saealib.DirectStrategy

class saealib.DirectStrategy(n_offspring=None)

Bases: OptimizationStrategy

Plain EA strategy without surrogate scoring.

Every candidate produced by ask() is evaluated with the true objective function. No surrogate manager is required.

Parameters:

n_offspring (int | None)

Methods

__init__
build_graph	Build the canonical direct strategy graph.
build_pipeline	Return the strategy's high-level Pipeline description.
contract	Return the strategy contract.
step	Perform one generation step: generate, score, evaluate, and update.

Method Details

DirectStrategy.__init__(n_offspring=None)

Parameters:

n_offspring (int | None)

Return type:

None

DirectStrategy.build_graph(provider)

Build the canonical direct strategy graph.

Parameters:

provider (ComponentProvider)

Return type:

ComponentGraph

DirectStrategy.build_pipeline(provider)

Return the strategy’s high-level Pipeline description.

Parameters:

provider (ComponentProvider)

Return type:

Pipeline

DirectStrategy.contract()

Return the strategy contract.

Return type:

ComponentContract

DirectStrategy.step(ctx, provider)

Perform one generation step: generate, score, evaluate, and update.

Parameters:

• ctx (OptimizationState) – Current optimization context.

• provider (ComponentProvider) – Component provider.

Returns:

Updated state returned by the pipeline.

Return type:

OptimizationState