I was try to use the “Brain Tumor Segmentation(GLI)” model in the MONAI Auto3DSEG to create segmentations on my own MRdata. It do worked once and then keep failing.
These are error messages:
Processing started
Writing input file to C:/Users/zcs/AppData/Local/Temp/Slicer/__SlicerTemp__2024-09-21_23+21+09.704/input-volume0.nrrd
Writing input file to C:/Users/zcs/AppData/Local/Temp/Slicer/__SlicerTemp__2024-09-21_23+21+09.704/input-volume1.nrrd
Writing input file to C:/Users/zcs/AppData/Local/Temp/Slicer/__SlicerTemp__2024-09-21_23+21+09.704/input-volume2.nrrd
Writing input file to C:/Users/zcs/AppData/Local/Temp/Slicer/__SlicerTemp__2024-09-21_23+21+09.704/input-volume3.nrrd
Creating segmentations with MONAIAuto3DSeg AI…
Auto3DSeg command: [‘C:/Users/zcs/AppData/Local/slicer.org/Slicer 5.7.0-2024-09-19/bin/…/bin\PythonSlicer.EXE’, ‘C:/Users/zcs/AppData/Local/slicer.org/Slicer 5.7.0-2024-09-19/slicer.org/Extensions-33018/MONAIAuto3DSeg/lib/Slicer-5.7/qt-scripted-modules\Scripts\auto3dseg_segresnet_inference.py’, ‘–model-file’, ‘C:\Users\zcs\.MONAIAuto3DSeg\models\brats-gli-v1.0.0\model.pt’, ‘–image-file’, ‘C:/Users/zcs/AppData/Local/Temp/Slicer/__SlicerTemp__2024-09-21_23+21+09.704/input-volume0.nrrd’, ‘–result-file’, ‘C:/Users/zcs/AppData/Local/Temp/Slicer/__SlicerTemp__2024-09-21_23+21+09.704/output-segmentation.nrrd’, ‘–image-file-2’, ‘C:/Users/zcs/AppData/Local/Temp/Slicer/__SlicerTemp__2024-09-21_23+21+09.704/input-volume1.nrrd’, ‘–image-file-3’, ‘C:/Users/zcs/AppData/Local/Temp/Slicer/__SlicerTemp__2024-09-21_23+21+09.704/input-volume2.nrrd’, ‘–image-file-4’, ‘C:/Users/zcs/AppData/Local/Temp/Slicer/__SlicerTemp__2024-09-21_23+21+09.704/input-volume3.nrrd’]
You are using torch.load with weights_only=False (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See pytorch/SECURITY.md at main · pytorch/pytorch · GitHub for more details). In a future release, the default value for weights_only will be flipped to True. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via torch.serialization.add_safe_globals. We recommend you start setting weights_only=True for any use case where you don’t have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
apex.normalization.InstanceNorm3dNVFuser is not installed properly, use nn.InstanceNorm3d instead.
Model epoch 262 metric 0.8930580615997314
Using crop_foreground
Using resample with resample_resolution [1.0, 1.0, 1.0]
Traceback (most recent call last):
File “C:\Users\zcs\AppData\Local\slicer.org\Slicer 5.7.0-2024-09-19\lib\Python\Lib\site-packages\monai\transforms\transform.py”, line 140, in apply_transform
return [_apply_transform(transform, item, unpack_items, lazy, overrides, log_stats) for item in data]
File “C:\Users\zcs\AppData\Local\slicer.org\Slicer 5.7.0-2024-09-19\lib\Python\Lib\site-packages\monai\transforms\transform.py”, line 140, in
return [_apply_transform(transform, item, unpack_items, lazy, overrides, log_stats) for item in data]
File “C:\Users\zcs\AppData\Local\slicer.org\Slicer 5.7.0-2024-09-19\lib\Python\Lib\site-packages\monai\transforms\transform.py”, line 98, in _apply_transform
return transform(data, lazy=lazy) if isinstance(transform, LazyTrait) else transform(data)
File “C:\Users\zcs\AppData\Local\slicer.org\Slicer 5.7.0-2024-09-19\lib\Python\Lib\site-packages\monai\transforms\io\dictionary.py”, line 162, in call
data = self._loader(d[key], reader)
File “C:\Users\zcs\AppData\Local\slicer.org\Slicer 5.7.0-2024-09-19\lib\Python\Lib\site-packages\monai\transforms\io\array.py”, line 282, in call
img_array, meta_data = reader.get_data(img)
File “C:\Users\zcs\AppData\Local\slicer.org\Slicer 5.7.0-2024-09-19\lib\Python\Lib\site-packages\monai\data\image_reader.py”, line 1343, in get_data
_copy_compatible_dict(header, compatible_meta)
File “C:\Users\zcs\AppData\Local\slicer.org\Slicer 5.7.0-2024-09-19\lib\Python\Lib\site-packages\monai\data\image_reader.py”, line 129, in _copy_compatible_dict
raise RuntimeError(
RuntimeError: affine matrix of all images should be the same for channel-wise concatenation. Got [[-4.57198656e-01 -1.03385621e-01 2.71376017e-03 1.41867333e+02]
[-1.01093522e-01 4.44159679e-01 -1.10584093e-01 -1.02964030e+02]
[ 3.02560651e-01 -1.50380787e+00 -6.31662069e+00 3.36919582e+01]
[ 0.00000000e+00 0.00000000e+00 0.00000000e+00 1.00000000e+00]] and [[-4.57248575e-01 -1.03391612e-01 2.71345400e-03 1.41872669e+02]
[-1.01099258e-01 4.44208168e-01 -1.10596074e-01 -1.02986774e+02]
[ 3.02538236e-01 -1.50374134e+00 -6.31632356e+00 3.36796764e+01]
[ 0.00000000e+00 0.00000000e+00 0.00000000e+00 1.00000000e+00]].
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
Processing failed with return code 1
Cleaning up temporary folder.
Processing failed after 4.64 seconds.
Processing finished.