Change Log

Change Log#

v0.2.9 (24/06/2024)#

CoFI Tools#

  • New tool: sparse matrix solvers

CoFI Core#

  • Allow sparse matrix input on BaseProblem.set_jacobian

v0.2.8 (21/03/2024)#

CoFI Tools#

  • Adapt CoFI to new BayesBay API

v0.2.7 (26/02/2024)#

CoFI Tools#

  • New tool: Neighpy

v0.2.6 (15/02/2024)#

CoFI Utils#

  • cofi.utils.GaussianPrior matrix generation aligned with textbook

    • sigma, instead of the square root of it is taken in now

CoFI Tools#

  • BayesBay adjusted according to API change

v0.2.5 (16/01/2024)#

CoFI Tools#

  • New tool: BayesBay

v0.2.4 (16/01/2024)#

CoFI Tools#

  • New tool: CoFI’s implementation of the Border Collie Optimization algorithm

v0.2.3 (31/07/2023)#

CoFI Utils#

  • A new class, cofi.utils.EnsembleOfInversions, has been introduced to replace cofi.utils.run_multiple_inversions

v0.2.2 (17/07/2023)#

CoFI Utils#

  • cofi.utils.run_multiple_inversions, sequential and parallel options

  • Make cofi.utils._reg_base.CompositeRegularization pickleable

v0.2.1 (07/07/2023)#

CoFI Tools#

  • Better cofi.simple_newton solver (more numerically stable; addition of stopping criteria)

v0.2.0 (05/06/2023)#

CoFI Core#

  • Rewrite BaseProblem.set_regularization

CoFI Utils#

  • Rewrite and implement regularization utils:

    • cofi.utils.BaseRegularization

    • cofi.utils.LpNormRegularization

    • cofi.utils.QuadraticReg

    • cofi.utils.ModelCovariance

    • cofi.utils.GaussianPrior

v0.1.3.dev2 (05/04/2023)#

CoFI Core#

  • Bug fix: avoid evaluating log_likelihood if prior is -np.inf

  • Enable properties set at BaseProblem constructor, e.g. cofi.BaseProblem(forward=my_fwd, model_shape=my_shape)

CoFI Tools#

  • Bug fix in numpy.linalg.lstsq

v0.1.3.dev1 (15/03/2023)#

CoFI Core#

  • Use try block for functools.update_wrapper

v0.1.3.dev0 (15/03/2023)#

CoFI Solvers -> CoFI Tools#

  • #110 BaseSolver -> BaseInferenceTool

CoFI Core#

  • _base_problem._FunctionWrapper improvements

v0.1.2.dev25 (06/02/2023)#

CoFI Solvers#

  • Bug fixes in BaseSolver

CoFI Utils#

  • #108 Utility regularization for flattening and smoothing in 1D cases

v0.1.2.dev24 (14/12/2022)#

Infrastructure#

  • #84 Use versioningit in build process

v0.1.2.dev23 (14/12/2022)#

CoFI Core#

  • #91 Raise warning when people set solver params that are not in optional list

  • #97 Make walkers_start_pos a property of InversionOptions instead of BaseProblem

  • #98 Typo, wording fixes; shorten error messages

v0.1.2.dev22 (23/11/2022)#

CoFI Core#

  • #90 Replaced BaseProblem.suggest_solvers with BaseProblem.suggest_tools

  • #89 Avoid importing third party modules on import cofi

CoFI Solvers#

  • #92 List pytorch optim algorithms dynamically

v0.1.2.dev21 (27/10/2022)#

CoFI Solvers#

  • torch.optim

    • return number of function evaluations

    • accept callback function

    • return better losses list

    • add this to docs tree

v0.1.2.dev20 (25/10/2022)#

CoFI Solvers#

  • Internal bug fix in PyTorch optimizers: adding “success” key in returned dictionary

v0.1.2.dev19 (25/10/2022)#

CoFI Solvers#

  • In solvers table: pytorch -> torch.optim

v0.1.2.dev18 (25/10/2022)#

CoFI Solvers#

  • Adding PyTorch.optim algorithms

v0.1.2.dev17 (19/10/2022)#

CoFI Solvers#

  • Simple newton

    • Fix dimension issue

    • return number of function evaluations

v0.1.2.dev16 (18/10/2022)#

CoFI Core#

  • #63 Minor restructure of BaseSolver._assign_options()

