def training_procedure(FLAGS):
"""
model definition
"""
encoder = Encoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
encoder.apply(weights_init)
decoder = Decoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
decoder.apply(weights_init)
# load saved models if load_saved flag is true
if FLAGS.load_saved:
encoder.load_state_dict(
torch.load(os.path.join('checkpoints', FLAGS.encoder_save)))
decoder.load_state_dict(
torch.load(os.path.join('checkpoints', FLAGS.decoder_save)))
"""
variable definition
"""
X_1 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels,
FLAGS.image_size, FLAGS.image_size)
X_2 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels,
FLAGS.image_size, FLAGS.image_size)
X_3 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels,
FLAGS.image_size, FLAGS.image_size)
style_latent_space = torch.FloatTensor(FLAGS.batch_size, FLAGS.style_dim)
"""
loss definitions
"""
cross_entropy_loss = nn.CrossEntropyLoss()
'''
add option to run on GPU
'''
if FLAGS.cuda:
encoder.cuda()
decoder.cuda()
cross_entropy_loss.cuda()
X_1 = X_1.cuda()
X_2 = X_2.cuda()
X_3 = X_3.cuda()
style_latent_space = style_latent_space.cuda()
"""
optimizer and scheduler definition
"""
auto_encoder_optimizer = optim.Adam(list(encoder.parameters()) +
list(decoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2))
reverse_cycle_optimizer = optim.Adam(list(encoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2))
# divide the learning rate by a factor of 10 after 80 epochs
auto_encoder_scheduler = optim.lr_scheduler.StepLR(auto_encoder_optimizer,
step_size=80,
gamma=0.1)
reverse_cycle_scheduler = optim.lr_scheduler.StepLR(
reverse_cycle_optimizer, step_size=80, gamma=0.1)
"""
training
"""
if torch.cuda.is_available() and not FLAGS.cuda:
print(
"WARNING: You have a CUDA device, so you should probably run with --cuda"
)
if not os.path.exists('checkpoints'):
os.makedirs('checkpoints')
if not os.path.exists('reconstructed_images'):
os.makedirs('reconstructed_images')
# load_saved is false when training is started from 0th iteration
if not FLAGS.load_saved:
with open(FLAGS.log_file, 'w') as log:
log.write(
'Epoch\tIteration\tReconstruction_loss\tKL_divergence_loss\tReverse_cycle_loss\n'
)
# load data set and create data loader instance
print('Loading MNIST paired dataset...')
paired_mnist = MNIST_Paired(root='mnist',
download=True,
train=True,
transform=transform_config)
loader = cycle(
DataLoader(paired_mnist,
batch_size=FLAGS.batch_size,
shuffle=True,
num_workers=0,
drop_last=True))
# initialize summary writer
writer = SummaryWriter()
for epoch in range(FLAGS.start_epoch, FLAGS.end_epoch):
print('')
print(
'Epoch #' + str(epoch) +
'..........................................................................'
)
# update the learning rate scheduler
auto_encoder_scheduler.step()
reverse_cycle_scheduler.step()
for iteration in range(int(len(paired_mnist) / FLAGS.batch_size)):
# A. run the auto-encoder reconstruction
image_batch_1, image_batch_2, _ = next(loader)
auto_encoder_optimizer.zero_grad()
X_1.copy_(image_batch_1)
X_2.copy_(image_batch_2)
style_mu_1, style_logvar_1, class_latent_space_1 = encoder(
Variable(X_1))
style_latent_space_1 = reparameterize(training=True,
mu=style_mu_1,
logvar=style_logvar_1)
kl_divergence_loss_1 = FLAGS.kl_divergence_coef * (
-0.5 * torch.sum(1 + style_logvar_1 - style_mu_1.pow(2) -
style_logvar_1.exp()))
kl_divergence_loss_1 /= (FLAGS.batch_size * FLAGS.num_channels *
FLAGS.image_size * FLAGS.image_size)
kl_divergence_loss_1.backward(retain_graph=True)
style_mu_2, style_logvar_2, class_latent_space_2 = encoder(
Variable(X_2))
style_latent_space_2 = reparameterize(training=True,
mu=style_mu_2,
logvar=style_logvar_2)
kl_divergence_loss_2 = FLAGS.kl_divergence_coef * (
-0.5 * torch.sum(1 + style_logvar_2 - style_mu_2.pow(2) -
style_logvar_2.exp()))
kl_divergence_loss_2 /= (FLAGS.batch_size * FLAGS.num_channels *
FLAGS.image_size * FLAGS.image_size)
kl_divergence_loss_2.backward(retain_graph=True)
reconstructed_X_1 = decoder(style_latent_space_1,
class_latent_space_2)
reconstructed_X_2 = decoder(style_latent_space_2,
class_latent_space_1)
reconstruction_error_1 = FLAGS.reconstruction_coef * mse_loss(
reconstructed_X_1, Variable(X_1))
reconstruction_error_1.backward(retain_graph=True)
reconstruction_error_2 = FLAGS.reconstruction_coef * mse_loss(
reconstructed_X_2, Variable(X_2))
reconstruction_error_2.backward()
reconstruction_error = (
reconstruction_error_1 +
reconstruction_error_2) / FLAGS.reconstruction_coef
kl_divergence_error = (kl_divergence_loss_1 + kl_divergence_loss_2
) / FLAGS.kl_divergence_coef
auto_encoder_optimizer.step()
# B. reverse cycle
image_batch_1, _, __ = next(loader)
image_batch_2, _, __ = next(loader)
reverse_cycle_optimizer.zero_grad()
X_1.copy_(image_batch_1)
X_2.copy_(image_batch_2)
style_latent_space.normal_(0., 1.)
_, __, class_latent_space_1 = encoder(Variable(X_1))
_, __, class_latent_space_2 = encoder(Variable(X_2))
reconstructed_X_1 = decoder(Variable(style_latent_space),
class_latent_space_1.detach())
reconstructed_X_2 = decoder(Variable(style_latent_space),
class_latent_space_2.detach())
style_mu_1, style_logvar_1, _ = encoder(reconstructed_X_1)
style_latent_space_1 = reparameterize(training=False,
mu=style_mu_1,
logvar=style_logvar_1)
style_mu_2, style_logvar_2, _ = encoder(reconstructed_X_2)
style_latent_space_2 = reparameterize(training=False,
mu=style_mu_2,
logvar=style_logvar_2)
reverse_cycle_loss = FLAGS.reverse_cycle_coef * l1_loss(
style_latent_space_1, style_latent_space_2)
reverse_cycle_loss.backward()
reverse_cycle_loss /= FLAGS.reverse_cycle_coef
reverse_cycle_optimizer.step()
if (iteration + 1) % 10 == 0:
print('')
print('Epoch #' + str(epoch))
print('Iteration #' + str(iteration))
print('')
print('Reconstruction loss: ' +
str(reconstruction_error.data.storage().tolist()[0]))
print('KL-Divergence loss: ' +
str(kl_divergence_error.data.storage().tolist()[0]))
print('Reverse cycle loss: ' +
str(reverse_cycle_loss.data.storage().tolist()[0]))
# write to log
with open(FLAGS.log_file, 'a') as log:
log.write('{0}\t{1}\t{2}\t{3}\t{4}\n'.format(
epoch, iteration,
reconstruction_error.data.storage().tolist()[0],
kl_divergence_error.data.storage().tolist()[0],
reverse_cycle_loss.data.storage().tolist()[0]))
# write to tensorboard
writer.add_scalar(
'Reconstruction loss',
reconstruction_error.data.storage().tolist()[0],
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) +
iteration)
writer.add_scalar(
'KL-Divergence loss',
kl_divergence_error.data.storage().tolist()[0],
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) +
iteration)
writer.add_scalar(
'Reverse cycle loss',
reverse_cycle_loss.data.storage().tolist()[0],
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) +
iteration)
# save model after every 5 epochs
if (epoch + 1) % 5 == 0 or (epoch + 1) == FLAGS.end_epoch:
torch.save(encoder.state_dict(),
os.path.join('checkpoints', FLAGS.encoder_save))
torch.save(decoder.state_dict(),
os.path.join('checkpoints', FLAGS.decoder_save))
"""
save reconstructed images and style swapped image generations to check progress
"""
image_batch_1, image_batch_2, _ = next(loader)
image_batch_3, _, __ = next(loader)
X_1.copy_(image_batch_1)
X_2.copy_(image_batch_2)
X_3.copy_(image_batch_3)
style_mu_1, style_logvar_1, _ = encoder(Variable(X_1))
_, __, class_latent_space_2 = encoder(Variable(X_2))
style_mu_3, style_logvar_3, _ = encoder(Variable(X_3))
style_latent_space_1 = reparameterize(training=False,
mu=style_mu_1,
logvar=style_logvar_1)
style_latent_space_3 = reparameterize(training=False,
mu=style_mu_3,
logvar=style_logvar_3)
reconstructed_X_1_2 = decoder(style_latent_space_1,
class_latent_space_2)
reconstructed_X_3_2 = decoder(style_latent_space_3,
class_latent_space_2)
# save input image batch
image_batch = np.transpose(X_1.cpu().numpy(), (0, 2, 3, 1))
image_batch = np.concatenate(
(image_batch, image_batch, image_batch), axis=3)
imshow_grid(image_batch, name=str(epoch) + '_original', save=True)
# save reconstructed batch
reconstructed_x = np.transpose(
reconstructed_X_1_2.cpu().data.numpy(), (0, 2, 3, 1))
reconstructed_x = np.concatenate(
(reconstructed_x, reconstructed_x, reconstructed_x), axis=3)
imshow_grid(reconstructed_x,
name=str(epoch) + '_target',
save=True)
style_batch = np.transpose(X_3.cpu().numpy(), (0, 2, 3, 1))
style_batch = np.concatenate(
(style_batch, style_batch, style_batch), axis=3)
imshow_grid(style_batch, name=str(epoch) + '_style', save=True)
# save style swapped reconstructed batch
reconstructed_style = np.transpose(
reconstructed_X_3_2.cpu().data.numpy(), (0, 2, 3, 1))
reconstructed_style = np.concatenate(
(reconstructed_style, reconstructed_style,
reconstructed_style),
axis=3)
imshow_grid(reconstructed_style,
name=str(epoch) + '_style_target',
save=True)
class LSGANs_Trainer(nn.Module):
def __init__(self, hyperparameters):
super(LSGANs_Trainer, self).__init__()
lr = hyperparameters['lr']
# Initiate the networks
self.encoder = Encoder(hyperparameters['input_dim_a'],
hyperparameters['gen'])
self.decoder = Decoder(hyperparameters['input_dim_a'],
hyperparameters['gen'])
self.dis_a = Discriminator()
self.dis_b = Discriminator()
self.interp_net_ab = Interpolator()
self.interp_net_ba = Interpolator()
self.instancenorm = nn.InstanceNorm2d(512, affine=False)
self.style_dim = hyperparameters['gen']['style_dim']
# Setup the optimizers
beta1 = hyperparameters['beta1']
beta2 = hyperparameters['beta2']
enc_params = list(self.encoder.parameters())
dec_params = list(self.decoder.parameters())
dis_a_params = list(self.dis_a.parameters())
dis_b_params = list(self.dis_b.parameters())
interperlator_ab_params = list(self.interp_net_ab.parameters())
interperlator_ba_params = list(self.interp_net_ba.parameters())
self.enc_opt = torch.optim.Adam(
[p for p in enc_params if p.requires_grad],
lr=lr,
betas=(beta1, beta2),
weight_decay=hyperparameters['weight_decay'])
self.dec_opt = torch.optim.Adam(
[p for p in dec_params if p.requires_grad],
lr=lr,
betas=(beta1, beta2),
weight_decay=hyperparameters['weight_decay'])
self.dis_a_opt = torch.optim.Adam(
[p for p in dis_a_params if p.requires_grad],
lr=lr,
betas=(beta1, beta2),
weight_decay=hyperparameters['weight_decay'])
self.dis_b_opt = torch.optim.Adam(
[p for p in dis_b_params if p.requires_grad],
lr=lr,
betas=(beta1, beta2),
weight_decay=hyperparameters['weight_decay'])
self.interp_ab_opt = torch.optim.Adam(
[p for p in interperlator_ab_params if p.requires_grad],
lr=lr,
betas=(beta1, beta2),
weight_decay=hyperparameters['weight_decay'])
self.interp_ba_opt = torch.optim.Adam(
[p for p in interperlator_ba_params if p.requires_grad],
lr=lr,
betas=(beta1, beta2),
