch.randn(batch_size, latentdim, 1, 1)).cuda(gpu_id)
fakeGen = G(latent)
fakeGenSource = D(fakeGen.detach())
fakeGenLoss = criterionSource(fakeGenSource, falseTensor.expand_as(fakeGenSource))
lossD = realLoss + fakeGenLoss
optimizerD.zero_grad()
lossD.backward()
torch.nn.utils.clip_grad_norm_(D.parameters(), 20)
optimizerD.step()
fakeGenSource = D(fakeGen)
lossG = criterionSource(fakeGenSource, trueTensor.expand_as(fakeGenSource))
optimizerG.zero_grad()
lossG.backward()
torch.nn.utils.clip_grad_norm_(G.parameters(), 20)
optimizerG.step()
board.add_scalar('realLoss', realLoss.item(), step)
board.add_scalar('fakeGenLoss', fakeGenLoss.item(), step)
board.add_scalar('lossD', lossD.item(), step)
board.add_scalar('lossG', lossG.item(), step)
if (epoch + 1) % 50 == 0:
torch.save(G.state_dict(), DIRNAME + "Gstage" + str(i) + 'epoch' + str(epoch) + ".pt")
torch.save(D.state_dict(), DIRNAME + "Dstage" + str(i) + 'epoch' + str(epoch) + ".pt")
if (epoch + 1) % 10 == 0:
with torch.no_grad():
G.eval()
sample_image(i, epoch)
G.train()
with torch.no_grad():
G.eval()
fsize = int((1 - (c ** (i + 1))) * n / c)
fakeSamples = G(Variable(torch.randn(fsize, latentdim, 1, 1)).cuda(gpu_id))
sums = fakeSamples.sum(dim=(1, 2, 3)).detach().cpu().numpy().argsort()[::-1].copy()
fake_name = DIRNAME + 'fake' + str(i + 1) + '.pt'
torch.save(fakeSamples.data[sums], fake_name)
del fakeSamples
G.train()