Ask questionsFeature Request: Add Pixel Unshuffle

I would like to propose the addition of an inverse operation for the current PixelShuffle function. We have been researching temporally connected feature cascades and have had a lot of success in hierarchically mixing residual information through this function.

The core operation is essentially the following: For feature map fm with batchsize b, channels c, downsize ratio of r, and original height and width of h,w, do:

out_channel = c*(r**2)
out_h = h//r
out_w = w//r
fm_view = fm.contiguous().view(b, c, out_h, r, out_w, r)
fm_prime = fm_view.permute(0,1,3,5,2,4).contiguous().view(b,out_channel, out_h, out_w)

cc @albanD @mruberry @fmassa @vfdev-5


Answer questions PiotrDabkowski

@alykhantejani Why should unshuffle qualify for the contrib while shuffle for the main repo? It is quite weird to implement just the upsample logic without the downsample counterpart. Lots of recent work relies on the unshuffle so it would be quite beneficial to put it just next to pixel_shuffle so that people do now waste time looking for it.


Related questions

TensorBoard logging requires TensorBoard with Python summary writer installed. This should be available in 1.14 or above hot 3
AttributeError: module 'torch.jit' has no attribute 'unused' hot 3
Script freezes with no output when using DistributedDataParallel hot 2
DataLoader leaking Semaphores. hot 2
[feature request] Add matrix exponential hot 2
cublas runtime error on torch.bmm() with CUDA10 and RTX2080Ti hot 2
libtorch does not initialize OpenMP/MKL by default hot 2
Use torch.device() with torch.load(..., map_location=torch.device()) hot 2
Cuda required when loading a TorchScript with map_location='cpu' hot 2
PyTorch 1.5 failed to import c:miniconda3-x64envs estlibsite-packages orchlibcaffe2_nvrtc.dll - pytorch hot 2
Error during import torch, NameError: name '_C' is not defined - pytorch hot 2
Quantisation of object detection models. hot 2
Problems with install python from source hot 2
torch.utils.tensorboard.SummaryWriter.add_graph do not support non-tensor inputs - pytorch hot 2
a retrained and saved jit module could not be reload. hot 2
Github User Rank List