weight_decay=hyperparameters['weight_decay'])
self.enc_scheduler = get_scheduler(self.enc_opt, hyperparameters)
self.dec_scheduler = get_scheduler(self.dec_opt, hyperparameters)
self.dis_a_scheduler = get_scheduler(self.dis_a_opt, hyperparameters)
self.dis_b_scheduler = get_scheduler(self.dis_b_opt, hyperparameters)
self.interp_ab_scheduler = get_scheduler(self.interp_ab_opt,
hyperparameters)
self.interp_ba_scheduler = get_scheduler(self.interp_ba_opt,
hyperparameters)
# Network weight initialization
self.apply(weights_init(hyperparameters['init']))
self.dis_a.apply(weights_init('gaussian'))
self.dis_b.apply(weights_init('gaussian'))
# Load VGG model if needed
if 'vgg_w' in hyperparameters.keys() and hyperparameters['vgg_w'] > 0:
self.vgg = load_vgg16(hyperparameters['vgg_model_path'] +
'/models')
self.vgg.eval()
for param in self.vgg.parameters():
param.requires_grad = False
self.total_loss = 0
self.best_iter = 0
self.perceptural_loss = Perceptural_loss()
def recon_criterion(self, input, target):
return torch.mean(torch.abs(input - target))
def forward(self, x_a, x_b):
self.eval()
c_a, s_a_fake = self.encoder(x_a)
c_b, s_b_fake = self.encoder(x_b)
# decode (cross domain)
s_ab_interp = self.interp_net_ab(s_a_fake, s_b_fake, self.v)
s_ba_interp = self.interp_net_ba(s_b_fake, s_a_fake, self.v)
x_ba = self.decoder(c_b, s_a_interp)
x_ab = selfdecoder(c_a, s_b_interp)
self.train()
return x_ab, x_ba
def zero_grad(self):
self.dis_a_opt.zero_grad()
self.dis_b_opt.zero_grad()
self.dec_opt.zero_grad()
self.enc_opt.zero_grad()
self.interp_ab_opt.zero_grad()
self.interp_ba_opt.zero_grad()
def dis_update(self, x_a, x_b, hyperparameters):
self.zero_grad()
# encode
c_a, s_a = self.encoder(x_a)
c_b, s_b = self.encoder(x_b)
# decode (cross domain)
self.v = torch.ones(s_a.size())
s_a_interp = self.interp_net_ba(s_b, s_a, self.v)
s_b_interp = self.interp_net_ab(s_a, s_b, self.v)
x_ba = self.decoder(c_b, s_a_interp)
x_ab = self.decoder(c_a, s_b_interp)
x_a_feature = self.dis_a(x_a)
x_ba_feature = self.dis_a(x_ba)
x_b_feature = self.dis_b(x_b)
x_ab_feature = self.dis_b(x_ab)
self.loss_dis_a = (x_ba_feature - x_a_feature).mean()
self.loss_dis_b = (x_ab_feature - x_b_feature).mean()
# gradient penality
self.loss_dis_a_gp = self.dis_a.calculate_gradient_penalty(x_ba, x_a)
self.loss_dis_b_gp = self.dis_b.calculate_gradient_penalty(x_ab, x_b)
self.loss_dis_total = hyperparameters['gan_w'] * self.loss_dis_a + \
hyperparameters['gan_w'] * self.loss_dis_b + \
hyperparameters['gan_w'] * self.loss_dis_a_gp + \
hyperparameters['gan_w'] * self.loss_dis_b_gp
self.loss_dis_total.backward()
self.total_loss += self.loss_dis_total.item()
self.dis_a_opt.step()
self.dis_b_opt.step()
def gen_update(self, x_a, x_b, hyperparameters):
self.zero_grad()
# encode
c_a, s_a = self.encoder(x_a)
c_b, s_b = self.encoder(x_b)
# decode (within domain)
x_a_recon = self.decoder(c_a, s_a)
x_b_recon = self.decoder(c_b, s_b)
# decode (cross domain)
self.v = torch.ones(s_a.size())
s_a_interp = self.interp_net_ba(s_b, s_a, self.v)
s_b_interp = self.interp_net_ab(s_a, s_b, self.v)
x_ba = self.decoder(c_b, s_a_interp)
x_ab = self.decoder(c_a, s_b_interp)
# encode again
c_b_recon, s_a_recon = self.encoder(x_ba)
c_a_recon, s_b_recon = self.encoder(x_ab)
# decode again
x_aa = self.decoder(
c_a_recon, s_a) if hyperparameters['recon_x_cyc_w'] > 0 else None
x_bb = self.decoder(
c_b_recon, s_b) if hyperparameters['recon_x_cyc_w'] > 0 else None
# reconstruction loss
self.loss_gen_recon_x_a = self.recon_criterion(x_a_recon, x_a)
self.loss_gen_recon_x_b = self.recon_criterion(x_b_recon, x_b)
self.loss_gen_recon_s_a = self.recon_criterion(s_a_recon, s_a)
self.loss_gen_recon_s_b = self.recon_criterion(s_b_recon, s_b)
self.loss_gen_recon_c_a = self.recon_criterion(c_a_recon, c_a)
self.loss_gen_recon_c_b = self.recon_criterion(c_b_recon, c_b)
self.loss_gen_cycrecon_x_a = self.recon_criterion(
x_aa, x_a) if hyperparameters['recon_x_cyc_w'] > 0 else 0
self.loss_gen_cycrecon_x_b = self.recon_criterion(
x_bb, x_b) if hyperparameters['recon_x_cyc_w'] > 0 else 0
# perceptual loss
self.loss_gen_vgg_a = self.perceptural_loss(
x_a_recon, x_a) if hyperparameters['vgg_w'] > 0 else 0
self.loss_gen_vgg_b = self.perceptural_loss(
x_b_recon, x_b) if hyperparameters['vgg_w'] > 0 else 0
self.loss_gen_vgg_aa = self.perceptural_loss(
x_aa, x_a) if hyperparameters['vgg_w'] > 0 else 0
self.loss_gen_vgg_bb = self.perceptural_loss(
x_bb, x_b) if hyperparameters['vgg_w'] > 0 else 0
# GAN loss
x_ba_feature = self.dis_a(x_ba)
x_ab_feature = self.dis_b(x_ab)
self.loss_gen_adv_a = -x_ba_feature.mean()
self.loss_gen_adv_b = -x_ab_feature.mean()
# total loss
self.loss_gen_total = hyperparameters['gan_w'] * self.loss_gen_adv_a + \
hyperparameters['gan_w'] * self.loss_gen_adv_b + \
hyperparameters['recon_x_w'] * self.loss_gen_recon_x_a + \
hyperparameters['recon_s_w'] * self.loss_gen_recon_s_a + \
hyperparameters['recon_c_w'] * self.loss_gen_recon_c_a + \
hyperparameters['recon_x_w'] * self.loss_gen_recon_x_b + \
hyperparameters['recon_s_w'] * self.loss_gen_recon_s_b + \
hyperparameters['recon_c_w'] * self.loss_gen_recon_c_b + \
hyperparameters['recon_x_cyc_w'] * self.loss_gen_cycrecon_x_a + \
hyperparameters['recon_x_cyc_w'] * self.loss_gen_cycrecon_x_b + \
hyperparameters['vgg_w'] * self.loss_gen_vgg_aa + \
hyperparameters['vgg_w'] * self.loss_gen_vgg_bb + \
hyperparameters['vgg_w'] * self.loss_gen_vgg_a + \
hyperparameters['vgg_w'] * self.loss_gen_vgg_b
self.loss_gen_total.backward()
self.total_loss += self.loss_gen_total.item()
self.dec_opt.step()
self.enc_opt.step()
self.interp_ab_opt.step()
self.interp_ba_opt.step()
def sample(self, x_a, x_b):
self.eval()
x_a_recon, x_b_recon, x_ab, x_ba, x_aa, x_bb = [], [], [], [], [], []
for i in range(x_a.size(0)):
c_a, s_a = self.encoder(x_a[i].unsqueeze(0))
c_b, s_b = self.encoder(x_b[i].unsqueeze(0))
x_a_recon.append(self.decoder(c_a, s_a))
x_b_recon.append(self.decoder(c_b, s_b))
self.v = torch.ones(s_a.size())
s_a_interp = self.interp_net_ba(s_b, s_a, self.v)
s_b_interp = self.interp_net_ab(s_a, s_b, self.v)
x_ab_i = self.decoder(c_a, s_b_interp)
x_ba_i = self.decoder(c_b, s_a_interp)
c_a_recon, s_b_recon = self.encoder(x_ab_i)
c_b_recon, s_a_recon = self.encoder(x_ba_i)
x_ab.append(self.decoder(c_a, s_b_interp.unsqueeze(0)))
x_ba.append(self.decoder(c_b, s_a_interp.unsqueeze(0)))
x_aa.append(self.decoder(c_a_recon, s_a.unsqueeze(0)))
x_bb.append(self.decoder(c_b_recon, s_b.unsqueeze(0)))
x_a_recon, x_b_recon = torch.cat(x_a_recon), torch.cat(x_b_recon)
x_ab, x_aa = torch.cat(x_ab), torch.cat(x_aa)
x_ba, x_bb = torch.cat(x_ba), torch.cat(x_bb)
self.train()
return x_a, x_a_recon, x_ab, x_aa, x_b, x_b_recon, x_ba, x_bb
def update_learning_rate(self):
if self.dis_a_scheduler is not None:
self.dis_a_scheduler.step()
if self.dis_b_scheduler is not None:
self.dis_b_scheduler.step()
if self.gen_scheduler is not None:
self.gen_scheduler.step()
if self.enc_scheduler is not None:
self.enc_scheduler.step()
if self.dec_scheduler is not None:
self.dec_scheduler.step()
if self.interpo_ab_scheduler is not None:
self.interpo_ab_scheduler.step()
if self.interpo_ba_scheduler is not None:
self.interpo_ba_scheduler.step()
def resume(self, checkpoint_dir, hyperparameters):
# Load encode
model_name = get_model(checkpoint_dir, "encoder")
state_dict = torch.load(model_name)
self.encoder.load_state_dict(state_dict)
# Load decode
model_name = get_model(checkpoint_dir, "decoder")
state_dict = torch.load(model_name)
self.decoder.load_state_dict(state_dict)
# Load discriminator a
model_name = get_model(checkpoint_dir, "dis_a")
state_dict = torch.load(model_name)
self.dis_a.load_state_dict(state_dict)
# Load discriminator a
model_name = get_model(checkpoint_dir, "dis_b")
state_dict = torch.load(model_name)
self.dis_b.load_state_dict(state_dict)
# Load interperlator ab
model_name = get_model(checkpoint_dir, "interp_ab")
state_dict = torch.load(model_name)
self.interp_net_ab.load_state_dict(state_dict)
# Load interperlator ba
model_name = get_model(checkpoint_dir, "interp_ba")
state_dict = torch.load(model_name)
self.interp_net_ba.load_state_dict(state_dict)
# Load optimizers
state_dict = torch.load(os.path.join(checkpoint_dir, 'optimizer.pt'))
self.enc_opt.load_state_dict(state_dict['enc_opt'])
self.dec_opt.load_state_dict(state_dict['dec_opt'])
self.dis_a_opt.load_state_dict(state_dict['dis_a_opt'])
self.dis_b_opt.load_state_dict(state_dict['dis_b_opt'])
self.interp_ab_opt.load_state_dict(state_dict['interp_ab_opt'])
self.interp_ba_opt.load_state_dict(state_dict['interp_ba_opt'])
self.best_iter = state_dict['best_iter']
self.total_loss = state_dict['total_loss']
# Reinitilize schedulers
self.dis_a_scheduler = get_scheduler(self.dis_a_opt, hyperparameters,
self.best_iter)
self.dis_b_scheduler = get_scheduler(self.dis_b_opt, hyperparameters,
self.best_iter)
self.enc_scheduler = get_scheduler(self.enc_opt, hyperparameters,
self.best_iter)
self.dec_scheduler = get_scheduler(self.dec_opt, hyperparameters,
self.best_iter)
self.interpo_ab_scheduler = get_scheduler(self.interp_ab_opt,
hyperparameters,
self.best_iter)
self.interpo_ba_scheduler = get_scheduler(self.interp_ba_opt,
hyperparameters,
self.best_iter)
print('Resume from iteration %d' % self.best_iter)
return self.best_iter, self.total_loss
def resume_iter(self, checkpoint_dir, surfix, hyperparameters):
# Load encode
state_dict = torch.load(
os.path.join(checkpoint_dir, 'encoder' + surfix + '.pt'))
self.encoder.load_state_dict(state_dict)
# Load decode
state_dict = torch.load(
os.path.join(checkpoint_dir, 'decoder' + surfix + '.pt'))
self.decoder.load_state_dict(state_dict)
# Load discriminator a
state_dict = torch.load(
os.path.join(checkpoint_dir, 'dis_a' + surfix + '.pt'))
self.dis_a.load_state_dict(state_dict)
# # Load discriminator b
state_dict = torch.load(
os.path.join(checkpoint_dir, 'dis_b' + surfix + '.pt'))
self.dis_b.load_state_dict(state_dict)
state_dict = torch.load(
os.path.join(checkpoint_dir, 'interp' + surfix + '.pt'))
# print(state_dict)
self.interp_net_ab.load_state_dict(state_dict['ab'])
self.interp_net_ba.load_state_dict(state_dict['ba'])
# Load interperlator ab
state_dict = torch.load(
os.path.join(checkpoint_dir, 'interp_ab' + surfix + '.pt'))
self.interp_net_ab.load_state_dict(state_dict)
# # Load interperlator ba
state_dict = torch.load(
os.path.join(checkpoint_dir, 'interp_ba' + surfix + '.pt'))
self.interp_net_ba.load_state_dict(state_dict)
# Load optimizers
state_dict = torch.load(
os.path.join(checkpoint_dir, 'optimizer' + surfix + '.pt'))
self.enc_opt.load_state_dict(state_dict['enc_opt'])
self.dec_opt.load_state_dict(state_dict['dec_opt'])
self.dis_a_opt.load_state_dict(state_dict['dis_a_opt'])
