saealib.ConstraintObjectiveSet#

class saealib.ConstraintObjectiveSet[ソース]#

ベースクラス: TrainingSet

Use the entire archive as training data with raw constraint values.

Analogue of ArchiveObjectiveSet for constraint surrogates. Returns (archive.x, archive.g) so that a surrogate can learn to predict the raw constraint values g(x).

例外:

ValueError -- If the archive has no constraint columns (archive.g.shape[1] == 0). Use this class only with problems that define at least one constraint.

Methods

__init__

build

Return all archive points with raw constraint values.

contract

Return the training-set contract.

Method Details

ConstraintObjectiveSet.__init__()#
ConstraintObjectiveSet.build(archive, population, ctx, candidate_x=None)[ソース]#

Return all archive points with raw constraint values.

パラメータ:

  • archive (Archive)

  • population (Population | None)

  • ctx (OptimizationState | None)

  • candidate_x (np.ndarray | None)

戻り値の型:

TrainingData

ConstraintObjectiveSet.contract()#

Return the training-set contract.

戻り値の型:

ComponentContract