from squid.loss import VGGLoss from squid.net import DnCnn target_net = DnCnn(layer_num=20) target_net = nn.DataParallel(target_net).cuda() model = SuperviseModel({ 'net': target_net, 'optimizer': torch.optim.Adam([{'name':'net_params', 'params':target_net.parameters(), 'base_lr':1e-4}], betas=(0.9, 0.999), weight_decay=0.0005), 'lr_step_ratio': 0.5, 'lr_step_size': 2, 'supervise':{ 'out': {'MSE_loss': {'obj': nn.MSELoss(size_average=True), 'factor':1.0, 'weight':1.0}}, }, 'metrics': {} }) train_dataset = Photo2PhotoData({ 'data_root': DATASET_DIR_TRAIN, 'desc_file_path': os.path.join(DATASET_TXT_DIR_TRAIN, DATASET_ID, '9_train.txt'), }) valid_dataset = Photo2PhotoData({ 'data_root': DATASET_DIR_VAL, 'desc_file_path': os.path.join(DATASET_TXT_DIR_VAL, DATASET_ID, 'val.txt'), })
5, 'supervise': { 'out': { 'MSE_loss': { 'obj': nn.MSELoss(size_average=True), 'factor': 0.1, 'weight': 1.0 } }, }, 'metrics': {} }) train_dataset = RandomCropPhoto2PhotoData({ 'crop_size': hr_size, 'crop_stride': 2, 'data_root': DATASET_DIR, 'desc_file_path': os.path.join(DATASET_TXT_DIR, DATASET_ID, 'train.txt'), }) valid_dataset = Photo2PhotoData({ 'data_root': DATASET_DIR, 'desc_file_path': os.path.join(DATASET_TXT_DIR, DATASET_ID, 'val.txt'), })