self.dis_b_opt.load_state_dict(state_dict['dis_b_opt'])
self.interp_ab_opt.load_state_dict(state_dict['interp_ab_opt'])
self.interp_ba_opt.load_state_dict(state_dict['interp_ba_opt'])
self.best_iter = state_dict['best_iter']
self.total_loss = state_dict['total_loss']
# Reinitilize schedulers
self.dis_a_scheduler = get_scheduler(self.dis_a_opt, hyperparameters,
self.best_iter)
self.dis_b_scheduler = get_scheduler(self.dis_b_opt, hyperparameters,
self.best_iter)
self.enc_scheduler = get_scheduler(self.enc_opt, hyperparameters,
self.best_iter)
self.dec_scheduler = get_scheduler(self.dec_opt, hyperparameters,
self.best_iter)
self.interpo_ab_scheduler = get_scheduler(self.interp_ab_opt,
hyperparameters,
self.best_iter)
self.interpo_ba_scheduler = get_scheduler(self.interp_ba_opt,
hyperparameters,
self.best_iter)
print('Resume from iteration %d' % self.best_iter)
return self.best_iter, self.total_loss
def save_better_model(self, snapshot_dir):
# remove sub_optimal models
files = glob.glob(snapshot_dir + '/*')
for f in files:
os.remove(f)
# Save encoder, decoder, interpolator, discriminators, and optimizers
encoder_name = os.path.join(snapshot_dir,
'encoder_%.4f.pt' % (self.total_loss))
decoder_name = os.path.join(snapshot_dir,
'decoder_%.4f.pt' % (self.total_loss))
interp_ab_name = os.path.join(snapshot_dir,
'interp_ab_%.4f.pt' % (self.total_loss))
interp_ba_name = os.path.join(snapshot_dir,
'interp_ba_%.4f.pt' % (self.total_loss))
dis_a_name = os.path.join(snapshot_dir,
'dis_a_%.4f.pt' % (self.total_loss))
dis_b_name = os.path.join(snapshot_dir,
'dis_b_%.4f.pt' % (self.total_loss))
opt_name = os.path.join(snapshot_dir, 'optimizer.pt')
torch.save(self.encoder.state_dict(), encoder_name)
torch.save(self.decoder.state_dict(), decoder_name)
torch.save(self.interp_net_ab.state_dict(), interp_ab_name)
torch.save(self.interp_net_ba.state_dict(), interp_ba_name)
torch.save(self.dis_a_opt.state_dict(), dis_a_name)
torch.save(self.dis_b_opt.state_dict(), dis_b_name)
torch.save(
{
'enc_opt': self.enc_opt.state_dict(),
'dec_opt': self.dec_opt.state_dict(),
'dis_a_opt': self.dis_a_opt.state_dict(),
'dis_b_opt': self.dis_b_opt.state_dict(),
'interp_ab_opt': self.interp_ab_opt.state_dict(),
'interp_ba_opt': self.interp_ba_opt.state_dict(),
'best_iter': self.best_iter,
'total_loss': self.total_loss
}, opt_name)
def save_at_iter(self, snapshot_dir, iterations):
encoder_name = os.path.join(snapshot_dir,
'encoder_%08d.pt' % (iterations + 1))
decoder_name = os.path.join(snapshot_dir,
'decoder_%08d.pt' % (iterations + 1))
interp_ab_name = os.path.join(snapshot_dir,
'interp_ab_%08d.pt' % (iterations + 1))
interp_ba_name = os.path.join(snapshot_dir,
'interp_ba_%08d.pt' % (iterations + 1))
dis_a_name = os.path.join(snapshot_dir,
'dis_a_%08d.pt' % (iterations + 1))
dis_b_name = os.path.join(snapshot_dir,
'dis_b_%08d.pt' % (iterations + 1))
opt_name = os.path.join(snapshot_dir,
'optimizer_%08d.pt' % (iterations + 1))
torch.save(self.encoder.state_dict(), encoder_name)
torch.save(self.decoder.state_dict(), decoder_name)
torch.save(self.interp_net_ab.state_dict(), interp_ab_name)
torch.save(self.interp_net_ba.state_dict(), interp_ba_name)
torch.save(self.dis_a_opt.state_dict(), dis_a_name)
torch.save(self.dis_b_opt.state_dict(), dis_b_name)
torch.save(
{
'enc_opt': self.enc_opt.state_dict(),
'dec_opt': self.dec_opt.state_dict(),
'dis_a_opt': self.dis_a_opt.state_dict(),
'dis_b_opt': self.dis_b_opt.state_dict(),
'interp_ab_opt': self.interp_ab_opt.state_dict(),
'interp_ba_opt': self.interp_ba_opt.state_dict(),
'best_iter': self.best_iter,
'total_loss': self.total_loss
}, opt_name)
def training_procedure(FLAGS):
"""
model definition
"""
encoder = Encoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
encoder.apply(weights_init)
decoder = Decoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
decoder.apply(weights_init)
discriminator = Discriminator()
discriminator.apply(weights_init)
# load saved models if load_saved flag is true
if FLAGS.load_saved:
raise Exception('This is not implemented')
encoder.load_state_dict(torch.load(os.path.join('checkpoints', FLAGS.encoder_save)))
decoder.load_state_dict(torch.load(os.path.join('checkpoints', FLAGS.decoder_save)))
"""
variable definition
"""
X_1 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels, FLAGS.image_size, FLAGS.image_size)
X_2 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels, FLAGS.image_size, FLAGS.image_size)
X_3 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels, FLAGS.image_size, FLAGS.image_size)
style_latent_space = torch.FloatTensor(FLAGS.batch_size, FLAGS.style_dim)
"""
loss definitions
"""
cross_entropy_loss = nn.CrossEntropyLoss()
adversarial_loss = nn.BCELoss()
'''
add option to run on GPU
'''
if FLAGS.cuda:
encoder.cuda()
decoder.cuda()
discriminator.cuda()
cross_entropy_loss.cuda()
adversarial_loss.cuda()
X_1 = X_1.cuda()
X_2 = X_2.cuda()
X_3 = X_3.cuda()
style_latent_space = style_latent_space.cuda()
"""
optimizer and scheduler definition
"""
auto_encoder_optimizer = optim.Adam(
list(encoder.parameters()) + list(decoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2)
)
reverse_cycle_optimizer = optim.Adam(
list(encoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2)
)
generator_optimizer = optim.Adam(
list(decoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2)
)
discriminator_optimizer = optim.Adam(
list(discriminator.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2)
)
# divide the learning rate by a factor of 10 after 80 epochs
auto_encoder_scheduler = optim.lr_scheduler.StepLR(auto_encoder_optimizer, step_size=80, gamma=0.1)
reverse_cycle_scheduler = optim.lr_scheduler.StepLR(reverse_cycle_optimizer, step_size=80, gamma=0.1)
generator_scheduler = optim.lr_scheduler.StepLR(generator_optimizer, step_size=80, gamma=0.1)
discriminator_scheduler = optim.lr_scheduler.StepLR(discriminator_optimizer, step_size=80, gamma=0.1)
# Used later to define discriminator ground truths
Tensor = torch.cuda.FloatTensor if FLAGS.cuda else torch.FloatTensor
"""
training
"""
if torch.cuda.is_available() and not FLAGS.cuda:
print("WARNING: You have a CUDA device, so you should probably run with --cuda")
if not os.path.exists('checkpoints'):
os.makedirs('checkpoints')
if not os.path.exists('reconstructed_images'):
os.makedirs('reconstructed_images')
# load_saved is false when training is started from 0th iteration
if not FLAGS.load_saved:
with open(FLAGS.log_file, 'w') as log:
headers = ['Epoch', 'Iteration', 'Reconstruction_loss', 'KL_divergence_loss', 'Reverse_cycle_loss']
if FLAGS.forward_gan:
headers.extend(['Generator_forward_loss', 'Discriminator_forward_loss'])
if FLAGS.reverse_gan:
headers.extend(['Generator_reverse_loss', 'Discriminator_reverse_loss'])
log.write('\t'.join(headers) + '\n')
# load data set and create data loader instance
print('Loading CIFAR paired dataset...')
paired_cifar = CIFAR_Paired(root='cifar', download=True, train=True, transform=transform_config)
loader = cycle(DataLoader(paired_cifar, batch_size=FLAGS.batch_size, shuffle=True, num_workers=0, drop_last=True))
# Save a batch of images to use for visualization
image_sample_1, image_sample_2, _ = next(loader)
image_sample_3, _, _ = next(loader)
# initialize summary writer
writer = SummaryWriter()
for epoch in range(FLAGS.start_epoch, FLAGS.end_epoch):
print('')
print('Epoch #' + str(epoch) + '..........................................................................')
# update the learning rate scheduler
auto_encoder_scheduler.step()
reverse_cycle_scheduler.step()
generator_scheduler.step()
discriminator_scheduler.step()
for iteration in range(int(len(paired_cifar) / FLAGS.batch_size)):
# Adversarial ground truths
valid = Variable(Tensor(FLAGS.batch_size, 1).fill_(1.0), requires_grad=False)
fake = Variable(Tensor(FLAGS.batch_size, 1).fill_(0.0), requires_grad=False)
# A. run the auto-encoder reconstruction
image_batch_1, image_batch_2, _ = next(loader)
auto_encoder_optimizer.zero_grad()
X_1.copy_(image_batch_1)
X_2.copy_(image_batch_2)
style_mu_1, style_logvar_1, class_latent_space_1 = encoder(Variable(X_1))
style_latent_space_1 = reparameterize(training=True, mu=style_mu_1, logvar=style_logvar_1)
kl_divergence_loss_1 = FLAGS.kl_divergence_coef * (
- 0.5 * torch.sum(1 + style_logvar_1 - style_mu_1.pow(2) - style_logvar_1.exp())
)
kl_divergence_loss_1 /= (FLAGS.batch_size * FLAGS.num_channels * FLAGS.image_size * FLAGS.image_size)
kl_divergence_loss_1.backward(retain_graph=True)
style_mu_2, style_logvar_2, class_latent_space_2 = encoder(Variable(X_2))
style_latent_space_2 = reparameterize(training=True, mu=style_mu_2, logvar=style_logvar_2)
kl_divergence_loss_2 = FLAGS.kl_divergence_coef * (
- 0.5 * torch.sum(1 + style_logvar_2 - style_mu_2.pow(2) - style_logvar_2.exp())
)
kl_divergence_loss_2 /= (FLAGS.batch_size * FLAGS.num_channels * FLAGS.image_size * FLAGS.image_size)
kl_divergence_loss_2.backward(retain_graph=True)
reconstructed_X_1 = decoder(style_latent_space_1, class_latent_space_2)
reconstructed_X_2 = decoder(style_latent_space_2, class_latent_space_1)
reconstruction_error_1 = FLAGS.reconstruction_coef * mse_loss(reconstructed_X_1, Variable(X_1))
reconstruction_error_1.backward(retain_graph=True)
reconstruction_error_2 = FLAGS.reconstruction_coef * mse_loss(reconstructed_X_2, Variable(X_2))
reconstruction_error_2.backward()
reconstruction_error = (reconstruction_error_1 + reconstruction_error_2) / FLAGS.reconstruction_coef
kl_divergence_error = (kl_divergence_loss_1 + kl_divergence_loss_2) / FLAGS.kl_divergence_coef
auto_encoder_optimizer.step()
# A-1. Discriminator training during forward cycle
if FLAGS.forward_gan:
# Training generator
generator_optimizer.zero_grad()
g_loss_1 = adversarial_loss(discriminator(Variable(reconstructed_X_1)), valid)
g_loss_2 = adversarial_loss(discriminator(Variable(reconstructed_X_2)), valid)
gen_f_loss = (g_loss_1 + g_loss_2) / 2.0
gen_f_loss.backward()
generator_optimizer.step()
# Training discriminator
discriminator_optimizer.zero_grad()
real_loss_1 = adversarial_loss(discriminator(Variable(X_1)), valid)
real_loss_2 = adversarial_loss(discriminator(Variable(X_2)), valid)
fake_loss_1 = adversarial_loss(discriminator(Variable(reconstructed_X_1)), fake)
fake_loss_2 = adversarial_loss(discriminator(Variable(reconstructed_X_2)), fake)
dis_f_loss = (real_loss_1 + real_loss_2 + fake_loss_1 + fake_loss_2) / 4.0
dis_f_loss.backward()
discriminator_optimizer.step()
# B. reverse cycle
image_batch_1, _, __ = next(loader)
image_batch_2, _, __ = next(loader)
reverse_cycle_optimizer.zero_grad()
X_1.copy_(image_batch_1)
X_2.copy_(image_batch_2)
style_latent_space.normal_(0., 1.)
