package gpr

  1. Overview
  2. Docs
Legend:
Library
Module
Module type
Parameter
Class
Class type

Module for sampling (multiple) points from the posterior distribution accounting for their covariance

type t

Type of covariance sampler

val calc : ?predictive:bool -> Means.t -> Covariances.t -> t

calc ?predictive mean variance

  • returns

    sampler given means and covariances. If predictive is true, the samples will be noisy.

val sample : ?rng:Gsl.Rng.t -> t -> Lacaml.D.vec

sample ?rng sampler

  • returns

    a sample vector from the posterior distribution given sampler and GSL random number generator rng.

  • parameter rng

    default = GSL default

val samples : ?rng:Gsl.Rng.t -> t -> n:int -> Lacaml.D.mat

samples ?rng sampler ~n

  • returns

    matrix of n sample vectors (stored row-wise) from the posterior distribution given sampler.

  • parameter rng

    default = GSL default