Lightning-pose

What's Changed
Major updates for this release include multi-GPU training, new imgaug transforms, and many small usability improvements
* update sleap2lp.py by themattinthehatt in https://github.com/paninski-lab/lightning-pose/pull/196
* Upgrade sphinx-rtd-theme to fix search 200 by ksikka in https://github.com/paninski-lab/lightning-pose/pull/201
* Fix search attempt 2. Fix 200 by ksikka in https://github.com/paninski-lab/lightning-pose/pull/202
* Log hydra job config to tensorboard by ksikka in https://github.com/paninski-lab/lightning-pose/pull/204
* Change backbone finetuning strategy to allow for DDP by ksikka in https://github.com/paninski-lab/lightning-pose/pull/205
* Support multi-GPU training for supervised training by ksikka in https://github.com/paninski-lab/lightning-pose/pull/206
* Multi-GPU support for unsupervised learning by ksikka in https://github.com/paninski-lab/lightning-pose/pull/207
* Continuous integration using github actions by ksikka in https://github.com/paninski-lab/lightning-pose/pull/208
* add coarse salt/pepper to imgaug transforms by themattinthehatt in https://github.com/paninski-lab/lightning-pose/pull/209
* Added option to choose colormap for labelled videos by hummuscience in https://github.com/paninski-lab/lightning-pose/pull/211
* complete config file documentation by themattinthehatt in https://github.com/paninski-lab/lightning-pose/pull/213
* Add PCA Centering method (experimental) by ksikka in https://github.com/paninski-lab/lightning-pose/pull/214
* Context frames can use any delimiter by ksikka in https://github.com/paninski-lab/lightning-pose/pull/216
* Replace relative paths with absolute paths in demo config by ksikka in https://github.com/paninski-lab/lightning-pose/pull/217
* Fix doc build failure by ksikka in https://github.com/paninski-lab/lightning-pose/pull/218
* Update README.md with downloads/week badge by ksikka in https://github.com/paninski-lab/lightning-pose/pull/219

New Contributors
* ksikka made their first contribution in https://github.com/paninski-lab/lightning-pose/pull/201

**Full Changelog**: https://github.com/paninski-lab/lightning-pose/compare/v1.5.1...v1.6.0

Several small updates, as well as reverting the default training setup to automatically save out the best model according to validation metrics, rather than the last model.

* make sure validation loader always has at least one sample
* allow for early stopping again, as well as saving out model weights every n epochs
* make training fully deterministic
* increase flexibility of LitDaliWrapper to work with custom dali video pipeline functions
* increase flexibility of keypoint correspondence array for true multiview datasets

What's Changed
* Passing vfr_check to fn.readers.video by hummuscience in https://github.com/danbider/lightning-pose/pull/179, https://github.com/danbider/lightning-pose/pull/180
* remove augmentations for val/test data by themattinthehatt in https://github.com/danbider/lightning-pose/pull/181
* implement pca object that handles NaNs by themattinthehatt in https://github.com/danbider/lightning-pose/pull/184
* support for multiple data streams by themattinthehatt farzad-ziaie Selmaan in https://github.com/danbider/lightning-pose/pull/187


**Full Changelog**: https://github.com/danbider/lightning-pose/compare/v1.4.0...v1.5.0

What's Changed
* [docs] avi vs mp4 faq by themattinthehatt in https://github.com/danbider/lightning-pose/pull/151
* WSL installation instructions by DavidGill159 in https://github.com/danbider/lightning-pose/pull/153
* generalize keypoint access from dataframes in metrics by themattinthehatt in https://github.com/danbider/lightning-pose/pull/154
* ignore hidden "_labeled" folders in dlc to lp conversion by hummuscience in https://github.com/danbider/lightning-pose/pull/157
* misc updates to docs and tests by themattinthehatt in https://github.com/danbider/lightning-pose/pull/162
* add new unit test fixture by themattinthehatt in https://github.com/danbider/lightning-pose/pull/163
* [fix] streamlit video page bugfix by themattinthehatt in https://github.com/danbider/lightning-pose/pull/167
* add sleap2lp file by Shmuel-columbia in https://github.com/danbider/lightning-pose/pull/168
* update dependencies
* add backbone pretrained on hand dataset
* save multiple checkpoint files

New Contributors
* DavidGill159 made their first contribution in https://github.com/danbider/lightning-pose/pull/153
* hummuscience made their first contribution in https://github.com/danbider/lightning-pose/pull/157
* Shmuel-columbia made their first contribution in https://github.com/danbider/lightning-pose/pull/168

**Full Changelog**: https://github.com/danbider/lightning-pose/compare/v1.3.0...v1.4.0

What's Changed
* FIX: predict_new_vids.py now outputs unsupervised metrics as well by EugenioBertolini in https://github.com/danbider/lightning-pose/pull/145
* [docs] update config file info by themattinthehatt in https://github.com/danbider/lightning-pose/pull/147
* v1.3.0 updates by themattinthehatt in https://github.com/danbider/lightning-pose/pull/149
* generalize method of accessing likelihood in apps
* generalize subdirectory names used by video app
* move version def from setup.py to \_\_init__.py
* refactor train_hydra script to import a train function for greater flexibility

New Contributors
* EugenioBertolini made their first contribution in https://github.com/danbider/lightning-pose/pull/145

**Full Changelog**: https://github.com/danbider/lightning-pose/compare/v1.2.3...v1.3.0

include example data in PyPI packaging