MaskDINO
MaskDINO copied to clipboard
loss weights clarification
Thank you for your amazing work!
Could you please guide how weights of losses (lambdas) were calculated? For example, in this config. I also noticed they differ from DINO, for example, here.
These weights are set with grid search. For example, class weight is set larger compared to DINO because mask losses will dilute the original class loss.