BayesianQuadratureWeighting

class otkerneldesign.BayesianQuadratureWeighting(kernel=None, distribution=None, distribution_sample_size=None, distribution_sample=None)

Optimally-weighting a sample for probabilistic integration.

Parameters:
kernelopenturns.CovarianceModel

Covariance kernel used to define potentials. By default a product of Matern kernels with smoothness 5/2.

distributionopenturns.Distribution

Distribution of the set of candidate set. If not specified, then distribution_sample must be specified instead. Even if distribution_sample is specified, can be useful if it allows the use of analytical formulas.

distribution_sample_sizepositive int

Size of the set of all candidate points. Unnecessary if distribution_sample is specified. Otherwise, 2^{12} by default.

distribution_sample2-d list of float

Large sample that empirically represents a distribution. If not specified, then distribution and distribution_sample_size must be in order to generate it automatically.

Methods

compute_bayesian_quadrature_weights(input_sample)

Compute optimal weights for probabilistic integration using a given sample.

Parameters:
input_sampleopenturns.Sample

Sample of points to be optimally weighted.