Hello!

I have created the following pipeline and run it on some NiFTI images here (https://drive.google.com/drive/folders/1r2_0RTHFN8XfSLnSgMGYZuQXuhEpP4NC?usp=sharing). I would like to save the processed images as NiFTI because I am processing >10,000 images before training a neural network. Instead of having to process all images every epoch, I would like to save them as temporary files until training is complete.

preprocess_transform = Compose([
    LoadImaged(keys=["img"]),
    AsChannelFirstd(keys=["img"], channel_dim=-1),
    Spacingd(keys=["img"], pixdim=(2, 2, 2), diagonal=True, mode="bilinear"),
    ScaleIntensityd(keys=["img"], minv=0, maxv=1),
    SaveImaged(
                    keys=["img"],
                    output_dir="./visualization",
                    output_ext=".nii",
                    resample=True,
     ),
    ToTensord(keys=["img"])
])

Without resampling, the saved images appear distorted and shorn, but with resampling, the following error is received. It appears that the affine transform requires >200,000 GB of memory?

Modifying image pixdim from [1.01821 1.01821 2.025 0. ] to [ 1.01821005 1.01821005 2.0250001 898.17486327]
input data information of the runtime error transform:
img statistics:
Type: <class 'numpy.ndarray'>
Shape: (1, 172, 172, 82)
Value range: (0.0, 1.0)
img_meta_dict statistics:
Type: <class 'dict'>
Value: {'sizeof_hdr': array(1543569408, dtype=int32), 'extents': array(4194304, dtype=int32), 'session_error': array(0, dtype=int16), 'dim_info': array(0, dtype=uint8), 'dim': array([ 1024, 20481, 20481, 20736, 256, 0, 0, 0],
dtype=int16), 'intent_p1': array(0., dtype=float32), 'intent_p2': array(0., dtype=float32), 'intent_p3': array(0., dtype=float32), 'intent_code': array(0, dtype=int16), 'datatype': array(4096, dtype=int16), 'bitpix': array(8192, dtype=int16), 'slice_start': array(0, dtype=int16), 'pixdim': array([ 4.6006030e-41, -7.9165687e-07, -7.9165687e-07, -6.3281336e-23,
4.0563766e-09, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00],
dtype=float32), 'vox_offset': array(0., dtype=float32), 'scl_slope': array(6.9055e-41, dtype=float32), 'scl_inter': array(6.9055e-41, dtype=float32), 'slice_end': array(0, dtype=int16), 'slice_code': array(0, dtype=uint8), 'xyzt_units': array(2, dtype=uint8), 'cal_max': array(0., dtype=float32), 'cal_min': array(0., dtype=float32), 'slice_duration': array(0., dtype=float32), 'toffset': array(0., dtype=float32), 'glmax': array(0, dtype=int32), 'glmin': array(0, dtype=int32), 'qform_code': array(256, dtype=int16), 'sform_code': array(0, dtype=int16), 'quatern_b': array(0., dtype=float32), 'quatern_c': array(0., dtype=float32), 'quatern_d': array(0., dtype=float32), 'qoffset_x': array(-3.216494e+31, dtype=float32), 'qoffset_y': array(1.5721015e+19, dtype=float32), 'qoffset_z': array(5.8155804e-15, dtype=float32), 'srow_x': array([0., 0., 0., 0.], dtype=float32), 'srow_y': array([0., 0., 0., 0.], dtype=float32), 'srow_z': array([0., 0., 0., 0.], dtype=float32), 'affine': array([[ 2. , 0. , 0. , 507.58554077],
[ 0. , 2. , 0. , 689.41204834],
[ 0. , 0. , 2. , 271.63400269],
[ 0. , 0. , 0. , 1. ]]), 'original_affine': array([[ 1.01821005, 0. , 0. , 507.58554077],
[ 0. , 1.01821005, 0. , 689.41204834],
[ 0. , 0. , 2.0250001 , 271.63400269],
[ 0. , 0. , 0. , 1. ]]), 'as_closest_canonical': False, 'spatial_shape': array([20481, 20481, 20736], dtype=int16), 'original_channel_dim': -1, 'filename_or_obj': '/scratch/sfan/AV45/I174732/ADNI_023_S_0376_PT_PET_Brain_AV45_recon_br_raw_20100527082453889_6_S85655_I174732.nii'}
img_transforms statistics:
Type: <class 'list'>
Value: [{'class': 'Spacingd', 'id': 23314182078984, 'orig_size': (336, 336, 81), 'extra_info': {'meta_data_key': 'img_meta_dict', 'old_affine': array([[ 1.01821005, 0. , 0. , 507.58554077],
[ 0. , 1.01821005, 0. , 689.41204834],
[ 0. , 0. , 2.0250001 , 271.63400269],
[ 0. , 0. , 0. , 1. ]]), 'mode': 'bilinear', 'padding_mode': 'border', 'align_corners': False}}]
Traceback (most recent call last):
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/transforms/transform.py", line 48, in apply_transform
return transform(data)
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/transforms/io/dictionary.py", line 231, in call
self.saver(img=d[key], meta_data=meta_data)
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/transforms/io/array.py", line 282, in call
self.saver.save(img, meta_data)
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/data/nifti_saver.py", line 157, in save
output_dtype=self.output_dtype,
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/data/nifti_writer.py", line 150, in write_nifti
spatial_size=output_spatial_shape[: len(data.shape)],
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/networks/layers/spatial_transforms.py", line 539, in forward
grid = nn.functional.affine_grid(theta=theta[:, :sr], size=list(dst_size), align_corners=self.align_corners)
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/torch/nn/functional.py", line 3933, in affine_grid
return torch.affine_grid_generator(theta, size, align_corners)
RuntimeError: [enforce fail at CPUAllocator.cpp:67] . DefaultCPUAllocator: can't allocate memory: you tried to allocate 278341060534272 bytes. Error code 12 (Cannot allocate memory)

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

Traceback (most recent call last):
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/transforms/transform.py", line 48, in apply_transform
return transform(data)
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/transforms/compose.py", line 144, in call
input_ = apply_transform(transform, input)
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/transforms/transform.py", line 71, in apply_transform
raise RuntimeError(f"applying transform {transform}") from e
RuntimeError: applying transform <monai.transforms.io.dictionary.SaveImaged object at 0x1534418a82e8>

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

Traceback (most recent call last):
File "preprocess.py", line 83, in
for i, batch in enumerate(check_loader):
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 517, in next
data = self._next_data()
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 557, in _next_data
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/torch/utils/data/_utils/fetch.py", line 44, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/data/dataset.py", line 93, in getitem
return self._transform(index)
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/data/dataset.py", line 79, in _transform
return apply_transform(self.transform, data_i) if self.transform is not None else data_i
File "/home/sfan/anaconda3/envs/torch_monai/lib/python3.6/site-packages/monai/transforms/transform.py", line 71, in apply_transform
raise RuntimeError(f"applying transform {transform}") from e
RuntimeError: applying transform <monai.transforms.compose.Compose object at 0x1534418a8400>