-
Notifications
You must be signed in to change notification settings - Fork 0
/
INFO_GAN_Manager.py
327 lines (270 loc) · 12.3 KB
/
INFO_GAN_Manager.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
import time
import matplotlib.pyplot as plt
import numpy as np
import torch
import torch.nn as nn
import torch.optim as optim
import torchvision.utils as vutils
import matplotlib.animation as animation
import utils
from CONSTANTS import Constants
from info_GAN_model_MNIST import Generator, Discriminator, DHead, QHead
class INFO_GAN_Manager:
def __init__(self, device, dataset_name, num_z, num_disc_c,
disc_c_dim, num_con_c, dataloader):
self.device = device
self.dataset_name = dataset_name
self.num_z = num_z # 62 for MNIST
self.num_dis_c = num_disc_c # 1 for MNIST
self.dis_c_dim = disc_c_dim # 10 for MNIST
self.num_con_c = num_con_c # 2 for MNIST
self.dataloader = dataloader
def train_info_GAN(self):
# Plot the training images.
utils.plot_train_images(self.device, self.dataloader,
'Info_GAN_Training Images {}'.format(self.dataset_name))
# Initialise the network.
netG = Generator().to(self.device)
netG.apply(utils.weights_init_InfoGAN)
print(netG)
discriminator = Discriminator().to(self.device)
discriminator.apply(utils.weights_init_InfoGAN)
print(discriminator)
netD = DHead().to(self.device)
netD.apply(utils.weights_init_InfoGAN)
print(netD)
netQ = QHead().to(self.device)
netQ.apply(utils.weights_init_InfoGAN)
print(netQ)
# Loss for discrimination between real and fake images.
criterionD = nn.BCELoss()
# Loss for discrete latent code.
criterionQ_dis = nn.CrossEntropyLoss()
# Loss for continuous latent code.
criterionQ_con = utils.NormalNLLLoss()
# Adam optimiser is used.
optimD = optim.Adam(
[
{'params': discriminator.parameters()},
{'params': netD.parameters()}
],
lr=Constants.INFO_GAN_LR,
betas=(0.5, 0.999)
)
optimG = optim.Adam(
[
{'params': netG.parameters()},
{'params': netQ.parameters()}
],
lr=Constants.INFO_GAN_LR,
betas=(0.5, 0.999))
# Fixed Noise
z = torch.randn(100, self.num_z, 1, 1, device=self.device)
fixed_noise = z
if self.num_dis_c != 0:
idx = np.arange(self.dis_c_dim).repeat(10)
dis_c = torch.zeros(100, self.num_dis_c, self.dis_c_dim, device=self.device)
for i in range(self.num_dis_c):
dis_c[torch.arange(0, 100), i, idx] = 1.0
dis_c = dis_c.view(100, -1, 1, 1)
fixed_noise = torch.cat((fixed_noise, dis_c), dim=1)
if self.num_con_c != 0:
con_c = torch.rand(100, self.num_con_c, 1, 1, device=self.device) * 2 - 1
fixed_noise = torch.cat((fixed_noise, con_c), dim=1)
real_label = 1.
fake_label = 0.
# List variables to store results pf training.
img_list = []
G_losses = []
D_losses = []
print("-" * 25)
print("Starting Training Loop...\n")
print('Epochs: %d\nDataset: {}\nBatch Size: %d\nLength of Data Loader: %d'.format(self.dataset_name) % (
Constants.INFO_GAN_EPOCH, 128, len(self.dataloader)))
print("-" * 25)
start_time = time.time()
iters = 0
for epoch in range(Constants.INFO_GAN_EPOCH):
epoch_start_time = time.time()
for i, (data, _) in enumerate(self.dataloader, 0):
# Get batch size
b_size = data.size(0)
# Transfer data tensor to GPU/CPU (device)
real_data = data.to(self.device)
# Updating discriminator and DHead
optimD.zero_grad()
# Real data
label = torch.full((b_size,), real_label, device=self.device)
output1 = discriminator(real_data)
probs_real = netD(output1).view(-1)
loss_real = criterionD(probs_real, label)
# Calculate gradients.
loss_real.backward()
# Fake data
label.fill_(fake_label)
noise, idx = utils.noise_sample(self.num_dis_c, self.dis_c_dim, self.num_con_c,
self.num_z, b_size, self.device)
fake_data = netG(noise)
output2 = discriminator(fake_data.detach())
probs_fake = netD(output2).view(-1)
loss_fake = criterionD(probs_fake, label)
# Calculate gradients.
loss_fake.backward()
# Net Loss for the discriminator
D_loss = loss_real + loss_fake
# Update parameters
optimD.step()
# Updating Generator and QHead
optimG.zero_grad()
# Fake data treated as real.
output = discriminator(fake_data)
label.fill_(real_label)
probs_fake = netD(output).view(-1)
gen_loss = criterionD(probs_fake, label)
q_logits, q_mu, q_var = netQ(output)
target = torch.LongTensor(idx).to(self.device)
# Calculating loss for discrete latent code.
dis_loss = 0
for j in range(self.num_dis_c):
dis_loss += criterionQ_dis(q_logits[:, j * 10: j * 10 + 10], target[j])
# Calculating loss for continuous latent code.
con_loss = 0
if self.num_con_c != 0:
con_loss = criterionQ_con(
noise[:, self.num_z + self.num_dis_c * self.dis_c_dim:]
.view(-1, self.num_con_c), q_mu, q_var) * 0.1
# Net loss for generator.
G_loss = gen_loss + dis_loss + con_loss
# Calculate gradients.
G_loss.backward()
# Update parameters.
optimG.step()
# Check progress of training.
if i != 0 and i % 100 == 0:
print('[%d/%d][%d/%d]\tLoss_D: %.4f\tLoss_G: %.4f'
% (epoch + 1, Constants.INFO_GAN_EPOCH, i, len(self.dataloader),
D_loss.item(), G_loss.item()))
# Save the losses for plotting.
G_losses.append(G_loss.item())
D_losses.append(D_loss.item())
iters += 1
epoch_time = time.time() - epoch_start_time
print("Time taken for Epoch %d: %.2fs" % (epoch + 1, epoch_time))
# Generate image after each epoch to check performance of the generator.
# Used for creating animated gif later.
with torch.no_grad():
gen_data = netG(fixed_noise).detach().cpu()
img_list.append(vutils.make_grid(gen_data, nrow=10, padding=2, normalize=True))
# Generate image to check performance of generator.
if (epoch + 1) == 1 or (epoch + 1) == Constants.INFO_GAN_EPOCH / 2:
with torch.no_grad():
gen_data = netG(fixed_noise).detach().cpu()
plt.figure(figsize=(10, 10))
plt.axis("off")
plt.imshow(np.transpose(vutils.make_grid(gen_data,
nrow=10, padding=2, normalize=True), (1, 2, 0)))
plt.savefig("Epoch_%d {}".format(self.dataset_name) % (epoch + 1))
plt.close('all')
# Save network weights.
if (epoch + 1) % 25 == 0:
torch.save({
'netG': netG.state_dict(),
'discriminator': discriminator.state_dict(),
'netD': netD.state_dict(),
'netQ': netQ.state_dict(),
'optimD': optimD.state_dict(),
'optimG': optimG.state_dict(),
'params': Constants.params
}, 'checkpoint/model_epoch_%d_{}'.format(self.dataset_name) % (epoch + 1))
training_time = time.time() - start_time
print("-" * 50)
print('Training finished!\nTotal Time for Training: %.2fm' % (training_time / 60))
print("-" * 50)
