Open Miraclerice opened 2 months ago
We followed the gMLP implementation used in MAXIM: https://github.com/vztu/maxim-pytorch/blob/main/maxim_pytorch/maxim_torch.py#L125
I think this might provide a residual connection on u, which might make the training more stable.
Why does the gMLP model put v+1, what does it do here? https://github.com/MungoMeng/Registration-CorrMLP/blob/da5ce37276a9a233bd85865e8dc487ac4a8047da/CorrMLP/networks.py#L419