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

833 Open 6,343 Closed
833 Open 6,343 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

Documentation Example of Truncated BPTT is not working; self.optimizer.step() makes no sense docs Documentation related needs triage Waiting to be triaged by maintainers
#20517 opened Dec 22, 2024 by simon-bachhuber
DistNetworkError when using multiprocessing_context parameter in pytorch dataloader bug Something isn't working needs triage Waiting to be triaged by maintainers ver: 2.4.x
#20516 opened Dec 21, 2024 by forestbat
The breadcrumb navigation is wrong. docs Documentation related needs triage Waiting to be triaged by maintainers
#20503 opened Dec 19, 2024 by slgxmh
Make logging_mode (on_step or on_epoch) available to loggers feature Is an improvement or enhancement needs triage Waiting to be triaged by maintainers
#20502 opened Dec 17, 2024 by SiddhantSadangi
DistributedTraining issue bug Something isn't working needs triage Waiting to be triaged by maintainers ver: 2.4.x
#20498 opened Dec 16, 2024 by moghadas76
MPS Mixed-precision Autocast feature Is an improvement or enhancement needs triage Waiting to be triaged by maintainers
#20497 opened Dec 13, 2024 by laclouis5
PaliGemma fine-tuning - error with distributed training bug Something isn't working needs triage Waiting to be triaged by maintainers ver: 2.4.x
#20496 opened Dec 13, 2024 by vdrvar
Error learning rate when load ckpt for cotinue training if check_val_every_n_epoch > 1 bug Something isn't working needs triage Waiting to be triaged by maintainers ver: 2.4.x
#20495 opened Dec 13, 2024 by razgzy
3
Activation config for callbacks feature Is an improvement or enhancement
#20494 opened Dec 13, 2024 by JohnHerry
1
Synchronize validation and test logging needs triage Waiting to be triaged by maintainers refactor
#20492 opened Dec 12, 2024 by jialiangZ
Loading checkpoint before fabric.setup(model) gets abnormal loss when using fabric.init_module() bug Something isn't working ver: 2.4.x
#20490 opened Dec 11, 2024 by kobenaxie
4
Allow user to use dataloader_idx in log name in LightningModule.log feature Is an improvement or enhancement
#20485 opened Dec 10, 2024 by lokesh-vr-17773
1
Model diverges or struggles to converge with complex-valued tensors in DDP 3rd party Related to a 3rd-party bug Something isn't working ver: 2.4.x
#20480 opened Dec 9, 2024 by ouioui199
2
Enforce port range for Apps on localhost (similar to cloudspace) feature Is an improvement or enhancement needs triage Waiting to be triaged by maintainers
#20470 opened Dec 5, 2024 by hummuscience
1
Pickling error when using cuda in collate_fn 3rd party Related to a 3rd-party ver: 2.4.x waiting on author Waiting on user action, correction, or update
#20469 opened Dec 5, 2024 by Richienb
1
A gracefull design to introduce third-party models as tool for validation design Includes a design discussion feature Is an improvement or enhancement
#20464 opened Dec 4, 2024 by JohnHerry
1
ModelCheckpointCallback is triggered by mistake after every validation stage when mannual optimization design Includes a design discussion working as intended Working as intended
#20459 opened Nov 29, 2024 by silverbulletmdc
@lantiga
4
Jax Support within Lightning design Includes a design discussion feature Is an improvement or enhancement
#20458 opened Nov 28, 2024 by ludwigwinkler
@lantiga
2
[BUG] (DataLoader) sanity check fails due to Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) 3rd party Related to a 3rd-party ver: 2.4.x
#20456 opened Nov 27, 2024 by MathiasBaumgartinger
6
Ask a Question and Chat with us are not working. bug Something isn't working ver: 2.4.x
#20455 opened Nov 27, 2024 by sorenwacker
@lantiga
1
"FileNotFoundError: The provided path is not a valid DeepSpeed checkpoint" when using strategy='deepspeed_stage_2' bug Something isn't working needs triage Waiting to be triaged by maintainers ver: 2.4.x
#20453 opened Nov 26, 2024 by ShiweiWu98
Error: Invalid value for '--accelerator': 'auto' is not one of 'cpu', 'gpu', 'cuda', 'mps', 'tpu'. fabric lightning.fabric.Fabric feature Is an improvement or enhancement help wanted Open to be worked on ver: 2.4.x
#20451 opened Nov 26, 2024 by wkd88
2
Make sure the upcoming change in the default for weights_only from False to True is handled correctly ver: 2.4.x ver: 2.6.x
#20450 opened Nov 26, 2024 by lantiga
BatchSizeFinder safety margin feature Is an improvement or enhancement needs triage Waiting to be triaged by maintainers
#20447 opened Nov 25, 2024 by edmcman
Add on_validation_model_train/eval to Callback API as well feature Is an improvement or enhancement needs triage Waiting to be triaged by maintainers
#20441 opened Nov 22, 2024 by yundai424
ProTip! Follow long discussions with comments:>50.