Pull Requests Lightning-AI/pytorch-lightning

Add adapt_checkpoint_hparams hook for customizing checkpoint hyperparameter loading pl

#21408 opened 2025-12-06 02:58 by arrdel

Move checkpoint for auto-requeue on SLURM clusters out of signal handlers pl

#21407 opened 2025-12-05 15:24 by martenlienen

Make the default SLURM requeue signal configurable fabric

#21405 opened 2025-12-05 15:14 by martenlienen

Gradient accumulation fix in cross entropy loss docs pl

#21386 opened 2025-11-27 18:33 by Sohaib-Ahmed21

feature(device): add MUSAAccelerator to support Musa device docs fabric pl

#21381 opened 2025-11-26 11:11 by chenkoucer

feat: add best_model_metrics to ModelCheckpoint callback docs ci fabric pl dependencies package

#21355 opened 2025-11-12 11:23 by Iruos8805

Update `pyproject.toml` `tool.ruff` target to py310 ready to be merged ci has conflicts fabric pl package exec-signoff-required

#21334 opened 2025-11-05 13:39 by matsumotosan

feat: allow configurable join_char for WandbLogger metric formatting logger: wandb pl

#21331 opened 2025-11-04 18:24 by Iruos8805

feat: add new callback hook `on_checkpoint_write_end` docs has conflicts pl

#21323 opened 2025-11-02 18:25 by Iruos8805

Fix progress bar deadlock on DDP metrics computation. has conflicts progress bar: rich pl

#21322 opened 2025-10-29 05:49 by tchittesh

[WIP] Compatibility with Pytorch 2.9 precision API bug fabric

#21306 opened 2025-10-20 10:04 by SkafteNicki

Feat/logger_map docs logger pl

#21297 opened 2025-10-19 09:36 by GdoongMathew

Do not replace custom sampler/dataloader when using `overfit_batches` bug pl

#21289 opened 2025-10-15 08:48 by SkafteNicki

Fix edgecase for early stopping bug callback: early stopping pl

#21279 opened 2025-10-10 05:11 by SkafteNicki

feat: add new callback hook `on_before_optimizer_setup` docs pl

#21272 opened 2025-10-06 17:22 by deependujha

refactor: try catch a more specific exception during log_metrics. ready to be merged refactor fabric pl exec-signoff-required

#21271 opened 2025-10-06 16:11 by GdoongMathew

Fix ReduceLROnPlateau scheduler error when validation doesn't run every epoch pl

#21266 opened 2025-10-04 21:05 by Copilot

docs: fix pip cmd.

#21242 opened 2025-09-22 13:55 by GdoongMathew

Remove Redundant Host & Device Synchronizations pl

#21233 opened 2025-09-21 05:42 by alpha0422

Docs/Chinese README has conflicts

#21221 opened 2025-09-15 16:28 by GdoongMathew

Autofix RET503 and RET504 fabric pl package

#21207 opened 2025-09-12 20:13 by matsumotosan

LR Finder: compute gradients w.r.t. log10(lr) in exponential mode (fix torch.gradient spacing) tuner has conflicts pl

#21171 opened 2025-09-07 18:06 by KAVYANSHTYAGI

Add `device_name` classmethod in `Accelerator`. has conflicts accelerator pl

#21112 opened 2025-08-24 08:30 by GdoongMathew

Document tensorboard experiment property docs logger has conflicts logger: tensorboard fabric pl dependencies

#21101 opened 2025-08-21 06:27 by SkafteNicki

Use mlflow variable to set max param length logger logger: mlflow pl dependencies

#21074 opened 2025-08-15 04:36 by SkafteNicki

feature: add `sub_dir` to CSV logger constructor logger has conflicts fabric logger: csv pl

#21045 opened 2025-08-11 01:04 by Shoumik-Gandre

Fix MLFlowLogger.save_dir Windows file URI handling logger logger: mlflow pl

#20988 opened 2025-07-19 15:25 by littlebullGit

docs(csv_logs): Clarify CSV and YAML logging distinction and improve examples docs logger logger: csv pl

#20951 opened 2025-07-01 09:08 by bhimrazy

update ModelSummary waiting on author pl

#20945 opened 2025-06-26 03:31 by YChienHung

Fix wrong behavior of `DDPStrategy` option with simple GAN training using DDP pl

#20936 opened 2025-06-25 02:45 by samsara-ku