LatinDesign

class probnum.quad.solvers.initial_designs.LatinDesign(n_nodes, measure)

Bases: InitialDesign

Initial design for Bayesian quadrature that samples from a Latin hypercube. 1

Parameters

  • n_nodes (IntLike) – The number of nodes to be designed.

  • measure (IntegrationMeasure) – The integration measure.

References

1

Mckay et al., A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code. Technometrics, 1979.

Attributes Summary

requires_rng

Whether the initial design requires a random number generator when called.

Methods Summary

__call__(rng)

Get the initial nodes.

Attributes Documentation

requires_rng

Methods Documentation

__call__(rng)[source]

Get the initial nodes.

Parameters

rng (Optional[Generator]) – A random number generator.

Returns

shape=(n_nodes, input_dim) – Initial design nodes.

Return type

nodes