Error with RandCropByPosNegId: The size of tensor a (19400862) must match the size of tensor b (25181628) at non-singleton dimension 0.

[ArpanGyawali]

I have read other issues related to this but couldnt solve it.

My pre transforms:

return [
            LoadImaged(keys=("image", "label")),
            EnsureChannelFirstd(keys=("image", "label")),
            Spacingd(
                keys=("image", "label"),
                pixdim=(1.0, 1.0, 1.0),
                mode=("bilinear", "nearest"),
            ),
            #SetWindowd(keys="image", level=150, width=450),
            ScaleIntensityRanged(keys="image", a_min=-75, a_max=375, b_min=0.0, b_max=1.0, clip=True),
            EnsureTyped(keys=("image", "label"), device=context.device),
            RandCropByPosNegLabeld(
                keys=("image", "label"),
                label_key="label",
                spatial_size=(128, 128, 128),
                pos=0.6,
                neg=0.4,
                num_samples=4,
                image_key="image",
                image_threshold=0,
                allow_smaller=True,
            ),
            RandShiftIntensityd(keys="image", offsets=0.05, prob=0.5),
            RandAffined(
                keys=['image', 'label'],
                mode=('bilinear', 'nearest'),
                padding_mode='zeros',
                prob=0.8,
                spatial_size=(128, 128, 128),
                rotate_range=(0, 0, np.pi/18),
                scale_range=(0.5, 0.5, 0.5)),
            #Rand3DElasticd(
             #   keys=['image', 'label'],
              #  sigma_range=(5,7),
               # magnitude_range=(0.05, 0.15),
               # prob=0.4,
             #   padding_mode='zeros',
            #    mode=['bilinear', 'nearest'],
            #),
            #RandRotated(keys=("image", "label"), prob=0.5, range_x=0.2, mode=('bilinear','nearest')),
            SelectItemsd(keys=("image", "label")),
        ]

I get the error:

raceback (most recent call last):
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/monai/transforms/transform.py", line 102, in apply_transform
    return _apply_transform(transform, data, unpack_items)
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/monai/transforms/transform.py", line 66, in _apply_transform
    return transform(parameters)
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/monai/transforms/croppad/dictionary.py", line 861, in __call__
    self.randomize(label, fg_indices, bg_indices, image)
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/monai/transforms/croppad/dictionary.py", line 852, in randomize
    self.cropper.randomize(label=label, fg_indices=fg_indices, bg_indices=bg_indices, image=image)
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/monai/transforms/croppad/array.py", line 1060, in randomize
    fg_indices_, bg_indices_ = map_binary_to_indices(label, image, self.image_threshold)
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/monai/transforms/utils.py", line 320, in map_binary_to_indices
    bg_indices = nonzero(img_flat & ~label_flat)
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/monai/data/meta_tensor.py", line 268, in __torch_function__
    ret = super().__torch_function__(func, types, args, kwargs)
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/torch/_tensor.py", line 1386, in __torch_function__
    ret = func(*args, **kwargs)
RuntimeError: The size of tensor a (19400862) must match the size of tensor b (25181628) at non-singleton dimension 0

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

Traceback (most recent call last):
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/ignite/engine/engine.py", line 1032, in _run_once_on_dataset_as_gen
    self.state.batch = next(self._dataloader_iter)
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/torch/utils/data/dataloader.py", line 630, in __next__
    data = self._next_data()
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/torch/utils/data/dataloader.py", line 674, in _next_data
    data = self._dataset_fetcher.fetch(index)  # may raise StopIteration
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/torch/utils/data/_utils/fetch.py", line 51, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/torch/utils/data/_utils/fetch.py", line 51, in <listcomp>
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/monai/data/dataset.py", line 107, in __getitem__
    return self._transform(index)
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/monai/data/dataset.py", line 921, in _transform
    data = apply_transform(_transform, data)
  File "/data/gpfs/projects/punim0406/Projects/skull-strip/monai-label-env/lib64/python3.9/site-packages/monai/transforms/transform.py", line 129, in apply_transform
    raise RuntimeError(f"applying transform {transform}") from e
RuntimeError: applying transform <monai.transforms.croppad.dictionary.RandCropByPosNegLabeld object at 0x14c6610606d0>

When i used CropForegroundd transform before RandCropByPosNegLabeld, everything worked fine. I removed it because my task is skull stripping and the region of interest is larger and covers significant part of the image.

Now when i remove it, why do i get this error?

[KumoLiu]

Hi @ArpanGyawali, please ensure your image and label are in the same shape. You can try to use ResizeWithPadOrCrop.

Hope it helps, thanks.

[ArpanGyawali]

@KumoLiu Hi I am sure initially the shape of image and label are the same, i manually labeled it by mayself. Does the transforms before the RandCropByPosNegLabeld change the size?

Also i have used Spacingd. All of my images have different number of slizes along z axis. So i cant give exact spatial size in ResizeWithPadOrCrop.
What should i do?

[KumoLiu]

Hi @ArpanGyawali, if you want to ensure the same shape, you can add ResizeWithPadOrCrop as a final transform to make sure you have the same shape. And also you can easily check the data shape after each transform by using DataStatsd or just print it.

MONAI/monai/transforms/utility/dictionary.py

Line 777 in 66a2fae

class DataStatsd(MapTransform):

Hope it helps, thanks.

[ArpanGyawali]

@KumoLiu
Can you help me with the above ques.