Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Add `step` parameter to `TensorBoardLogger.log_hyperparams` (#20176)
* Add step to TensorBoardLogger.log_hyperparams The metrics that get logged using this method are always logged to step 0, so no meaningful graph is made * Change expected step value to None --------- Co-authored-by: Jirka Borovec <[email protected]> Co-authored-by: Luca Antiga <[email protected]>
M
Matthew Hoffman committed
1129d4cecf2bcbf78d9340655c3950e744a019e1
Parent: a9125c2
Committed by GitHub <[email protected]>
on 12/11/2024, 12:09:05 PM