Closed EmKarr closed 2 years ago
Hello @EmKarr,
Could you elaborate a bit what is missing to cover your need that is not available between
torch::torch_flatten()
and tensor$flatten()
?
Thanks
Hi cregouby, sorry, seems like I went to the wrong package to bring up the issue, should have been "torch", not "torchvision".
In torch, this issue already had beed addressed: https://github.com/mlverse/torch/issues/716#issuecomment-946117545 Thanks, best wishes, Mike
Hello,
thank you for the brilliant torch package for R, I really like it :-)
Nevertheless, I am running into problems setting up a model via nn_sequential(), for me it seems like the R counterpart of PyTorch's nn.Flatten() is missing. If so, would it be possible to include such a function? Sequential models currently are a prerequisite for using the "innsight" package, which I would like to do, so using nn_module() is not a solution.
All the best, stay well and safe, Mike