-
Notifications
You must be signed in to change notification settings - Fork 3.8k
Pull requests: Lightning-AI/pytorch-lightning
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
docs: add :emphasize-lines: directives to code blocks in code_structure.rst
#21950
opened Sep 12, 2026 by
CJstate
Loading…
chore: update changelogs and bump to 2.7.0.dev
#21949
opened Sep 10, 2026 by
bhimrazy
Collaborator
Loading…
Docs: Pytorch: Optimization: correct return type of configure_optimizers
#21947
opened Sep 10, 2026 by
sohang3112
Loading…
4 of 7 tasks
Fix batch size tuning after validation errors
#21946
opened Sep 9, 2026 by
WenJing95
Loading…
3 of 7 tasks
Remove the call tracker's forward hooks and reset its flag on every call
#21945
opened Sep 9, 2026 by
vineethsaivs
Contributor
Loading…
Do not run the wrapped module's property getter when setting an attribute
#21944
opened Sep 9, 2026 by
vineethsaivs
Contributor
Loading…
feat(TQDMProgressBar): add warmup_batches to exclude torch.compile overhead from ETA
#21943
opened Sep 9, 2026 by
piyushumate
Loading…
feat(callbacks): add
min_delta parameter to ModelCheckpoint (#14353)
#21942
opened Sep 9, 2026 by
piyushumate
Loading…
7 tasks done
Fix resumed training drawing the first epoch with sampler epoch 0 when the dataloader uses workers
#21941
opened Sep 9, 2026 by
VihaanAgarwal
Loading…
3 tasks done
tests: skip test_tensorboard_with_symlink on Windows without symlink privilege
#21939
opened Sep 8, 2026 by
CJstate
Loading…
test: skip symlink case without Windows privilege
#21937
opened Sep 8, 2026 by
mikemikimike
Loading…
Keep the protocol on last-checkpoint paths from a remote dirpath
#21936
opened Sep 7, 2026 by
vineethsaivs
Contributor
Loading…
refactor: use public
LRScheduler instead of private _LRScheduler in annotations
#21934
opened Sep 7, 2026 by
bhimrazy
Collaborator
Loading…
docs: Add :emphasize-lines: to code blocks in code_structure.rst (fixes #12119)
#21933
opened Sep 7, 2026 by
CJstate
Loading…
Fix the uneven first step in the lr_finder learning-rate ladder
#21931
opened Sep 5, 2026 by
vineethsaivs
Contributor
Loading…
Fix FSDP non-finite gradient norm handling
#21929
opened Sep 3, 2026 by
bwang-pplx
Loading…
6 tasks done
docs: fix linkcheck by replacing dead www.pytorchlightning.ai community links
#21927
opened Sep 1, 2026 by
bhimrazy
Collaborator
Loading…
6 of 7 tasks
Fix step scheduler frequency across epochs
#21926
opened Aug 30, 2026 by
Alphaxiaoteng
Loading…
5 tasks done
Fix lr_finder suggestion discarding every sample when skip_end=0
#21923
opened Aug 29, 2026 by
vineethsaivs
Contributor
Loading…
5 tasks done
Negate the deprecated offline flag in CometLogger
#21921
opened Aug 26, 2026 by
VenishPaneliya
Loading…
docs: explain optimizer access in manual optimization
#21920
opened Aug 26, 2026 by
nightcityblade
Loading…
3 of 7 tasks
fix: drop references to wandb's removed RunDisabled class
#21919
opened Aug 26, 2026 by
dmitryduev
Loading…
Fix scale_batch_size running zero batches when the search starts mid-training
#21916
opened Aug 22, 2026 by
vineethsaivs
Contributor
Loading…
7 tasks done
fix(core): pin explicit Any type for yaml.full_load result in load_hparams_from_yaml
#21912
opened Aug 20, 2026 by
mayuriphad
Loading…
Prevent ModelCheckpoint from deleting resumed remote checkpoints
#21910
opened Aug 18, 2026 by
tandede
Loading…
4 tasks done
Previous Next
ProTip!
Add no:assignee to see everything that’s not assigned.