saealib.BatchExpectedImprovement#

class saealib.BatchExpectedImprovement(n_draws=4096, direction=None)[source]#

Bases: AcquisitionFunction

Monte Carlo qEI with candidate-joint covariance and greedy set utility.

Parameters:

n_draws (int)
direction (np.ndarray | None)

Methods

`__init__`
`contract`	Return the batch expected-improvement contract.
`evaluate`	Return greedy qEI marginal scores from joint normal draws.
`prepare`	Compute a generation-scoped reference value shared across `evaluate()` calls.

Method Details

BatchExpectedImprovement.__init__(n_draws=4096, direction=None)[source]#

Parameters:

n_draws (int)
direction (ndarray | None)

BatchExpectedImprovement.contract()[source]#

Return the batch expected-improvement contract.

Return type:: ComponentContract

BatchExpectedImprovement.evaluate(candidates_x, prediction, archive, ctx=None, *, prepared=None)[source]#: Return greedy qEI marginal scores from joint normal draws.

BatchExpectedImprovement.prepare(archive, ctx=None)#

Compute a generation-scoped reference value shared across evaluate() calls.

Concrete hook returning None by default. PointwiseAcquisition overrides it with the compute_reference() behavior; archive-only acquisitions need not override it.

Parameters:

archive (Archive) – Archive of evaluated solutions.
ctx (OptimizationState or None, optional) – Optimization state, used here only for ctx.rng.

Returns:

A prepared reference value, passed to evaluate(..., prepared=...).

Return type:

Any