torch.utils.tensorboard import #84712Description
🚀 The feature, motivation and pitch
Importing the tensorboard writer yields a deprecation warning from distutils in python 3.10, i.e.,
DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
This can be easily fixed as mentioned in the deprecation warning. One needs to replace the current lines 2 - 9 (https://github.com/pytorch/pytorch/blob/master/torch/utils/tensorboard/__init__.py#L2)
from distutils.version import LooseVersion
if not hasattr(tensorboard, "__version__") or LooseVersion(
tensorboard.__version__
) < LooseVersion("1.15"):
raise ImportError("TensorBoard logging requires TensorBoard version 1.15 or above")
del LooseVersion
with
import packaging
if not hasattr(tensorboard, "__version__") or packaging.version.parse(
tensorboard.__version__
) < packaging.version.Version("1.15"):
raise ImportError("TensorBoard logging requires TensorBoard version 1.15 or above")
del packaging
I am happy to submit the PR but I do not know how to test against all supported python versions and changing all requirements correctly in all different build targets.
Alternatives
No response
Additional context
No response