Repository Issues
Lightning-AI/pytorch-lightning
Pretrain, finetune and deploy AI models on multiple GPUs, TPUs with zero code changes.
Issues
Open
Variable length number of batches is not supported
bugdata handlinghelp wantedver: 2.0.x
No assignee yetHas a beginner-friendly labelContributing guide available
2 comments2 reactions0 assignees
Open
Edge case causes incorrect filesystem to be selected for finding cloud checkpoints
3rd partybugcheckpointinghelp wantedver: 2.0.x
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
6 comments1 reaction0 assignees
Open
Graceful termination doesn't work when DataLoaders have num_workers > 0
bughelp wantedver: 2.0.x
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
3 comments0 reactions0 assignees
Open
self.log: `strategy.reduce` vs `reduce_fx`
featurehelp wantedloggingpl
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
1 comment0 reactions0 assignees
Open
[WandbLogger] Call `wandb.finish()` to ensure all artifacts are uploaded before training ends.
bughelp wantedlogger: wandbver: 2.1.x
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
2 comments2 reactions0 assignees
Open
Update deepspeed activation checkpointing docs
docshelp wantedstrategy: deepspeed
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
1 comment0 reactions0 assignees
Open
Refactor LearningRateMonitor
callback: lr monitorfeaturehelp wantedrefactor
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
2 comments0 reactions0 assignees
Open
TensorboardLogger not handling remote filesystem paths correctly on Windows environment
bughelp wantedloggerlogger: tensorboardver: 2.0.x
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
1 comment1 reaction0 assignees
Open
An example for knowledge distillation. Especially for the load the teacher model's weights , and train the student model
examplefeaturehelp wanted
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
2 comments1 reaction0 assignees
Open
WandbLogger doesn't format config correctly
bughelp wantedlogger: wandbver: 2.0.x
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
18 comments1 reaction0 assignees
Open
Stepwise LR-Scheduler not working across epochs
bughelp wantedlr schedulerver: 2.0.x
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
3 comments0 reactions0 assignees
Open
Init arguments upon multiple inheritance from LightningDataModule and other parent class
bughelp wantedlightningdatamodulever: 2.0.x
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
2 comments0 reactions0 assignees
Open
Resume from checkpoint not working on multi-gpu instance
bughelp wantedrepro neededver: 1.9.x
Why recommendedNo assignee yet · No comments yet
No assignee yetNo comments yetHas a beginner-friendly labelContributing guide available
0 comments0 reactions0 assignees
Open
If ModelCheckpoint and Validation occur on the same step, validation should run first, reopen
bugcallback: model checkpointhelp wantedrepro needed
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
1 comment0 reactions0 assignees
Open
Add device information to the accelerator config message
acceleratorfeaturehelp wantedtrainer
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
4 comments0 reactions0 assignees
Open
Why recommendedNo assignee yet · No comments yet
No assignee yetNo comments yetHas a beginner-friendly labelContributing guide available
0 comments2 reactions0 assignees
Open
Step when validation happens drifts for `val_check_interval` when gradient accumulation turned on
bughelp wantedloopsver: 2.0.x
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
5 comments2 reactions0 assignees
Open
Incompatibility with MLFlow 1.30
3rd partybughelp wantedlogger: mlflow
Why recommendedHas a beginner-friendly label · Contributing guide available
Has a beginner-friendly labelContributing guide available
4 comments0 reactions1 assignee
Open
`TensorBoardLogger` does not save logs correctly on GCS
bughelp wantedlogger: tensorboard
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
5 comments2 reactions0 assignees
Open
`ProgressBar` created in `LightningModule.configure_callbacks` does not override default `ProgressBar`
bugcallbackdiscussionhelp wanted
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelContributing guide available
3 comments2 reactions0 assignees