SKLearnClusteringRecommender

class baybe.recommenders.pure.nonpredictive.clustering.SKLearnClusteringRecommender

Bases: NonPredictiveRecommender, ABC

Intermediate class for cluster-based selection of discrete candidates.

Suitable for sklearn-like models that have a fit and predict method. Specific model parameters and cluster sub-selection techniques can be declared in the derived classes.

Public methods

__init__([model_params, ...])	Method generated by attrs for class SKLearnClusteringRecommender.
recommend(batch_size, searchspace[, ...])	Recommend a batch of points from the given search space.

Public attributes and properties

model_params	Optional model parameter that will be passed to the surrogate constructor.
compatibility	Class variable reflecting the search space compatibility.
model_cluster_num_parameter_name	Class variable describing the name of the clustering parameter.
allow_repeated_recommendations	Allow to make recommendations that were already recommended earlier.
allow_recommending_already_measured	Allow to make recommendations that were measured previously.
allow_recommending_pending_experiments	Allow pending_experiments to be part of the recommendations.

__init__(model_params: dict = NOTHING, *, allow_repeated_recommendations: bool = False, allow_recommending_already_measured: bool = True, allow_recommending_pending_experiments: bool = False)

Method generated by attrs for class SKLearnClusteringRecommender.

For details on the parameters, see Public attributes and properties.

recommend(batch_size: int, searchspace: SearchSpace, objective: Objective | None = None, measurements: DataFrame | None = None, pending_experiments: DataFrame | None = None)

Recommend a batch of points from the given search space.

Parameters:

• batch_size (int) – The number of points to be recommended.

• searchspace (SearchSpace) – The search space from which to recommend the points.

• objective (Optional[Objective]) – An optional objective to be optimized.

• measurements (Optional[DataFrame]) – Optional experimentation data that can be used for model training. The data is to be provided in “experimental representation”: It needs to contain one column for each parameter spanning the search space (column name matching the parameter name) and one column for each target tracked by the objective (column name matching the target name). Each row corresponds to one conducted experiment, where the parameter columns define the experimental setting and the target columns report the measured outcomes.

• pending_experiments (Optional[DataFrame]) – Parameter configurations in “experimental representation” specifying experiments that are currently pending.

Return type:

DataFrame

Returns:

A dataframe containing the recommendations in experimental representation as individual rows.

allow_recommending_already_measured: bool

Allow to make recommendations that were measured previously. This only has an influence in discrete search spaces.

allow_recommending_pending_experiments: bool

Allow pending_experiments to be part of the recommendations. If set to False, the corresponding points will be removed from the candidates. This only has an influence in discrete search spaces.

allow_repeated_recommendations: bool

Allow to make recommendations that were already recommended earlier. This only has an influence in discrete search spaces.

compatibility: ClassVar[SearchSpaceType] = 'DISCRETE'

Class variable reflecting the search space compatibility.

model_cluster_num_parameter_name: ClassVar[str]

Class variable describing the name of the clustering parameter.

model_params: dict

Optional model parameter that will be passed to the surrogate constructor. This is initialized with reasonable default values for the derived child classes.