  • Wording change in BaseProblem.summary()

v0.1.2.dev15 (14/10/2022)#

CoFI Core#

  • Further explanation in BaseProblem.summary()

v0.1.2.dev14 (14/10/2022)#

CoFI Core#

  • Made CoFI pure Python, requires >=3.7

v0.1.2.dev13 (13/10/2022)#

CoFI Core#

  • Fix BaseProblem.hessian_times_vector and BaseProblem.jacobian_times_vector that are generated from provided hessian / jacobian functions, by squeezing the results to ensure 1D dimensions

CoFI Solvers#

  • Fix InversionResult keys to include underscores (so that attributes can be accessed easily)

v0.1.2.dev12 (10/10/2022)#

CoFI Solvers#

  • cofi.simple_newton

    • hide options of line search (until line search is implemented)

    • prevent initial_model from being modified inplace

v0.1.2.dev11 (10/10/2022)#

CoFI Core#

  • Minor fix (removing debug prints)

  • #72 set_data_misfit error message fix

CoFI Utils#

  • Add util to cofi namespace by importing it

CoFI Solvers#

  • matrix-based solvers -> matrix solvers

v0.1.2.dev10 (03/10/2022)#

CoFI Core#

  • #56 Modify BaseProblem.data_misfit to include data covariance matrix

  • #70 Words renaming optimise -> optimize, etc.

CoFI Utils#

  • #54 Utility functions using findiff to generate the difference matrices

CoFI Solvers#

  • #68 Optimise special cases in linear system solver

v0.1.2.dev9 (13/09/2022)#

CoFI Core#

  • Fixed potential problem in auto generated “times vector” functions when input matrix might be of type numpy.matrix

  • Enabled possibility for parallelism with emcee, by making user defined functions pickleable

  • #53 Add set_regularisation(reg, reg_matrix, lamda)

  • #57 Create our own exception class

  • #59 Optimize import cofi by not importing cofi.solvers

  • #61 Remove lambda function from BaseProblem to avoid error in multiprocessing

CoFI Solvers#

  • #55 Linear solvers with Tikhonov regularisations

v0.1.2.dev8 (13/07/2022)#

  • Bug fix in _FunctionWrapper, for functions with extra arguments like BaseProblem.hessian_times_vector(m, v) and BaseProblem.jacobian_times_vector(m, v)

v0.1.2.dev7 (15/06/2022)#

  • Bug fix in BaseSolver.model_covariance

v0.1.2.dev6 (15/06/2022)#

  • Bug fix in BaseSolver._assign_options

  • BaseProblem.model_covariance_inv and BaseProblem.model_covariance

v0.1.2.dev5 (09/06/2022)#

  • BaseProblem.set_data_covariance and more general linear system solver

v0.1.2.dev4 (07/06/2022)#

  • Bugs fix in EmceeSolver and result summary

v0.1.2.dev3 (06/06/2022)#

  • Bug fixed in BaseProblem.set_regularisation

v0.1.2.dev2 (03/06/2022)#

  • Added emcee as new solver, with the following new APIs

    • BaseProblem.set_log_prior

    • BaseProblem.set_log_likelihood,

    • BaseProblem.set_log_posterior,

    • BaseProblem.set_log_posterior_with_blobs,

    • BaseProblem.set_blobs_dtype

  • Process sampler output by converting to arviz.InferenceData, with the new API:

    • class SamplingResult

    • SamplingResult.to_arviz()

  • Removed BaseProblem.set_dataset(x,y), added BaseProblem.set_data(y)

  • Added args and kwargs for all setting functions in BaseProblem

    • _FunctionWrapper

  • Relaxed python version >=3.8 to >=3.6

  • Docs improvement, updated with emcee

v0.1.2.dev1 (16/05/2022)#

  • bug fixes

v0.1.2.dev0 (13/05/2022)#

  • InversionRunner has been changed into Inversion

  • Added a references list for each backend tool

  • How much information in BaseProblem is used for the inversion run now displayed through Inversion.summary()

  • numpy and scipy versions relaxed

  • Set objective function to be equal to data misfit if regularisation is not set

  • Better error message when building failed

Contents