parser.add_argument('--seed', type=float, default=0)
args = parser.parse_args()
device = torch.device('cuda:' +
str(args.gpu) if torch.cuda.is_available() else 'cpu')
# Data
torch.manual_seed(args.seed)
if device == 'cuda':
cudnn.benchmark = True
torch.cuda.manual_seed(args.seed)
print('load data: ', args.dataset)
train_loader, test_loader = data_loader.getDataSet(args.dataset,
args.batch_size,
args.test_batch_size,
args.imageSize)
# Model
print('==> Building model..')
net = models.Resnet()
net = net.to(device)
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(),
lr=args.lr,
momentum=0.9,
weight_decay=5e-4)
# Training
parser.add_argument('--decreasing_lr2', default=[15, 30], nargs='+', help='decreasing strategy')
args = parser.parse_args()
device = torch.device('cuda:' + str(args.gpu) if torch.cuda.is_available() else 'cpu')
torch.manual_seed(args.seed)
random.seed(args.seed)
if device == 'cuda':
cudnn.benchmark = True
torch.cuda.manual_seed(args.seed)
print('load in-domain data: ',args.dataset_inDomain)
train_loader_inDomain, test_loader_inDomain = data_loader.getDataSet(args.dataset_inDomain, args.batch_size, args.test_batch_size, args.imageSize)
# Model
print('==> Building model..')
net = models.SDENet_mnist(layer_depth=6, num_classes=10, dim=64)
net = net.to(device)
real_label = 0
fake_label = 1
criterion = nn.CrossEntropyLoss()
criterion2 = nn.BCELoss()
optimizer_F = optim.SGD([ {'params': net.downsampling_layers.parameters()}, {'params': net.drift.parameters()},
{'params': net.fc_layers.parameters()}], lr=args.lr, momentum=0.9, weight_decay=5e-4)