Daniel@0: function [cov, covf] = gpcovar(net, x)
Daniel@0: %GPCOVAR Calculate the covariance for a Gaussian Process.
Daniel@0: %
Daniel@0: %	Description
Daniel@0: %
Daniel@0: %	COV = GPCOVAR(NET, X) takes  a Gaussian Process data structure NET
Daniel@0: %	together with a matrix X of input vectors, and computes the
Daniel@0: %	covariance matrix COV.  The inverse of this matrix is used when
Daniel@0: %	calculating the mean and variance of the predictions made by NET.
Daniel@0: %
Daniel@0: %	[COV, COVF] = GPCOVAR(NET, X) also generates the covariance matrix
Daniel@0: %	due to the covariance function specified by NET.COVARFN as calculated
Daniel@0: %	by GPCOVARF.
Daniel@0: %
Daniel@0: %	See also
Daniel@0: %	GP, GPPAK, GPUNPAK, GPCOVARP, GPCOVARF, GPFWD, GPERR, GPGRAD
Daniel@0: %
Daniel@0: 
Daniel@0: %	Copyright (c) Ian T Nabney (1996-2001)
Daniel@0: 
Daniel@0: % Check arguments for consistency
Daniel@0: errstring = consist(net, 'gp', x);
Daniel@0: if ~isempty(errstring);
Daniel@0:   error(errstring);
Daniel@0: end
Daniel@0: 
Daniel@0: ndata = size(x, 1);
Daniel@0: 
Daniel@0: % Compute prior covariance
Daniel@0: if nargout >= 2
Daniel@0:   [covp, covf] = gpcovarp(net, x, x);
Daniel@0: else
Daniel@0:   covp = gpcovarp(net, x, x);
Daniel@0: end
Daniel@0: 
Daniel@0: % Add output noise variance
Daniel@0: cov = covp + (net.min_noise + exp(net.noise))*eye(ndata);
Daniel@0: