wolffd@0: function net = gpunpak(net, hp)
wolffd@0: %GPUNPAK Separates hyperparameter vector into components. 
wolffd@0: %
wolffd@0: %	Description
wolffd@0: %	NET = GPUNPAK(NET, HP) takes an Gaussian Process data structure NET
wolffd@0: %	and  a hyperparameter vector HP, and returns a Gaussian Process data
wolffd@0: %	structure  identical to the input model, except that the covariance
wolffd@0: %	bias BIAS, output noise NOISE, the input weight vector INWEIGHTS and
wolffd@0: %	the vector of covariance function specific parameters  FPAR have all
wolffd@0: %	been set to the corresponding elements of HP.
wolffd@0: %
wolffd@0: %	See also
wolffd@0: %	GP, GPPAK, GPFWD, GPERR, GPGRAD
wolffd@0: %
wolffd@0: 
wolffd@0: %	Copyright (c) Ian T Nabney (1996-2001)
wolffd@0: 
wolffd@0: % Check arguments for consistency
wolffd@0: errstring = consist(net, 'gp');
wolffd@0: if ~isempty(errstring);
wolffd@0:   error(errstring);
wolffd@0: end
wolffd@0: if net.nwts ~= length(hp)
wolffd@0:   error('Invalid weight vector length');
wolffd@0: end
wolffd@0: 
wolffd@0: net.bias = hp(1);
wolffd@0: net.noise = hp(2);
wolffd@0: 
wolffd@0: % Unpack input weights
wolffd@0: mark1 = 2 + net.nin;
wolffd@0: net.inweights = hp(3:mark1);
wolffd@0: 
wolffd@0: % Unpack function specific parameters
wolffd@0: net.fpar = hp(mark1 + 1:size(hp, 2));
wolffd@0: