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 FSDPPrecision rejecting a scaler for 16-mixed precision
- Adopt typing_extensions.override in SpikeDetection callback
- fix: prevent ModelCheckpoint from deleting remote checkpoints from previous runs
- fix: use self.root_device.type instead of hardcoded cuda for device mesh initialization
- `ModelCheckpoint` deletes *previous run's* checkpoint when remote filesystem
- Device mesh initialization fails when running FSDP on CPU
- Mixed precision autocast is disabled when running FSDP on CPU
- Fix FSDP device mesh initialization on CPU
- feat: add num_nodes() to ClusterEnvironment
- Fix FSDP mixed precision autocast on CPU
- Docs
- Python not yet supported