Using GradCAM for model volo_d2_224 #380
Unanswered
stevenlimcorn
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I have noticed the available support for ViT model for GradCAM and the exmaple is well structured.
I was wondering if it is possible to use other architectures such as
volo_d2_224
. In the paper they mentioned about using outlook attention. Unlike ViT which has192 channels
, this model has512 channels
as the output. Usingtimm
library it will look something like this.Did I use the wrong layer for this? or is this still not supported?
Thanks!
Beta Was this translation helpful? Give feedback.
All reactions