Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[0.14.0] - 2025-09-10

Added

• transformations subpackage

• New NumericalTarget interface, including advanced machinery for defining and manipulating target transformations based on the new Transformation class hierarchy

• match_bell and match_triangular convenience constructors to NumericalTarget for reproducing the legacy MATCH modes

• normalized_ramp convenience constructor to NumericalTarget for reproducing the legacy behavior when imposing bounds on MIN/MAX targets

• normalized_sigmoid, match_absolute, match_quadratic and match_power convenience constructors to NumericalTarget enabling additional matching/normalizing behaviors

• Full support for accessing posterior information of NumericalTarget, i.e. now including settings considered MATCH mode in the legacy interface, as well as targets used in DesirabilityObjective

• as_pre_transformation flag to DesirabilityObjective for controlling whether the desirability transformation is applied before or after model fitting

• supports_partial_measurements property to Objective

• is_normalized property to NumericalTarget

• negate, normalize, abs, clamp, log, exp and power methods to NumericalTarget for easy creation of transformed targets from existing ones

• Addition and multiplication dunder methods (for scalar values) to NumericalTarget

• get_image method to NumericalTarget for computing the images of transformed target value ranges

• Support for non-normalized targets in DesirabilityObjective

• Objective.to_botorch method for converting objectives to BoTorch

• qEHVI and qLogEHVI acquisition functions (in addition to their noisy variants)

• Tests for migrating to new NumericalTarget interface

• API diagram in user guide

• Metadata and MeasurableMetadata classes providing optional information for BayBE objects

• Objective now has a metadata attribute as well as a description property

• Target and Parameter now have a metadata attribute as well as description and unit properties

• FPSRecommender now optionally uses the fpsample library (if installed) with fallback to internal implementation. The use of fpsample can be deactivated by setting the environment variable BAYBE_USE_FPSAMPLE

• farthest_point_sampling now also supports a collection of integers for initialization, using them for pre-selecting points

• Benchmarks can now run in different modes, store detailed machine information, and local usage got new file naming and execution options

Changed

• The behavior of NumericalTarget is no longer defined via a mode (i.e. MIN, MAX, MATCH in legacy interface) but controlled using a minimization flag and corresponding target transformations. This allows for more flexible target definitions, makes invalid target configurations unrepresentable, and is in line with the definition of mathematical optimization problems. Also, it avoids the need to explicitly specify an irrelevant optimization direction in the context of active learning.

• By default, DesirabilityObjective now fits separate models for each target, rather than modeling only the scalarized desirability value (see new DesirabilityObjective.as_pre_transformation flag). As a result, posterior evaluations now return information for each target individually, instead of just for the desirability value.

• Objective transformations (both tensor and dataframe based) now always use the Torch computation route, avoiding the need for duplicated transformation logic

• Specifying bounds for Interval is now optional

• unstructure_base and get_base_structure_hook (de-)serialization utilities have been replaced with unstructure_with_type and make_base_structure_hook

Fixed

• It is no longer possible to use identical names between parameters and targets

• Random seed context is correctly set within benchmarks

• Measurement input validation now respects typical tolerances associated with floating point representation inaccuracy

• Exotic serialization issues with constraints and conditions arising from missing converters for floats

• MetaRecommender’s no longer expose their private attributes via the constructor

Removed

• Telemetry

• Option to specify reference values for add_fake_measurements

• convert_bounds utility (since now equivalent to Interval.create)

• geom_mean utility

Deprecations

• Creating NumericalTargets using a mode argument

• TargetMode and TargetTransformation enums

• linear_transform, triangular_transform and bell_transform functions

Expired Deprecations (from 0.10.*)

• SequentialGreedyRecommender class

• SubspaceContinuous.samples_random method

• SubspaceContinuous.samples_full_factorial method

• data argument to transform method of search space classes

• Automatically setting allow_extra=True in transform methods of search space classes when left unspecified

[0.13.2] - 2025-07-09

Changed

• Lockfiles are now generated using uv lock and consumed using uv sync

Fixed

• The Python version specifier now also allows patch versions of Python 3.13

[0.13.1] - 2025-06-06

• Support for Python 3.13

• random_tie_break flag to farthest_point_sampling to toggle between random or deterministic sampling for equidistant cases

• random_tie_break and initialization attributes to FPSRecommender to control sampling in farthest_point_sampling

• Flag for toggling parallel computation in simulate_scenarios

• Additional transfer learning and synthetic benchmarks

• Utility normalize_input_dtypes for ensuring all input dataframe columns have the expected dtypes

• CompositeSurrogate now has a _posterior_comp method similar to Surrogate

• SHAPInsight.explain_target method for computing explanations for only a specific target

Changed

• CategoricalParameter and TaskParameter now also allow Boolean entries as values and active_values

• SubspaceDiscrete.from_dataframe now handles purely Boolean columns differently, inferring a CategoricalParameter with INT encoding for them

• SHAPInsight.explain now returns a tuple of explanations that contains one explanation for each surrogate model used by the (possibly multi-output) objective

• SHAPInsight.plot now has the optional target_index argument, enabling users to select for which target they want to plot the shap assessment (default is the first target)

• add_measurements, update_measurements, fuzzy_row_match and some .recommend calls now operate on dtype-normalized copies of the input if it contained unexpected dtypes for a parameter or target

• scikit-learn and scipy are now lazy-loaded

• Validity of optional model_params attribute of RandomForestSurrogate, NGBoostSurrogate and BayesianLinearSurrogate is now checked via a hardcoded TypedDict instead of dynamically retrieved specifications, required for lazy-loading related packages

• CustomONNXSurrogate.onnx_str is no longer validated before being used

Fixed

• Using PosteriorStandardDeviation with MIN targets no longer results in minimization of the acquisition function

• Added missing garbage collection call to pareto.py, potentially solving serialization issues in certain cases

• catch_constant_targets decorator is now properly typed

• Incorrect normalization of explanation shapes for SHAPInsight

Removed

• SHAPInsight.uses_shap_explainer

[0.13.0] - 2025-04-16

Added

• extras group for installing all dependencies required for optional features

• Support for NumPy 2.0+

• ParetoObjective class for Pareto optimization of multiple targets and corresponding qNoisyExpectedHypervolumeImprovement / qLogNoisyExpectedHypervolumeImprovement / qLogNParEGO acquisition functions

• Composite surrogates now drop rows containing NaNs (separately for each target), effectively enabling partial measurements

• SubstanceParameter, CustomDiscreteParameter and CategoricalParameter now also support restricting the search space via active_values, while values continue to identify allowed measurement inputs

• Campaign.posterior_stats and Surrogate.posterior_stats as convenience methods for providing statistical measures about the target predictions of a given set of candidates

• acquisition_values and joint_acquisition_value convenience methods to Campaign and BayesianRecommender for computing acquisition values

• Campaign.get_acquisition_function and BayesianRecommender.get_acquisition_function convenience methods for retrieving the underlying acquisition function

• AcquisitionFunction.evaluate convenience method for computing acquisition values from candidates in experimental representation

• qPSTD acquisition function

• BCUT2D encoding for SubstanceParameter

• SHAPInsight now supports the waterfall plot type

• Cardinality constraints sections to the user guide

• ContinuousCardinalityConstraint is now compatible with BotorchRecommender

• Attribute max_n_subspaces to BotorchRecommender, allowing to control optimization behavior in the presence of cardinality constraints

• Surrogate.replicate method for making single-target surrogate models multi-target compatible

• CompositeSurrogate class for composing multi-target surrogates from single-target surrogates

• is_multi_output attribute to Objective

• supports_multi_output attribute/property to Surrogate/AcquisitionFunction

• n_outputs property to Objective

• Attribute relative_threshold and method get_absolute_thresholds to ContinuousCardinalityConstraint for handling inactivity ranges

• Utilities inactive_parameter_combinations andn_inactive_parameter_combinations to both ContinuousCardinalityConstraintand SubspaceContinuous for iterating over cardinality-constrained parameter sets

• Utilities activate_parameter and is_cardinality_fulfilled for enforcing and validating cardinality constraints

• Utility is_inactive for determining if parameters are inactive

• A MinimumCardinalityViolatedWarning is triggered when minimum cardinality constraints are violated

• Stored benchmarking results now include the Python environment and version

Changed

• Targets are now allowed to contain NaN, deferring potential failure to attempted recommendation instead of data ingestion

• For label-like parameters, SubspaceDiscrete now only includes parameter values that are in active_values

• The default value for sequential_continuous in BotorchRecommender has been changed to True

• SHAPInsight now allows explanation input that has additional columns compared to the background data (will be ignored)

• Acquisition function indicator is_mc has been removed in favor of new indicators supports_batching and supports_pending_experiments

• fuzzy_row_match now uses vectorized operations, resulting in a speedup of matching measurements to the search space between 4x and 40x

• Model scaling now uses the parameter bounds instead of the search space bounds

• benchmarks module now accepts a list of domains to be executed

• Construction of BoTorch acquisition functions has been redesigned from ground up

• ngboost is now an optional dependency

• setuptools-scm is now an optional dependency, used for improved version inference

• create_example_plots, to_string and indent have been relocated within utils

Fixed

• Incorrect optimization direction with PSTD with a single minimization target

• Provide version fallback in case scm fails to infer version during installation

Removed

• fuzzy_row_match will no longer warn about entries not matching to the search space

• funcy dependency

• scikit-learn-extra dependency by integrating relevant code parts into baybe

Expired Deprecations (from 0.9.*)

• baybe.objective namespace

• acquisition_function_cls constructor parameter for BayesianRecommender

• VarUCB and qVarUCB acquisition functions

[0.12.2] - 2025-01-31

Changed

• More robust settings for the GP fitting

• The beta parameter of UCB and qUCB can now also take negative values

[0.12.1] - 2025-01-29

Changed

• Default of allow_recommending_already_recommended is changed back to False to avoid exploitation cycles (#468)

[0.12.0] - 2025-01-28

Breaking Changes

• Lookup callables for simulation are now expected to accept/return dataframes with the corresponding parameter/target column labels

Added

• SHAP explanations via the new SHAPInsight class

• Optional insights dependency group

• Insights user guide

• Example for a traditional mixture

• allow_missing and allow_extra keyword arguments to Objective.transform

• add_noise_to_perturb_degenerate_rows utility

• benchmarks subpackage for defining and running performance tests – Campaign.toggle_discrete_candidates to dynamically in-/exclude discrete candidates

• filter_df utility for filtering dataframe content

• arrays_to_dataframes decorator to create lookups from array-based callables

• DiscreteConstraint.get_valid to conveniently access valid candidates

• Functionality for persisting benchmarking results on S3 from a manual pipeline run

• remain_switched option to TwoPhaseMetaRecommender

• is_stateful class variable to MetaRecommender

• get_non_meta_recommender method to MetaRecommender

Changed

• SubstanceParameter encodings are now computed exclusively with the scikit-fingerprints package, granting access to all fingerprints available therein

• Example for slot-based mixtures has been revised and grouped together with the new traditional mixture example

• Memory caching is now non-verbose

• CustomDiscreteParameter does not allow duplicated rows in data anymore

• De-/activating Polars via BAYBE_DEACTIVATE_POLARS now requires passing values compatible with strtobool

• All arguments to MetaRecommender.select_recommender are now optional

• MetaRecommenders can now be composed of other MetaRecommenders

• For performance reasons, search space manipulation using polars is no longer guaranteed to produce the same row order as the corresponding pandas operations

• allow_repeated_recommendations has been renamed to allow_recommending_already_recommended and is now True by default

Fixed

• Rare bug arising from degenerate SubstanceParameter.comp_df rows that caused wrong number of recommendations being returned

• ContinuousConstraints can now be used in single point precision

• Search spaces are now stateless, preventing unintended side effects that could lead to incorrect candidate sets when reused in different optimization contexts

• qNIPV not working with single MIN targets

• Passing a TargetTransformation without passing bounds when creating a NumericalTarget now raises an error

• Crash when using ContinuousCardinalityConstraint caused by an unintended interplay between constraints and dropped parameters yielding empty parameter sets

• Minimizing a single NumericalTarget with specified bounds/transformation via SingleTargetObjective no longer erroneously maximizes it

• allow_* flags are now context-aware, i.e. setting them in a context where they are irrelevant now raises an error instead of passing silently

Removed

• botorch_function_wrapper utility for creating lookup callables

Deprecations

• Passing a dataframe via the data argument to Objective.transform is no longer possible. The dataframe must now be passed as positional argument.

• The new allow_extra flag is automatically set to True in Objective.transform when left unspecified

• get_transform_parameters has been replaced with get_transform_objects

• Passing a dataframe via the data argument to Target.transform is no longer possible. The data must now be passed as a series as first positional argument.

• SubstanceEncoding value MORGAN_FP. As a replacement, ECFP with 1024 bits and radius of 4 can be used.

• SubstanceEncoding value RDKIT. As a replacement, RDKIT2DDESCRIPTORS can be used.

• The metadata attribute of SubspaceDiscrete no longer exists. Metadata is now exclusively handled by the Campaign class.

• get_current_recommender and get_next_recommender of MetaRecommender have become obsolete and calling them is no longer possible

• Passing allow_* flags to recommenders is no longer supported since the necessary metadata required for the flags is no longer available at that level. The functionality has been taken over by Campaign.

[0.11.4] - 2025-01-27

Changed

• Polars lazy streaming has been deactivated due to instabilities

Fixed

• Improvement-based Monte Carlo acquisition functions now use the correct reference value for single-target minimization

[0.11.3] - 2024-11-06

Fixed

• protobuf dependency issue, version pin was removed

[0.11.2] - 2024-10-11

Added

• n_restarts and n_raw_samples keywords to configure continuous optimization behavior for BotorchRecommender

• User guide for utilities

• mypy rule expecting explicit override markers for method overrides

Changed

• Utility add_fake_results renamed to add_fake_measurements

• Utilities add_fake_measurements and add_parameter_noise now also return the dataframe they modified in-place

Fixed

• Leftover attrs-decorated classes are garbage collected before the subclass tree is traversed, avoiding sporadic serialization problems

[0.11.1] - 2024-10-01

Added

• Continuous linear constraints have been consolidated in the new ContinuousLinearConstraint class

Changed

• get_surrogate now also returns the model for transformed single targets or desirability objectives

Fixed

• Unsafe name-based matching of columns in get_comp_rep_parameter_indices

Deprecations

• ContinuousLinearEqualityConstraint and ContinuousLinearInequalityConstraint replaced by ContinuousLinearConstraint with the corresponding operator keyword

[0.11.0] - 2024-09-09

Breaking Changes

• The public methods of Surrogate models now operate on dataframes in experimental representation instead of tensors in computational representation

• Surrogate.posterior models now returns a Posterior object

• param_bounds_comp of SearchSpace, SubspaceDiscrete and SubspaceContinuous has been replaced with comp_rep_bounds, which returns a dataframe

Added

• py.typed file to enable the use of type checkers on the user side

• IndependentGaussianSurrogate base class for surrogate models providing independent Gaussian posteriors for all candidates (cannot be used for batch prediction)

• comp_rep_columns property for Parameter, SearchSpace, SubspaceDiscrete and SubspaceContinuous classes

• New mechanisms for surrogate input/output scaling configurable per class

• SurrogateProtocol as an interface for user-defined surrogate architectures

• Support for binary targets via BinaryTarget class

• Support for bandit optimization via BetaBernoulliMultiArmedBanditSurrogate class

• Bandit optimization example

• qThompsonSampling acquisition function

• BetaPrior class

• recommend now accepts the pending_experiments argument, informing the algorithm about points that were already selected for evaluation

• Pure recommenders now have the allow_recommending_pending_experiments flag, controlling whether pending experiments are excluded from candidates in purely discrete search spaces

• get_surrogate and posterior methods to Campaign

• tenacity test dependency

• Multi-version documentation

Changed

• The transition from experimental to computational representation no longer happens in the recommender but in the surrogate

• Fallback models created by catch_constant_targets are stored outside the surrogate

• to_tensor now also handles numpy arrays

• MIN mode of NumericalTarget is now implemented via the acquisition function instead of negating the computational representation

• Search spaces now store their parameters in alphabetical order by name

• Improvement-based acquisition functions now consider the maximum posterior mean instead of the maximum noisy measurement as reference value

• Iteration tests now attempt up to 5 repeated executions if they fail due to numerical reasons

Fixed

• CategoricalParameter and TaskParameter no longer incorrectly coerce a single string input to categories/tasks

• farthest_point_sampling no longer depends on the provided point order

• Batch predictions for RandomForestSurrogate

• Surrogates providing only marginal posterior information can no longer be used for batch recommendation

• SearchSpace.from_dataframe now creates a proper empty discrete subspace without index when called with continuous parameters only

• Metadata updates are now only triggered when a discrete subspace is present

• Unintended reordering of discrete search space parts for recommendations obtained with BotorchRecommender

Removed

• register_custom_architecture decorator

• Scalar and DefaultScaler classes

Deprecations

• The role of register_custom_architecture has been taken over by baybe.surrogates.base.SurrogateProtocol

• BayesianRecommender.surrogate_model has been replaced with get_surrogate

[0.10.0] - 2024-08-02

Breaking Changes

• Providing an explicit batch_size is now mandatory when asking for recommendations

• RecommenderProtocol.recommend now accepts an optional Objective

• RecommenderProtocol.recommend now expects training data to be provided as a single dataframe in experimental representation instead of two separate dataframes in computational representation

• Parameter.is_numeric has been replaced with Parameter.is_numerical

• DiscreteParameter.transform_rep_exp2comp has been replaced with DiscreteParameter.transform

• filter_attributes has been replaced with match_attributes

Added

• Surrogate base class now exposes a to_botorch method

• SubspaceDiscrete.to_searchspace and SubspaceContinuous.to_searchspace convenience constructor

• Validators for Campaign attributes

• _optional subpackage for managing optional dependencies

• New acquisition functions for active learning: qNIPV (negative integrated posterior variance) and PSTD (posterior standard deviation)

• Acquisition function: qKG (knowledge gradient)

• Abstract ContinuousNonlinearConstraint class

• Abstract CardinalityConstraint class and DiscreteCardinalityConstraint/ContinuousCardinalityConstraint subclasses

• Uniform sampling mechanism for continuous spaces with cardinality constraints

• register_hooks utility enabling user-defined augmentation of arbitrary callables

• transform methods of SearchSpace, SubspaceDiscrete and SubspaceContinuous now take additional allow_missing and allow_extra keyword arguments

• More details to the transfer learning user guide

• Activated doctests

• SubspaceDiscrete.from_parameter, SubspaceContinuous.from_parameter, SubspaceContinuous.from_product and SearchSpace.from_parameter convenience constructors

• DiscreteParameter.to_subspace, ContinuousParameter.to_subspace and Parameter.to_searchspace convenience constructors

• Utilities for permutation and dependency data augmentation

• Validation and translation tests for kernels

• BasicKernel and CompositeKernel base classes

• Activated pre-commit.ci with auto-update

• User guide for active learning

• Polars expressions for DiscreteSumConstraint, DiscreteProductConstraint, DiscreteExcludeConstraint, DiscreteLinkedParametersConstraint and DiscreteNoLabelDuplicatesConstraint

• Discrete search space Cartesian product can be created lazily via Polars

• Examples demonstrating the register_hooks utility: basic registration mechanism, monitoring the probability of improvement, and automatic campaign stopping

• Documentation building now uses a lockfile to fix the exact environment

Changed

• Passing an Objective to Campaign is now optional

• GaussianProcessSurrogate models are no longer wrapped when cast to BoTorch

• Restrict upper versions of main dependencies, motivated by major numpy release

• Sampling methods in qNIPV and BotorchRecommender are now specified via DiscreteSamplingMethod enum

• Interval class now supports degenerate intervals containing only one element

• add_fake_results now directly processes Target objects instead of a Campaign

• path argument in plotting utility is now optional and defaults to Path(".")

