projects / platform / upstream / pytorch.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 54abfda) | patch
Adds Cyclical Learning Rate and Momentum (#18001)
authorSam Pepose <sampepose@fb.com>
Thu, 28 Mar 2019 02:47:43 +0000 (19:47 -0700)
committerFacebook Github Bot <facebook-github-bot@users.noreply.github.com>
Thu, 28 Mar 2019 02:56:04 +0000 (19:56 -0700)
commit8635078d9ed47266e6df89ad5ef887b598a76f25
tree78704f54331627703849b5342182d22df217d77atree | snapshot
parent54abfda12434692ea5902ce5f94062fdac7fde61commit | diff
Adds Cyclical Learning Rate and Momentum (#18001)

Summary:
This implements a cyclical learning rate (CLR) schedule with an optional inverse cyclical momentum. More info about CLR: https://github.com/bckenstler/CLR

This is finishing what #2016 started. Resolves #1909.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/18001

Differential Revision: D14451845

Pulled By: sampepose

fbshipit-source-id: 8f682e0c3dee3a73bd2b14cc93fcf5f0e836b8c9
docs/source/optim.rst diff | blob | history
test/test_optim.py diff | blob | history
torch/optim/lr_scheduler.py diff | blob | history
Domain: Machine Learning / ML Framework;