Cuda 11.6 deprecated about torch

[dfalbel]

LibTorch is very finicky to support older CUDA versions (actually also for newer versions too) so we have a very small range of versions supported.

I recommend installing torch using the pre-buiilt binaries from. Those bundle CUDA and CUDNNN versions required by torch, so you don't need to manage a CUDA installation.

You can install them with somehting like:

kind <- "cu118"
version <- "0.12.0"
options(repos = c(
  torch = sprintf("https://torch-cdn.mlverse.org/packages/%s/%s/", kind, version),
  CRAN = "https://cloud.r-project.org" # or any other from which you want to install the other R dependencies.
))
install.packages("torch", type = "binary")

[AGPatriota]

I got it with:

install.packages("https://cran.r-project.org/src/contrib/Archive/torch/torch_0.11.0.tar.gz", repos = NULL, type = "source")

It is installing the additional packages.

require('torch')
Carregando pacotes exigidos: torch
ℹ Additional software needs to be downloaded and installed for torch to work correctly.
Do you want to continue? (Sim/não/cancelar) Sim
tentando a URL 'https://download.pytorch.org/libtorch/cu116/libtorch-cxx11-abi-shared-with-deps-1.13.1%2Bcu116.zip'

I have to change my GPU.

Thanks