Learn2Learn's TieredImageNet taking long time to load, QuickDraw errors during transformation stage?

[patricks-lab]

(Reposted here from learnables slack since slack history is limited & isn't a permanent record)

Two questions (this is on a project I'm working w/ Brando):

1. I was wondering why TieredImageNet takes a long time to load (>10 mins, longer than other learn2learn datasets I've worked with). (FYI I had to manually download the dataset from kaggle - https://www.kaggle.com/datasets/andrijdavid/fsl-imagenet since the original gdrive link was broken). I just want to make sure this long time to load tieredImageNet is expected and normal, and not something that's wrong.

2. I was trying to use learn2learn's QuickDraw but it seems like I'm getting errors when I attempt to apply transformations (resizing to 84x84, random crop) on it. The issue stems from when l2l's quickdraw library attempts to apply transformations onto a quickdraw image (that is apparently in the form of a np.memmap/.npy record that PIL can't understand) and so I get the following error:

   Traceback (most recent call last):
   File "/home/pzy2/diversity-for-predictive-success-of-meta-learning/div_src/diversity_src/dataloaders/maml_patricks_l2l.py", line 2300, in <module>
   loop_through_l2l_indexable_benchmark_with_model_test()
   File "/home/pzy2/diversity-for-predictive-success-of-meta-learning/div_src/diversity_src/dataloaders/maml_patricks_l2l.py", line 2259, in loop_through_l2l_indexable_benchmark_with_model_test
   for benchmark in [quickdraw_l2l_tasksets()]: #hdb8_l2l_tasksets(),hdb9_l2l_tasksets(), delaunay_l2l_tasksets()]:#[dtd_l2l_tasksets(), cu_birds_l2l_tasksets(), fc100_l2l_tasksets()]:
   File "/home/pzy2/diversity-for-predictive-success-of-meta-learning/div_src/diversity_src/dataloaders/maml_patricks_l2l.py", line 2216, in quickdraw_l2l_tasksets
   _transforms: tuple[TaskTransform, TaskTransform, TaskTransform] = get_task_transforms_quickdraw(_datasets,
   File "/home/pzy2/diversity-for-predictive-success-of-meta-learning/div_src/diversity_src/dataloaders/maml_patricks_l2l.py", line 2184, in get_task_transforms_quickdraw
   train_transforms: TaskTransform = DifferentTaskTransformIndexableForEachDataset(train_dataset,
   File "/home/pzy2/diversity-for-predictive-success-of-meta-learning/div_src/diversity_src/dataloaders/common.py", line 130, in __init__
   self.indexable_dataset = MetaDataset(indexable_dataset)
   File "learn2learn/data/meta_dataset.pyx", line 59, in learn2learn.data.meta_dataset.MetaDataset.__init__
   File "learn2learn/data/meta_dataset.pyx", line 96, in learn2learn.data.meta_dataset.MetaDataset.create_bookkeeping
   File "learn2learn/data/meta_dataset.pyx", line 65, in learn2learn.data.meta_dataset.MetaDataset.__getitem__
   File "/home/pzy2/miniconda3/envs/metalearning3.9/lib/python3.9/site-packages/learn2learn/vision/datasets/quickdraw.py", line 511, in __getitem__
   image = self.transform(image)
   File "/home/pzy2/miniconda3/envs/metalearning3.9/lib/python3.9/site-packages/torchvision/transforms/transforms.py", line 60, in __call__
   img = t(img)
   File "/home/pzy2/miniconda3/envs/metalearning3.9/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
   return forward_call(*input, **kwargs)
   File "/home/pzy2/miniconda3/envs/metalearning3.9/lib/python3.9/site-packages/torchvision/transforms/transforms.py", line 900, in forward
   i, j, h, w = self.get_params(img, self.scale, self.ratio)
   File "/home/pzy2/miniconda3/envs/metalearning3.9/lib/python3.9/site-packages/torchvision/transforms/transforms.py", line 859, in get_params
   width, height = F._get_image_size(img)
   File "/home/pzy2/miniconda3/envs/metalearning3.9/lib/python3.9/site-packages/torchvision/transforms/functional.py", line 67, in _get_image_size
   return F_pil._get_image_size(img)
   File "/home/pzy2/miniconda3/envs/metalearning3.9/lib/python3.9/site-packages/torchvision/transforms/functional_pil.py", line 26, in _get_image_size
   raise TypeError("Unexpected type {}".format(type(img)))
   TypeError: Unexpected type <class 'numpy.memmap'>

[seba-1511]

Hello @patricks-lab,

1. Tiered-ImageNet is larger than some of the other datasets, but even so 10 minutes seems excessive. Maybe something is off with the Kaggle version of the dataset. I've been meaning to upload the learn2learn versions to Zenodo.
2. Could you provide a short colab replicating the issue? The fix might be as simple as parsing the memmap to an image with PIL.

Thanks,

[brando90]

cross: https://stackoverflow.com/questions/76082746/how-to-transform-a-quickdraw-image-to-84-by-84-in-pytorch-using-the-learn2learn and https://discuss.pytorch.org/t/how-to-transform-a-quickdraw-image-to-84-by-84-in-pytorch-using-the-learn2learn-library/178292

[seba-1511]

Closing since we'll be moving datasets to Zenodo in #400.