Skip to content

Issues: Lightning-AI/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

515 Open 4,946 Closed
515 Open 4,946 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Implicit gradient accumulation? optimization optimizer question Further information is requested
#15045 opened Oct 8, 2022 by celsofranssa
1
module 'distutils' has no attribute 'version' needs triage Waiting to be triaged by maintainers
#15041 opened Oct 8, 2022 by yipliu
2
lightning.app.storage.Path tool fails to share a files between works needs triage Waiting to be triaged by maintainers
#15039 opened Oct 7, 2022 by DaniDapena
1
Calling add_argument inside LightningArgumentParser __init__ breaks custom functionality bug Something isn't working lightningcli pl.cli.LightningCLI
#15038 opened Oct 7, 2022 by Erotemic
unable to import/install pytorch_lightning pl Generic label for PyTorch Lightning package question Further information is requested
#15037 opened Oct 7, 2022 by kiristern
1
Callback not invoked for the validation set with DDP needs triage Waiting to be triaged by maintainers
#15028 opened Oct 7, 2022 by athn-nik
Keep User-Defined Order of Callbacks callback discussion In a discussion stage trainer: argument
#15026 opened Oct 7, 2022 by wistuba
Slurm Autorequeue Doesn't Work for Job Arrays bug Something isn't working environment: slurm
#15022 opened Oct 7, 2022 by Queuecumber pl:1.7.x
1
4
Overfit batches parameter gives a validation batch needs triage Waiting to be triaged by maintainers
#15021 opened Oct 6, 2022 by HekpoMaH
Keep a LightningWork "Hot" Until Run Method is Called app:lightningwork lightning_app.LightningWork app Generic label for Lightning App package feature Is an improvement or enhancement
#15015 opened Oct 6, 2022 by alecmerdler
3
Split out CloudRuntime.dispatch into multiple methods app Generic label for Lightning App package good first issue Good for newcomers priority: 2 Low priority task refactor
#15012 opened Oct 5, 2022 by awaelchli
Re-evaluate CUDA NVML and fork checks after torch 1.13 release accelerator: cuda Compute Unified Device Architecture GPU
#15010 opened Oct 5, 2022 by awaelchli pl:future
@awaelchli
pytorch lightning causes slurm nodes to drain environment: slurm question Further information is requested
#15008 opened Oct 5, 2022 by meshghi
4
Problem with mixed precision and transformer in validation bug Something isn't working precision: native amp Native Automatic Mixed Precision
#15006 opened Oct 5, 2022 by catalys1
@rohitgr7
Multiple GPU Training Very Slow performance strategy: ddp DistributedDataParallel
#15004 opened Oct 5, 2022 by samuelstevens
trainer.validate() will not load optimizer properly, different behavior from trainer.fit() strategy: deepspeed
#14993 opened Oct 4, 2022 by MattYoon
1
Add option to not flatten config in WandbLogger.log_hyperparams feature Is an improvement or enhancement logger: wandb Weights & Biases
#14988 opened Oct 4, 2022 by zephyap
2
Store last.ckpt as symlink when appropriate to save space checkpointing Related to checkpointing feature Is an improvement or enhancement
#14973 opened Oct 3, 2022 by ZhaofengWu
3
IndexError on multi-node multi-gpu SLURM script with DDP. environment: slurm question Further information is requested
#14949 opened Sep 29, 2022 by tsikup
1
CUDA OOM when running trainer.validate() with deepspeed at optimizer initialization (?) bug Something isn't working strategy: deepspeed
#14928 opened Sep 28, 2022 by DrMatters
5 tasks done
1
@rohitgr7
3
Lr_scheduler step where to use , warning optimization question Further information is requested
#14903 opened Sep 27, 2022 by msverma101
1
Tensors not on the same device when using FSDP auto-wrapping bug Something isn't working strategy: fsdp Fully Sharded Data Parallel
#14900 opened Sep 27, 2022 by awaelchli
@rohitgr7
2
Support FSDP in Lite feature Is an improvement or enhancement lightninglite pl.lite.LightningLite
#14898 opened Sep 26, 2022 by awaelchli pl:future
Allow to specify the value of the trainer/global_step metric on W&B during validation needs triage Waiting to be triaged by maintainers
#14892 opened Sep 25, 2022 by andreapesare
4
Allow strict=False in resuming trainer.fit() from a checkpoint duplicate This issue or pull request already exists feature Is an improvement or enhancement trainer: fit
#14879 opened Sep 24, 2022 by Jose-Bastos
1
1
ProTip! Add no:assignee to see everything that’s not assigned.