It all in the title:
torch.Tensor.cuda
has an async
parameter that allows asynchronous transfer from pinned memory to GPU. Does torch.Tensor.to
automatically uses async when possible?
I wasn’t able to find the relevant piece of source code.
It all in the title:
torch.Tensor.cuda
has an async
parameter that allows asynchronous transfer from pinned memory to GPU. Does torch.Tensor.to
automatically uses async when possible?
I wasn’t able to find the relevant piece of source code.
It was added in the current master branch as non_blocking
, since asynch
will be a keyword in Python 3.7.
See the docs.
If you want to use it, you would have to build PyTorch from source. You can find the build instructions here.
Nice, thank you very much.