ignite
https://github.com/pytorch/ignite
Python
High-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently.
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 supported3 Subscribers
Add a CodeTriage badge to ignite
Help out
- Issues
- Fix Perplexity overflow for half-precision logits
- ReduceLROnPlateauScheduler can't be restored from a checkpoint: state_dict contains a bound method
- Fix ReduceLROnPlateauScheduler state_dict containing a bound method
- Test distributed confusion matrices with rank-dependent inputs
- Preserve subgroup metric state in nested checkpoints
- FastaiLRFinder: first iteration ignores start_lr and get_results() pairs each loss with the next iteration's lr
- Fix FastaiLRFinder ignoring start_lr in the first iteration and logging shifted learning rates
- Fix loading list-backed metric state
- fix: clamp HitRate top_k to number of items
- Fix SSIM padding for 3D inputs with non-cubic kernel_size
- Docs
- Python not yet supported