_, __, class_latent_space_1 = encoder(Variable(X_1))
_, __, class_latent_space_2 = encoder(Variable(X_2))
reconstructed_X_1 = decoder(Variable(style_latent_space), class_latent_space_1.detach())
reconstructed_X_2 = decoder(Variable(style_latent_space), class_latent_space_2.detach())
style_mu_1, style_logvar_1, _ = encoder(reconstructed_X_1)
style_latent_space_1 = reparameterize(training=False, mu=style_mu_1, logvar=style_logvar_1)
style_mu_2, style_logvar_2, _ = encoder(reconstructed_X_2)
style_latent_space_2 = reparameterize(training=False, mu=style_mu_2, logvar=style_logvar_2)
reverse_cycle_loss = FLAGS.reverse_cycle_coef * l1_loss(style_latent_space_1, style_latent_space_2)
reverse_cycle_loss.backward()
reverse_cycle_loss /= FLAGS.reverse_cycle_coef
reverse_cycle_optimizer.step()
# B-1. Discriminator training during reverse cycle
if FLAGS.reverse_gan:
# Training generator
generator_optimizer.zero_grad()
g_loss_1 = adversarial_loss(discriminator(Variable(reconstructed_X_1)), valid)
g_loss_2 = adversarial_loss(discriminator(Variable(reconstructed_X_2)), valid)
gen_r_loss = (g_loss_1 + g_loss_2) / 2.0
gen_r_loss.backward()
generator_optimizer.step()
# Training discriminator
discriminator_optimizer.zero_grad()
real_loss_1 = adversarial_loss(discriminator(Variable(X_1)), valid)
real_loss_2 = adversarial_loss(discriminator(Variable(X_2)), valid)
fake_loss_1 = adversarial_loss(discriminator(Variable(reconstructed_X_1)), fake)
fake_loss_2 = adversarial_loss(discriminator(Variable(reconstructed_X_2)), fake)
dis_r_loss = (real_loss_1 + real_loss_2 + fake_loss_1 + fake_loss_2) / 4.0
dis_r_loss.backward()
discriminator_optimizer.step()
if (iteration + 1) % 10 == 0:
print('')
print('Epoch #' + str(epoch))
print('Iteration #' + str(iteration))
print('')
print('Reconstruction loss: ' + str(reconstruction_error.data.storage().tolist()[0]))
print('KL-Divergence loss: ' + str(kl_divergence_error.data.storage().tolist()[0]))
print('Reverse cycle loss: ' + str(reverse_cycle_loss.data.storage().tolist()[0]))
if FLAGS.forward_gan:
print('Generator F loss: ' + str(gen_f_loss.data.storage().tolist()[0]))
print('Discriminator F loss: ' + str(dis_f_loss.data.storage().tolist()[0]))
if FLAGS.reverse_gan:
print('Generator R loss: ' + str(gen_r_loss.data.storage().tolist()[0]))
print('Discriminator R loss: ' + str(dis_r_loss.data.storage().tolist()[0]))
# write to log
with open(FLAGS.log_file, 'a') as log:
row = []
row.append(epoch)
row.append(iteration)
row.append(reconstruction_error.data.storage().tolist()[0])
row.append(kl_divergence_error.data.storage().tolist()[0])
row.append(reverse_cycle_loss.data.storage().tolist()[0])
if FLAGS.forward_gan:
row.append(gen_f_loss.data.storage().tolist()[0])
row.append(dis_f_loss.data.storage().tolist()[0])
if FLAGS.reverse_gan:
row.append(gen_r_loss.data.storage().tolist()[0])
row.append(dis_r_loss.data.storage().tolist()[0])
row = [str(x) for x in row]
log.write('\t'.join(row) + '\n')
# write to tensorboard
writer.add_scalar('Reconstruction loss', reconstruction_error.data.storage().tolist()[0],
epoch * (int(len(paired_cifar) / FLAGS.batch_size) + 1) + iteration)
writer.add_scalar('KL-Divergence loss', kl_divergence_error.data.storage().tolist()[0],
epoch * (int(len(paired_cifar) / FLAGS.batch_size) + 1) + iteration)
writer.add_scalar('Reverse cycle loss', reverse_cycle_loss.data.storage().tolist()[0],
epoch * (int(len(paired_cifar) / FLAGS.batch_size) + 1) + iteration)
if FLAGS.forward_gan:
writer.add_scalar('Generator F loss', gen_f_loss.data.storage().tolist()[0],
epoch * (int(len(paired_cifar) / FLAGS.batch_size) + 1) + iteration)
writer.add_scalar('Discriminator F loss', dis_f_loss.data.storage().tolist()[0],
epoch * (int(len(paired_cifar) / FLAGS.batch_size) + 1) + iteration)
if FLAGS.reverse_gan:
writer.add_scalar('Generator R loss', gen_r_loss.data.storage().tolist()[0],
epoch * (int(len(paired_cifar) / FLAGS.batch_size) + 1) + iteration)
writer.add_scalar('Discriminator R loss', dis_r_loss.data.storage().tolist()[0],
epoch * (int(len(paired_cifar) / FLAGS.batch_size) + 1) + iteration)
# save model after every 5 epochs
if (epoch + 1) % 5 == 0 or (epoch + 1) == FLAGS.end_epoch:
torch.save(encoder.state_dict(), os.path.join('checkpoints', FLAGS.encoder_save))
torch.save(decoder.state_dict(), os.path.join('checkpoints', FLAGS.decoder_save))
"""
save reconstructed images and style swapped image generations to check progress
"""
X_1.copy_(image_sample_1)
X_2.copy_(image_sample_2)
X_3.copy_(image_sample_3)
style_mu_1, style_logvar_1, _ = encoder(Variable(X_1))
_, __, class_latent_space_2 = encoder(Variable(X_2))
style_mu_3, style_logvar_3, _ = encoder(Variable(X_3))
style_latent_space_1 = reparameterize(training=False, mu=style_mu_1, logvar=style_logvar_1)
style_latent_space_3 = reparameterize(training=False, mu=style_mu_3, logvar=style_logvar_3)
reconstructed_X_1_2 = decoder(style_latent_space_1, class_latent_space_2)
reconstructed_X_3_2 = decoder(style_latent_space_3, class_latent_space_2)
# save input image batch
image_batch = np.transpose(X_1.cpu().numpy(), (0, 2, 3, 1))
if FLAGS.num_channels == 1:
image_batch = np.concatenate((image_batch, image_batch, image_batch), axis=3)
imshow_grid(image_batch, name=str(epoch) + '_original', save=True)
# save reconstructed batch
reconstructed_x = np.transpose(reconstructed_X_1_2.cpu().data.numpy(), (0, 2, 3, 1))
if FLAGS.num_channels == 1:
reconstructed_x = np.concatenate((reconstructed_x, reconstructed_x, reconstructed_x), axis=3)
imshow_grid(reconstructed_x, name=str(epoch) + '_target', save=True)
style_batch = np.transpose(X_3.cpu().numpy(), (0, 2, 3, 1))
if FLAGS.num_channels == 1:
style_batch = np.concatenate((style_batch, style_batch, style_batch), axis=3)
imshow_grid(style_batch, name=str(epoch) + '_style', save=True)
# save style swapped reconstructed batch
reconstructed_style = np.transpose(reconstructed_X_3_2.cpu().data.numpy(), (0, 2, 3, 1))
if FLAGS.num_channels == 1:
reconstructed_style = np.concatenate((reconstructed_style, reconstructed_style, reconstructed_style), axis=3)
imshow_grid(reconstructed_style, name=str(epoch) + '_style_target', save=True)
save_image(original_sample,
OUTPUT_PATH + '/epoch={}_original.png'.format(str(epoch)),
nrow=NUM_FRAMES,
normalize=True)
save_image(decoded_sample,
OUTPUT_PATH + '/epoch={}_recon.png'.format(str(epoch)),
nrow=NUM_FRAMES,
normalize=True)
epoch_loss /= 3
if epoch_loss < lowest_loss:
lowest_loss = epoch_loss
# save checkpoints
torch.save(encoder.state_dict(),
os.path.join('checkpoints', ENCODER_SAVE))
torch.save(decoder.state_dict(),
os.path.join('checkpoints', DECODER_SAVE))
print('Model Saved! Epoch loss at {}'.format(lowest_loss))
encoder_test.load_state_dict(
torch.load(os.path.join('checkpoints', ENCODER_SAVE)))
decoder_test.load_state_dict(
torch.load(os.path.join('checkpoints', DECODER_SAVE)))
video1 = next(loader).float().cuda()[0].unsqueeze(0)
video2 = next(loader).float().cuda()[0].unsqueeze(0)
X1_v1, KL1_v1, muL1_v1, det_q1_v1 = encoder_test(video1,
BATCH_SIZE=1)
def training_procedure(FLAGS):
"""
model definition
"""
encoder = Encoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
encoder.apply(weights_init)
decoder = Decoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
decoder.apply(weights_init)
# load saved models if load_saved flag is true
if FLAGS.load_saved:
encoder.load_state_dict(
torch.load(os.path.join('checkpoints', FLAGS.encoder_save)))
decoder.load_state_dict(
torch.load(os.path.join('checkpoints', FLAGS.decoder_save)))
"""
variable definition
"""
X = torch.FloatTensor(FLAGS.batch_size, 1, FLAGS.image_size,
FLAGS.image_size)
'''
add option to run on GPU
'''
if FLAGS.cuda:
encoder.cuda()
decoder.cuda()
X = X.cuda()
"""
optimizer definition
"""
auto_encoder_optimizer = optim.Adam(list(encoder.parameters()) +
list(decoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2))
"""
training
"""
if torch.cuda.is_available() and not FLAGS.cuda:
print(
"WARNING: You have a CUDA device, so you should probably run with --cuda"
)
if not os.path.exists('checkpoints'):
os.makedirs('checkpoints')
# load_saved is false when training is started from 0th iteration
if not FLAGS.load_saved:
with open(FLAGS.log_file, 'w') as log:
log.write(
'Epoch\tIteration\tReconstruction_loss\tStyle_KL_divergence_loss\tClass_KL_divergence_loss\n'
)
# load data set and create data loader instance
print('Loading MNIST dataset...')
mnist = datasets.MNIST(root='mnist',
download=True,
train=True,
transform=transform_config)
loader = cycle(
DataLoader(mnist,
batch_size=FLAGS.batch_size,
shuffle=True,
num_workers=0,
drop_last=True))
# initialize summary writer
writer = SummaryWriter()
for epoch in range(FLAGS.start_epoch, FLAGS.end_epoch):
print('')
print(
'Epoch #' + str(epoch) +
'..........................................................................'
)
for iteration in range(int(len(mnist) / FLAGS.batch_size)):
# load a mini-batch
image_batch, labels_batch = next(loader)
# set zero_grad for the optimizer
auto_encoder_optimizer.zero_grad()
X.copy_(image_batch)
style_mu, style_logvar, class_mu, class_logvar = encoder(
Variable(X))
grouped_mu, grouped_logvar = accumulate_group_evidence(
class_mu.data, class_logvar.data, labels_batch, FLAGS.cuda)
# kl-divergence error for style latent space
style_kl_divergence_loss = FLAGS.kl_divergence_coef * (
-0.5 * torch.sum(1 + style_logvar - style_mu.pow(2) -
style_logvar.exp()))
style_kl_divergence_loss /= (FLAGS.batch_size *
FLAGS.num_channels *
FLAGS.image_size * FLAGS.image_size)
style_kl_divergence_loss.backward(retain_graph=True)
# kl-divergence error for class latent space
class_kl_divergence_loss = FLAGS.kl_divergence_coef * (
-0.5 * torch.sum(1 + grouped_logvar - grouped_mu.pow(2) -
grouped_logvar.exp()))
class_kl_divergence_loss /= (FLAGS.batch_size *
FLAGS.num_channels *
FLAGS.image_size * FLAGS.image_size)
class_kl_divergence_loss.backward(retain_graph=True)
# reconstruct samples
"""
sampling from group mu and logvar for each image in mini-batch differently makes
the decoder consider class latent embeddings as random noise and ignore them
"""
style_latent_embeddings = reparameterize(training=True,
mu=style_mu,
logvar=style_logvar)
class_latent_embeddings = group_wise_reparameterize(
training=True,
mu=grouped_mu,
logvar=grouped_logvar,
labels_batch=labels_batch,
cuda=FLAGS.cuda)
reconstructed_images = decoder(style_latent_embeddings,
class_latent_embeddings)
reconstruction_error = FLAGS.reconstruction_coef * mse_loss(
reconstructed_images, Variable(X))
reconstruction_error.backward()
auto_encoder_optimizer.step()
if (iteration + 1) % 50 == 0:
print('')
print('Epoch #' + str(epoch))
print('Iteration #' + str(iteration))
print('')
print('Reconstruction loss: ' +
str(reconstruction_error.data.storage().tolist()[0]))
print('Style KL-Divergence loss: ' +
str(style_kl_divergence_loss.data.storage().tolist()[0]))
print('Class KL-Divergence loss: ' +
str(class_kl_divergence_loss.data.storage().tolist()[0]))
# write to log
with open(FLAGS.log_file, 'a') as log:
log.write('{0}\t{1}\t{2}\t{3}\t{4}\n'.format(
epoch, iteration,
reconstruction_error.data.storage().tolist()[0],
style_kl_divergence_loss.data.storage().tolist()[0],
class_kl_divergence_loss.data.storage().tolist()[0]))
# write to tensorboard
writer.add_scalar(
'Reconstruction loss',
reconstruction_error.data.storage().tolist()[0],
epoch * (int(len(mnist) / FLAGS.batch_size) + 1) + iteration)
writer.add_scalar(
'Style KL-Divergence loss',
style_kl_divergence_loss.data.storage().tolist()[0],
epoch * (int(len(mnist) / FLAGS.batch_size) + 1) + iteration)
writer.add_scalar(
'Class KL-Divergence loss',
class_kl_divergence_loss.data.storage().tolist()[0],
epoch * (int(len(mnist) / FLAGS.batch_size) + 1) + iteration)
# save checkpoints after every 5 epochs
if (epoch + 1) % 5 == 0 or (epoch + 1) == FLAGS.end_epoch:
torch.save(encoder.state_dict(),
os.path.join('checkpoints', FLAGS.encoder_save))
torch.save(decoder.state_dict(),
os.path.join('checkpoints', FLAGS.decoder_save))
def train(opt):
#### device
device = torch.device('cuda:{}'.format(opt.gpu_id)
if opt.gpu_id >= 0 else torch.device('cpu'))
#### dataset
data_loader = UnAlignedDataLoader()
data_loader.initialize(opt)
data_set = data_loader.load_data()
print("The number of training images = %d." % len(data_set))
#### initialize models
## declaration
E_a2Zb = Encoder(input_nc=opt.input_nc,
ngf=opt.ngf,
norm_type=opt.norm_type,
use_dropout=not opt.no_dropout,
n_blocks=9)
G_Zb2b = Decoder(output_nc=opt.output_nc,
ngf=opt.ngf,
norm_type=opt.norm_type)
T_Zb2Za = LatentTranslator(n_channels=256,
norm_type=opt.norm_type,
use_dropout=not opt.no_dropout)
D_b = Discriminator(input_nc=opt.input_nc,
ndf=opt.ndf,
n_layers=opt.n_layers,
norm_type=opt.norm_type)
E_b2Za = Encoder(input_nc=opt.input_nc,
ngf=opt.ngf,
norm_type=opt.norm_type,
use_dropout=not opt.no_dropout,
n_blocks=9)
G_Za2a = Decoder(output_nc=opt.output_nc,
ngf=opt.ngf,
norm_type=opt.norm_type)
T_Za2Zb = LatentTranslator(n_channels=256,
norm_type=opt.norm_type,
use_dropout=not opt.no_dropout)
D_a = Discriminator(input_nc=opt.input_nc,
ndf=opt.ndf,
n_layers=opt.n_layers,
norm_type=opt.norm_type)
## initialization
E_a2Zb = init_net(E_a2Zb, init_type=opt.init_type).to(device)
G_Zb2b = init_net(G_Zb2b, init_type=opt.init_type).to(device)
T_Zb2Za = init_net(T_Zb2Za, init_type=opt.init_type).to(device)
D_b = init_net(D_b, init_type=opt.init_type).to(device)
E_b2Za = init_net(E_b2Za, init_type=opt.init_type).to(device)
G_Za2a = init_net(G_Za2a, init_type=opt.init_type).to(device)
T_Za2Zb = init_net(T_Za2Zb, init_type=opt.init_type).to(device)
D_a = init_net(D_a, init_type=opt.init_type).to(device)
print(
"+------------------------------------------------------+\nFinish initializing networks."
