Daniel@0: Daniel@0:
Daniel@0:Daniel@0: Daniel@0: [y, extra] = netevfwd(w, net, x, t, x_test) Daniel@0: [y, extra, invhess] = netevfwd(w, net, x, t, x_test, invhess) Daniel@0:Daniel@0: Daniel@0: Daniel@0:
[y, extra] = netevfwd(w, net, x, t, x_test)
takes a network data
Daniel@0: structure
Daniel@0: net
together with the input x
and target t
training data
Daniel@0: and input test data x_test
.
Daniel@0: It returns the normal forward propagation through the network y
Daniel@0: together with a matrix extra
which consists of error bars (variance)
Daniel@0: for a regression problem or moderated outputs for a classification problem.
Daniel@0:
Daniel@0: The optional argument (and return value)
Daniel@0: invhess
is the inverse of the network Hessian
Daniel@0: computed on the training data inputs and targets. Passing it in avoids
Daniel@0: recomputing it, which can be a significant saving for large training sets.
Daniel@0:
Daniel@0:
mlpevfwd
, rbfevfwd
, glmevfwd
, fevbayes
Copyright (c) Ian T Nabney (1996-9) Daniel@0: Daniel@0: Daniel@0: Daniel@0: