saealib.PreSelectionStrategy

class saealib.PreSelectionStrategy(n_candidates, n_select)

ベースクラス: OptimizationStrategy

Pre-selection strategy.

Generates n_candidates offspring, scores them with the surrogate, and true-evaluates only the top n_select candidates.

パラメータ:

• n_candidates (int) -- Number of candidates to generate and score with the surrogate.

• n_select (int) -- Number of top-scoring candidates to evaluate with the true objective function.

Methods

__init__	Initialize PreSelectionStrategy.
build_graph	Build the canonical pre-selection strategy graph.
build_pipeline	Return the strategy's high-level Pipeline description.
contract	Return the strategy contract.
step	Generate a large candidate pool, screen with surrogate, true-evaluate top-k.

Method Details

PreSelectionStrategy.__init__(n_candidates, n_select)

Initialize PreSelectionStrategy.

パラメータ:

• n_candidates (int) -- Number of candidates to generate and score with the surrogate.

• n_select (int) -- Number of top-scoring candidates to evaluate with the true objective function.

PreSelectionStrategy.build_graph(provider)

Build the canonical pre-selection strategy graph.

パラメータ:

provider (ComponentProvider)

戻り値の型:

ComponentGraph

PreSelectionStrategy.build_pipeline(provider)

Return the strategy's high-level Pipeline description.

パラメータ:

provider (ComponentProvider)

戻り値の型:

Pipeline

PreSelectionStrategy.contract()

Return the strategy contract.

戻り値の型:

ComponentContract

PreSelectionStrategy.step(ctx, provider)

Generate a large candidate pool, screen with surrogate, true-evaluate top-k.

Rebuilds the pipeline each call so component/parameter changes take effect immediately.

パラメータ:

• ctx (OptimizationState) -- Current optimization context.

• provider (ComponentProvider) -- Component provider.

戻り値:

Updated state after one generation step.

戻り値の型:

OptimizationState