def __getitem__(self, index):
datafiles = self.files[index]
image = Image.open(datafiles["img"]).convert('RGB')
label = Image.open(datafiles["label"])
size = np.asarray(image).shape
name = datafiles["name"]
composed_transforms = transforms.Compose([
tr.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),
tr.ToTensor()])
sample = {'image': image, 'label': label}
sampled = composed_transforms(sample)
image, label = sampled['image'], sampled['label']
return image, label, np.array(size), name
def __getitem__(self, index):
datafiles = self.files[index]
image = Image.open(datafiles["img"]).convert('RGB')
label = Image.open(datafiles["label"])
size = np.asarray(image).shape
name = datafiles["name"]
composed_transforms = transforms.Compose([
tr.RandomHorizontalFlip(),
# tr.RandomRotate(180),
tr.RandomScaleCrop(base_size=self.base_size, crop_size=self.crop_size, fill=255),
tr.RandomGaussianBlur(),
tr.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),
tr.ToTensor()])
sample = {'image': image, 'label': label}
sampled = composed_transforms(sample)
image, label = sampled['image'], sampled['label']
return image, label, np.array(size), name
elif args.model_name == 'FlowNet':
model = FlowNet(args)
elif args.model_name == 'CircleFlowSRNet':
model = CircleFlowSRNet(args)
elif args.model_name == 'TestSRNet':
model = TestSRNet(args)
elif args.model_name == 'TestCircleSRNet':
model = TestCircleSRNet(args)
elif args.model_name == 'TestGNet':
model = TestGNet(args)
elif args.model_name == 'TestContinueNet':
model = TestContinueNet(args)
else:
raise ValueError('no model named: {}'.format(args.model_name))
transform = transforms.Compose([transforms.Normalize()])
transform = transforms.Compose([transforms.Normalize()])
if args.model_name.find('Un') >= 0:
dataset = Unsupervised_Dataset_Loader(args.test_dir, args.batch_size,
args.scale, args.im_crop_H,
args.im_crop_W, transform,
args.random_crop)
else:
#dataset = Dataset_Loader(args.test_dir, args.adv_data_dir,args.batch_size,args.im_crop_H,args.im_crop_W,transform,args.random_crop)
dataset = Dataset_Loader(args.test_dir, args.batch_size,
args.im_crop_H, args.im_crop_W, transform,
args.random_crop)
if args.test_new:
#dataset = Dataset_Loader(args.test_dir, args.adv_data_dir,args.batch_size,args.im_crop_H,args.im_crop_W,transform,args.random_crop)
test_dataloader = torch.utils.data.DataLoader(