Grad-CAM or attention map

[s9021025292140]

Would it be possible to add functionality for Grad-CAM or attention map similar to those used in DINO?

Thank you!

[guarin]

Hi! For Grad-CAM it is probably best if you just use the grad cam library: https://github.com/jacobgil/pytorch-grad-cam For the attention map visualization from DINO (here) the issue is mostly on how to make it work with other models than DINO. I guess it should be possible for any TIMM vision transformer, would definitely be a nice addition!