Torch-uncertainty

What's Changed

- We improve the handling of optional packages.
- We fix AURC metrics in the multi-GPU setting
- We improve the mc-dropout wrapper and its documentation

* :shirt: Small fixes and improvements by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/109

**Full Changelog**: https://github.com/ENSTA-U2IS-AI/torch-uncertainty/compare/v0.2.1...v0.2.1.post0

What's Changed
* :sparkles: Add LPBNN in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/90
* :sparkles: Implement Adaptive ECE metric by qbouniot in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/92
* :sparkles: Finalize Depth Estimation, add DeeplabV3, & Add Selective Classification in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/88
* :sparkles: Add LPBNN, Adaptive ECE, start supporting Depth estimation & Improve segmentation in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/93
* :bug: Fix documentation in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/95
* :sparkles: Add a Laplace wrapper in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/96
* :sparkles: Add trajectory models, including Snapshot Ensembles in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/101
* :sparkles: Refactor wrappers & PP, Add Checkpoint Ensembles, EMA, SWA, & SWAG, Add LaplaceApprox & ABNN in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/98

Thanks to alafage for the review.

**Full Changelog**: https://github.com/ENSTA-U2IS-AI/torch-uncertainty/compare/v0.2.0...v0.2.1

What's Changed
* :sparkles: Add LPBNN by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/90
* :sparkles: Implement Adaptive ECE metric by qbouniot in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/92
* :sparkles: Finalize Depth Estimation, add DeeplabV3, & Add Selective Classification by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/88
* :sparkles: Add LPBNN, Adaptive ECE, start supporting Depth estimation & Improve segmentation by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/93
* :bug: Fix documentation by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/95
* :sparkles: Add a Laplace wrapper by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/96
* :sparkles: Add trajectory models including Snapshot Ensembles by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/101
* :sparkles: Refactor wrappers & PP, Add Checkpoint Ensembles, EMA, SWA, & SWAG, Add LaplaceApprox & ABNN by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/98
* :zap: Bump version by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/108


**Full Changelog**: https://github.com/ENSTA-U2IS-AI/torch-uncertainty/compare/v0.2.0...v0.2.1

This update brings a complete overhaul reconstruction around our uncertainty-aware routines. Highlights include:

- Lightning 2.0: Support and a complete overhaul of the command-line interface.

- RegressionRoutine: Fully functional, now supporting probabilistic regression with PyTorch distributions.

- SegmentationRoutine: Introduces semantic segmentation support for datasets like Cityscapes and MUAD.

Stay tuned for even more (Monocular depth estimation!) in TorchUncertainty 0.2.1!

Breaking Changes

As we are still in pre-release, this version breaks a large part of the routine and CLI components of TorchUncertainty 0.1.6.

CLI

The behavior of the CLI has completely changed and is now based on the configuration files from Lightning 2.0. We provide a new [page](https://torch-uncertainty.github.io/cli_guide.html) that explains how to leverage Baselines using the CLI for easy benchmarking.

Routines

Notably, there is no more distinction between ensemble and single routines to reduce code entropy: single routines are ensemble routines with 1 estimator. Furthermore, the routines' loss parameters now take an instantiated loss instead of a type, the optimization_procedure is renamed optim_recipe and is now a dictionary and not a callable. The ood_criterion and the calibration sets are now strings.

Metrics

The NegativeLogLikelihood metric is renamed CategoricalNLL.

Baselines

All baselines have been renamed to explicitly contain "Baselines" in their name.

Tutorials

We have rewritten and updated the [tutorials](https://torch-uncertainty.github.io/auto_tutorials/index.html) should now be clearer. Send us feedback!

What's Changed
* :heavy_minus_sign: Avoid using Argvcontext in tutorials by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/82
* 🚀 Upgrade to Lightning 2.0 by alafage in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/79
* :rocket: Update to Lightning 2.0, Add Segmentation, & Rework Regression by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/85


**Full Changelog**: https://github.com/ENSTA-U2IS-AI/torch-uncertainty/compare/v0.1.6...v0.2.0

What's Changed
* ⬆️ Bump tj-actions/changed-files from 34 to 41 in /.github/workflows by dependabot in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/75
* :sparkles: Add ResNet-20, corruptions and improve docs by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/76
* :sparkles: Add the grouping loss to single model training by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/77
* :sparkles: Add Monte-Carlo Batch Normalization by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/78
* :sparkles: Add grouping loss, Monte-Carlo Batch Normalization, OpenImage-O, MUAD & Improve code quality by o-laurent in https://github.com/ENSTA-U2IS-AI/torch-uncertainty/pull/80


**Full Changelog**: https://github.com/ENSTA-U2IS-AI/torch-uncertainty/compare/v0.1.5...v0.1.6

What's Changed
* :sparkles: Improve documentation, Enable BNN on GPUs, improve BNN code & code overall quality by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/44
* :sparkles: Continue improving Bayesian layers, & Fix MI in rare cases by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/45
* :sparkles: Add Deep Evidential Regression, generalize Packed layers, and refactor the datasets by badrmarani in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/46 and o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/48
* :sparkles: Add visualization tools by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/50
* ⬆️ Bump urllib3 from 2.0.6 to 2.0.7 by dependabot in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/51
* :sparkles: Add MC-Dropout & Visualization tools by by badrmarani in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/49 and o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/53
* ⬆️ Bump werkzeug from 3.0.0 to 3.0.1 by dependabot in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/62
* :wrench: Switch to ruff instead of black + isort + flake8 by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/64
* :sparkles: Add Evidential Classification, switch to ruff, update packages by xuanlongORZ in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/56 and o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/65
* :sparkles: add beta nll, a modified GaussianNLL by xuanlongORZ in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/66
* ✨ Mixup variants + Cross validation + Temperature scaling in routines by qbouniot in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/63
* :shirt: Extend ruff rules by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/67
* :fire: Remove poetry & add flit by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/69
* :shirt: Improve the code and the documentation by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/71
* ✨ Sparsification metric and plot methods for Calibration Error by alafage in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/68
* :bug: Fix Monte-Carlo Dropout by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/73
* :sparkles: Remove Poetry, add Mixup, rework metrics, & improve code quality by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/70
* :wrench: Update sphinx dependency by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/74

New Contributors
* badrmarani made their first contribution in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/46
* dependabot made their first contribution in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/51
* xuanlongORZ made their first contribution in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/56
* qbouniot made their first contribution in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/63

Huge thanks to them!

**Full Changelog**: https://github.com/ENSTA-U2IS/torch-uncertainty/compare/v0.1.4...v0.1.5

MIMO, Scalers, & more datasets

In this PR, we added a new preprocessing function for MIMO-like networks. We added different post-hoc scaling methods to improve the calibration. We added MNIST-C and TinyImageNet-C to the corrupted datasets. Furthermore, we refined the GitHub workflow to improve the development chain. Finally, we also added one tutorial for temperature scaling.

What's Changed
* Add scalers, TinyImageNet-C & various improvements by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/37
* Add MNIST-C, a tutorial, & small fixes by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/38
* :book: Improve the scaling tutorial & misc by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/39
* :sparkles: Add MIMO by alafage in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/41
* Add MIMO, NotMNIST, improve coverage, and Misc by o-laurent in https://github.com/ENSTA-U2IS/torch-uncertainty/pull/42


**Full Changelog**: https://github.com/ENSTA-U2IS/torch-uncertainty/compare/v0.1.3...v0.1.4