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(pytorch): replace recursive __next__ in _Sequential CombinedLoader with iterative loop
- Fix `CSVLogger` resetting auto-incremented `step` to 0 after `save()` flush
- CPU `ddp_spawn` workers SIGABRT at exit: `_init_dist_connection` only destroys the process group for NCCL
- Fix ModelCheckpoint manual optimization snapshot restoration
- Fix(fabric): always register process-group atexit destroy for spawn backends
- perf(loggers): single-stat checkpoint scanning and deterministic merge_dicts
- Avoid mutating caller metrics dict in ModelCheckpoint._format_checkpoint_name
- Add `enable_device_summary` flag to Trainer
- Fix inverted inequality operator in `EarlyStopping` `divergence_threshold` message
- Support TorchTPU (torch_tpu) accelerator in Fabric and PyTorch Lightning
- Docs
- Python not yet supported