pytorch-lightning
https://github.com/pytorchlightning/pytorch-lightning
Python
The lightweight PyTorch wrapper for high-performance AI research. Scale your models, not the boilerplate.
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported12 Subscribers
Add a CodeTriage badge to pytorch-lightning
Help out
- Issues
- Fix resumed training drawing the first epoch with sampler epoch 0 when the dataloader uses workers
- feat(callbacks): add `min_delta` parameter to `ModelCheckpoint` (#14353)
- feat(TQDMProgressBar): add warmup_batches to exclude torch.compile overhead from ETA
- Do not run the wrapped module's property getter when setting an attribute
- Remove the call tracker's forward hooks and reset its flag on every call
- Fix batch size tuning after validation errors
- Keep the protocol on last-checkpoint paths from a remote dirpath
- First epoch after resuming from checkpoint uses sampler epoch 0 (regression from #20775, num_workers > 0)
- tests: skip test_tensorboard_with_symlink on Windows without symlink privilege
- test_tensorboard_with_symlink fails on Windows without symlink privilege
- Docs
- Python not yet supported