saymrwulf/pytorch
1
0
Fork 0
mirror of https://github.com/saymrwulf/pytorch.git synced 2026-05-14 20:57:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

torch.jit doc link for nvfuser readme.md (#77780)

Browse source 
adding a quick link to nvfuser README.md in jit doc

Note that for 1.12 release, we probably want to have the link pointed to the doc in the release code base. I don't know if we have a tag for 1.12 release candidate yet, so we might want to update that.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/77780
Approved by: https://github.com/davidberard98
This commit is contained in:
jjsjann123 2022-07-07 23:25:34 +00:00 committed by PyTorch MergeBot
parent ae6dd20ba7
commit d2c726d43c
1 changed files with 1 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
docs/source/jit.rst
View file

@ -876,7 +876,7 @@ now supported.
Fusion Backends
~~~~~~~~~~~~~~~
There are a couple of fusion backends available to optimize TorchScript execution. The default fuser on CPUs is NNC, which can perform fusions for both CPUs and GPUs. The default fuser on GPUs is NVFuser, which supports a wider range of operators and has demonstrated generated kernels with improved throughput. See the `NVFuser documentation <https://github.com/pytorch/pytorch/blob/release/1.12/torch/csrc/jit/codegen/cuda/README.md>`_ for more details on usage and debugging.
There are a couple of fusion backends available to optimize TorchScript execution. The default fuser on CPUs is NNC, which can perform fusions for both CPUs and GPUs. The default fuser on GPUs is NVFuser, which supports a wider range of operators and has demonstrated generated kernels with improved throughput. See the `NVFuser documentation <https://github.com/pytorch/pytorch/blob/master/torch/csrc/jit/codegen/cuda/README.md>`_ for more details on usage and debugging.
References