Ashwin Vaidya

Closing this as MLFlow logger PR is merged

@Shakib-IO are you still working on this?

@RitikaxShakya currently we override the number of devices to 1 in Engine and the CLI. To start with, we should remove these lines. https://github.com/openvinotoolkit/anomalib/blob/debdae70bc6e089958eaefa066b4bcd79711bb23/src/anomalib/engine/engine.py#L305 https://github.com/openvinotoolkit/anomalib/blob/debdae70bc6e089958eaefa066b4bcd79711bb23/src/anomalib/utils/config.py#L130 Doing this will break a...

@haimat unfortunately we don't have an exact timeline for this. Currently, we are busy with some other high-priority tasks.

@1094558673 I am closing this as @samet-akcay 's answer should give you some pointers. Feel free to continue in the discussion page if you have any more questions.

This is something we need to address. We move the metrics to CPU for computation to avoid OOM on GPU. But this breaks multi-gpu training. For now the solution is...

We also have these lines here https://github.com/openvinotoolkit/anomalib/blob/main/src/anomalib/utils/callbacks/post_processing_configuration.py#L75. Did you try removing these as well?

I am closing this as `benchmark.py` has been replaced by pipelines. I've taken a note of this issue for when third-party logging is restored https://github.com/openvinotoolkit/anomalib/issues/2056

Blocked by: https://github.com/Lightning-AI/metrics/issues/1526

@jpcbertoldo is this still relevant with v1.0?