# Generate image to check performance of trained generator.
with torch.no_grad():
gen_data = netG(fixed_noise).detach().cpu()
plt.figure(figsize=(10, 10))
plt.axis("off")
plt.imshow(np.transpose(vutils.make_grid(gen_data, nrow=10, padding=2, normalize=True), (1, 2, 0)))
plt.savefig("Epoch_%d_{}".format(self.dataset_name) % (Constants.INFO_GAN_EPOCH))
# Save network weights.
torch.save({
'netG': netG.state_dict(),
'discriminator': discriminator.state_dict(),
'netD': netD.state_dict(),
'netQ': netQ.state_dict(),
'optimD': optimD.state_dict(),
'optimG': optimG.state_dict(),
'params': Constants
}, 'checkpoint/model_final_{}'.format(self.dataset_name))
# Plot the training losses.
plt.figure(figsize=(10, 5))
plt.title("Generator and Discriminator Loss During Training")
plt.plot(G_losses, label="G")
plt.plot(D_losses, label="D")
plt.xlabel("iterations")
plt.ylabel("Loss")
plt.legend()
plt.savefig("Loss Curve {}".format(self.dataset_name))
# Animation showing the improvements of the generator.
fig = plt.figure(figsize=(10, 10))
plt.axis("off")
ims = [[plt.imshow(np.transpose(i, (1, 2, 0)), animated=True)] for i in img_list]
anim = animation.ArtistAnimation(fig, ims, interval=1000, repeat_delay=1000, blit=True)
anim.save('infoGAN_{}.gif'.format(self.dataset_name), dpi=80, writer='imagemagick')
plt.show()
def generate_images(self):
state_dict = torch.load("./checkpoint/InfoGAN/MNIST/model_final_MNIST",
map_location=torch.device('cpu'))
# Set the device to run on: GPU or CPU.
device = torch.device("cuda:0" if (torch.cuda.is_available())
else "cpu")
# Get the 'params' dictionary from the loaded state_dict.
params = state_dict['params']
# Create the generator network.
netG = Generator().to(device)
# Load the trained generator weights.
netG.load_state_dict(state_dict['netG'])
print(netG)
c = np.linspace(-2, 2, 10).reshape(1, -1)
c = np.repeat(c, 3000, 0).reshape(-1, 1)
# print(c.shape)
# print(c)
c = torch.from_numpy(c).float().to(device)
c = c.view(-1, 1, 1, 1)
# print(c.size())
zeros = torch.zeros(30000, 1, 1, 1, device=device)
# Continuous latent code.
c2 = torch.cat((c, zeros), dim=1)
c3 = torch.cat((zeros, c), dim=1)
idx = np.arange(10).repeat(3000)
print(idx)
dis_c = torch.zeros(30000, 10, 1, 1, device=device)
dis_c[torch.arange(0, 30000), idx] = 1.0
# Discrete latent code.
c1 = dis_c.view(30000, -1, 1, 1)
z = torch.randn(30000, 62, 1, 1, device=device)
# To see variation along c2 (Horizontally) and c1 (Vertically)
noise1 = torch.cat((z, c1, c2), dim=1)
# To see variation along c3 (Horizontally) and c1 (Vertically)
noise2 = torch.cat((z, c1, c3), dim=1)
# Generate image.
with torch.no_grad():
generated_img1 = netG(noise1).detach().cpu()
print(generated_img1.size())
torch.save(generated_img1, 'C_GAN_generate_datasets/30k_image_set_MNIST_noise_1.pt')
# Display the generated image.
fig = plt.figure(figsize=(10, 10))
plt.axis("off")
plt.imshow(np.transpose(vutils.make_grid(generated_img1, nrow=10, padding=2, normalize=True), (1, 2, 0)))
plt.show()
# Generate image.
with torch.no_grad():
generated_img2 = netG(noise2).detach().cpu()
torch.save(generated_img2, '30k_image_set_MNIST_noise_2.pt')
# # Display the generated image.
fig = plt.figure(figsize=(10, 10))
plt.axis("off")
plt.imshow(np.transpose(vutils.make_grid(generated_img2,
nrow=10, padding=2, normalize=True), (1, 2, 0)))
plt.show()