emukit.experimental_design.acquisitions package

Submodules

Contains integrated variance acquisition

class emukit.experimental_design.acquisitions.integrated_variance.IntegratedVarianceReduction(model, space, x_monte_carlo=None, num_monte_carlo_points=100000)

Bases: Acquisition

Acquisition function for integrated variance reduction

evaluate(x)

Parameters:: x (ndarray) – The new training point(s) to evaluate with shape (n_points x 1)
Return type:: ndarray
Returns:: A numpy array with shape (n_points x 1) containing the values of the acquisition evaluated at each x row

property has_gradients

Abstract property. Whether acquisition value has analytical gradient calculation available.

Returns:: True if gradients are available

class emukit.experimental_design.acquisitions.model_variance.ModelVariance(model)

Bases: Acquisition

This acquisition selects the point in the domain where the predictive variance is the highest

evaluate(x)

Abstract method. Evaluates the acquisition function.

Parameters:: x (ndarray) – (n_points x n_dims) array of points at which to calculate acquisition function values
Return type:: ndarray
Returns:: (n_points x 1) array of acquisition function values

evaluate_with_gradients(x)

Optional abstract method that must be implemented if has_gradients returns True. Evaluates value and gradient of acquisition function at x.

Parameters:: x (ndarray) – (n_points x n_dims) array of points at which to calculate acquisition function values and gradient
Return type:: Tuple
Returns:: Tuple contains an (n_points x 1) array of acquisition function values and (n_points x n_dims) array of acquisition function gradients with respect to x

property has_gradients

Abstract property. Whether acquisition value has analytical gradient calculation available.

Returns:: True if gradients are available

emukit.experimental_design.acquisitions package

Submodules

Module contents