Fastxtend

**Full Changelog**: https://github.com/warner-benjamin/fastxtend/compare/v0.0.16...v0.0.17

What's Changed
* Easily Access Original Transformers Model https://github.com/warner-benjamin/fastxtend/pull/22
* Bug Fix: Remove delegates for log_wandb_table https://github.com/warner-benjamin/fastxtend/pull/20


**Full Changelog**: https://github.com/warner-benjamin/fastxtend/compare/v0.1.6...v0.1.7

What's Changed
* Audio Fixes for Current Versions of fastcore & torchaudio
* Foreach EMA fixed to work with multiple types


**Full Changelog**: https://github.com/warner-benjamin/fastxtend/compare/v0.1.5...v0.1.6

Fixes a few doc strings.

- Adds support for training Hugging Face Transformers models via `HuggingFaceCallback` & `HuggingFaceLoader`
- Profiler Callbacks updated to work with `HuggingFaceCallback` & `HuggingFaceLoader`
- Compile Callback improvements
- Support LR Finder
- Support PyTorch 2.1 Compile format
- Improve saved model compatibility
- `GradientAccumulation` & `GradientAccumulationSchedule` improvements
- Auto-detect micro-batch size if using fastai dataloader
- Drop last macro-batch to match full size training
- Add `DataLoaderMixin` for adding fastai dataloader features to non-fastai dataloaders
- Switch FFCV `Loader` to use `DataLoaderMixin`

**Full Changelog**: https://github.com/warner-benjamin/fastxtend/compare/v0.1.4...v0.1.5