Changes
PyTorch Lightning
Changed
- Allow LightningCLI to use a customized argument parser class (#20596)
- Change
wandbdefault x-axis totensorboard'sglobal_stepwhensync_tensorboard=True(#20611) - Added a new
checkpoint_path_prefixparameter to the MLflow logger which can control the path to where the MLflow artifacts for the model checkpoints are stored (#20538) - CometML logger was updated to support the recent Comet SDK (#20275)
- bump: testing with latest
torch2.6 (#20509)
Fixed
- Fixed CSVLogger logging hyperparameter at every write which increases latency (#20594)
- Fixed OverflowError when resuming from checkpoint with an iterable dataset (#20565)
- Fixed swapped
_R_coand_Pto prevent type error (#20508) - Always call
WandbLogger.experimentfirst in_call_setup_hookto ensuretensorboardlogs can sync towandb(#20610) - Fixed TBPTT example (#20528)
- Fixed test compatibility as AdamW became a subclass of Adam (#20574)
- Fixed file extension of model checkpoints uploaded by NeptuneLogger (#20581)
- Reset trainer variable
should_stopwhenfitis called (#19177) - Fixed making
WandbLoggerupload models from allModelCheckpointcallbacks, not just one (#20191) - Error when logging to MLFlow deleted experiment (#20556)
Lightning Fabric
Changed
Removed
- Removed legacy support for
lightning run model; usefabric runinstead. (#20588)
Full commit list: 2.5.0 -> 2.5.1
Contributors
We thank all folks who submitted issues, features, fixes and doc changes. It's the only way we can collectively make Lightning ⚡ better for everyone, nice job!
In particular, we would like to thank the authors of the pull-requests above, in no particular order:
@benglewis, @Borda, @cgebbe, @duydl, @haifeng-jin, @japdubengsub, @justusschock, @lantiga, @mauvilsa, @millskyle, @ringohoffman, @ryan597, @senarvi, @TresYap
Thank you ❤️ and we hope you'll keep them coming!