OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-07 00:21:07 +01:00
Code Issues Actions 37 Packages Projects Releases Wiki Activity

update mps note with more details (#78669)

Browse Source 
Follow up to the comments in https://github.com/pytorch/pytorch/pull/77767#pullrequestreview-978807521
Pull Request resolved: https://github.com/pytorch/pytorch/pull/78669
Approved by: https://github.com/kulinseth, https://github.com/anjali411
This commit is contained in:
albanD 2022-06-02 20:53:19 +00:00 committed by PyTorch MergeBot
parent 3e0f1a8a32
commit b30b1f3dec
1 changed files with 22 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
docs/source/notes/mps.rst
View File

@ -14,14 +14,18 @@ capabilities to setup and run operations on GPU.
To get started, simply move your Tensor and Module to the ``mps`` device: To get started, simply move your Tensor and Module to the ``mps`` device:
.. code:: .. code:: python
# Make sure the current PyTorch binary was built with MPS enabled # Check that MPS is available
print(torch.backends.mps.is_built()) if not torch.backends.mps.is_available():
# And that the current hardware and MacOS version are sufficient to if not torch.backends.mps.is_built():
# be able to use MPS print("MPS not available because the current PyTorch install was not "
print(torch.backends.mps.is_available()) "built with MPS enabled.")
else:
print("MPS not available because the current MacOS version is not 12.3+ "
"and/or you do not have an MPS-enabled device on this machine.")
else:
mps_device = torch.device("mps") mps_device = torch.device("mps")
# Create a Tensor directly on the mps device # Create a Tensor directly on the mps device