Consider setting persistent_workers=True in 'train_dataloader' to speed up the dataloader worker initialization.

[ghost]

Anomalib library is installed.

Copy-pasted the code listed in "Training" section of Anomalib webpage. Programming is running however there is no output.
Suggestions on screen suggest setting persistent_workers=True in 'train_dataloader' to speed up the dataloader worker initialization. Screen shot attached.

Checking the "Task Manager" display the CPU and Memory consumption by the Python Kernel to be low. Initially at the start of the program the consumption are high however the consumption values falls low.

Appreciate help to locate the python/configuration file for setting persistent_workers=True.

[samet-akcay]

This is where the dataloaders are initialized. If you want to add persistent_workers=True, you could add them here
https://github.com/openvinotoolkit/anomalib/blob/6c9fd4e28f9df3d2aea7fede77ca0b0c77601e08/src/anomalib/data/base/datamodule.py#L215-L242

[ghost]

Thanks Samet.

Added a line of code to the train_dataloader function.
persistent_workers=True,

Suggestion message is not displayed anymore. Code is running. However we are not able to figure out progress in model training. A progress indicator or verbose status messages could help.

Checking the "Task Manager" display the CPU and Memory consumption by the Python Kernel to be low. Initially at the start of the program the consumption are high however the consumption values falls low.

.

[samet-akcay]

However we are not able to figure out progress in model training. A progress indicator or verbose status messages could help.

What do you mean by this? What sort of indicator do you want to see?

[ghost]

Could it be possible to indicate task completion in percentage.