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 supported8 Subscribers
Add a CodeTriage badge to pytorch-lightning
Help out
- Issues
- DirectML backend implementation prototype
- Support multi-run with hydra + DDP
- TPU v3-8 deadlocks when using datasets larger than 2^15 on 8 devices
- Mixed CPU/GPU Fabric instances
- CUDA OOM with DeepSpeed ZeRO Stage 3 Offload
- Unable to train on v3-8 TPUs with lightning. Training is stuck/deadlocked ?
- Prepare data in GPU device in DataModule/DataSet
- Enable multiple loggers to save to the same version directory
- Gradients seem not synchronized with manual optimization and DDPStrategy with `static_graph=True`
- Example of gradient clipping with manual optimization does not handle gradient unscaling properly
- Docs
- Python not yet supported