0.2.1

What's Changed

• Add py.typed to the package to enable the type annotations by @kourbou in #211
• Functional Laplace Updated by @Ludvins in #192
• Fix sigma_noise gradient propagation in BaseLaplace.log_likelihood by @kourbou in #212
• Add pytest-mock for FunctionalLaplace Tests by @Ludvins in #215
• Fix FullSubnetLaplace.sample() by @elcorto in #216
• Document pros and cons of subnet selection methods by @elcorto in #218
• Update README by @wiseodd in #219
• Bump version number for new release by @runame in #220
• Update README re. ASDL by @wiseodd in #223
• Make importing asdfghjkl conditional to the installed optional dependency by @wiseodd in #225

New Contributors

• @kourbou made their first contribution in #211
• @Ludvins made their first contribution in #192

Full Changelog: 0.2...0.2.1

0.2

TL;DR

• This release is all about laying down the groundwork for Laplace on foundation models. The main target for this version is LLMs. Support for other foundation models will added in versions >= 0.3.
• Additionally, this release adds support for Laplace for sequential decision making e.g. Bayesian optimization.
• A new backend, curvlinops, is added.
• Added native serialization support.
• Updated the code base to conform modern Python development standard (typehinting, enums, ruff, etc.).

Thanks to all contributors!

What's Changed

LLM-related

• Bringing laplace-torch to foundation-model era by @wiseodd in #144
• Add support for cross entropy loss inputs with multiple leading dimensions by @runame in #132
• Add an option to reduce LLM features in LLLaplace by @wiseodd in #172
• Update huggingface_example.md by @wiseodd in #204
• Refactor reward-modeling likelihood by @runame in #200
• Doc & example for reward modeling by @wiseodd in #167
• Make the dict keys for models with dict-like inputs general by @wiseodd in #168
• Feature caching mechanism in LLLA by @wiseodd in #170
• Added more test coverage to the subset of params functionality by @wiseodd in #185

Bayesian optimization

• Enable BayesOpt by @wiseodd in #120

Curvlinops

• Add Curvlinops backend & add default functorch implementations of many curvature quantities by @wiseodd in #146
• Point to curvlinops master branch in setup.cfg by @wiseodd in #151

Serialization

• Add native serialization support by @wiseodd in #148
• Enable torch.save() + torch.load() for different map_locations by @elcorto in #159

Devs

• Typehinting by @wiseodd in #180
• Add "Contributing" to the README by @wiseodd in #198
• Add ruff check, ruff format --check, and pytest Github actions by @wiseodd in #176
• Remove depreciated loss_average argument of KFACLinearOperator and add makefile for ruff by @runame in #197

Etc

• Fixes and features for SubnetLaplace by @edaxberger in #87
• Improve predictive samples by @aleximmer in #95
• Add experiments repo reference by @runame in #99
• Use ASDL as the default for classification by @edaxberger in #96
• Update Laplace bridge predictive by @runame in #101
• Make last-layer Jacobians agnostic to NN output shape by @runame in #112
• Fix device and dtype of block_diag used for Kron.to_matrix() by @runame in #117
• Add KronDecomposed.diag() feature by @aleximmer in #121
• Replacing torch.einsum() with opt_einsum by @Heatdh in #125
• Computing the classification BMA, i.e. average of softmaxes, online. by @wiseodd in #133
• Add asdl_fisher_kwargs argument by @runame in #134
• Running metrics by @wiseodd in #135
• Add support for diagonal Kronecker factors in Kron matrix class by @runame in #136
• Add prior_structure argument to optimize_prior_precision by @runame in #123
• Move back to ASDL's main repo as dependency by @aleximmer in #183
• fix typo in Jacobian dimension by @ruili-pml in #190
• Prevent computing posterior precision in KronLaplace when it's not fitted by @wiseodd in #173
• Use backend-native Jacobians if available by @wiseodd in #187
• Update docs by @wiseodd in #184
• Remove try-except from gridsearch by @wiseodd in #199
• Add fast computation of functional_variance for DiagLLLaplace and KronLLLaplace by @wiseodd in #145
• Caveats by @wiseodd in #202
• Add some checks in optimize_prior_precision by @wiseodd in #205
• Add backend 'flowchart' by @wiseodd in #207

New Contributors

• @Heatdh made their first contribution in #125
• @elcorto made their first contribution in #159
• @ruili-pml made their first contribution in #190

Full Changelog: 0.1a2...0.2

laplace 0.1a2

What's Changed

• Remove abc by @metodj in #41
• Calibration example by @wiseodd in #54
• Enable cross validation support by @wiseodd in #38
• Minor improvements by @runame in #60
• Update symeig method based on new torch interface by @aleximmer in #61
• Online marglik as a method by @aleximmer in #64
• Support for refitting LA with option to override or update by @aleximmer in #62
• Low rank LA by @aleximmer in #65
• Fix device of eye in symeig by @aleximmer in #66
• Keep initialization of H for all-weights and last-layer separate by @runame in #72
• Fix device bug in eig_lowrank by @runame in #74
• Subnetwork Laplace by @edaxberger in #58

laplace 0.1a1

Initial release of the laplace package.