Error replicating 3D classification tutorial

[dgmato]

Hi all,

I want to replicate the tutorial titled "3D classification example based on DenseNet". I've downloaded the IXI dataset and tested the jupyter notebook.

However, I got an error message when creating the data loaders and I cannot train the network.

It seems to be related with the "EnsureChannelFirst()" transform: "ValueError: meta_dict not available, EnsureChannelFirst is not in use". I've tried with different versions of MONAI and PyTorch, and I was not able to solve this issue.

Could anyone help with this? How could I solve this issue?

Below you can find the entire error message:

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
Input In [4], in <cell line: 10>()
      7 check_ds = ImageDataset(image_files=images, labels=labels, transform=train_transforms)
      8 check_loader = DataLoader(check_ds, batch_size=3, num_workers=2, pin_memory=pin_memory)
---> 10 im, label = monai.utils.misc.first(check_loader)
     11 print(type(im), im.shape, label, label.shape)
     13 # create a training data loader

File ~\anaconda3\lib\site-packages\monai\utils\misc.py:72, in first(iterable, default)
     68 def first(iterable, default=None):
     69     """
     70     Returns the first item in the given iterable or `default` if empty, meaningful mostly with 'for' expressions.
     71     """
---> 72     for i in iterable:
     73         return i
     74     return default

File ~\anaconda3\lib\site-packages\torch\utils\data\dataloader.py:521, in _BaseDataLoaderIter.__next__(self)
    519 if self._sampler_iter is None:
    520     self._reset()
--> 521 data = self._next_data()
    522 self._num_yielded += 1
    523 if self._dataset_kind == _DatasetKind.Iterable and \
    524         self._IterableDataset_len_called is not None and \
    525         self._num_yielded > self._IterableDataset_len_called:

File ~\anaconda3\lib\site-packages\torch\utils\data\dataloader.py:1203, in _MultiProcessingDataLoaderIter._next_data(self)
   1201 else:
   1202     del self._task_info[idx]
-> 1203     return self._process_data(data)

File ~\anaconda3\lib\site-packages\torch\utils\data\dataloader.py:1229, in _MultiProcessingDataLoaderIter._process_data(self, data)
   1227 self._try_put_index()
   1228 if isinstance(data, ExceptionWrapper):
-> 1229     data.reraise()
   1230 return data

File ~\anaconda3\lib\site-packages\torch\_utils.py:434, in ExceptionWrapper.reraise(self)
    430 except TypeError:
    431     # If the exception takes multiple arguments, don't try to
    432     # instantiate since we don't know how to
    433     raise RuntimeError(msg) from None
--> 434 raise exception

RuntimeError: Caught RuntimeError in DataLoader worker process 0.
Original Traceback (most recent call last):
  File "C:\Users\david\anaconda3\lib\site-packages\monai\transforms\transform.py", line 92, in apply_transform
    return _apply_transform(transform, data, unpack_items)
  File "C:\Users\david\anaconda3\lib\site-packages\monai\transforms\transform.py", line 60, in _apply_transform
    return transform(parameters)
  File "C:\Users\david\anaconda3\lib\site-packages\monai\transforms\utility\array.py", line 208, in __call__
    raise ValueError(msg)
ValueError: meta_dict not available, EnsureChannelFirst is not in use.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "C:\Users\david\anaconda3\lib\site-packages\monai\transforms\transform.py", line 92, in apply_transform
    return _apply_transform(transform, data, unpack_items)
  File "C:\Users\david\anaconda3\lib\site-packages\monai\transforms\transform.py", line 60, in _apply_transform
    return transform(parameters)
  File "C:\Users\david\anaconda3\lib\site-packages\monai\transforms\compose.py", line 160, in __call__
    input_ = apply_transform(_transform, input_, self.map_items, self.unpack_items)
  File "C:\Users\david\anaconda3\lib\site-packages\monai\transforms\transform.py", line 116, in apply_transform
    raise RuntimeError(f"applying transform {transform}") from e
RuntimeError: applying transform <monai.transforms.utility.array.EnsureChannelFirst object at 0x000002B5A3AC0E80>

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "C:\Users\david\anaconda3\lib\site-packages\torch\utils\data\_utils\worker.py", line 287, in _worker_loop
    data = fetcher.fetch(index)
  File "C:\Users\david\anaconda3\lib\site-packages\torch\utils\data\_utils\fetch.py", line 49, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "C:\Users\david\anaconda3\lib\site-packages\torch\utils\data\_utils\fetch.py", line 49, in <listcomp>
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "C:\Users\david\anaconda3\lib\site-packages\monai\data\image_dataset.py", line 118, in __getitem__
    img = apply_transform(self.transform, img, map_items=False)
  File "C:\Users\david\anaconda3\lib\site-packages\monai\transforms\transform.py", line 116, in apply_transform
    raise RuntimeError(f"applying transform {transform}") from e
RuntimeError: applying transform <monai.transforms.compose.Compose object at 0x000002B5A3DF5310>

[KumoLiu]

Hi @dgmato,
May I know your MONAI version? I think maybe your should update your MONAI version to 0.9.1.
Hope it will helps you!

[dgmato]

Hi @KumoLiu. Thanks for the quick reply. I've tried with the latest version of MONAI and it did not work either. Then, I decided to install the same versions used for the tutorial. Therefore, I install the version indicated in the output of the tutorial notebook.

I am using the following versions right now:

MONAI version: 0.7.0
Numpy version: 1.21.2
Pytorch version: 1.10.2+cu111
MONAI flags: HAS_EXT = False, USE_COMPILED = False
MONAI rev id: bfa054b9c3064628a21f4c35bbe3132964e91f43

Optional dependencies:
Pytorch Ignite version: 0.4.8
Nibabel version: 3.2.1
scikit-image version: 0.18.3
Pillow version: 8.4.0
Tensorboard version: 2.9.0
gdown version: 4.4.0
TorchVision version: NOT INSTALLED or UNKNOWN VERSION.
tqdm version: 4.64.0
lmdb version: 1.3.0
psutil version: 5.8.0
pandas version: 1.4.2
einops version: 0.3.2
transformers version: 4.14.1

I will try updating MONAI to 0.9.1.

[KumoLiu]

You could be misled by mismatch between the tutorial's output version and the actual Monai version. I will try to fix it as soon as possible. Theoretically, updating to version0.9.1 should not report an error. But just in case, check if the type of your data output is metatensor, if you still have problems after updating your version, feel free to ask me!

im, label = monai.utils.misc.first(check_loader)
print(type(im))  # check if the type is MetaTensor

[dgmato]

@KumoLiu it works with version 0.9.1. Thanks!