Runtime error on Windows

[lassoan]

When trying to run the TotalSegmentator modle with latest Slicer Preview Release I get the following error:

[Python] Failed to compute results.
[Python] Command '['docker', 'run', '--rm', '--volume', 'C:/Users/andra/AppData/Local/Temp/Slicer/__SlicerTemp__2024-11-02_12+34+09.373:/app/data/input_data', '--volume', 'C:/Users/andra/AppData/Local/Temp/Slicer/__SlicerTemp__2024-11-02_12+34+09.374:/app/data/output_data', 'mhubai/totalsegmentator:latest', 'python3', '-m', 'mhubio.run', '--config', '/app/models/totalsegmentator/config/slicer.yml']' returned non-zero exit status 1.
Traceback (most recent call last):
  File "C:/Users/andra/AppData/Local/slicer.org/Slicer 5.7.0-2024-11-01/slicer.org/Extensions-33087/MHubRunner/lib/Slicer-5.7/qt-scripted-modules/MRunner.py", line 432, in onApplyButton
    self.logic.process(
  File "C:/Users/andra/AppData/Local/slicer.org/Slicer 5.7.0-2024-11-01/slicer.org/Extensions-33087/MHubRunner/lib/Slicer-5.7/qt-scripted-modules/MRunner.py", line 891, in process
    self.runContainerSync(model, tmpDirIn, tmpDirOut, useGPU=useGPU)
  File "C:/Users/andra/AppData/Local/slicer.org/Slicer 5.7.0-2024-11-01/slicer.org/Extensions-33087/MHubRunner/lib/Slicer-5.7/qt-scripted-modules/MRunner.py", line 756, in runContainerSync
    self.logProcessOutput(proc)
  File "C:/Users/andra/AppData/Local/slicer.org/Slicer 5.7.0-2024-11-01/slicer.org/Extensions-33087/MHubRunner/lib/Slicer-5.7/qt-scripted-modules/MRunner.py", line 526, in logProcessOutput
    raise CalledProcessError(retcode, proc.args, output=proc.stdout, stderr=proc.stderr)
subprocess.CalledProcessError: Command '['docker', 'run', '--rm', '--volume', 'C:/Users/andra/AppData/Local/Temp/Slicer/__SlicerTemp__2024-11-02_12+34+09.373:/app/data/input_data', '--volume', 'C:/Users/andra/AppData/Local/Temp/Slicer/__SlicerTemp__2024-11-02_12+34+09.374:/app/data/output_data', 'mhubai/totalsegmentator:latest', 'python3', '-m', 'mhubio.run', '--config', '/app/models/totalsegmentator/config/slicer.yml']' returned non-zero exit status 1.

Docker container log:

Processing started
�[95m**************************************************************�[0m
�[95m**                      CITATION NOTICE                     **�[0m
�[95m** -------------------------------------------------------- **�[0m
�[95m** �[0m                                                        �[95m **�[0m
�[95m** �[0mYou are running an MHub model. Cite the MHub platform in�[95m **�[0m
�[95m** �[0m your work. Visit https://mhub.ai/cite for more.        �[95m **�[0m
�[95m** �[0m                                                        �[95m **�[0m
�[95m** �[0mYou are running a model. Please cite the authors of the �[95m **�[0m
�[95m** �[0mmodel and the MHub model url in your work:              �[95m **�[0m
�[95m** �[0m The model card and run instructions for this model are �[95m **�[0m
�[95m** �[0m publicly available on the MHub platform under https://m�[95m **�[0m
�[95m** �[0m hub.ai/models/totalsegmentator.                        �[95m **�[0m
�[95m** �[0m                                                        �[95m **�[0m
�[95m** �[0mThank you for using MHub!                               �[95m **�[0m
�[95m** �[0m                                                        �[95m **�[0m
�[95m**************************************************************�[0m

Traceback (most recent call last):
  File "/usr/lib/python3.8/runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/usr/lib/python3.8/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/usr/local/lib/python3.8/dist-packages/mhubio/run.py", line 613, in <module>
    run(config_file)
  File "/usr/local/lib/python3.8/dist-packages/mhubio/run.py", line 374, in run
    config = Config(config_file=config_file, config=config_data)
  File "/usr/local/lib/python3.8/dist-packages/mhubio/core/Config.py", line 135, in __init__
    self._config = dict_merge(arg_config, self._config.copy())
AttributeError: 'NoneType' object has no attribute 'copy'

I get the same error when running the docker run command from command-line. This is the output:

Traceback (most recent call last):
  File "/usr/lib/python3.8/runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/usr/lib/python3.8/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/usr/local/lib/python3.8/dist-packages/mhubio/run.py", line 613, in <module>
    run(config_file)
  File "/usr/local/lib/python3.8/dist-packages/mhubio/run.py", line 374, in run
    config = Config(config_file=config_file, config=config_data)
  File "/usr/local/lib/python3.8/dist-packages/mhubio/core/Config.py", line 135, in __init__
    self._config = dict_merge(arg_config, self._config.copy())
AttributeError: 'NoneType' object has no attribute 'copy'
**************************************************************
**                      CITATION NOTICE                     **
** -------------------------------------------------------- **
**                                                          **
** You are running an MHub model. Cite the MHub platform in **
**  your work. Visit https://mhub.ai/cite for more.         **
**                                                          **
** You are running a model. Please cite the authors of the  **
** model and the MHub model url in your work:               **
**  The model card and run instructions for this model are  **
**  publicly available on the MHub platform under https://m **
**  hub.ai/models/totalsegmentator.                         **
**                                                          **
** Thank you for using MHub!                                **
**                                                          **
**************************************************************

The error seems to be that self._config is None.

[LennyN95]

Thank you @lassoan for sharing. I couldn't reproduce the issue on Mac or Linux but I have a guess what could happen here and am looking into it. Meanwhile, have you tired any other model?

[lassoan]

Yes, all other models that I've tried fail with the exact same error. I've tried with TotalSegmentator, BAMF CT Kidney Seg, and CaSuSt so far.

[LennyN95]

@lassoan Does this issue still occur under the latest version of mhub TotalSegmentator (e.g., after manually updating the image from the cli via pull mhubai/totalsegmentator:latest)?