Question about bringing specific class folder in torchvision.ImageFolder

vision
1

say there is directory , and there are query, reference folder inside test data directory.
what I want to do is make different dataset_loader for each query and reference images.
Need advice on this work. thanks

13513

2

I think the easiest way would be to write a custom Dataset and just load all images from the specified folder.
Here is a dummy snippet you could adapt to your use case:

class MyDataset(Dataset):
    def __init__(self, path, transform=None):
        self.image_paths = glob.glob(os.path.join(path, '*.jpg'))
        self.transform = transform
        
    def __getitem__(self, index):
        x = Image.open(self.image_paths[index])
        if self.transform:
            x = self.transform(x)
        return x
    
    def __len__(self):
        return len(self.image_paths)
    
query_folder = 'QUERY_PATH'
reference_folder = 'REFERENCE_PATH'
dataset_query = MyDataset(query_folder, transform=transforms.ToTensor())
dataset_reference = MyDataset(reference_folder, transform=transforms.ToTensor())

Have a look at the data loading tutorial for more information.

1 Like
3

thanks for the help!! It helped me a lot!!

1 Like