)
#### optimizer and criterion
## criterion
criterionGAN = GANLoss(opt.gan_mode).to(device)
criterionZId = nn.L1Loss()
criterionIdt = nn.L1Loss()
criterionCTC = nn.L1Loss()
criterionZCyc = nn.L1Loss()
## optimizer
optimizer_G = torch.optim.Adam(itertools.chain(E_a2Zb.parameters(),
G_Zb2b.parameters(),
T_Zb2Za.parameters(),
E_b2Za.parameters(),
G_Za2a.parameters(),
T_Za2Zb.parameters()),
lr=opt.lr,
betas=(opt.beta1, opt.beta2))
optimizer_D = torch.optim.Adam(itertools.chain(D_a.parameters(),
D_b.parameters()),
lr=opt.lr,
betas=(opt.beta1, opt.beta2))
## scheduler
scheduler = [
get_scheduler(optimizer_G, opt),
get_scheduler(optimizer_D, opt)
]
print(
"+------------------------------------------------------+\nFinish initializing the optimizers and criterions."
)
#### global variables
checkpoints_pth = os.path.join(opt.checkpoints, opt.name)
if os.path.exists(checkpoints_pth) is not True:
os.mkdir(checkpoints_pth)
os.mkdir(os.path.join(checkpoints_pth, 'images'))
record_fh = open(os.path.join(checkpoints_pth, 'records.txt'),
'w',
encoding='utf-8')
loss_names = [
'GAN_A', 'Adv_A', 'Idt_A', 'CTC_A', 'ZId_A', 'ZCyc_A', 'GAN_B',
'Adv_B', 'Idt_B', 'CTC_B', 'ZId_B', 'ZCyc_B'
]
fake_A_pool = ImagePool(
opt.pool_size
) # create image buffer to store previously generated images
fake_B_pool = ImagePool(
opt.pool_size
) # create image buffer to store previously generated images
print(
"+------------------------------------------------------+\nFinish preparing the other works."
)
print(
"+------------------------------------------------------+\nNow training is beginning .."
)
#### training
cur_iter = 0
for epoch in range(opt.epoch_count, opt.niter + opt.niter_decay + 1):
epoch_start_time = time.time() # timer for entire epoch
for i, data in enumerate(data_set):
## setup inputs
real_A = data['A'].to(device)
real_B = data['B'].to(device)
## forward
# image cycle / GAN
latent_B = E_a2Zb(real_A) #-> a -> Zb : E_a2b(a)
fake_B = G_Zb2b(latent_B) #-> Zb -> b' : G_b(E_a2b(a))
latent_A = E_b2Za(real_B) #-> b -> Za : E_b2a(b)
fake_A = G_Za2a(latent_A) #-> Za -> a' : G_a(E_b2a(b))
# Idt
'''
rec_A = G_Za2a(E_b2Za(fake_B)) #-> b' -> Za' -> rec_a : G_a(E_b2a(fake_b))
rec_B = G_Zb2b(E_a2Zb(fake_A)) #-> a' -> Zb' -> rec_b : G_b(E_a2b(fake_a))
'''
idt_latent_A = E_b2Za(real_A) #-> a -> Za : E_b2a(a)
idt_A = G_Za2a(idt_latent_A) #-> Za -> idt_a : G_a(E_b2a(a))
idt_latent_B = E_a2Zb(real_B) #-> b -> Zb : E_a2b(b)
idt_B = G_Zb2b(idt_latent_B) #-> Zb -> idt_b : G_b(E_a2b(b))
# ZIdt
T_latent_A = T_Zb2Za(latent_B) #-> Zb -> Za'' : T_b2a(E_a2b(a))
T_rec_A = G_Za2a(
T_latent_A) #-> Za'' -> a'' : G_a(T_b2a(E_a2b(a)))
T_latent_B = T_Za2Zb(latent_A) #-> Za -> Zb'' : T_a2b(E_b2a(b))
T_rec_B = G_Zb2b(
T_latent_B) #-> Zb'' -> b'' : G_b(T_a2b(E_b2a(b)))
# CTC
T_idt_latent_B = T_Za2Zb(idt_latent_A) #-> a -> T_a2b(E_b2a(a))
T_idt_latent_A = T_Zb2Za(idt_latent_B) #-> b -> T_b2a(E_a2b(b))
# ZCyc
TT_latent_B = T_Za2Zb(T_latent_A) #-> T_a2b(T_b2a(E_a2b(a)))
TT_latent_A = T_Zb2Za(T_latent_B) #-> T_b2a(T_a2b(E_b2a(b)))
### optimize parameters
## Generator updating
set_requires_grad(
[D_b, D_a],
False) #-> set Discriminator to require no gradient
optimizer_G.zero_grad()
# GAN loss
loss_G_A = criterionGAN(D_b(fake_B), True)
loss_G_B = criterionGAN(D_a(fake_A), True)
loss_GAN = loss_G_A + loss_G_B
# Idt loss
loss_idt_A = criterionIdt(idt_A, real_A)
loss_idt_B = criterionIdt(idt_B, real_B)
loss_Idt = loss_idt_A + loss_idt_B
# Latent cross-identity loss
loss_Zid_A = criterionZId(T_rec_A, real_A)
loss_Zid_B = criterionZId(T_rec_B, real_B)
loss_Zid = loss_Zid_A + loss_Zid_B
# Latent cross-translation consistency
loss_CTC_A = criterionCTC(T_idt_latent_A, latent_A)
loss_CTC_B = criterionCTC(T_idt_latent_B, latent_B)
loss_CTC = loss_CTC_B + loss_CTC_A
# Latent cycle consistency
loss_ZCyc_A = criterionZCyc(TT_latent_A, latent_A)
loss_ZCyc_B = criterionZCyc(TT_latent_B, latent_B)
loss_ZCyc = loss_ZCyc_B + loss_ZCyc_A
loss_G = opt.lambda_gan * loss_GAN + opt.lambda_idt * loss_Idt + opt.lambda_zid * loss_Zid + opt.lambda_ctc * loss_CTC + opt.lambda_zcyc * loss_ZCyc
# backward and gradient updating
loss_G.backward()
optimizer_G.step()
## Discriminator updating
set_requires_grad([D_b, D_a],
True) # -> set Discriminator to require gradient
optimizer_D.zero_grad()
# backward D_b
fake_B_ = fake_B_pool.query(fake_B)
#-> real_B, fake_B
pred_real_B = D_b(real_B)
loss_D_real_B = criterionGAN(pred_real_B, True)
pred_fake_B = D_b(fake_B_)
loss_D_fake_B = criterionGAN(pred_fake_B, False)
loss_D_B = (loss_D_real_B + loss_D_fake_B) * 0.5
loss_D_B.backward()
# backward D_a
fake_A_ = fake_A_pool.query(fake_A)
#-> real_A, fake_A
pred_real_A = D_a(real_A)
loss_D_real_A = criterionGAN(pred_real_A, True)
pred_fake_A = D_a(fake_A_)
loss_D_fake_A = criterionGAN(pred_fake_A, False)
loss_D_A = (loss_D_real_A + loss_D_fake_A) * 0.5
loss_D_A.backward()
# update the gradients
optimizer_D.step()
### validate here, both qualitively and quantitatively
## record the losses
if cur_iter % opt.log_freq == 0:
# loss_names = ['GAN_A', 'Adv_A', 'Idt_A', 'CTC_A', 'ZId_A', 'ZCyc_A', 'GAN_B', 'Adv_B', 'Idt_B', 'CTC_B', 'ZId_B', 'ZCyc_B']
losses = [
loss_G_A.item(),
loss_D_A.item(),
loss_idt_A.item(),
loss_CTC_A.item(),
loss_Zid_A.item(),
loss_ZCyc_A.item(),
loss_G_B.item(),
loss_D_B.item(),
loss_idt_B.item(),
loss_CTC_B.item(),
loss_Zid_B.item(),
loss_ZCyc_B.item()
]
# record
line = ''
for loss in losses:
line += '{} '.format(loss)
record_fh.write(line[:-1] + '\n')
# print out
print('Epoch: %3d/%3dIter: %9d--------------------------+' %
(epoch, opt.epoch, i))
field_names = loss_names[:len(loss_names) // 2]
table = PrettyTable(field_names=field_names)
for l_n in field_names:
table.align[l_n] = 'm'
table.add_row(losses[:len(field_names)])
print(table.get_string(reversesort=True))
field_names = loss_names[len(loss_names) // 2:]
table = PrettyTable(field_names=field_names)
for l_n in field_names:
table.align[l_n] = 'm'
table.add_row(losses[-len(field_names):])
print(table.get_string(reversesort=True))
## visualize
if cur_iter % opt.vis_freq == 0:
if opt.gpu_id >= 0:
real_A = real_A.cpu().data
real_B = real_B.cpu().data
fake_A = fake_A.cpu().data
fake_B = fake_B.cpu().data
idt_A = idt_A.cpu().data
idt_B = idt_B.cpu().data
T_rec_A = T_rec_A.cpu().data
T_rec_B = T_rec_B.cpu().data
plt.subplot(241), plt.title('real_A'), plt.imshow(
tensor2image_RGB(real_A[0, ...]))
plt.subplot(242), plt.title('fake_B'), plt.imshow(
tensor2image_RGB(fake_B[0, ...]))
plt.subplot(243), plt.title('idt_A'), plt.imshow(
tensor2image_RGB(idt_A[0, ...]))
plt.subplot(244), plt.title('L_idt_A'), plt.imshow(
tensor2image_RGB(T_rec_A[0, ...]))
plt.subplot(245), plt.title('real_B'), plt.imshow(
tensor2image_RGB(real_B[0, ...]))
plt.subplot(246), plt.title('fake_A'), plt.imshow(
tensor2image_RGB(fake_A[0, ...]))
plt.subplot(247), plt.title('idt_B'), plt.imshow(
tensor2image_RGB(idt_B[0, ...]))
plt.subplot(248), plt.title('L_idt_B'), plt.imshow(
tensor2image_RGB(T_rec_B[0, ...]))
