poli.objective_repository.white_noise.register.WhiteNoiseBlackBox

class poli.objective_repository.white_noise.register.WhiteNoiseBlackBox(batch_size: Optional[int] = None, parallelize: bool = False, num_workers: Optional[int] = None, evaluation_budget: int = inf)

A toy black box function that generates standard Gaussian noise.

Parameters

• batch_size (int, optional) – The batch size for vectorized evaluation.

• parallelize (bool, optional) – Whether to parallelize the evaluation.

• num_workers (int, optional) – The number of workers for parallel evaluation.

• evaluation_budget (int, optional) – The maximum number of evaluations.

_black_box(x, context=None)

Returns standard Gaussian noise.

__init__(batch_size: Optional[int] = None, parallelize: bool = False, num_workers: Optional[int] = None, evaluation_budget: int = inf)

Initializes a WhiteNoiseBlackBox.

Parameters

• info (ProblemSetupInformation) – The problem setup information.

• batch_size (int, optional) – The batch size for vectorized evaluation, by default None (i.e. all of the input).

• parallelize (bool, optional) – Whether to parallelize the evaluation, by default False.

• num_workers (int, optional) – The number of workers for parallel evaluation, by default None (which corresponds to half the CPUs available, rounded downwards).

• evaluation_budget (int, optional) – The maximum number of evaluations, by default float(“inf”).

Methods

__init__([batch_size, parallelize, ...])	Initializes a WhiteNoiseBlackBox.
reset_evaluation_budget()	Resets the evaluation budget by setting the number of evaluations made to 0.
set_observer(observer)	Set the observer object for recording observations during evaluation.
terminate()	Terminate the black box optimization problem.