Skip to content

Tags: lucidrains/byol-pytorch

Tags

0.8.2

Toggle 0.8.2's commit message
address #93

0.8.1

Toggle 0.8.1's commit message
address #92

0.8.0

Toggle 0.8.0's commit message
add BYOLTrainer for huggingface accelerate distributed training

0.7.2

Toggle 0.7.2's commit message
concat augmented images and only call online encoder once

0.7.1

Toggle 0.7.1's commit message
remove some inplace operations

0.7.0

Toggle 0.7.0's commit message
just automatically sync batchnorm if more than one machine detected, …

…allow for overriding with sync_batchnorm keyword argument

0.6.0

Toggle 0.6.0's commit message
fix simsiam, thanks to @chingisooinar

0.5.7

Toggle 0.5.7's commit message
add more informative error message

0.5.6

Toggle 0.5.6's commit message
default to returning projections, but can be turned off with return_p…

…rojection = False on forward

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.