Skip to content

Issues: Lightning-AI/pytorch-lightning

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Clear current search query, filters, and sorts
49 Open 1,225 Closed
49 Open 1,225 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

feat: add support for cloud files and non-local checkpoints pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#20701 opened Apr 7, 2025 by parambharat • Draft
4 of 8 tasks
1
2
self.all_gather does not work on on_train_epoch_end bug Something isn't working needs triage Waiting to be triaged by maintainers ver: 2.5.x won't fix This will not be worked on
#20683 opened Mar 28, 2025 by yoniaflalo
1
Add support for Comet experiment data flush after saving checkpoint into CometLogger feature Is an improvement or enhancement needs triage Waiting to be triaged by maintainers won't fix This will not be worked on
#20681 opened Mar 27, 2025 by yaricom
1
2
Non-reproducible results with num_workers=0 bug Something isn't working needs triage Waiting to be triaged by maintainers ver: 2.5.x won't fix This will not be worked on
#20679 opened Mar 27, 2025 by mrava87
1
Allow Using Specific Dataloaders in LightningCLI (Instead of a DataModule) feature Is an improvement or enhancement needs triage Waiting to be triaged by maintainers won't fix This will not be worked on
#20673 opened Mar 26, 2025 by stuartthomson
1
Add fp8 (torchao)/fsdp2/torch_compile handlers and tests pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#20445 opened Nov 25, 2024 by qingquansong • Draft
1 of 7 tasks
2
Better registration support for a wide range of third-party hardware accelerator docs Documentation related fabric lightning.fabric.Fabric pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#20349 opened Oct 19, 2024 by zhiyuan1i Loading…
7 of 8 tasks
20
Support variable batch size in throughput callback pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#20236 opened Aug 29, 2024 by alex-hh • Draft
7 tasks
1
2
fix: allow loading of nested states in Fabric.load [wip] fabric lightning.fabric.Fabric waiting on author Waiting on user action, correction, or update won't fix This will not be worked on
#20210 opened Aug 18, 2024 by Markus28 • Draft
4 of 11 tasks
1
2
Add email callback on train complete pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#20162 opened Aug 4, 2024 by loucaspapalazarou Loading…
5 of 7 tasks
1
2
Test optimizer to device fabric lightning.fabric.Fabric pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#20062 opened Jul 9, 2024 by corwinjoy • Draft
7 tasks
1
5
WIP: Disable _optimizer_to_device logic fabric lightning.fabric.Fabric has conflicts pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#20036 opened Jul 2, 2024 by corwinjoy Loading…
7 tasks
4
modified num_replicas=self.world_size pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#19962 opened Jun 9, 2024 by arjunagarwal899 • Draft
2 of 7 tasks
1
1
Include VertexAI cluster environment for Fabric docs Documentation related fabric lightning.fabric.Fabric has conflicts pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#19911 opened May 27, 2024 by miguelalba96 Loading…
4 tasks
1
WIP: Integrate Collective into strategies fabric lightning.fabric.Fabric refactor won't fix This will not be worked on
#19881 opened May 17, 2024 by awaelchli • Draft
1
Added some more potentially robust ways to do learning rate tuning has conflicts pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#19867 opened May 13, 2024 by varchasgopalaswamy Loading…
3 of 7 tasks
1
Log TensorBoard histograms fabric lightning.fabric.Fabric has conflicts pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#19851 opened May 6, 2024 by dominicgkerr Loading…
6 of 10 tasks
1
2
[WAITING FOR PL CORE MAINTAINER OPINION] Bugfix/17958 multi optimizer step count behaviour docs Documentation related has conflicts pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#19589 opened Mar 7, 2024 by Anner-deJong Loading…
4 of 8 tasks
1
4
Enable support for Intel XPU devices (AKA Intel GPUs) data (external) litdata package fabric lightning.fabric.Fabric pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#19443 opened Feb 9, 2024 by coreyjadams • Draft
4
Delete stub forward() method in LightningModule community This PR is from the community has conflicts pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#19423 opened Feb 6, 2024 by BlackHC Loading…
7 tasks
5
CPU-Memory keeps accumulating during trainer.predict bug Something isn't working performance trainer: predict ver: 2.1.x won't fix This will not be worked on
#19398 opened Feb 2, 2024 by surajpaib
1
5
[NEPTUNE] Optimize uploading k-best model checkpoints pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#19071 opened Nov 27, 2023 by AleksanderWWW • Draft
7 tasks
2
Add support to saving.py for loading GPU-trained models on CPU-only machines pl Generic label for PyTorch Lightning package won't fix This will not be worked on
#19024 opened Nov 17, 2023 by amorehead Loading…
9 of 12 tasks
14
Adding defaults to hparams in base class LightningModule causes incorrect _hparams_name in checkpoints bug Something isn't working question Further information is requested ver: 2.0.x won't fix This will not be worked on
#17889 opened Jun 21, 2023 by LarsKue
1
4
Logged monitor value for train_loss after early stopping does not match the stopping criteria bug Something isn't working needs triage Waiting to be triaged by maintainers ver: 2.0.x won't fix This will not be worked on
#17644 opened May 16, 2023 by jlperla
2
ProTip! Add no:assignee to see everything that’s not assigned.