code/train_shape.py
@@ -116,7 +116,7 @@ transformations = transforms.Compose([transforms.Scale(224),transforms.RandomCrop(224), transforms.ToTensor()]) pose_dataset = datasets.300W_LP(args.data_dir, args.filename_list, pose_dataset = datasets.Pose_300W_LP(args.data_dir, args.filename_list, transformations) train_loader = torch.utils.data.DataLoader(dataset=pose_dataset, batch_size=batch_size,
