Daniel@0: function [y, extra, invhess] = netevfwd(w, net, x, t, x_test, invhess)
Daniel@0: %NETEVFWD Generic forward propagation with evidence for network
Daniel@0: %
Daniel@0: %	Description
Daniel@0: %	[Y, EXTRA] = NETEVFWD(W, NET, X, T, X_TEST) takes a network data
Daniel@0: %	structure  NET together with the input X and target T training data
Daniel@0: %	and input test data X_TEST. It returns the normal forward propagation
Daniel@0: %	through the network Y together with a matrix EXTRA which consists of
Daniel@0: %	error bars (variance) for a regression problem or moderated outputs
Daniel@0: %	for a classification problem.
Daniel@0: %
Daniel@0: %	The optional argument (and return value)  INVHESS is the inverse of
Daniel@0: %	the network Hessian computed on the training data inputs and targets.
Daniel@0: %	Passing it in avoids recomputing it, which can be a significant
Daniel@0: %	saving for large training sets.
Daniel@0: %
Daniel@0: %	See also
Daniel@0: %	MLPEVFWD, RBFEVFWD, GLMEVFWD, FEVBAYES
Daniel@0: %
Daniel@0: 
Daniel@0: %	Copyright (c) Ian T Nabney (1996-2001)
Daniel@0: 
Daniel@0: func = [net.type, 'evfwd'];
Daniel@0: net = netunpak(net, w);
Daniel@0: if nargin == 5
Daniel@0:   [y, extra, invhess] = feval(func, net, x, t, x_test);
Daniel@0: else
Daniel@0:   [y, extra, invhess] = feval(func, net, x, t, x_test, invhess);
Daniel@0: end