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 supported10 Subscribers
Add a CodeTriage badge to pytorch-lightning
Help out
- Issues
- Checkpoint every_n_steps reruns epoch on restore
- Multi-node Training with DDP stuck at "Initialize distributed..." on SLURM cluster
- Differentiate testing multiple sets/models when logging
- Current FSDPPrecision does not support custom scaler for 16-mixed precision
- Please make it simple!
- Multi-gpu training is much lower than single gpu (due to additional processes?)
- Enable batch size finder for distributed strategies
- TensorBoardLogger does not document .add_image() function
- Huge metrics jump between epochs && Step and epoch log not matched, when accumulate_grad_batches > 1
- Turn off hpc checkpoint saving in SLURM environment if trainer.fit(..., ckpt_path="last")
- Docs
- Python not yet supported