Daniel@0: function [net, options, varargout] = netopt_weighted(net, options, x, t, eso_w, alg);
Daniel@0: %NETOPT	Optimize the weights in a network model. 
Daniel@0: %
Daniel@0: %	Description
Daniel@0: %
Daniel@0: %	NETOPT is a helper function which facilitates the training of
Daniel@0: %	networks using the general purpose optimizers as well as sampling
Daniel@0: %	from the posterior distribution of parameters using general purpose
Daniel@0: %	Markov chain Monte Carlo sampling algorithms. It can be used with any
Daniel@0: %	function that searches in parameter space using error and gradient
Daniel@0: %	functions.
Daniel@0: %
Daniel@0: %	[NET, OPTIONS] = NETOPT(NET, OPTIONS, X, T, ALG) takes a network
Daniel@0: %	data structure NET, together with a vector OPTIONS of parameters
Daniel@0: %	governing the behaviour of the optimization algorithm, a matrix X of
Daniel@0: %	input vectors and a matrix T of target vectors, and returns the
Daniel@0: %	trained network as well as an updated OPTIONS vector. The string ALG
Daniel@0: %	determines which optimization algorithm (CONJGRAD, QUASINEW, SCG,
Daniel@0: %	etc.) or Monte Carlo algorithm (such as HMC) will be used.
Daniel@0: %
Daniel@0: %	[NET, OPTIONS, VARARGOUT] = NETOPT(NET, OPTIONS, X, T, ALG) also
Daniel@0: %	returns any additional return values from the optimisation algorithm.
Daniel@0: %
Daniel@0: %	See also
Daniel@0: %	NETGRAD, BFGS, CONJGRAD, GRADDESC, HMC, SCG
Daniel@0: %
Daniel@0: 
Daniel@0: %	Copyright (c) Ian T Nabney (1996-9)
Daniel@0: 
Daniel@0: optstring = [alg, '(''neterr_weighted'', w, options, ''netgrad_weighted'', net, x, t, eso_w)'];
Daniel@0: 
Daniel@0: % Extract weights from network as single vector
Daniel@0: w = netpak(net);
Daniel@0: 
Daniel@0: % Carry out optimisation
Daniel@0: [s{1:nargout}] = eval(optstring);
Daniel@0: w = s{1};
Daniel@0: 
Daniel@0: if nargout > 1
Daniel@0:   options = s{2};
Daniel@0: 
Daniel@0:   % If there are additional arguments, extract them
Daniel@0:   nextra = nargout - 2;
Daniel@0:   if nextra > 0
Daniel@0:     for i = 1:nextra
Daniel@0:       varargout{i} = s{i+2};
Daniel@0:     end
Daniel@0:   end
Daniel@0: end
Daniel@0: 
Daniel@0: % Pack the weights back into the network
Daniel@0: net = netunpak(net, w);