tanzhenyu
tanzhenyu
I think all options are important: 1) per batch 2) every 1000 (user configured) batches 3) every epoch Do remind that sometimes there is no concept of an epoch, and...
follow-up comment here, we should support YOLOv3 as it is a great demonstration for "reusable components", and YOLOv3 is definitely still heavily used, compared to more recent models (2022 is...
> I am willing to work on this as well. Thanks for providing the help! This is currently assigned, would you be open to contributing other transformer based vision models?...
> Sure! https://github.com/keras-team/keras-cv/issues/671 Somehow I can't assign to you through user name...
> Hey @LukeWood, I and @soumik12345 would like to take this up. Since there is no `contribution-welcome` label added to this issue, would first love to get a green light...
It seems I created a duplicate issue. Yes we'd like to accept the contribution as a keras-application style model. Is there any way to split the work, for example, porting...
> Sure I understand it is not possible to review a large PR like this one. I will split it in several PR ( see the complete list in #518...
> Will this support SOTA Transformer anchor models like: > > https://github.com/IDEACVR/DINO I'd say it's highly desired the components to be generic enough to handle this model. If you know...
> Hey @tanzhenyu > > I got around reading the contriution guideline and the API design document. I was unable to find a guide to contribute a model inside of...
Yep I think it should be other way around, I will make a change