saealib.GlobalSurrogateManager¶

class saealib.GlobalSurrogateManager(surrogate, acquisition, training_set=None)[source]¶

Bases: SurrogateManager

Surrogate manager that fits once on the full archive.

Fits the surrogate on all archived solutions, then predicts and scores all candidates in a single batch. Suitable when global approximation quality is sufficient.

Parameters:

surrogate (Surrogate) – Surrogate model instance.
acquisition (AcquisitionFunction) – Acquisition function used to score predictions.
training_set (TrainingSet or None) – Strategy object for building training data. Defaults to ArchiveObjectiveSet(), which preserves the previous behaviour.

Methods

`__init__`
`score_candidates`	Fit on full archive, predict and score all candidates at once.

Method Details

GlobalSurrogateManager.__init__(surrogate, acquisition, training_set=None)[source]¶

Parameters:

surrogate (Surrogate)
acquisition (AcquisitionFunction)
training_set (TrainingSet | None)

GlobalSurrogateManager.score_candidates(candidates_x, archive, provider=None, ctx=None)[source]¶

Fit on full archive, predict and score all candidates at once.

Parameters:

candidates_x (np.ndarray)
archive (Archive)
provider (ComponentProvider | None)
ctx (OptimizationContext | None)

Return type:

tuple[np.ndarray, list[SurrogatePrediction]]