• UnusedObjectWarning by non-predictive recommenders is now ignored during simulations

• The default kernel factory now avoids strong jumps by linearly interpolating between two fixed low and high dimensional prior regimes

• The previous default kernel factory has been renamed to EDBOKernelFactory and now fully reflects the original logic

• The default acquisition function has been changed from qEI to qLogEI for improved numerical stability

Removed

• Support for Python 3.9 removed due to new BoTorch requirements and guidelines from Scientific Python

• Linter typos for spellchecking

Fixed

• sequential flag of SequentialGreedyRecommender is now set to True

• Serialization bug related to class layout of SKLearnClusteringRecommender

• MetaRecommenders no longer trigger warnings about non-empty objectives or measurements when calling a NonPredictiveRecommender

• Bug introduced in 0.9.0 (PR #221, commit 3078f3), where arguments to to_gpytorch are not passed on to the GPyTorch kernels

• Positive-valued kernel attributes are now correctly handled by validators and hypothesis strategies

• As a temporary workaround to compensate for missing IndexKernel priors, fit_gpytorch_mll_torch is used instead of fit_gpytorch_mll when a TaskParameter is present, which acts as regularization via early stopping during model fitting

Deprecations

• SequentialGreedyRecommender class replaced with BotorchRecommender

• SubspaceContinuous.samples_random has been replaced with SubspaceContinuous.sample_uniform

• SubspaceContinuous.samples_full_factorial has been replaced with SubspaceContinuous.sample_from_full_factorial

• Passing a dataframe via the data argument to the transform methods of SearchSpace, SubspaceDiscrete and SubspaceContinuous is no longer possible. The dataframe must now be passed as positional argument.

• The new allow_extra flag is automatically set to True in transform methods of search space classes when left unspecified

Expired Deprecations (from 0.7.*)

• Interval.is_finite property

• Specifying target configs without type information

• Specifying parameters/constraints at the top level of a campaign configs

• Passing numerical_measurements_must_be_within_tolerance to Campaign

• batch_quantity argument

• Passing allow_repeated_recommendations or allow_recommending_already_measured to MetaRecommender (or former Strategy)

• *Strategy classes and baybe.strategies subpackage

• Specifying MetaRecommender (or former Strategy) configs without type information

[0.9.1] - 2024-06-04

Changed

• Discrete searchspace memory estimate is now natively represented in bytes

Fixed

• Non-GP surrogates not working with deepcopy and the simulation package due to slotted base class

• Datatype inconsistencies for various parameters’ values and comp_df and SubSelectionCondition’s selection related to floating point precision

[0.9.0] - 2024-05-21

Added

• Class hierarchy for objectives

• AdditiveKernel, LinearKernel, MaternKernel, PeriodicKernel, PiecewisePolynomialKernel, PolynomialKernel, ProductKernel, RBFKernel, RFFKernel, RQKernel, ScaleKernel classes

• KernelFactory protocol enabling context-dependent construction of kernels

• Preset mechanism for GaussianProcessSurrogate

• hypothesis strategies and roundtrip test for kernels, constraints, objectives, priors and acquisition functions

• New acquisition functions: qSR, qNEI, LogEI, qLogEI, qLogNEI

• GammaPrior, HalfCauchyPrior, NormalPrior, HalfNormalPrior, LogNormalPrior and SmoothedBoxPrior classes

• Possibility to deserialize classes from optional class name abbreviations

• Basic deserialization tests using different class type specifiers

• Serialization user guide

• Environment variables user guide

• Utility for estimating memory requirements of discrete product search space

• mypy for search space and objectives

Changed

• Reorganized acquisition.py into acquisition subpackage

• Reorganized simulation.py into simulation subpackage

• Reorganized gaussian_process.py into gaussian_process subpackage

• Acquisition functions are now their own objects

• acquisition_function_cls constructor parameter renamed to acquisition_function

• User guide now explains the new objective classes

• Telemetry deactivation warning is only shown to developers

• torch, gpytorch and botorch are lazy-loaded for improved startup time

• If an exception is encountered during simulation, incomplete results are returned with a warning instead of passing through the uncaught exception

• Environment variables BAYBE_NUMPY_USE_SINGLE_PRECISION and BAYBE_TORCH_USE_SINGLE_PRECISION to enforce single point precision usage

Removed

• model_params attribute from Surrogate base class, GaussianProcessSurrogate and CustomONNXSurrogate

• Dependency on requests package

Fixed

• n_task_params now evaluates to 1 if task_idx == 0

• Simulation no longer fails in ignore mode when lookup dataframe contains duplicate parameter configurations

• Simulation no longer fails for targets in MATCH mode

• closest_element now works for array-like input of all kinds

• Structuring concrete subclasses no longer requires providing an explicit type field

• _target(s) attributes of Objectives are now (de-)serialized without leading underscore to support user-friendly serialization strings

• Telemetry does not execute any code if it was disabled

• Running simulations no longer alters the states of the global random number generators

Deprecations

• The former baybe.objective.Objective class has been replaced with SingleTargetObjective and DesirabilityObjective

• acquisition_function_cls constructor parameter for BayesianRecommender

• VarUCB and qVarUCB acquisition functions

Expired Deprecations (from 0.6.*)

• BayBE class

• baybe.surrogate module

• baybe.targets.Objective class

• baybe.strategies.Strategy class

[0.8.2] - 2024-03-27

Added

• Simulation user guide

• Example for transfer learning backtesting utility

• pyupgrade pre-commit hook

• Better human readable __str__ representation of objective and targets

• Alternative dataframe deserialization from pd.DataFrame constructors

Changed

• More detailed and sophisticated search space user guide

• Support for Python 3.12

• Upgraded syntax to Python 3.9

• Bumped onnx version to fix vulnerability

• Increased threshold for low-dimensional GP priors

• Replaced fit_gpytorch_mll_torch with fit_gpytorch_mll

• Use tox-uv in pipelines

Fixed

• telemetry dependency is no longer a group (enables Poetry installation)

