mgrankin / over9000

Over9000 optimizer

mgrankin/over9000 Issues

Please update README with latest optimezers like MADGRAD and AdaBelief
Closed 3 years ago2
Performance mismatch between notebook and readme
Closed 4 years ago2
If I use RangerLars should I use any additional lr_scheduler
Closed 4 years ago1
Recommendation on optimal LR
Closed 4 years ago3
RangerLars
Closed 4 years ago2
Update benchmarcks now that Ranger supports gradient centralization
Closed 4 years ago1
Port to fastai v2?
Closed 4 years ago6
How about RAdam + LAMB + Lookahead?
Closed 4 years ago2
What do you think about weight decay?
Closed 5 years ago1
Failure to execute on Windows.
Closed 5 years ago12
AdaBound
Closed 5 years ago6
question about LAMB implementation.
Closed 5 years ago7
Citing the repository
Closed 5 years ago2
What is the meaning of Imagenette and Imagewoof ?
Closed 5 years ago2
LR Scheduler
Closed 5 years ago2
Merge them all?
Closed 5 years ago1
How to use over9000 in common pytorch code?
Closed 5 years ago10
Lookahead has no attribute 'state'
Closed 5 years ago2
Testing that uses LAMB only
Closed 5 years ago2