Lightning-AI/pytorch-lightning
Resuming should allow to differentiate what to resume (steps/opti/weights)
オープン
#5,339 opened on 2021/01/03
featurehelp wantedpriority: 1
Repository metrics
- Stars
- (26,687 個のスター)
- PR merge metrics
- (平均マージ 9d 15h) (30d で 3 merged PRs)
説明
Currently it is possible to either resume only the full training state (epoch/global steps / optimizer / scheduler options / and weights), or only the weights.
I would like to be able to switch the optimizer at some point, i.e. skip restoring optimizer/scheduler, but still load the epoch/global steps. I only see a way to do this with hacks at the moment. Any other way? Could this be a feature, to specify in the trainer.init function specifically what to restore/what not to restore?