Spinkoo
Spinkoo
You can use it as a backbone for a segmentation model just like a regular resnet for example : [Mask RCNN with resnet as backbone](https://pytorch.org/vision/stable/_modules/torchvision/models/detection/mask_rcnn.html)
I do indeed share the same question, Like since the shortest path of the gradient is O(1) in a resnet due to the sum of all the previous feature maps...
Foor anyone else facing these sort of issues, you can check the build manual I put in place [Open3D on WSL + CUDA](https://github.com/Spinkoo/Open3DWSL)