lucidrains/ema-pytorch
A simple way to keep track of an Exponential Moving Average (EMA) version of your pytorch model
PythonMIT
Issues
- 0
- 2
torch.no_grad on net?
#25 opened by TKassis - 0
Leaf/computed variables requires_grad error
#20 opened by crlandsc - 11
Post-hoc ema?
#14 opened by elyxlz - 1
- 4
question about usage
#11 opened by fire717 - 5
Support for non-deepcopyable models
#6 opened by stefan-baumann - 1
Using EMA in a DDP setup
#9 opened by JamesMcCullochDickens - 1
Implementation error
#8 opened by sbl1996 - 4
Saving EMA also saves online model?
#7 opened by Netruk44 - 0
about the training speed
#5 opened by winnechan - 1
Question about EMA
#1 opened by slerman12