v0.1.16

What's Changed

• Pin MyST-Parser Version by @JonathanWenger in #637
• Clean up ProbNum's Development Dependencies by @marvinpfoertner in #636
• Update the isort config and bump black version by @marvinpfoertner in #638
• Row selection policy for the PLS by @JonathanWenger in #588
• codecov/codecov-action instead of bash uploader by @marvinpfoertner in #640
• Update jax[minimum-jaxlib] requirement from <0.2.29 to <0.3.1 by @dependabot in #642
• Add Function class and refactor RandomProcess.{input,output}_{dim => shape} by @marvinpfoertner in #639
• Refactor UKF Components to use linearize() by @pnkraemer in #635
• Split nonlinear filtering notebooks by @pnkraemer in #644
• [Bugfix] Missing rng argument in sample function of random variable rv1 {+,-,*,/,//,@,%} rv2 by @marvinpfoertner in #648
• Simplified TaylorMode implementation by @pnkraemer in #647
• [Bugfix] Scalar arguments in Kernel.__call__ by @marvinpfoertner in #649
• Remove input_shape broadcasting capabilities from Kernel by @marvinpfoertner in #650
• Update jax[minimum-jaxlib] requirement from <0.3.1 to <0.3.2 by @dependabot in #651

Full Changelog: v0.1.15...v0.1.16

v0.1.15

What's Changed

• __(r)matmul__ of Matrix linear operator outputs np.matrix by @marvinpfoertner in #602
• Default ODE filter arguments by @pnkraemer in #603
• Correct links in footer of documentation by @JonathanWenger in #604
• Fix pylint errors for stepsize by @pnkraemer in #608
• Fix pylint errors for _perturbsolve_ivp.py by @pnkraemer in #606
• Fix pylint errors for _probsolve_ivp.py by @pnkraemer in #607
• Update sphinx requirement from <4.3.3,>=3.0 to >=3.0,<4.4.1 by @dependabot in #612
• Fix pylint errors for _odesolution.py and _odesolver.py by @pnkraemer in #609
• Filter warning for importlib by @pnkraemer in #613
• pylint cleanup quad by @mmahsereci in #610
• Additions to the global pylint configuration by @marvinpfoertner in #616
• Another additions to the global pylint configuration by @mmahsereci in #617
• Fix a bunch of pylint errors in filtsmooth by @pnkraemer in #611
• Update jax requirement from <0.2.27 to <0.2.28 by @dependabot in #619
• Cholesky decompositions of LinearOperators by @marvinpfoertner in #618
• Update jaxlib requirement from <0.1.76 to <0.1.77 by @dependabot in #621
• Process noise in discrete transitions by @pnkraemer in #615
• Small documentation fixes and updates to roles by @JonathanWenger in #623
• Remove some abstractmethod-decorators from RandomProcess by @pnkraemer in #629
• Pin the pytest version and replace jaxlib dependency by @pnkraemer in #626
• Fix typos in code chunks in docs by @pitmonticone in #631
• Drop Python 3.7 Support by @marvinpfoertner in #632
• Scipy intersphinx mapping update by @pnkraemer in #630
• Split filtsmooth notebooks by @pnkraemer in #628
• Improve ODE filter initialisation and add more strategies by @pnkraemer in #614
• Enabled pylint: hidden-method in Markov process submodule by @pnkraemer in #634
• Reorthogonalization for a stable implementation of the probabilistic linear solver by @JonathanWenger in #580

Full Changelog: v0.1.14...v0.1.15

v0.1.14

Bug Fixes

• MaxIterations stopping criterion for the linear solver is now stateless (#579)

Refactoring

• ProbabilisticLinearSolver base class as proposal for general PN method implementation (#565)
• Cleanup in probnum.typing (#599)
• Replace NumPy Array Scalars by scalar arrays (#600)

Other

• ProbNum is now citable, so we added a CITATION.cff file (#587, #596)
• We now have CODEOWNERS (#589, #592)
• Changes to the documentation (#584, #586, #595)

v0.1.13

New Features

• Base class StoppingCriterion for probabilistic numerical methods which supports boolean arithmetic (#561, #570)
• Bayesian quadrature can be run in an iterative fashion (#420)

Improvements

• Efficient arithmetic for linear operators (#505)

Bug Fixes

• Optimized memory consumption of Constant random variables (#551, #553)
• Matrix-variate normal matrix multiplication now works for all possible cases (#558, #572)

Refactoring

• Belief updates for probabilistic linear solvers (#546)
• Package structure improved (#560, #568)
• Removed support for complex dtypes of LinearOperators (#575)

Testing

• Optional arguments can now be passed to pytest via tox (#563)

Other

• Dropped support for Python 3.6 (#548)

v0.1.12

New Features

• Kernel broadcasting (#539): Allows efficient evaluation of kernels without forming a full kernel matrix.

Improvements

• Updated quickstart guides for linear solvers (#511) and linear operators (#534)

Refactoring

• Probabilistic linear solvers: Information operators (#494) and stopping criteria (#499)

Documentation

• Corrected documentation of some ODE example problems (#544)

v0.1.11

Improvements

• New Lorenz systems as ODE example problems (#515)
• Improved package build process (78cd1ed, 993a275)

Refactoring

• Improvements to the SDE API (#501)
• Removal of global pylint disables (#522)
• Restructured .gitignore (#525)

Documentation

• Fixes to tutorials and Sphinx build (#516)

v0.1.10

New Features

• Event handling for ODE solvers (#498)

Improvements

• Interface function for perturbation-based ODE solvers (#481)
• Use of LinearOperators in statespace.IBM (#479)

Refactoring

• Reorganization of diffeq subpackage (#477)
• Restructuring of filtsmooth subpackage (#478)
• Refactoring of statespace subpackage (#492)
• Components (state, belief, policy) of probabilistic linear solvers as separate classes (#460)

Documentation

• Automatically generated documentation for config options (#480)
• Various updates to tutorials (#495, #506, #510)

v0.1.9

New Features

• Perturbed step ODE solvers (#382)
• Global configuration module to set package-wide defaults (#465)

Improvements

• Random sparse matrices are now generated in a memory efficient way (#455, #457)
• On-the-fly construction of filtering posteriors (#403)

Refactoring

• Renamed type module to probnum.typing (#454)
• Removal of random_state attribute from random variables (#387)
• ODE examples moved into problem zoo (#475)

Documentation

• New website and logo (#433)

v0.1.8

New Features

• Abstract base class ProbabilisticNumericalMethod #410

Refactoring

• Simplification of RandomVariable instantiation code #425

Bug Fixes

• Fixed tests for extended Kalman filters #409
• LinearOperator.inv() no longer throws warnings for sparse matrices #434

Documentation

• updated tutorials #419 #423
• restructured contribution guide #414

v0.1.7

New Features

• Step size perturbation for differential equation solvers (#376)
• Wrappers for SciPy Runge-Kutta methods with manual step size selection (#380)

Benchmarks

• Filtering and smoothing (#373)
• Linear operators (#385, #390)

Refactoring

• Linear operators now use a compositional pattern (#381)

Bug Fixes

• Stabilization of linear solves in discrete transitions (#386)
• Performance of Kronecker matrix matrix multiplication (#391,#392)