def test(args):
# ---------- load model_real_cartoon ---------- #
rc_e1 = E1(args.sep, int((args.resize / 64)))
rc_e2 = E2(args.sep, int((args.resize / 64)))
rc_decoder = Decoder(int((args.resize / 64)))
if torch.cuda.is_available():
rc_e1 = rc_e1.cuda()
rc_e2 = rc_e2.cuda()
rc_decoder = rc_decoder.cuda()
if args.load_rc != '':
save_file = os.path.join(args.load_rc)
load_model_for_eval(save_file, rc_e1, rc_e2, rc_decoder)
rc_e1 = rc_e1.eval()
rc_e2 = rc_e2.eval()
rc_decoder = rc_decoder.eval()
# ---------- load model_cartoon ---------- #
c_e1 = E1(args.sep, int((args.resize / 64)))
c_e2 = E2(args.sep, int((args.resize / 64)))
c_decoder = Decoder(int((args.resize / 64)))
if torch.cuda.is_available():
c_e1 = c_e1.cuda()
c_e2 = c_e2.cuda()
c_decoder = c_decoder.cuda()
if args.load_c != '':
save_file = os.path.join(args.load_c)
load_model_for_eval(save_file, c_e1, c_e2, c_decoder)
c_e1 = c_e1.eval()
c_e2 = c_e2.eval()
c_decoder = c_decoder.eval()
# -------------- running -------------- #
if not os.path.exists(args.out) and args.out != "":
os.mkdir(args.out)
# trans(args, rc_e1, rc_e2, rc_decoder, c_e1, c_e2, c_decoder)
test_domA_cluster, test_domB_cluster = my_get_test_imgs(args)
for idx, (test_domA, test_domB) in enumerate(
list(zip(test_domA_cluster, test_domB_cluster))):
trans(args, idx, test_domA, test_domB, rc_e1, rc_e2, rc_decoder, c_e1,
c_e2, c_decoder)
def test(args):
# ---------- load model_real_cartoon ---------- #
rc_e1 = E1(args.sep, int((args.resize / 64)))
rc_e2 = E2(args.sep, int((args.resize / 64)))
rc_decoder = Decoder(int((args.resize / 64)))
if torch.cuda.is_available():
rc_e1 = rc_e1.cuda()
rc_e2 = rc_e2.cuda()
rc_decoder = rc_decoder.cuda()
if args.load_rc != '':
save_file = os.path.join(args.load_rc)
load_model_for_eval(save_file, rc_e1, rc_e2, rc_decoder)
rc_e1 = rc_e1.eval()
rc_e2 = rc_e2.eval()
rc_decoder = rc_decoder.eval()
# ---------- load model_cartoon ---------- #
c_e1 = E1(args.sep, int((args.resize / 64)))
c_e2 = E2(args.sep, int((args.resize / 64)))
c_decoder = Decoder(int((args.resize / 64)))
if torch.cuda.is_available():
c_e1 = c_e1.cuda()
c_e2 = c_e2.cuda()
c_decoder = c_decoder.cuda()
if args.load_c != '':
save_file = os.path.join(args.load_c)
load_model_for_eval(save_file, c_e1, c_e2, c_decoder)
c_e1 = c_e1.eval()
c_e2 = c_e2.eval()
c_decoder = c_decoder.eval()
# -------------- running -------------- #
if not os.path.exists(args.out) and args.out != "":
os.mkdir(args.out)
trans(args, rc_e1, rc_e2, rc_decoder, c_e1, c_e2, c_decoder)
def get_eval_model(load, sep, resize):
e1 = E1(sep, int((resize / 64)))
e2 = E2(sep, int((resize / 64)))
decoder = Decoder(int((resize / 64)))
if torch.cuda.is_available():
e1 = e1.cuda()
e2 = e2.cuda()
decoder = decoder.cuda()
_iter = load_model_for_eval(load, e1, e2, decoder)
e1 = e1.eval()
e2 = e2.eval()
decoder = decoder.eval()
return e1, e2, decoder
def eval(args):
e1 = E1(args.sep, int((args.resize / 64)))
e2 = E2(args.sep, int((args.resize / 64)))
decoder = Decoder(int((args.resize / 64)))
if torch.cuda.is_available():
e1 = e1.cuda()
e2 = e2.cuda()
decoder = decoder.cuda()
if args.load != '':
save_file = os.path.join(args.load, 'checkpoint')
_iter = load_model_for_eval(save_file, e1, e2, decoder)
e1 = e1.eval()
e2 = e2.eval()
decoder = decoder.eval()
if not os.path.exists(args.out) and args.out != "":
os.mkdir(args.out)
save_imgs(args, e1, e2, decoder, _iter)
def train(args):
if not os.path.exists(args.out):
os.makedirs(args.out)
_iter = 0
comp_transformA = transforms.Compose([
transforms.CenterCrop(args.cropA),
transforms.Resize(args.resize),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])
comp_transformB = transforms.Compose([
transforms.CenterCrop(args.cropB),
transforms.Resize(args.resize),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])
domA_train = CustomDataset(os.path.join(args.root, 'trainA.txt'), transform=comp_transformA)