plt.savefig(
os.path.join(checkpoints_pth, 'images',
'%03d_%09d.jpg' % (epoch, i)))
cur_iter += 1
#break #-> debug
## till now, we finish one epoch, try to update the learning rate
update_learning_rate(schedulers=scheduler,
opt=opt,
optimizer=optimizer_D)
## save the model
if epoch % opt.ckp_freq == 0:
#-> save models
# torch.save(model.state_dict(), PATH)
#-> load in models
# model.load_state_dict(torch.load(PATH))
# model.eval()
if opt.gpu_id >= 0:
E_a2Zb = E_a2Zb.cpu()
G_Zb2b = G_Zb2b.cpu()
T_Zb2Za = T_Zb2Za.cpu()
D_b = D_b.cpu()
E_b2Za = E_b2Za.cpu()
G_Za2a = G_Za2a.cpu()
T_Za2Zb = T_Za2Zb.cpu()
D_a = D_a.cpu()
'''
torch.save( E_a2Zb.cpu().state_dict(), os.path.join(checkpoints_pth, 'epoch_%3d-E_a2b.pth' % epoch))
torch.save( G_Zb2b.cpu().state_dict(), os.path.join(checkpoints_pth, 'epoch_%3d-G_b.pth' % epoch))
torch.save(T_Zb2Za.cpu().state_dict(), os.path.join(checkpoints_pth, 'epoch_%3d-T_b2a.pth' % epoch))
torch.save( D_b.cpu().state_dict(), os.path.join(checkpoints_pth, 'epoch_%3d-D_b.pth' % epoch))
torch.save( E_b2Za.cpu().state_dict(), os.path.join(checkpoints_pth, 'epoch_%3d-E_b2a.pth' % epoch))
torch.save( G_Za2a.cpu().state_dict(), os.path.join(checkpoints_pth, 'epoch_%3d-G_a.pth' % epoch))
torch.save(T_Za2Zb.cpu().state_dict(), os.path.join(checkpoints_pth, 'epoch_%3d-T_a2b.pth' % epoch))
torch.save( D_a.cpu().state_dict(), os.path.join(checkpoints_pth, 'epoch_%3d-D_a.pth' % epoch))
'''
torch.save(
E_a2Zb.state_dict(),
os.path.join(checkpoints_pth, 'epoch_%3d-E_a2b.pth' % epoch))
torch.save(
G_Zb2b.state_dict(),
os.path.join(checkpoints_pth, 'epoch_%3d-G_b.pth' % epoch))
torch.save(
T_Zb2Za.state_dict(),
os.path.join(checkpoints_pth, 'epoch_%3d-T_b2a.pth' % epoch))
torch.save(
D_b.state_dict(),
os.path.join(checkpoints_pth, 'epoch_%3d-D_b.pth' % epoch))
torch.save(
E_b2Za.state_dict(),
os.path.join(checkpoints_pth, 'epoch_%3d-E_b2a.pth' % epoch))
torch.save(
G_Za2a.state_dict(),
os.path.join(checkpoints_pth, 'epoch_%3d-G_a.pth' % epoch))
torch.save(
T_Za2Zb.state_dict(),
os.path.join(checkpoints_pth, 'epoch_%3d-T_a2b.pth' % epoch))
torch.save(
D_a.state_dict(),
os.path.join(checkpoints_pth, 'epoch_%3d-D_a.pth' % epoch))
if opt.gpu_id >= 0:
E_a2Zb = E_a2Zb.to(device)
G_Zb2b = G_Zb2b.to(device)
T_Zb2Za = T_Zb2Za.to(device)
D_b = D_b.to(device)
E_b2Za = E_b2Za.to(device)
G_Za2a = G_Za2a.to(device)
T_Za2Zb = T_Za2Zb.to(device)
D_a = D_a.to(device)
print("+Successfully saving models in epoch: %3d.-------------+" %
epoch)
#break #-> debug
record_fh.close()
print("≧◔◡◔≦ Congratulation! Finishing the training!")
def training_procedure(FLAGS):
"""
model definition
"""
encoder = Encoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
encoder.apply(weights_init)
decoder = Decoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
decoder.apply(weights_init)
# load saved models if load_saved flag is true
if FLAGS.load_saved:
encoder.load_state_dict(
torch.load(os.path.join('checkpoints', FLAGS.encoder_save)))
decoder.load_state_dict(
torch.load(os.path.join('checkpoints', FLAGS.decoder_save)))
"""
variable definition
"""
X = torch.FloatTensor(FLAGS.batch_size, 784)
'''
run on GPU if GPU is available
'''
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
encoder.to(device=device)
decoder.to(device=device)
X = X.to(device=device)
"""
optimizer definition
"""
auto_encoder_optimizer = optim.Adam(list(encoder.parameters()) +
list(decoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2))
"""
"""
if torch.cuda.is_available() and not FLAGS.cuda:
print(
"WARNING: You have a CUDA device, so you should probably run with --cuda"
)
if not os.path.exists('checkpoints'):
os.makedirs('checkpoints')
# load_saved is false when training is started from 0th iteration
if not FLAGS.load_saved:
with open(FLAGS.log_file, 'w') as log:
log.write(
'Epoch\tIteration\tReconstruction_loss\tStyle_KL_divergence_loss\tClass_KL_divergence_loss\n'
)
# load data set and create data loader instance
dirs = os.listdir(os.path.join(os.getcwd(), 'data'))
print('Loading double multivariate normal time series data...')
for dsname in dirs:
params = dsname.split('_')
if params[2] in ('theta=-1'):
print('Running dataset ', dsname)
ds = DoubleMulNormal(dsname)
# ds = experiment3(1000, 50, 3)
loader = cycle(
DataLoader(ds,
batch_size=FLAGS.batch_size,
shuffle=True,
drop_last=True))
# initialize summary writer
writer = SummaryWriter()
for epoch in range(FLAGS.start_epoch, FLAGS.end_epoch):
print()
print(
'Epoch #' + str(epoch) +
'........................................................')
# the total loss at each epoch after running iterations of batches
total_loss = 0
for iteration in range(int(len(ds) / FLAGS.batch_size)):
# load a mini-batch
image_batch, labels_batch = next(loader)
# set zero_grad for the optimizer
auto_encoder_optimizer.zero_grad()
X.copy_(image_batch)
style_mu, style_logvar, class_mu, class_logvar = encoder(
Variable(X))
grouped_mu, grouped_logvar = accumulate_group_evidence(
class_mu.data, class_logvar.data, labels_batch,
FLAGS.cuda)
# kl-divergence error for style latent space
style_kl_divergence_loss = FLAGS.kl_divergence_coef * (
-0.5 * torch.sum(1 + style_logvar - style_mu.pow(2) -
style_logvar.exp()))
style_kl_divergence_loss /= (FLAGS.batch_size *
FLAGS.num_channels *
FLAGS.image_size *
FLAGS.image_size)
style_kl_divergence_loss.backward(retain_graph=True)
# kl-divergence error for class latent space
class_kl_divergence_loss = FLAGS.kl_divergence_coef * (
-0.5 *
torch.sum(1 + grouped_logvar - grouped_mu.pow(2) -
grouped_logvar.exp()))
class_kl_divergence_loss /= (FLAGS.batch_size *
FLAGS.num_channels *
FLAGS.image_size *
FLAGS.image_size)
class_kl_divergence_loss.backward(retain_graph=True)
# reconstruct samples
"""
sampling from group mu and logvar for each image in mini-batch differently makes
the decoder consider class latent embeddings as random noise and ignore them
"""
style_latent_embeddings = reparameterize(
training=True, mu=style_mu, logvar=style_logvar)
class_latent_embeddings = group_wise_reparameterize(
training=True,
mu=grouped_mu,
logvar=grouped_logvar,
labels_batch=labels_batch,
cuda=FLAGS.cuda)
reconstructed_images = decoder(style_latent_embeddings,
class_latent_embeddings)
reconstruction_error = FLAGS.reconstruction_coef * mse_loss(
reconstructed_images, Variable(X))
reconstruction_error.backward()
total_loss += style_kl_divergence_loss + class_kl_divergence_loss + reconstruction_error
auto_encoder_optimizer.step()
if (iteration + 1) % 50 == 0:
print('\tIteration #' + str(iteration))
print('Reconstruction loss: ' + str(
reconstruction_error.data.storage().tolist()[0]))
print('Style KL loss: ' +
str(style_kl_divergence_loss.data.storage().
tolist()[0]))
print('Class KL loss: ' +
str(class_kl_divergence_loss.data.storage().
tolist()[0]))
# write to log
with open(FLAGS.log_file, 'a') as log:
log.write('{0}\t{1}\t{2}\t{3}\t{4}\n'.format(
epoch, iteration,
reconstruction_error.data.storage().tolist()[0],
style_kl_divergence_loss.data.storage().tolist()
[0],
class_kl_divergence_loss.data.storage().tolist()
[0]))
# write to tensorboard
writer.add_scalar(
'Reconstruction loss',
reconstruction_error.data.storage().tolist()[0],
epoch * (int(len(ds) / FLAGS.batch_size) + 1) +
iteration)
writer.add_scalar(
'Style KL-Divergence loss',
style_kl_divergence_loss.data.storage().tolist()[0],
epoch * (int(len(ds) / FLAGS.batch_size) + 1) +
iteration)
writer.add_scalar(
'Class KL-Divergence loss',
class_kl_divergence_loss.data.storage().tolist()[0],
epoch * (int(len(ds) / FLAGS.batch_size) + 1) +
iteration)
if epoch == 0 and (iteration + 1) % 50 == 0:
torch.save(
encoder.state_dict(),
os.path.join('checkpoints', 'encoder_' + dsname))
torch.save(
decoder.state_dict(),
os.path.join('checkpoints', 'decoder_' + dsname))
# save checkpoints after every 10 epochs
if (epoch + 1) % 10 == 0 or (epoch + 1) == FLAGS.end_epoch:
torch.save(
encoder.state_dict(),
os.path.join('checkpoints', 'encoder_' + dsname))
torch.save(
decoder.state_dict(),
os.path.join('checkpoints', 'decoder_' + dsname))
print('Total loss at current epoch: ', total_loss.item())
def training_procedure(FLAGS):
"""
model definition
"""
encoder = Encoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
encoder.apply(weights_init)
decoder = Decoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
decoder.apply(weights_init)
# load saved models if load_saved flag is true
if FLAGS.load_saved:
encoder.load_state_dict(
torch.load(os.path.join(savedir, FLAGS.encoder_save)))
decoder.load_state_dict(
torch.load(os.path.join(savedir, FLAGS.decoder_save)))
'''
add option to run on GPU
'''
if FLAGS.cuda:
encoder.cuda()
decoder.cuda()
"""
optimizer definition
"""
auto_encoder_optimizer = optim.Adam(list(encoder.parameters()) +
list(decoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2))
"""
training
"""
if torch.cuda.is_available() and not FLAGS.cuda:
print(
"WARNING: You have a CUDA device, so you should probably run with --cuda"
)
savedir = 'checkpoints_%d' % (FLAGS.batch_size)
if not os.path.exists(savedir):
os.makedirs(savedir)
# load_saved is false when training is started from 0th iteration
if not FLAGS.load_saved:
with open(FLAGS.log_file, 'w') as log:
log.write(
'Epoch\tIteration\tReconstruction_loss\tStyle_KL_divergence_loss\tClass_KL_divergence_loss\n'
)
# load data set and create data loader instance
print('Loading MNIST dataset...')
mnist = datasets.MNIST(root='mnist',
download=True,
train=True,
transform=transform_config)
# Creating data indices for training and validation splits:
dataset_size = len(mnist)
indices = list(range(dataset_size))
split = 10000
np.random.seed(0)
np.random.shuffle(indices)
train_indices, val_indices = indices[split:], indices[:split]
train_mnist, val_mnist = torch.utils.data.random_split(
mnist, [dataset_size - split, split])
# Creating PT data samplers and loaders:
weights_train = torch.ones(len(mnist))
weights_test = torch.ones(len(mnist))
weights_train[val_mnist.indices] = 0
weights_test[train_mnist.indices] = 0
counts = torch.zeros(10)
for i in range(10):
idx_label = mnist.targets[train_mnist.indices].eq(i)
counts[i] = idx_label.sum()
max = float(counts.max())
sum_counts = float(counts.sum())
for i in range(10):
idx_label = mnist.targets[train_mnist.indices].eq(
i).nonzero().squeeze()
weights_train[train_mnist.indices[idx_label]] = (sum_counts /
counts[i])
train_sampler = SubsetRandomSampler(train_mnist.indices)
valid_sampler = SubsetRandomSampler(val_mnist.indices)
kwargs = {'num_workers': 1, 'pin_memory': True} if FLAGS.cuda else {}
loader = DataLoader(mnist,
batch_size=FLAGS.batch_size,
sampler=train_sampler,
**kwargs)
valid_loader = DataLoader(mnist,
batch_size=FLAGS.batch_size,
sampler=valid_sampler,
**kwargs)
monitor = torch.zeros(FLAGS.end_epoch - FLAGS.start_epoch, 4)
# initialize summary writer
writer = SummaryWriter()
for epoch in range(FLAGS.start_epoch, FLAGS.end_epoch):
print('')
print(
'Epoch #' + str(epoch) +
'..........................................................................'
