A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch
BSD 3-Clause "New" or "Revised" License
8.2k
stars
1.36k
forks
source link
Add type hints to distributed Adam optimizer #1699
Closed
timmoon10 closed 11 months ago
This PR makes some minor stylistic changes to the distributed Adam optimizer:
GradientBucket.sync_wait
andParameterBucket.sync_wait
,ParameterFragment
class a dataclass.None of these changes should affect functionality.