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.
pkg_resources has been deprecated a while ago, and the deprecation warning turned into an error with a recent setuptools update. This PR replaces pkg_resources.packaging.Version in pytorch/attention.py with packaging.version.Version and adds packaging as an install dependency in `setup.py.
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
Replace pkg_sources.packaging.Version with packaging.version.Version in pytorch/attention.py.
Add packaging as an install dependency in setup.py.
Description
pkg_resources
has been deprecated a while ago, and the deprecation warning turned into an error with a recentsetuptools
update. This PR replacespkg_resources.packaging.Version
inpytorch/attention.py
withpackaging.version.Version
and addspackaging
as an install dependency in `setup.py.Type of change
Changes
pkg_sources.packaging.Version
withpackaging.version.Version
inpytorch/attention.py
.packaging
as an install dependency insetup.py
.Checklist: