GPU memory requirements increases when using DCNv3_pytorch rather than DCNv3

[jonassoebro]

Thank you for this awesome implementation!

I've been using the InternImage as a backbone of my Detectron2 Faster-R-CNN implementation. My trained models need to run on CPU, so I tried making them compatible using the DCNv3_pytorch operator rather than the DCNv3. However the pytorch implementation seems to require a lot more GPU memory - is this expected or am I missing something?

[rentainhe]

Maybe you can refer to InternImage's official repo to see if some issues can help u with this problem.

• InternImage

[jonassoebro]

I looked but could not find anyone with similar issues. I made a duplicate issue there if anyone wants to follow. OpenGVLab/InternImage#260