[0.8.1] - 2024-03-11

Added

• Better human readable __str__ representation of campaign

• README now contains an example on substance encoding results

• Transfer learning user guide

• from_simplex constructor now also takes and applies optional constraints

Changed

• Full lookup backtesting example now tests different substance encodings

• Replaced unmaintained mordred dependency by mordredcommunity

• SearchSpaces now use ndarray instead of Tensor

Fixed

• from_simplex now efficiently validated in Campaign.validate_config

[0.8.0] - 2024-02-29

Changed

• BoTorch dependency bumped to >=0.9.3

Removed

• Workaround for BoTorch hybrid recommender data type

• Support for Python 3.8

[0.7.4] - 2024-02-28

Added

• Subpackages for the available recommender types

• Multi-style plotting capabilities for generated example plots

• JSON file for plotting themes

• Smoke testing in relevant tox environments

• ContinuousParameter base class

• New environment variable BAYBE_CACHE_DIR that can customize the disk cache directory or turn off disk caching entirely

• Options to control the number of nonzero parameters in SubspaceDiscrete.from_simplex

• Temporarily ignore ONNX vulnerabilities

• Better human readable __str__ representation of search spaces

• pretty_print_df function for printing shortened versions of dataframes

• Basic Transfer Learning example

• Repo now has reminders (https://github.com/marketplace/actions/issue-reminder) enabled

• mypy for recommenders

Changed

• Recommenders now share their core logic via their base class

• Remove progress bars in examples

• Strategies are now called MetaRecommender’s and part of the recommenders.meta module

• Recommender’s are now called PureRecommender’s and part of the recommenders.pure module

• strategy keyword of Campaign renamed to recommender

• NaiveHybridRecommender renamed to NaiveHybridSpaceRecommender

Fixed

• Unhandled exception in telemetry when username could not be inferred on Windows

• Metadata is now correctly updated for hybrid spaces

• Unintended deactivation of telemetry due to import problem

• Line wrapping in examples

Deprecations

• TwoPhaseStrategy, SequentialStrategy and StreamingSequentialStrategy have been replaced with their new MetaRecommender versions

[0.7.3] - 2024-02-09

Added

• Copy button for code blocks in documentation

• mypy for campaign, constraints and telemetry

• Top-level example summaries

• RecommenderProtocol as common interface for Strategy and Recommender

• SubspaceDiscrete.from_simplex convenience constructor

Changed

• Order of README sections

• Imports from top level baybe.utils no longer possible

• Renamed utils.numeric to utils.numerical

• Optional chem dependencies are lazily imported, improving startup time

Fixed

• Several minor issues in documentation

• Visibility and constructor exposure of Campaign attributes that should be private

• TaskParameters no longer disappear from computational representation when the search space contains only one task parameter value

• Failing baybe import from environments containing only core dependencies caused by eagerly loading chem dependencies

• tox coretest now uses correct environment and skips unavailable tests

• Basic serialization example no longer requires optional chem dependencies

Removed

• Detailed headings in table of contents of examples

Deprecations

• Passing numerical_measurements_must_be_within_tolerance to the Campaign constructor is no longer supported. Instead, Campaign.add_measurements now takes an additional parameter to control the behavior.

• batch_quantity replaced with batch_size

• allow_repeated_recommendations and allow_recommending_already_measured are now attributes of Recommender and no longer attributes of Strategy

[0.7.2] - 2024-01-24

Added

• Target enums

• mypy for targets and intervals

• Tests for code blocks in README and user guides

• hypothesis strategies and roundtrip tests for targets, intervals, and dataframes

• (De-)serialization of target subclasses via base class

• Docs building check now part of CI

• Automatic formatting checks for code examples in documentation

• Deserialization of classes with classmethod constructors can now be customized by providing an optional constructor field

• SearchSpace.from_dataframe convenience constructor

Changed

• Renamed bounds_transform_func target attribute to transformation

• Interval.is_bounded now implements the mathematical definition of boundedness

• Moved and renamed target transform utility functions

• Examples have two levels of headings in the table of content

• Fix orders of examples in table of content

• DiscreteCustomConstraint validator now expects dataframe instead of series

• ignore_example flag builds but does not execute examples when building documentation

• New user guide versions for campaigns, targets and objectives

• Binarization of dataframes now happens via pickling

Fixed

• Wrong use of tolerance argument in constraints user guide

• Errors with generics and type aliases in documentation

• Deduplication bug in substance_data hypothesis strategy

• Use pydoclint as flake8 plugin and not as a stand-alone linter

• Margins in documentation for desktop and mobile version

• Intervals can now also be deserialized from a bounds iterable

• SubspaceDiscrete and SubspaceContinuous now have (de-)serialization methods

Removed

• Conda install instructions and version badge

• Early fail for different Python versions in regular pipeline

Deprecations

• Interval.is_finite replaced with Interval.is_bounded

• Specifying target configs without explicit type information is deprecated

• Specifying parameters/constraints at the top level of a campaign configuration JSON is deprecated. Instead, an explicit searchspace field must be provided with an optional constructor entry

