v1.0.0

What's Changed

• Pin orbax-checkpoint version for Windows by @stoprightthere in #174
• Stable release by @stoprightthere in #175

Full Changelog: v0.4.2...v1.0.0

v0.4.2

What's Changed

• Maintenance by @stoprightthere in #173
  • Fix compatibility with modern plum-dispatch/numpy.
  • Replace legacy plum typing aliases with modern Python type syntax.
  • Drop Python 3.9 support.
  • Discontinue testing for gpflow due to incompatibility with the latest setuptools.

Full Changelog: v0.4.1...v0.4.2

v0.4.1

What's Changed

• Added a link to a pull request example in README by @vabor112 in #171
• HammingGraph space by @colmont in #170

New Contributors

• @colmont made their first contribution in #170

Full Changelog: v0.4...v0.4.1

v0.4.0

What's Changed

• Add support for Python 3.12
• Support for Python 3.8 discontinued
• Guide on implementing a custom space/kernel by @vabor112 in #159
• GPJax>=0.12.2 support by @stoprightthere in #162 and @vabor112 and @stoprightthere in #166
• Comprehensible exceptions by @stoprightthere in #160
• CI improvements by @stoprightthere in #161 and #163

Full Changelog: v0.3.1...v0.4.0

v0.3.1

What's Changed

• Example Applications by @vabor112 in #157

Full Changelog: v0.3.0...v0.3.1

v0.3.0

What's Changed

• Implementation of kernels on the edge space of graphs or simplicial complexes by @cookbook-ms in #139

New Contributors

• @cookbook-ms made their first contribution in #139

Full Changelog: v0.2.3...v0.3

v0.2.3

Constraint version of plum-dispatch because of wesselb/lab#23

Full Changelog: v0.2.2...v0.2.3

v0.2.2

What's Changed

• Replace opt_einsum's contract with lab's einsum for better backend-independence by @vabor112 in #145
• Hypersphere space small improvements by @vabor112 in #142
• The Hypercube space for binary vectors and labeled unweighted graphs by @vabor112 in #141
• Fix algorithm selecting signatures and add precomputed characters for SO(9), SU(7), SU(8), SU(9) by @imbirik in #151
• Revise tests and numerous fixes by @vabor112 in #149

v0.2.1

What's Changed

• Add "If you have a question" section to README.md by @vabor112 in #131
• Github cosmetics by @stoprightthere in #133
• Replace all references to "gpflow" organization with "geometric-kernels" organization by @vabor112 in #134
• Use fit_gpytorch_model or fit.fit_gpytorch_mll depening on the botorсh version by @vabor112 in #137
• Add a missing type cast and fix a typo in kernels/karhunen_loeve.py by @vabor112 in #136
• Minor documentation improvements by @vabor112 in #135
• Add citation to the preprint of the GeometricKernels paper by @vabor112 in #138
• Add citation file by @aterenin in #140
• Fix dependencies (Version 0.2.1) by @stoprightthere in #143

Full Changelog: v0.2...v0.2.1

v0.2

New geometric kernel that just works, kernels.MaternGeometricKernel. Relies on (hopefully) sensible defaults we defined. Mostly by @stoprightthere.

New spaces, based on Azangulov et al. (2022, 2023), mostly by @imbirik and @stoprightthere:

• hyperbolic spaces $\mathbb{H}_n$ in spaces.Hyperbolic,
• manifolds of symmetric positive definite matrices $\mathrm{SPD}(n)$ endowed with the affine-invariant Riemannian metric in spaces.SymmetricPositiveDefiniteMatrices,
• special orthogonal groups $\mathrm{SO}(n)$ in spaces.SpecialOrthogonal.
• special unitary groups $\mathrm{SU}(n)$ in spaces.SpecialUnitary.

New package geometric_kernels.feature_maps for (approximate) finite-dimensional feature maps. Mostly by @stoprightthere.

New small package geometric_kernels.sampling for efficient sampling from geometric Gaussian process priors. Based on the (approximate) finite-dimensional feature maps. Mostly by @stoprightthere.

Examples/Tutorials improvements, mostly by @vabor112:

• new Jupyter notebooks Graph.ipynb, Hyperbolic.ipynb, Hypersphere.ipynb, Mesh.ipynb, SPD.ipynb, SpecialOrthogonal.ipynb, SpecialUnitary.ipynb, Torus.ipynb featuring tutorials on all the spaces in the library,
• new Jupyter notebooks backends/JAX_Graph.ipynb, backends/PyTorch_Graph.ipynb, backends/TensorFlow_Graph.ipynb showcasing how to use all the backends supported by the library,
• new Jupyter notebooks frontends/GPflow.ipynb, frontends/GPJax.ipynb, frontends/GPyTorch.ipynb showcasing how to use all the frontends supported by the library,
• other notebooks updated and grouped together in other/ folder.

Documentation improvements, mostly by @vabor112:

• all docstrings throughout the library revised,
• added new documentation pages describing the basic theoretical concepts, in docs/theory,
• notebooks are now rendered as part of the documentation, you can refer to them from the docstrings and other documentation pages,
• introduced a more or less unified style for docstrings.

Other:

• refactoring and bug fixes,
• added type hints throughout the library and enabled mypy,
• updated frontends (with limited suppot for GPJax due to conflicting dependencies),
• improved spaces.ProductDiscreteSpectrumSpace and kernels.ProductGeometricKernel,
• filtered out or fixed some annoying external warnings,
• added a new banner for README.md and for our landing page, courtesy of @aterenin,
• example notebooks are now run as tests,
• we now support Python 3.8, 3.9, 3.10, 3.11 and have test workflows for all the supported Python versions,
• we now provide a PyPI package,
• LAB is now a lightweight dependency, thanks to @wesselb,
• kernels are now normalized to have unit outputscale by default.