-
Notifications
You must be signed in to change notification settings - Fork 348
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Extend tch::Device to support check for/using Apple M1 MPS GPU? #501
Comments
fn main() {
println!("{:?}", tch::Device::cuda_if_available());
}
looks like brew install libtorch
export LIBTORCH=/opt/homebrew/Cellar/libtorch/1.11.0_1
export LD_LIBRARY_PATH=${LIBTORCH}/lib:$LD_LIBRARY_PATH
cargo run does not yield a solution that automatically uses the Apple M1 GPU at the moment |
option 1. libtorch from brew does not come with option 2. https://download.pytorch.org/libtorch/nightly/cpu/libtorch-macos-latest.zip is x86_64, not arm64 option 3.
|
|
pytorch/pytorch#68625 (comment) Looks like max_unpool2d_backward was removed? |
probably a separate issue of trying to use |
https://docs.rs/tch/0.7.2/tch/enum.Device.html#method.is_cuda
https://github.com/pytorch/pytorch/pull/77767/files
https://github.com/pytorch/pytorch/blob/master/docs/source/notes/mps.rst
The text was updated successfully, but these errors were encountered: