print(args)
if args.custom_img:
output_path = join('output', args.experiment_name, 'custom_testing')
from data import Custom
test_dataset = Custom(args.custom_data, args.custom_attr, args.img_size, 'test', args.attrs)
else:
output_path = join('output', args.experiment_name, 'sample_testing')
if args.data == 'CelebA':
from data import CelebA
test_dataset = CelebA(args.data_path, args.attr_path, args.img_size, 'test', args.attrs)
if args.data == 'CelebA-HQ':
from data import CelebA_HQ
test_dataset = CelebA_HQ(args.data_path, args.attr_path, args.image_list_path, args.img_size, 'test', args.attrs)
os.makedirs(output_path, exist_ok=True)
test_dataloader = data.DataLoader(
test_dataset, batch_size=1, num_workers=args.num_workers,
shuffle=False, drop_last=False
)
if args.num_test is None:
print('Testing images:', len(test_dataset))
else:
print('Testing images:', min(len(test_dataset), args.num_test))
attgan = AttGAN(args)
attgan.load(find_model(join('output', args.experiment_name, 'checkpoint'), args.load_epoch))
progressbar = Progressbar()
os.makedirs(join('output', args.experiment_name, 'checkpoint'), exist_ok=True)
os.makedirs(join('output', args.experiment_name, 'sample_training'),
exist_ok=True)
with open(join('output', args.experiment_name, 'setting.txt'), 'w') as f:
f.write(json.dumps(vars(args), indent=4, separators=(',', ':')))
if args.data == 'CelebA':
from data import CelebA
train_dataset = CelebA(args.data_path, args.attr_path, args.img_size,
'train', args.attrs)
valid_dataset = CelebA(args.data_path, args.attr_path, args.img_size,
'valid', args.attrs)
if args.data == 'CelebA-HQ':
from data import CelebA_HQ
train_dataset = CelebA_HQ(args.data_path, args.attr_path,
args.image_list_path, args.img_size, 'train',
args.attrs)
valid_dataset = CelebA_HQ(args.data_path, args.attr_path,
args.image_list_path, args.img_size, 'valid',
args.attrs)
train_dataloader = data.DataLoader(train_dataset,
batch_size=args.batch_size,
num_workers=args.num_workers,
shuffle=True,
drop_last=True)
valid_dataloader = data.DataLoader(valid_dataset,
batch_size=args.n_samples,
num_workers=args.num_workers,
shuffle=False,
drop_last=False)
print('Training images:', len(train_dataset), '/', 'Validating images:',