Accelerate

**Full Changelog**: https://github.com/Chris-hughes10/pytorch-accelerated/compare/v0.1.43...v0.1.44

Minor updates to make an edge case error more readable

What's Changed
* Bump accelerate version to 1.17 by Chris-hughes10 in https://github.com/Chris-hughes10/pytorch-accelerated/pull/48


**Full Changelog**: https://github.com/Chris-hughes10/pytorch-accelerated/compare/v0.1.41...v0.1.42

What's Changed
* Update examples format by Chris-hughes10 in https://github.com/Chris-hughes10/pytorch-accelerated/pull/42
* Refactor loss tracking by Chris-hughes10 in https://github.com/Chris-hughes10/pytorch-accelerated/pull/44
* bump accelerate version by Chris-hughes10 in https://github.com/Chris-hughes10/pytorch-accelerated/pull/45
* Enables distributed evaluation on uneven inputs by Chris-hughes10 in https://github.com/Chris-hughes10/pytorch-accelerated/pull/46


**Full Changelog**: https://github.com/Chris-hughes10/pytorch-accelerated/compare/v0.1.40...v0.1.41

What's Changed
* Add option to execute callbacks during ModelEma evaluation loop by Chris-hughes10 in https://github.com/Chris-hughes10/pytorch-accelerated/pull/41


**Full Changelog**: https://github.com/Chris-hughes10/pytorch-accelerated/compare/v0.1.39...v0.1.40

What's Changed
* Improve gathering to automatically pad tensors across processes
* Add `get_model` method in Trainer by bepuca in https://github.com/Chris-hughes10/pytorch-accelerated/pull/39


**Full Changelog**: https://github.com/Chris-hughes10/pytorch-accelerated/compare/v0.1.38...v0.1.39