[0.7.1] - 2023-12-07

Added

• Release pipeline now also publishes source distributions

• hypothesis strategies and tests for parameters package

Changed

• Reworked validation tests for parameters package

• SubstanceParameter now collects inconsistent user input in an ExceptionGroup

Fixed

• Link handling in documentation

[0.7.0] - 2023-12-04

Added

• GitHub CI pipelines

• GitHub documentation pipeline

• Optional --force option for building the documentation despite errors

• Enabled passing optional arguments to tox -e docs calls

• Logo and banner images

• Project metadata for pyproject.toml

• PyPI release pipeline

• Favicon for homepage

• More literature references

• First drafts of first user guides

Changed

• Reworked README for GitHub landing page

• Now has concise contribution guidelines

• Use Furo theme for documentation

Removed

• --debug flag for documentation building

[0.6.1] - 2023-11-27

Added

• Script for building HTML documentation and corresponding tox environment

• Linter typos for spellchecking

• Parameter encoding enums

• mypy for parameters package

• tox environments for mypy

Changed

• Replacing pylint, flake8, µfmt and usort with ruff

• Markdown based documentation replaced with HTML based documentation

Fixed

• encoding is no longer a class variable

• Now installed with correct pandas dependency flag

• comp_df column names for CustomDiscreteParameter are now safe

[0.6.0] - 2023-11-17

Added

• Raises section for validators and corresponding contributing guideline

• Bring your own model: surrogate classes for custom model architectures and pre-trained ONNX models

• Test module for deprecation warnings

• Option to control the switching point of TwoPhaseStrategy (former Strategy)

• SequentialStrategy and StreamingSequentialStrategy classes

• Telemetry env variable BAYBE_TELEMETRY_VPN_CHECK turning the initial connectivity check on/off

• Telemetry env variable BAYBE_TELEMETRY_VPN_CHECK_TIMEOUT for setting the connectivity check timeout

Changed

• Reorganized modules into subpackages

• Serialization no longer relies on cattrs’ global converter

• Refined (un-)structuring logic

• Telemetry env variable BAYBE_TELEMETRY_HOST renamed to BAYBE_TELEMETRY_ENDPOINT

• Telemetry env variable BAYBE_DEBUG_FAKE_USERHASH renamed to BAYBE_TELEMETRY_USERNAME

• Telemetry env variable BAYBE_DEBUG_FAKE_HOSTHASH renamed to BAYBE_TELEMETRY_HOSTNAME

• Bumped cattrs version

Fixed

• Now supports Python 3.11

• Removed pyarrow version pin

• TaskParameter added to serialization test

• Deserialization (e.g. from config) no longer silently drops unknown arguments

Deprecations

• BayBE class replaced with Campaign

• baybe.surrogate replaced with baybe.surrogates

• baybe.targets.Objective replaced with baybe.objective.Objective

• baybe.strategies.Strategy replaced with baybe.strategies.TwoPhaseStrategy

[0.5.1] - 2023-10-19

Added

• Linear in-/equality constraints over continuous parameters

• Constrained optimization for SequentialGreedyRecommender

• RandomRecommender now supports linear in-/equality constraints via polytope sampling

Changed

• Include linting for all functions

• Rewrite functions to distinguish between private and public ones

• Unreachable telemetry endpoints now automatically disables telemetry and no longer cause any data submission loops

• add_fake_results utility now considers potential target bounds

• Constraint names have been refactored to indicate whether they operate on discrete or continuous parameters

Fixed

• Random recommendation failing for small discrete (sub-)spaces

• Deserialization issue with TaskParameter

[0.5.0] - 2023-09-15

Added

• TaskParameter for multitask modelling

• Basic transfer learning capability using multitask kernels

• Advanced simulation mechanisms for transfer learning and search space partitioning

• Extensive docstring documentation in all files

• Autodoc using sphinx

• Script for automatic code documentation

• New tox environments for a full and a core-only pytest run

Changed

• Discrete subspaces require unique indices

• Simulation function signatures are redesigned (but largely backwards compatible)

• Docstring contents and style (numpy -> google)

• Regrouped additional dependencies

[0.4.2] - 2023-08-29

Added

• Test environments for multiple Python versions via tox

Changed

• Removed environment.yml

• Telemetry host endpoint is now flexible via the environment variable BAYBE_TELEMETRY_HOST

Fixed

• Inference for __version__

[0.4.1] - 2023-08-23

Added

• Vulnerability check via pip-audit

• tests dependency group

Changed

• Removed no longer required fsspec dependency

Fixed

• Scipy vulnerability by bumping version to 1.10.1

• Missing pyarrow dependency

[0.4.0] - 2023-08-16

Added

• from_dataframe convenience constructors for discrete and continuous subspaces

• from_bounds convenience constructor for continuous subspaces

• empty convenience constructors discrete and continuous subspaces

• baybe, strategies and utils namespace for convenient imports

• Simple test for config validation

• VarUCB and qVarUCB acquisition functions emulating maximum variance for active learning

