Class CovSum

java.lang.Object
jgpml.covariancefunctions.CovSum
All Implemented Interfaces:
CovarianceFunction

public class CovSum extends Object implements CovarianceFunction
Composes a covariance function as the sum of other covariance functions. This function doesn't actually compute very much on its own, it merely calls other covariance functions with the right parameters.
  • Constructor Summary

    Constructors
    Constructor
    Description
    CovSum(int inputDimensions, CovarianceFunction... f)
    Create a new CovarianceFunction as sum of the CovarianceFunctions passed as input.
  • Method Summary

    Modifier and Type
    Method
    Description
    Jama.Matrix
    compute(Jama.Matrix loghyper, Jama.Matrix X)
    Compute covariance matrix of a dataset X
    Jama.Matrix[]
    compute(Jama.Matrix loghyper, Jama.Matrix X, Jama.Matrix Xstar)
    Compute compute test set covariances
    Jama.Matrix
    computeDerivatives(Jama.Matrix loghyper, Jama.Matrix X, int index)
    Coompute the derivatives of this CovarianceFunction with respect to the hyperparameter with index idx
    int
    Returns the number of hyperparameters of thisCovarianceFunction

    Methods inherited from class java.lang.Object

    clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
  • Constructor Details

    • CovSum

      public CovSum(int inputDimensions, CovarianceFunction... f)
      Create a new CovarianceFunction as sum of the CovarianceFunctions passed as input.
      Parameters:
      inputDimensions - input dimension of the dataset
      f - array of CovarianceFunction
      See Also:
  • Method Details

    • numParameters

      public int numParameters()
      Returns the number of hyperparameters of thisCovarianceFunction
      Specified by:
      numParameters in interface CovarianceFunction
      Returns:
      number of hyperparameters
    • compute

      public Jama.Matrix compute(Jama.Matrix loghyper, Jama.Matrix X)
      Compute covariance matrix of a dataset X
      Specified by:
      compute in interface CovarianceFunction
      Parameters:
      loghyper - column Matrix of hyperparameters
      X - input dataset
      Returns:
      K covariance Matrix
    • compute

      public Jama.Matrix[] compute(Jama.Matrix loghyper, Jama.Matrix X, Jama.Matrix Xstar)
      Compute compute test set covariances
      Specified by:
      compute in interface CovarianceFunction
      Parameters:
      loghyper - column Matrix of hyperparameters
      X - input dataset
      Xstar - test set
      Returns:
      [K(Xstar, Xstar) K(X,Xstar)]
    • computeDerivatives

      public Jama.Matrix computeDerivatives(Jama.Matrix loghyper, Jama.Matrix X, int index)
      Coompute the derivatives of this CovarianceFunction with respect to the hyperparameter with index idx
      Specified by:
      computeDerivatives in interface CovarianceFunction
      Parameters:
      loghyper - hyperparameters
      X - input dataset
      index - hyperparameter index
      Returns:
      Matrix of derivatives