frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-21 05:34:18 +08:00
Code Issues Packages Projects Releases Wiki Activity

Add "mps" device to PyTorch framework.

Browse Source 
Remove the "mlc" device for Mac platforms.

This commit will be followed up with:

* adding MPS runtime components
* PyTorch ops for MPS device

Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/76291
Approved by: https://github.com/albanD
This commit is contained in:
Kulin Seth
2022-04-27 19:21:57 +00:00
committed by PyTorch MergeBot
parent a0bf0f5611
commit 54c75e1e8f
37 changed files with 138 additions and 141 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
torch/_python_dispatcher.py
View File

@ -21,7 +21,7 @@ keys for a single example of each use case. These use cases are listed below:
- XLA/AutogradXLA: represents out-of-tree backends which we don't have either inference or autograd
kernel defined in pytorch core library. Backend owner is responsible for registering both
inference & autograd kernels in their extensions(e.g. torch-xla) for the operators they support.
E.g. XLA, XPU, MLC
E.g. XLA, XPU, MPS
- CompositeExplicitAutograd: alias key mapped to inference kernels of all backends like CPU, CUDA, XLA etc.
Kernels registered to this key MUST work for inference for all backends.
- Autograd: alias key mapped to autograd of all backends like AutogradCPU, AutogradXLA, AutogradOther.