NVIDIA / TransformerEngine

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper and Ada GPUs, to provide better performance with lower memory utilization in both training and inference.

Apache License 2.0

1.61k stars 256 forks source link

Description

Nvfuser is depracated starting torch 2.2 so disabling that here

Fixes # (issue)

Type of change

[ ] Documentation change (change only to the documentation, either a fix or a new content)
[x] Bug fix (non-breaking change which fixes an issue)
[ ] New feature (non-breaking change which adds functionality)
[ ] Breaking change (fix or feature that would cause existing functionality to not work as expected)

Changes

Please list the changes introduced in this PR: Ignore nvfuser options when torch version >= 2.2

NVIDIA / TransformerEngine

disable using nvfuser when pytorch version >= 2.2 #905

Description

Type of change

Changes