When I run
vak prep clark_train_better.toml
I get this error:
(venv) ellie@ElliePC:~/vakFishCrowEngine$ vak prep clark_train_better.toml
2026-07-24 22:55:46,527 - vak.prep.frame_classification.frame_classification - INFO - vak version: 1.1.0
2026-07-24 22:55:46,527 - vak.prep.frame_classification.frame_classification - INFO - Will prepare dataset as directory: /home/ellie/vakFishCrowEngine/prep/train/data-vak-frame-classification-dataset-generated-260724_225546
2026-07-24 22:55:47,193 - vak.prep.spectrogram_dataset.prep - INFO - making array files containing spectrograms from audio files in: /home/ellie/vakFishCrowEngine/clark/data
2026-07-24 22:55:47,195 - vak.prep.spectrogram_dataset.audio_helper - INFO - creating array files with spectrograms
[ ] | 0% Completed | 53.91 sTerminated
(venv) ellie@ElliePC:~/vakFishCrowEngine$ /usr/lib/python3.12/multiprocessing/resource_tracker.py:254: UserWarning: resource_tracker: There appear to be 5 leaked semaphore objects to clean up at shutdown
warnings.warn('resource_tracker: There appear to be %d '
[process exited with code 1 (0x00000001)]
You can now close this terminal with Ctrl+D, or press Enter to restart.
The error stays the same every time I run it, and it always happens during the same step in prepping my training dataset. Sometimes, getting this error multiple times in a row leads to my terminal crashing with this error:
Catastrophic failure
Error code: Wsl/Service/E_UNEXPECTED
The whole process went smoothly when I ran the vak tutorial in the same command line interface, so I'm not sure what the problem is with my own dataset. It seems to be a pytorch issue, so I've tried avoiding using multiple threads. In the toml file, I specified workers = 1 (even if it’s in the training config section and not prep), accelerator = "cpu", and device = 1.
- Operating System: Windows Subsystem for Linux, version 2
- Distribution: Ubuntu
- vak Version: 1.1.0
- Python Version: 3.12.3
Thank you for your help!
When I run
I get this error:
The error stays the same every time I run it, and it always happens during the same step in prepping my training dataset. Sometimes, getting this error multiple times in a row leads to my terminal crashing with this error:
The whole process went smoothly when I ran the vak tutorial in the same command line interface, so I'm not sure what the problem is with my own dataset. It seems to be a pytorch issue, so I've tried avoiding using multiple threads. In the toml file, I specified workers = 1 (even if it’s in the training config section and not prep), accelerator = "cpu", and device = 1.
Thank you for your help!