Isaac Robinson

I don't have strong feelings about this! Although I would like there to be a mechanism to load a model with a set number of classes without having to have...

I mean I have very often built my own data pipeline. useful for implementing certain augmentations, plugging into large datasets that can't be entirely stored locally, plugging into existing data...

Yeah there's a bug, it's accidentally keeping the 365 classes from the o365 pretrain. @probicheaux is working on it iirc but he's been ill for a few days

Feel free to raise a new issue btw so it's easier for us to keep track

Good insight @panagiotamoraiti ! :)

@SkalskiP amy thoughts ?

Nice debugging! Yeah I think in our research repo there's a sync that happens, guess it didn't make it here. Feel free to submit a PR? A workaround is to...

We don't use that loss. I think it's unlikely it'll help your problem. Feel free to submit a PR with a fix though! :)

RF-DETR already uses a variant of focal loss

I can't think of anything that would be especially useful for RF-DETR vs dealing with class imbalances for other models. I might suggest building a custom dataloader to oversample from...