)
elbo_epoch = 0
term1_epoch = 0
term2_epoch = 0
term3_epoch = 0
for it, (image_batch, labels_batch) in enumerate(loader):
# set zero_grad for the optimizer
auto_encoder_optimizer.zero_grad()
X = image_batch.cuda().detach().clone()
elbo, reconstruction_proba, style_kl_divergence_loss, class_kl_divergence_loss = process(
FLAGS, X, labels_batch, encoder, decoder)
(-elbo).backward()
auto_encoder_optimizer.step()
elbo_epoch += elbo
term1_epoch += reconstruction_proba
term2_epoch += style_kl_divergence_loss
term3_epoch += class_kl_divergence_loss
print("Elbo epoch %.2f" % (elbo_epoch / (it + 1)))
print("Rec. Proba %.2f" % (term1_epoch / (it + 1)))
print("KL style %.2f" % (term2_epoch / (it + 1)))
print("KL content %.2f" % (term3_epoch / (it + 1)))
# save checkpoints after every 5 epochs
if (epoch + 1) % 5 == 0 or (epoch + 1) == FLAGS.end_epoch:
monitor[epoch, :] = eval(FLAGS, valid_loader, encoder, decoder)
torch.save(
encoder.state_dict(),
os.path.join(savedir, FLAGS.encoder_save + '_e%d' % epoch))
torch.save(
decoder.state_dict(),
os.path.join(savedir, FLAGS.decoder_save + '_e%d' % epoch))
print("VAL elbo %.2f" % (monitor[epoch, 0]))
print("VAL Rec. Proba %.2f" % (monitor[epoch, 1]))
print("VAL KL style %.2f" % (monitor[epoch, 2]))
print("VAL KL content %.2f" % (monitor[epoch, 3]))
torch.save(monitor, os.path.join(savedir, 'monitor_e%d' % epoch))
range=(-1., 1.))
it += 1
# Sample images
if (ep + 1) % args.sample_epochs == 0:
E.eval()
G.eval()
with torch.no_grad():
mu, logvar = E(fixed_reals)
latents = sample_latent(mu, logvar)
samples = G(latents, fixed_annos_onehot)
vutils.save_image(samples,
join(sample_path,
'{:03d}_fake.jpg'.format(ep)),
nrow=4,
padding=0,
normalize=True,
range=(-1., 1.))
# Checkpoints
if (ep + 1) % args.save_epochs == 0:
torch.save(E.state_dict(),
join(checkpoint_path, '{:03}.E.pth'.format(ep)))
torch.save(G.state_dict(),
join(checkpoint_path, '{:03}.G.pth'.format(ep)))
torch.save(D.state_dict(),
join(checkpoint_path, '{:03}.D.pth'.format(ep)))
torch.save(G_opt.state_dict(),
join(checkpoint_path, '{:03}.G_opt.pth'.format(ep)))
torch.save(D_opt.state_dict(),
join(checkpoint_path, '{:03}.D_opt.pth'.format(ep)))
def training_procedure(FLAGS):
"""
model definition
"""
encoder = Encoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
encoder.apply(weights_init)
decoder = Decoder(style_dim=FLAGS.style_dim, class_dim=FLAGS.class_dim)
decoder.apply(weights_init)
discriminator = Discriminator()
discriminator.apply(weights_init)
# load saved models if load_saved flag is true
if FLAGS.load_saved:
encoder.load_state_dict(torch.load(os.path.join('checkpoints', FLAGS.encoder_save)))
decoder.load_state_dict(torch.load(os.path.join('checkpoints', FLAGS.decoder_save)))
discriminator.load_state_dict(torch.load(os.path.join('checkpoints', FLAGS.discriminator_save)))
"""
variable definition
"""
real_domain_labels = 1
fake_domain_labels = 0
X_1 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels, FLAGS.image_size, FLAGS.image_size)
X_2 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels, FLAGS.image_size, FLAGS.image_size)
X_3 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels, FLAGS.image_size, FLAGS.image_size)
domain_labels = torch.LongTensor(FLAGS.batch_size)
style_latent_space = torch.FloatTensor(FLAGS.batch_size, FLAGS.style_dim)
"""
loss definitions
"""
cross_entropy_loss = nn.CrossEntropyLoss()
'''
add option to run on GPU
'''
if FLAGS.cuda:
encoder.cuda()
decoder.cuda()
discriminator.cuda()
cross_entropy_loss.cuda()
X_1 = X_1.cuda()
X_2 = X_2.cuda()
X_3 = X_3.cuda()
domain_labels = domain_labels.cuda()
style_latent_space = style_latent_space.cuda()
"""
optimizer definition
"""
auto_encoder_optimizer = optim.Adam(
list(encoder.parameters()) + list(decoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2)
)
discriminator_optimizer = optim.Adam(
list(discriminator.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2)
)
generator_optimizer = optim.Adam(
list(encoder.parameters()) + list(decoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2)
)
"""
training
"""
if torch.cuda.is_available() and not FLAGS.cuda:
print("WARNING: You have a CUDA device, so you should probably run with --cuda")
if not os.path.exists('checkpoints'):
os.makedirs('checkpoints')
# load_saved is false when training is started from 0th iteration
if not FLAGS.load_saved:
with open(FLAGS.log_file, 'w') as log:
log.write('Epoch\tIteration\tReconstruction_loss\tKL_divergence_loss\t')
log.write('Generator_loss\tDiscriminator_loss\tDiscriminator_accuracy\n')
# load data set and create data loader instance
print('Loading MNIST paired dataset...')
paired_mnist = MNIST_Paired(root='mnist', download=True, train=True, transform=transform_config)
loader = cycle(DataLoader(paired_mnist, batch_size=FLAGS.batch_size, shuffle=True, num_workers=0, drop_last=True))
# initialise variables
discriminator_accuracy = 0.
# initialize summary writer
writer = SummaryWriter()
for epoch in range(FLAGS.start_epoch, FLAGS.end_epoch):
print('')
print('Epoch #' + str(epoch) + '..........................................................................')
for iteration in range(int(len(paired_mnist) / FLAGS.batch_size)):
# A. run the auto-encoder reconstruction
image_batch_1, image_batch_2, _ = next(loader)
auto_encoder_optimizer.zero_grad()
X_1.copy_(image_batch_1)
X_2.copy_(image_batch_2)
style_mu_1, style_logvar_1, class_1 = encoder(Variable(X_1))
style_1 = reparameterize(training=True, mu=style_mu_1, logvar=style_logvar_1)
kl_divergence_loss_1 = - 0.5 * torch.sum(1 + style_logvar_1 - style_mu_1.pow(2) - style_logvar_1.exp())
kl_divergence_loss_1 /= (FLAGS.batch_size * FLAGS.num_channels * FLAGS.image_size * FLAGS.image_size)
kl_divergence_loss_1.backward(retain_graph=True)
_, __, class_2 = encoder(Variable(X_2))
reconstructed_X_1 = decoder(style_1, class_1)
reconstructed_X_2 = decoder(style_1, class_2)
reconstruction_error_1 = mse_loss(reconstructed_X_1, Variable(X_1))
reconstruction_error_1.backward(retain_graph=True)
reconstruction_error_2 = mse_loss(reconstructed_X_2, Variable(X_1))
reconstruction_error_2.backward()
reconstruction_error = reconstruction_error_1 + reconstruction_error_2
kl_divergence_error = kl_divergence_loss_1
auto_encoder_optimizer.step()
# B. run the generator
for i in range(FLAGS.generator_times):
generator_optimizer.zero_grad()
image_batch_1, _, __ = next(loader)
image_batch_3, _, __ = next(loader)
domain_labels.fill_(real_domain_labels)
X_1.copy_(image_batch_1)
X_3.copy_(image_batch_3)
style_mu_1, style_logvar_1, _ = encoder(Variable(X_1))
style_1 = reparameterize(training=True, mu=style_mu_1, logvar=style_logvar_1)
kl_divergence_loss_1 = - 0.5 * torch.sum(1 + style_logvar_1 - style_mu_1.pow(2) - style_logvar_1.exp())
kl_divergence_loss_1 /= (FLAGS.batch_size * FLAGS.num_channels * FLAGS.image_size * FLAGS.image_size)
kl_divergence_loss_1.backward(retain_graph=True)
_, __, class_3 = encoder(Variable(X_3))
reconstructed_X_1_3 = decoder(style_1, class_3)
output_1 = discriminator(Variable(X_3), reconstructed_X_1_3)
generator_error_1 = cross_entropy_loss(output_1, Variable(domain_labels))
generator_error_1.backward(retain_graph=True)
style_latent_space.normal_(0., 1.)
reconstructed_X_latent_3 = decoder(Variable(style_latent_space), class_3)
output_2 = discriminator(Variable(X_3), reconstructed_X_latent_3)
generator_error_2 = cross_entropy_loss(output_2, Variable(domain_labels))
generator_error_2.backward()
generator_error = generator_error_1 + generator_error_2
kl_divergence_error += kl_divergence_loss_1
generator_optimizer.step()
# C. run the discriminator
for i in range(FLAGS.discriminator_times):
discriminator_optimizer.zero_grad()
# train discriminator on real data
domain_labels.fill_(real_domain_labels)
image_batch_1, _, __ = next(loader)
image_batch_2, image_batch_3, _ = next(loader)
X_1.copy_(image_batch_1)
X_2.copy_(image_batch_2)
X_3.copy_(image_batch_3)
real_output = discriminator(Variable(X_2), Variable(X_3))
discriminator_real_error = cross_entropy_loss(real_output, Variable(domain_labels))
discriminator_real_error.backward()
# train discriminator on fake data
domain_labels.fill_(fake_domain_labels)
style_mu_1, style_logvar_1, _ = encoder(Variable(X_1))
style_1 = reparameterize(training=False, mu=style_mu_1, logvar=style_logvar_1)
_, __, class_3 = encoder(Variable(X_3))
reconstructed_X_1_3 = decoder(style_1, class_3)
fake_output = discriminator(Variable(X_3), reconstructed_X_1_3)
discriminator_fake_error = cross_entropy_loss(fake_output, Variable(domain_labels))
discriminator_fake_error.backward()
# total discriminator error
discriminator_error = discriminator_real_error + discriminator_fake_error
# calculate discriminator accuracy for this step
target_true_labels = torch.cat((torch.ones(FLAGS.batch_size), torch.zeros(FLAGS.batch_size)), dim=0)
if FLAGS.cuda:
target_true_labels = target_true_labels.cuda()
discriminator_predictions = torch.cat((real_output, fake_output), dim=0)
_, discriminator_predictions = torch.max(discriminator_predictions, 1)
discriminator_accuracy = (discriminator_predictions.data == target_true_labels.long()
).sum().item() / (FLAGS.batch_size * 2)
if discriminator_accuracy < FLAGS.discriminator_limiting_accuracy:
discriminator_optimizer.step()
if (iteration + 1) % 50 == 0:
print('')
print('Epoch #' + str(epoch))
print('Iteration #' + str(iteration))
print('')
print('Reconstruction loss: ' + str(reconstruction_error.data.storage().tolist()[0]))
print('KL-Divergence loss: ' + str(kl_divergence_error.data.storage().tolist()[0]))
print('')
print('Generator loss: ' + str(generator_error.data.storage().tolist()[0]))
print('Discriminator loss: ' + str(discriminator_error.data.storage().tolist()[0]))
print('Discriminator accuracy: ' + str(discriminator_accuracy))
print('..........')
# write to log
with open(FLAGS.log_file, 'a') as log:
log.write('{0}\t{1}\t{2}\t{3}\t{4}\t{5}\t{6}\n'.format(
epoch,
iteration,
reconstruction_error.data.storage().tolist()[0],
kl_divergence_error.data.storage().tolist()[0],
generator_error.data.storage().tolist()[0],
discriminator_error.data.storage().tolist()[0],
discriminator_accuracy
))
# write to tensorboard
writer.add_scalar('Reconstruction loss', reconstruction_error.data.storage().tolist()[0],
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) + iteration)
writer.add_scalar('KL-Divergence loss', kl_divergence_error.data.storage().tolist()[0],
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) + iteration)
writer.add_scalar('Generator loss', generator_error.data.storage().tolist()[0],
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) + iteration)
writer.add_scalar('Discriminator loss', discriminator_error.data.storage().tolist()[0],
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) + iteration)
writer.add_scalar('Discriminator accuracy', discriminator_accuracy * 100,
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) + iteration)
# save model after every 5 epochs
if (epoch + 1) % 5 == 0 or (epoch + 1) == FLAGS.end_epoch:
torch.save(encoder.state_dict(), os.path.join('checkpoints', FLAGS.encoder_save))
torch.save(decoder.state_dict(), os.path.join('checkpoints', FLAGS.decoder_save))
torch.save(discriminator.state_dict(), os.path.join('checkpoints', FLAGS.discriminator_save))
def training_procedure(FLAGS):
"""
model definition
"""
encoder = Encoder(nv_dim=FLAGS.nv_dim, nc_dim=FLAGS.nc_dim)
encoder.apply(weights_init)
decoder = Decoder(nv_dim=FLAGS.nv_dim, nc_dim=FLAGS.nc_dim)
decoder.apply(weights_init)
discriminator = Discriminator()
discriminator.apply(weights_init)
# load saved models if load_saved flag is true
if FLAGS.load_saved:
encoder.load_state_dict(
torch.load(os.path.join('checkpoints', FLAGS.encoder_save)))
decoder.load_state_dict(
torch.load(os.path.join('checkpoints', FLAGS.decoder_save)))
discriminator.load_state_dict(
torch.load(os.path.join('checkpoints', FLAGS.discriminator_save)))
"""
variable definition
"""
real_domain_labels = 1
fake_domain_labels = 0
X_1 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels,
FLAGS.image_size, FLAGS.image_size)
X_2 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels,
FLAGS.image_size, FLAGS.image_size)
X_3 = torch.FloatTensor(FLAGS.batch_size, FLAGS.num_channels,
FLAGS.image_size, FLAGS.image_size)
domain_labels = torch.LongTensor(FLAGS.batch_size)
"""
loss definitions
"""
cross_entropy_loss = nn.CrossEntropyLoss()
'''
add option to run on GPU
'''
if FLAGS.cuda:
encoder.cuda()
decoder.cuda()
discriminator.cuda()
cross_entropy_loss.cuda()
X_1 = X_1.cuda()
X_2 = X_2.cuda()
X_3 = X_3.cuda()
domain_labels = domain_labels.cuda()
"""
optimizer definition
"""
auto_encoder_optimizer = optim.Adam(list(encoder.parameters()) +
list(decoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2))
discriminator_optimizer = optim.Adam(list(discriminator.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2))
generator_optimizer = optim.Adam(list(encoder.parameters()) +
list(decoder.parameters()),
lr=FLAGS.initial_learning_rate,
betas=(FLAGS.beta_1, FLAGS.beta_2))
"""
training
"""
if torch.cuda.is_available() and not FLAGS.cuda:
print(
"WARNING: You have a CUDA device, so you should probably run with --cuda"
)
if not os.path.exists('checkpoints'):
os.makedirs('checkpoints')
if not os.path.exists('reconstructed_images'):
os.makedirs('reconstructed_images')
# load_saved is false when training is started from 0th iteration
if not FLAGS.load_saved:
with open(FLAGS.log_file, 'w') as log:
log.write('Epoch\tIteration\tReconstruction_loss\t')
log.write(
'Generator_loss\tDiscriminator_loss\tDiscriminator_accuracy\n')
# load data set and create data loader instance
print('Loading MNIST paired dataset...')
