result = open(
os.path.join(
args.result, "Office31_HAFN_" + args.task + '_' + args.post + '.' +
args.repeat + "_score.txt"), "a")
t_root = os.path.join(args.data_root, args.target, "images")
t_label = os.path.join(args.data_root, args.target, "label.txt")
data_transform = transforms.Compose([
transforms.Scale((256, 256)),
transforms.CenterCrop((224, 224)),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
])
t_set = OfficeImage(t_root, t_label, data_transform)
# assert len(t_set) == 795
t_loader = torch.utils.data.DataLoader(t_set,
batch_size=args.batch_size,
shuffle=args.shuffle,
num_workers=args.num_workers)
netG = ResBase50().cuda()
netF = ResClassifier(class_num=args.class_num,
extract=False,
dropout_p=args.dropout_p).cuda()
netG.eval()
netF.eval()
for epoch in range(args.epoch / 2, args.epoch + 1):
if epoch % 10 != 0:
s2_weight = args.s2_weight
lr = args.lr
beta1 = args.beta1
beta2 = args.beta2
gpu_id = args.gpu_id
num_classes = args.num_classes
threshold = args.threshold
log_interval = args.log_interval
cls_epoches = args.cls_epoches
gan_epoches = args.gan_epoches
s1_label = "/home/silvia/MSDA/A_W_2_D_Open/bvlc_A_W_2_D/data/amazon.txt"
s2_label = "/home/silvia/MSDA/A_W_2_D_Open/bvlc_A_W_2_D/data/webcam.txt"
t_label = "/home/silvia/MSDA/A_W_2_D_Open/bvlc_A_W_2_D/data/dslr.txt"
s1_set = OfficeImage(s1_label, split="train")
s2_set = OfficeImage(s2_label, split="train")
t_set = OfficeImage(t_label, split="train")
t_set_test = OfficeImage(t_label, split="test")
assert len(s1_set) == 2817
assert len(s2_set) == 795
assert len(t_set) == 498
assert len(t_set_test) == 498
s1_loader_raw = torch.utils.data.DataLoader(s1_set,
batch_size=batch_size,
shuffle=shuffle,
num_workers=num_workers)
s2_loader_raw = torch.utils.data.DataLoader(s2_set,
batch_size=batch_size,
log_file = osp.join(
log_dir,
os.path.abspath(__file__).split('/')[-1].split('.')[0] + '.txt')
if os.path.isfile(log_file):
os.remove(log_file)
logger = get_log(log_file)
s1_root = os.path.join(data_root, args.s1)
s1_label = os.path.join(data_root, args.s1 + ".txt")
s2_root = os.path.join(data_root, args.s2)
s2_label = os.path.join(data_root, args.s2 + ".txt")
s3_root = os.path.join(data_root, args.s3)
s3_label = os.path.join(data_root, args.s3 + ".txt")
t_root = os.path.join(data_root, args.t)
t_label = os.path.join(data_root, args.t + ".txt")
s1_set = OfficeImage(s1_root, s1_label, split="train")
s2_set = OfficeImage(s2_root, s2_label, split="train")
s3_set = OfficeImage(s3_root, s3_label, split="train")
t_set = OfficeImage(t_root, t_label, split="train")
t_set_test = OfficeImage(t_root, t_label, split="test")
assert len(s1_set) == 2427
assert len(s2_set) == 4439
assert len(s3_set) == 4357
assert len(t_set) == 4365
assert len(t_set_test) == 4365
s1_loader_raw = torch.utils.data.DataLoader(s1_set,
batch_size=batch_size,
shuffle=shuffle,
num_workers=num_workers,
drop_last=True)
s2_loader_raw = torch.utils.data.DataLoader(s2_set,
print_args(args)
source_root = os.path.join(args.data_root, args.source, "images")
source_label = os.path.join(args.data_root, args.source, "label.txt")
target_root = os.path.join(args.data_root, args.target, "images")
target_label = os.path.join(args.data_root, args.target, "label.txt")
train_transform = transforms.Compose([
transforms.Scale((256, 256)),
transforms.RandomCrop((224, 224)),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
])
source_set = OfficeImage(source_root, source_label, train_transform)
target_set = OfficeImage(target_root, target_label, train_transform)
source_loader = torch.utils.data.DataLoader(source_set,
batch_size=args.batch_size,
shuffle=args.shuffle,
num_workers=args.num_workers)
target_loader = torch.utils.data.DataLoader(target_set,
batch_size=args.batch_size,
shuffle=args.shuffle,
num_workers=args.num_workers)
netG = ResBase50().cuda()
netF = ResClassifier(class_num=args.class_num,
extract=args.extract,
dropout_p=args.dropout_p).cuda()