domB_train = CustomDataset(os.path.join(args.root, 'trainB.txt'), transform=comp_transformB)
A_label = torch.full((args.bs,), 1)
B_label = torch.full((args.bs,), 0)
B_separate = torch.full((args.bs, args.sep * (args.resize // 64) * (args.resize // 64)), 0)
e1 = E1(args.sep, args.resize // 64)
e2 = E2(args.sep, args.resize // 64)
decoder = Decoder(args.resize // 64)
disc = Disc(args.sep, args.resize // 64)
mse = nn.MSELoss()
bce = nn.BCELoss()
if torch.cuda.is_available():
e1 = e1.cuda()
e2 = e2.cuda()
decoder = decoder.cuda()
disc = disc.cuda()
A_label = A_label.cuda()
B_label = B_label.cuda()
B_separate = B_separate.cuda()
mse = mse.cuda()
bce = bce.cuda()
ae_params = list(e1.parameters()) + list(e2.parameters()) + list(decoder.parameters())
ae_optimizer = optim.Adam(ae_params, lr=args.lr, betas=(0.5, 0.999))
disc_params = disc.parameters()
disc_optimizer = optim.Adam(disc_params, lr=args.disclr, betas=(0.5, 0.999))
if args.load != '':
save_file = os.path.join(args.load, 'checkpoint')
_iter = load_model(save_file, e1, e2, decoder, ae_optimizer, disc, disc_optimizer)
e1 = e1.train()
e2 = e2.train()
decoder = decoder.train()
disc = disc.train()
print('Started training...')
while True:
domA_loader = torch.utils.data.DataLoader(domA_train, batch_size=args.bs,
shuffle=True, num_workers=6)
domB_loader = torch.utils.data.DataLoader(domB_train, batch_size=args.bs,
shuffle=True, num_workers=6)
if _iter >= args.iters:
break
for domA_img, domB_img in zip(domA_loader, domB_loader):
if domA_img.size(0) != args.bs or domB_img.size(0) != args.bs:
break
domA_img = Variable(domA_img)
domB_img = Variable(domB_img)
if torch.cuda.is_available():
domA_img = domA_img.cuda()
domB_img = domB_img.cuda()
domA_img = domA_img.view((-1, 3, args.resize, args.resize))
domB_img = domB_img.view((-1, 3, args.resize, args.resize))
ae_optimizer.zero_grad()
A_common = e1(domA_img)
A_separate = e2(domA_img)
A_encoding = torch.cat([A_common, A_separate], dim=1)
B_common = e1(domB_img)
B_encoding = torch.cat([B_common, B_separate], dim=1)
A_decoding = decoder(A_encoding)
B_decoding = decoder(B_encoding)
loss = mse(A_decoding, domA_img) + mse(B_decoding, domB_img)
if args.discweight > 0:
preds_A = disc(A_common)
preds_B = disc(B_common)
loss += args.discweight * (bce(preds_A, B_label) + bce(preds_B, B_label))
loss.backward()
torch.nn.utils.clip_grad_norm_(ae_params, 5)
ae_optimizer.step()
if args.discweight > 0:
disc_optimizer.zero_grad()
A_common = e1(domA_img)
B_common = e1(domB_img)
disc_A = disc(A_common)
disc_B = disc(B_common)
loss2 = bce(disc_A, A_label) + bce(disc_B, B_label)
loss2.backward()
torch.nn.utils.clip_grad_norm_(disc_params, 5)
disc_optimizer.step()
if _iter % args.progress_iter == 0:
print('Outfile: %s | Iteration %d | loss %.6f | loss1: %.6f | loss2: %.6f' % (args.out, _iter, loss+loss2, loss, loss2))
if _iter % args.display_iter == 0:
e1 = e1.eval()
e2 = e2.eval()
decoder = decoder.eval()
save_imgs(args, e1, e2, decoder, _iter)
e1 = e1.train()
e2 = e2.train()
decoder = decoder.train()
if _iter % args.save_iter == 0:
save_file = os.path.join(args.out, 'checkpoint_%d' % _iter)
save_model(save_file, e1, e2, decoder, ae_optimizer, disc, disc_optimizer, _iter)
_iter += 1
def train(args):
args.out = args.out + '_size_' + str(args.resize)
if args.sep > 0:
args.out = args.out + '_sep_' + str(args.sep)
if args.disc_weight > 0:
args.out = args.out + '_disc-weight_' + str(args.disc_weight)
if args.disc_lr != 0.0002:
args.out = args.out + '_disc-lr_' + str(args.disc_lr)
_iter = 0
comp_transform = transforms.Compose([
transforms.CenterCrop(args.crop),
transforms.Resize(args.resize),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])
domA_train = CustomImageFolder(root=os.path.join(args.dataroot, 'trainA'),
transform=comp_transform)
domB_train = CustomImageFolder(root=os.path.join(args.dataroot, 'trainB'),
transform=comp_transform)
A_label = torch.full((args.bs, ), 1)
B_label = torch.full((args.bs, ), 0)
B_separate = torch.full(
(args.bs, args.sep * (args.resize / 64) * (args.resize / 64)), 0)
e1 = E1(args.sep, int((args.resize / 64)))
e2 = E2(args.sep, int((args.resize / 64)))
decoder = Decoder(int((args.resize / 64)))
disc = Disc(args.sep, int((args.resize / 64)))
mse = nn.MSELoss()
bce = nn.BCELoss()
if torch.cuda.is_available():
e1 = e1.cuda()
e2 = e2.cuda()
decoder = decoder.cuda()
disc = disc.cuda()
A_label = A_label.cuda()
B_label = B_label.cuda()
B_separate = B_separate.cuda()
mse = mse.cuda()
bce = bce.cuda()
ae_params = list(e1.parameters()) + list(e2.parameters()) + list(
decoder.parameters())
ae_optimizer = optim.Adam(ae_params, lr=args.lr, betas=(0.5, 0.999))
disc_params = disc.parameters()
disc_optimizer = optim.Adam(disc_params,
lr=args.disc_lr,
betas=(0.5, 0.999))
if args.load != '':
save_file = os.path.join(args.load, 'checkpoint')
_iter = load_model(save_file, e1, e2, decoder, ae_optimizer, disc,
disc_optimizer)
e1 = e1.train()
e2 = e2.train()
decoder = decoder.train()
disc = disc.train()
while True:
domA_loader = torch.utils.data.DataLoader(domA_train,
batch_size=args.bs,
shuffle=True,
num_workers=6)
domB_loader = torch.utils.data.DataLoader(domB_train,
batch_size=args.bs,
shuffle=True,
num_workers=6)
if _iter >= args.iters:
break
for domA_img, domB_img in zip(domA_loader, domB_loader):
domA_img = Variable(domA_img)
domB_img = Variable(domB_img)
if torch.cuda.is_available():
domA_img = domA_img.cuda()
domB_img = domB_img.cuda()
domA_img = domA_img.view((-1, 3, args.resize, args.resize))
domB_img = domB_img.view((-1, 3, args.resize, args.resize))
ae_optimizer.zero_grad()
A_common = e1(domA_img)
A_separate = e2(domA_img)
A_encoding = torch.cat([A_common, A_separate], dim=1)
B_common = e1(domB_img)
B_encoding = torch.cat([B_common, B_separate], dim=1)
A_decoding = decoder(A_encoding)
B_decoding = decoder(B_encoding)
loss = mse(A_decoding, domA_img) + mse(B_decoding, domB_img)
if args.disc_weight > 0:
preds_A = disc(A_common)
preds_B = disc(B_common)
loss += args.disc_weight * (bce(preds_A, B_label) +
bce(preds_B, B_label))
loss.backward()
torch.nn.utils.clip_grad_norm_(ae_params, 5)
ae_optimizer.step()
if args.disc_weight > 0:
disc_optimizer.zero_grad()
A_common = e1(domA_img)
B_common = e1(domB_img)
disc_A = disc(A_common)
disc_B = disc(B_common)
loss = bce(disc_A, A_label) + bce(disc_B, B_label)
loss.backward()
torch.nn.utils.clip_grad_norm_(disc_params, 5)
disc_optimizer.step()
if _iter % args.progress_iter == 0:
print('Outfile: %s <<>> Iteration %d' % (args.out, _iter))
e1 = e1.eval()
e2 = e2.eval()
decoder = decoder.eval()
if _iter % args.display_iter == 0:
save_imgs(args, e1, e2, decoder)
e1 = e1.train()
e2 = e2.train()
decoder = decoder.train()
if _iter % args.save_iter == 0:
save_file = os.path.join(args.out, 'checkpoint')
save_model(save_file, e1, e2, decoder, ae_optimizer, disc,
disc_optimizer, _iter)
_iter += 1
def train(config):
if not os.path.exists(config.out):
os.makedirs(config.out)
comp_transform = transforms.Compose([
transforms.CenterCrop(config.crop),
transforms.Resize(config.resize),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])
domain_a_train = CustomDataset(os.path.join(config.root, 'trainA.txt'), transform=comp_transform)
domain_b_train = CustomDataset(os.path.join(config.root, 'trainB.txt'), transform=comp_transform)
a_label = torch.full((config.bs,), 1)
b_label = torch.full((config.bs,), 0)
b_separate = torch.full((config.bs,
config.sep,
config.resize // (2 ** (config.n_blocks + 1)),
config.resize // (2 ** (config.n_blocks + 1))), 0)
# build networks
e1 = E1(sep=config.sep, size=config.resize)
e2 = E2(n_feats=config.n_tot_feats, sep=config.sep)
decoder = Decoder(n_feats=config.n_tot_feats)
disc = Disc(size=config.resize, sep=config.sep)
rho_clipper = RhoClipper(0., 1.)
mse = nn.MSELoss()
bce = nn.BCELoss()
if torch.cuda.is_available():
e1 = e1.cuda()
e2 = e2.cuda()
decoder = decoder.cuda()
disc = disc.cuda()
a_label = a_label.cuda()
b_label = b_label.cuda()
b_separate = b_separate.cuda()
mse = mse.cuda()
bce = bce.cuda()
ae_params = list(e1.parameters()) + list(e2.parameters()) + list(decoder.parameters())
ae_optimizer = optim.Adam(ae_params, lr=config.lr,
betas=(config.beta1, config.beta2), eps=config.eps)
disc_params = disc.parameters()
disc_optimizer = optim.Adam(disc_params, lr=config.d_lr,
betas=(config.beta1, config.beta2), eps=config.eps)
_iter: int = 0
if config.load != '':
save_file = os.path.join(config.load, 'checkpoint')
_iter = load_model(save_file, e1, e2, decoder, ae_optimizer, disc, disc_optimizer)
e1 = e1.train()
e2 = e2.train()
decoder = decoder.train()
disc = disc.train()
print('[*] Started training...')
while True:
domain_a_loader = torch.utils.data.DataLoader(domain_a_train, batch_size=config.bs,
shuffle=True, num_workers=config.n_threads)
domain_b_loader = torch.utils.data.DataLoader(domain_b_train, batch_size=config.bs,
shuffle=True, num_workers=config.n_threads)
if _iter >= config.iters:
break
for domain_a_img, domain_b_img in zip(domain_a_loader, domain_b_loader):
if domain_a_img.size(0) != config.bs or domain_b_img.size(0) != config.bs:
break
domain_a_img = Variable(domain_a_img)
domain_b_img = Variable(domain_b_img)
if torch.cuda.is_available():
domain_a_img = domain_a_img.cuda()
domain_b_img = domain_b_img.cuda()
domain_a_img = domain_a_img.view((-1, 3, config.resize, config.resize))
domain_b_img = domain_b_img.view((-1, 3, config.resize, config.resize))
ae_optimizer.zero_grad()
a_common = e1(domain_a_img)
a_separate = e2(domain_a_img)
a_encoding = torch.cat([a_common, a_separate], dim=1)
b_common = e1(domain_b_img)
b_encoding = torch.cat([b_common, b_separate], dim=1)
a_decoding = decoder(a_encoding)
b_decoding = decoder(b_encoding)
g_loss = mse(a_decoding, domain_a_img) + mse(b_decoding, domain_b_img)
preds_a = disc(a_common)
preds_b = disc(b_common)
g_loss += config.adv_weight * (bce(preds_a, b_label) + bce(preds_b, b_label))
g_loss.backward()
torch.nn.utils.clip_grad_norm_(ae_params, 5.)
ae_optimizer.step()
disc_optimizer.zero_grad()
a_common = e1(domain_a_img)
b_common = e1(domain_b_img)
disc_a = disc(a_common)
disc_b = disc(b_common)
d_loss = bce(disc_a, a_label) + bce(disc_b, b_label)
d_loss.backward()
torch.nn.utils.clip_grad_norm_(disc_params, 5.)
disc_optimizer.step()
decoder.apply(rho_clipper)
if _iter % config.progress_iter == 0:
print('[*] [%07d/%07d] d_loss : %.4f, g_loss : %.4f' %
(_iter, config.iters, d_loss, g_loss))
if _iter % config.display_iter == 0:
e1 = e1.eval()
e2 = e2.eval()
decoder = decoder.eval()
save_images(config, e1, e2, decoder, _iter)
e1 = e1.train()
e2 = e2.train()
decoder = decoder.train()
if _iter % config.save_iter == 0:
save_file = os.path.join(config.out, 'checkpoint')
save_model(save_file, e1, e2, decoder, ae_optimizer, disc, disc_optimizer, _iter)
_iter += 1