paired_mnist = MNIST_Paired(root='mnist',
download=True,
train=True,
transform=transform_config)
loader = cycle(
DataLoader(paired_mnist,
batch_size=FLAGS.batch_size,
shuffle=True,
num_workers=0,
drop_last=True))
# initialise variables
discriminator_accuracy = 0.
# initialize summary writer
writer = SummaryWriter()
for epoch in range(FLAGS.start_epoch, FLAGS.end_epoch):
print('')
print(
'Epoch #' + str(epoch) +
'..........................................................................'
)
for iteration in range(int(len(paired_mnist) / FLAGS.batch_size)):
# A. run the auto-encoder reconstruction
image_batch_1, image_batch_2, labels_batch_1 = next(loader)
auto_encoder_optimizer.zero_grad()
X_1.copy_(image_batch_1)
X_2.copy_(image_batch_2)
nv_1, nc_1 = encoder(Variable(X_1))
nv_2, nc_2 = encoder(Variable(X_2))
reconstructed_X_1 = decoder(nv_1, nc_2)
reconstructed_X_2 = decoder(nv_2, nc_1)
reconstruction_error_1 = mse_loss(reconstructed_X_1, Variable(X_1))
reconstruction_error_1.backward(retain_graph=True)
reconstruction_error_2 = mse_loss(reconstructed_X_2, Variable(X_2))
reconstruction_error_2.backward()
reconstruction_error = reconstruction_error_1 + reconstruction_error_2
if FLAGS.train_auto_encoder:
auto_encoder_optimizer.step()
# B. run the adversarial part of the architecture
# B. a) run the discriminator
for i in range(FLAGS.discriminator_times):
discriminator_optimizer.zero_grad()
# train discriminator on real data
domain_labels.fill_(real_domain_labels)
image_batch_1, image_batch_2, labels_batch_1 = next(loader)
X_1.copy_(image_batch_1)
X_2.copy_(image_batch_2)
real_output = discriminator(Variable(X_1), Variable(X_2))
discriminator_real_error = FLAGS.disc_coef * cross_entropy_loss(
real_output, Variable(domain_labels))
discriminator_real_error.backward()
# train discriminator on fake data
domain_labels.fill_(fake_domain_labels)
image_batch_3, _, labels_batch_3 = next(loader)
X_3.copy_(image_batch_3)
nv_3, nc_3 = encoder(Variable(X_3))
# reconstruction is taking common factor from X_1 and varying factor from X_3
reconstructed_X_3_1 = decoder(nv_3, encoder(Variable(X_1))[1])
fake_output = discriminator(Variable(X_1), reconstructed_X_3_1)
discriminator_fake_error = FLAGS.disc_coef * cross_entropy_loss(
fake_output, Variable(domain_labels))
discriminator_fake_error.backward()
# total discriminator error
discriminator_error = discriminator_real_error + discriminator_fake_error
# calculate discriminator accuracy for this step
target_true_labels = torch.cat((torch.ones(
FLAGS.batch_size), torch.zeros(FLAGS.batch_size)),
dim=0)
if FLAGS.cuda:
target_true_labels = target_true_labels.cuda()
discriminator_predictions = torch.cat(
(real_output, fake_output), dim=0)
_, discriminator_predictions = torch.max(
discriminator_predictions, 1)
discriminator_accuracy = (discriminator_predictions.data
== target_true_labels.long()).sum(
).item() / (FLAGS.batch_size * 2)
if discriminator_accuracy < FLAGS.discriminator_limiting_accuracy and FLAGS.train_discriminator:
discriminator_optimizer.step()
# B. b) run the generator
for i in range(FLAGS.generator_times):
generator_optimizer.zero_grad()
image_batch_1, _, labels_batch_1 = next(loader)
image_batch_3, __, labels_batch_3 = next(loader)
domain_labels.fill_(real_domain_labels)
X_1.copy_(image_batch_1)
X_3.copy_(image_batch_3)
nv_3, nc_3 = encoder(Variable(X_3))
# reconstruction is taking common factor from X_1 and varying factor from X_3
reconstructed_X_3_1 = decoder(nv_3, encoder(Variable(X_1))[1])
output = discriminator(Variable(X_1), reconstructed_X_3_1)
generator_error = FLAGS.gen_coef * cross_entropy_loss(
output, Variable(domain_labels))
generator_error.backward()
if FLAGS.train_generator:
generator_optimizer.step()
# print progress after 10 iterations
if (iteration + 1) % 10 == 0:
print('')
print('Epoch #' + str(epoch))
print('Iteration #' + str(iteration))
print('')
print('Reconstruction loss: ' +
str(reconstruction_error.data.storage().tolist()[0]))
print('Generator loss: ' +
str(generator_error.data.storage().tolist()[0]))
print('')
print('Discriminator loss: ' +
str(discriminator_error.data.storage().tolist()[0]))
print('Discriminator accuracy: ' + str(discriminator_accuracy))
print('..........')
# write to log
with open(FLAGS.log_file, 'a') as log:
log.write('{0}\t{1}\t{2}\t{3}\t{4}\t{5}\n'.format(
epoch, iteration,
reconstruction_error.data.storage().tolist()[0],
generator_error.data.storage().tolist()[0],
discriminator_error.data.storage().tolist()[0],
discriminator_accuracy))
# write to tensorboard
writer.add_scalar(
'Reconstruction loss',
reconstruction_error.data.storage().tolist()[0],
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) +
iteration)
writer.add_scalar(
'Generator loss',
generator_error.data.storage().tolist()[0],
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) +
iteration)
writer.add_scalar(
'Discriminator loss',
discriminator_error.data.storage().tolist()[0],
epoch * (int(len(paired_mnist) / FLAGS.batch_size) + 1) +
iteration)
# save model after every 5 epochs
if (epoch + 1) % 5 == 0 or (epoch + 1) == FLAGS.end_epoch:
torch.save(encoder.state_dict(),
os.path.join('checkpoints', FLAGS.encoder_save))
torch.save(decoder.state_dict(),
os.path.join('checkpoints', FLAGS.decoder_save))
torch.save(discriminator.state_dict(),
os.path.join('checkpoints', FLAGS.discriminator_save))
"""
save reconstructed images and style swapped image generations to check progress
"""
image_batch_1, image_batch_2, labels_batch_1 = next(loader)
image_batch_3, _, __ = next(loader)
X_1.copy_(image_batch_1)
X_2.copy_(image_batch_2)
X_3.copy_(image_batch_3)
nv_1, nc_1 = encoder(Variable(X_1))
nv_2, nc_2 = encoder(Variable(X_2))
nv_3, nc_3 = encoder(Variable(X_3))
reconstructed_X_1 = decoder(nv_1, nc_2)
reconstructed_X_3_2 = decoder(nv_3, nc_2)
# save input image batch
image_batch = np.transpose(X_1.cpu().numpy(), (0, 2, 3, 1))
image_batch = np.concatenate(
(image_batch, image_batch, image_batch), axis=3)
imshow_grid(image_batch, name=str(epoch) + '_original', save=True)
# save reconstructed batch
reconstructed_x = np.transpose(
reconstructed_X_1.cpu().data.numpy(), (0, 2, 3, 1))
reconstructed_x = np.concatenate(
(reconstructed_x, reconstructed_x, reconstructed_x), axis=3)
imshow_grid(reconstructed_x,
name=str(epoch) + '_target',
save=True)
# save cross reconstructed batch
style_batch = np.transpose(X_3.cpu().numpy(), (0, 2, 3, 1))
style_batch = np.concatenate(
(style_batch, style_batch, style_batch), axis=3)
imshow_grid(style_batch, name=str(epoch) + '_style', save=True)
reconstructed_style = np.transpose(
reconstructed_X_3_2.cpu().data.numpy(), (0, 2, 3, 1))
reconstructed_style = np.concatenate(
(reconstructed_style, reconstructed_style,
reconstructed_style),
axis=3)
imshow_grid(reconstructed_style,
name=str(epoch) + '_style_target',
save=True)
class DoubleDQN:
def __init__(self, env, tau=0.1, gamma=0.9, epsilon=1.0):
self.env = env
self.tau = tau
self.gamma = gamma
self.embedding_size = 30
self.hidden_size = 30
self.obs_shape = self.env.get_obs().shape
self.action_shape = 40 // 5
if args.encoding == "onehot":
self.encoder = OneHot(
args.bins,
self.env.all_questions + self.env.held_out_questions,
self.hidden_size).to(DEVICE)
else:
self.encoder = Encoder(self.embedding_size,
self.hidden_size).to(DEVICE)
self.model = DQN(self.obs_shape, self.action_shape,
self.encoder).to(DEVICE)
self.target_model = DQN(self.obs_shape, self.action_shape,
self.encoder).to(DEVICE)
self.optimizer = torch.optim.Adam(self.model.parameters())
self.epsilon = epsilon
if os.path.exists(MODEL_FILE):
checkpoint = torch.load(MODEL_FILE)
self.encoder.load_state_dict(checkpoint['encoder_state_dict'])
self.model.load_state_dict(checkpoint['model_state_dict'])
self.target_model.load_state_dict(
checkpoint['target_model_state_dict'])
self.optimizer.load_state_dict(checkpoint['optimizer_state_dict'])
self.epsilon = checkpoint['epsilon']
# hard copy model parameters to target model parameters
for target_param, param in zip(self.model.parameters(),
self.target_model.parameters()):
target_param.data.copy_(param)
def get_action(self, state, goal):
assert len(state.shape
) == 2 # This function should not be called during update
if (np.random.rand() > self.epsilon):
q_values = self.model.forward(state, goal)
idx = torch.argmax(q_values).detach()
obj_selection = idx // 8
direction_selection = idx % 8
else:
action = self.env.sample_random_action()
obj_selection = action[0]
direction_selection = action[1]
return int(obj_selection), int(direction_selection)
def compute_loss(self, batch):
states, actions, goals, rewards, next_states, satisfied_goals, dones = batch
rewards = torch.FloatTensor(rewards).to(DEVICE)
dones = torch.FloatTensor(dones).to(DEVICE)
curr_Q = self.model(states, goals)
curr_Q_prev_actions = [
curr_Q[batch, actions[batch][0], actions[batch][1]]
for batch in range(len(states))
] # TODO: Use pytorch gather
curr_Q_prev_actions = torch.stack(curr_Q_prev_actions)
next_Q = self.target_model(next_states, goals)
next_Q_max_actions = torch.max(next_Q, -1).values
next_Q_max_actions = torch.max(next_Q_max_actions, -1).values
next_Q_max_actions = rewards + (
1 - dones) * self.gamma * next_Q_max_actions
loss = F.mse_loss(curr_Q_prev_actions, next_Q_max_actions.detach())
return loss
def update(self, replay_buffer, batch_size):
for _ in range(UPDATE_STEPS):
batch = replay_buffer.sample(batch_size)
loss = self.compute_loss(batch)
self.optimizer.zero_grad()
loss.backward()
self.optimizer.step()
def update_target_net(self): # TODO: Check this function
# target network update
for target_param, param in zip(self.target_model.parameters(),
self.model.parameters()):
target_param.data.copy_(self.tau * param +
(1 - self.tau) * target_param)
def save_model(self):
torch.save(
{
'model_state_dict': self.model.state_dict(),
'target_model_state_dict': self.target_model.state_dict(),
'encoder_state_dict': self.encoder.state_dict(),
'optimizer_state_dict': self.optimizer.state_dict(),
'epsilon': self.epsilon
}, MODEL_FILE)