• Surrogate model serialization

• Surrogate model parameter passing

Changed

• Renamed create constructors to from_product

• Renamed empty checks for subspaces to is_empty

• Fixed inconsistent class names in surrogate.py

• Fixed inconsistent class names in parameters.py

• Cached recommendations are now private

• Parameters, targets and objectives are now immutable

• Adjusted comments in example files

• Accelerated the slowest tests

• Removed try blocks from config examples

• Upgraded numpy requirement to >= 1.24.1

• Requires protobuf<=3.20.3

• SearchSpace parameters in surrogate models are now handled in fit

• Dataframes are encoded in binary for serialization

• comp_rep is loaded directly from the serialization string

Fixed

• Include scaling in FPS recommender

• Support for pandas>=2.0.0

[0.3.2] - 2023-07-24

Added

• Constraints serialization

Changed

• A maximum of one DependenciesConstraint is allowed

• Bumped numpy and matplotlib versions

[0.3.1] - 2023-07-17

Added

• Code coverage check with pytest-cov

• Hybrid mode for SequentialGreedyRecommender

Changed

• Removed support for infinite parameter bounds

• Removed not yet implemented MULTI objective mode

Fixed

• Changelog assert in Azure pipeline

• Bug: telemetry could not be fully deactivated

[0.3.0] - 2023-06-27

Added

• Interval class for representing parameter/target bounds

• Activated mypy for the first few modules and fixed their type issues

• Automatic (de-)serialization and SerialMixin class

• Basic serialization example, demo and tests

• Mechanisms for loading and validating config files

• Telemetry via OpenTelemetry

• More detailed package installation info

• Fallback mechanism for NonPredictiveRecommender

• Introduce naive hybrid recommender

Changed

• Switched from pydantic to attrs in all modules except constraints.py

• Removed subclass initialization hooks and type attribute

• Refactored class attributes and their conversion/validation/initialization

• Removed no longer needed HashableDict class

• Refactored strategy and recommendation module structures

• Replaced dict-based configuration logic with object-based logic

• Overall versioning scheme and version inference for telemetry

• No longer using private telemetry imports

• Fixed package versions for dev tools

• Revised “Getting Started” section in README.md

• Revised examples

Fixed

• Telemetry no longer crashing when package was not installed

[0.2.4] - 2023-03-24

Added

• Tests for different search space types and their compatible recommenders

Changed

• Initial strategies converted to recommenders

• Config keyword initial_strategy replaced by initial_recommender_cls

• Config keywords for the clustering recommenders changed from x to CLUSTERING_x

• skicit-learn-extra is now optional dependency in the [extra] group

• Type identifiers of greedy recommenders changed to ‘SEQUENTIAL_GREEDY_x’

Fixed

• Parameter bounds now only contain dimensions that actually appear in the search space

[0.2.3] - 2023-03-14

Added

• Parsing for continuous parameters

• Caching of recommendations to avoid unnecessary computations

• Strategy support for hybrid spaces

• Custom discrete constraint with user-provided validator

Changed

• Parameter class hierarchy

• SearchSpace has now a discrete and continuous subspace

• Model fit now done upon requesting recommendations

Fixed

• Updated BoTorch and GPyTorch versions are also used in pyproject.toml

[0.2.2] - 2023-01-13

Added

• SearchSpace class

• Code testing with pytest

• Option to specify initial data for backtesting simulations

• SequentialGreedyRecommender class

Changed

• Switched from miniconda to micromamba in Azure pipeline

Fixed

• BoTorch version upgrade to fix critical bug (https://github.com/pytorch/botorch/pull/1454)

[0.2.1] - 2022-12-01

Fixed

• Parameters cannot be initialized with duplicate values

[0.2.0] - 2022-11-10

Added

• Initial strategy: Farthest Point Sampling

• Initial strategy: Partitioning Around Medoids

• Initial strategy: K-means

• Initial strategy: Gaussian Mixture Model

• Constraints and conditions for discrete parameters

• Data scaling functionality

• Decorator for automatic model scaling

• Decorator for handling constant targets

• Decorator for handling batched model input

• Surrogate model: Mean prediction

• Surrogate model: Random forrest

• Surrogate model: NGBoost

• Surrogate model: Bayesian linear

• Save/load functionality for BayBE objects

Fixed

• UCB now usable as acquisition function, hard-set beta parameter to 1.0

• Temporary GP priors now exactly reproduce EDBO setting

[0.1.0] - 2022-10-01

Added

• Code skeleton with a central object to access functionality

• Basic parser for categorical parameters with one-hot encoding

• Basic parser for discrete numerical parameters

• Azure pipeline for code formatting and linting

• Single-task Gaussian process strategy

• Streamlit dashboard for comparing single-task strategies

• Input functionality to read measurements including automatic matching to search space

• Integer encoding for categorical parameters

• Parser for numerical discrete parameters

• Single numerical target with Min and Max mode

• Recommendation functionality

• Parameter scaling depending on parameter types and user-chosen scalers

• Noise and fake-measurement utilities

• Internal metadata storing various info about datapoints in the search space

• BayBE options controlling recommendation and data addition behavior

• Config parsing and validation using pydantic

• Global random seed control

• Strategy connection with BayBE object

• Custom parameters as labels with user-provided encodings

• Substance parameters which are encoded via cheminformatics descriptors

• Data cleaning utilities useful for descriptors

• Simulation capabilities for testing the package on existing data

• Parsing and preprocessing for multiple targets / desirability ansatz

• Basic README file

• Automatic publishing of tagged versions

• Caching of experimental parameters and chemical descriptors

• Choices for acquisition functions and their usage with arbitrary surrogate models

• Temporary logic for selecting GP priors