wolffd@0: wolffd@0: wolffd@0: wolffd@0: Netlab Reference Manual mlppak wolffd@0: wolffd@0: wolffd@0: wolffd@0:

mlppak wolffd@0:

wolffd@0:

wolffd@0: Purpose wolffd@0:

wolffd@0: Combines weights and biases into one weights vector. wolffd@0: wolffd@0:

wolffd@0: Synopsis wolffd@0:

wolffd@0:

wolffd@0: w = mlppak(net)
wolffd@0:

wolffd@0: wolffd@0: wolffd@0:

wolffd@0: Description wolffd@0:

wolffd@0: w = mlppak(net) takes a network data structure net and wolffd@0: combines the component weight matrices bias vectors into a single row wolffd@0: vector w. The facility to switch between these two wolffd@0: representations for the network parameters is useful, for example, in wolffd@0: training a network by error function minimization, since a single wolffd@0: vector of parameters can be handled by general-purpose optimization wolffd@0: routines. wolffd@0: wolffd@0:

The ordering of the paramters in w is defined by wolffd@0:

wolffd@0: 
wolffd@0:   w = [net.w1(:)', net.b1, net.w2(:)', net.b2];
wolffd@0:

wolffd@0: wolffd@0: where w1 is the first-layer weight matrix, b1 is the wolffd@0: first-layer bias vector, w2 is the second-layer weight matrix, wolffd@0: and b2 is the second-layer bias vector. wolffd@0: wolffd@0:

wolffd@0: See Also wolffd@0:

wolffd@0: mlp, mlpunpak, mlpfwd, mlperr, mlpbkp, mlpgrad

wolffd@0: Pages: wolffd@0: Index wolffd@0:

wolffd@0: