])), batch_size=args.batchsize, shuffle=True, num_workers=args.nworkers
)
test_loader = torch.utils.data.DataLoader(
datasets.Imagenet32(train=False, transform=transforms.Compose([
add_noise,
])), batch_size=args.val_batchsize, shuffle=False, num_workers=args.nworkers
)
elif args.data == 'imagenet64':
im_dim = 3
init_layer = layers.LogitTransform(0.05)
if args.imagesize != 64:
logger.info('Changing image size to 64.')
args.imagesize = 64
train_loader = torch.utils.data.DataLoader(
datasets.Imagenet64(train=True, transform=transforms.Compose([
add_noise,
])), batch_size=args.batchsize, shuffle=True, num_workers=args.nworkers
)
test_loader = torch.utils.data.DataLoader(
datasets.Imagenet64(train=False, transform=transforms.Compose([
add_noise,
])), batch_size=args.val_batchsize, shuffle=False, num_workers=args.nworkers
)
if args.task in ['classification', 'hybrid']:
try:
n_classes
except NameError:
raise ValueError('Cannot perform classification with {}'.format(args.data))
else:
n_classes = 1
num_workers=args.nworkers)
test_loader = torch.utils.data.DataLoader(datasets.Imagenet32(
train=False, transform=transforms.Compose([
add_noise,
])),
batch_size=args.val_batchsize,
shuffle=False,
num_workers=args.nworkers)
elif args.data == 'imagenet64':
im_dim = 3
init_layer = layers.LogitTransform(0.05)
if args.imagesize != 64:
logger.info('Changing image size to 64.')
args.imagesize = 64
train_loader = torch.utils.data.DataLoader(datasets.Imagenet64(
train=True, transform=transforms.Compose([
add_noise,
])),
batch_size=args.batchsize,
shuffle=True,
num_workers=args.nworkers)
test_loader = torch.utils.data.DataLoader(datasets.Imagenet64(
train=False, transform=transforms.Compose([
add_noise,
])),
batch_size=args.val_batchsize,
shuffle=False,
num_workers=args.nworkers)
if args.task in ['classification', 'hybrid']:
try:
n_classes
