Blaž Rolih

Current main branch does have the latest version of torchmetrics, so this should work (I didn't get the chance to try it). Maybe the problem is that Precision is inside...

> @jpcbertoldo, @phcarval, @blaz-r , do you guys have any concerns about this pr I can check it tomorrow.

Hello. Now that I am looking at exporting, it seems like it is saving the last model, not the best one. @ashwinvaidya17 can you confirm this please? While this will...

Hello. You can still train without the labels, but you won't be able to test the model.

Sorry I understood the question incorrectly. All models implemented in Anomalib only utilize normal data for training, so you should be able to train the model using your good data,...

Hello, that is correct. All models in currently Anomalib are unsupervised, using only good data for training. Labeled anomalous images are only used in test phase, where you require ground...

Hi, I haven't encountered this issue before, but it is possible it is due to software if updates are done too rapidly, though I'm not sure. Could you please send...

Thank you both for this. As I said in previous comment, it'd be great if you could give a minimal code that can reproduce this issue. Another question I have...

Thank you. I'll look into it when I get some time. I suspect it has something to do with threading and timing.

I'm really busy right now and don't have access to my pico. I'll try to test this as soon as possible, but might still take a few days.