fpderiv

Forward propagation derivative function

Syntax

fpderiv('dperf_dwb',net,X,T,Xi,Ai,EW)
fpderiv('de_dwb',net,X,T,Xi,Ai,EW)


Description

This function calculates derivatives using the chain rule from inputs to outputs, and in the case of dynamic networks, forward through time.

fpderiv('dperf_dwb',net,X,T,Xi,Ai,EW) takes these arguments,

net	Neural network
X	Inputs, an R-by-Q matrix (or N-by-TS cell array of Ri-by-Q matrices)
T	Targets, an S-by-Q matrix (or M-by-TS cell array of Si-by-Q matrices)
Xi	Initial input delay states (optional)
Ai	Initial layer delay states (optional)
EW	Error weights (optional)

and returns the gradient of performance with respect to the network’s weights and biases, where R and S are the number of input and output elements and Q is the number of samples (or N and M are the number of input and output signals, Ri and Si are the number of each input and outputs elements, and TS is the number of timesteps).

fpderiv('de_dwb',net,X,T,Xi,Ai,EW) returns the Jacobian of errors with respect to the network’s weights and biases.

Examples

Here a feedforward network is trained and both the gradient and Jacobian are calculated.

[x,t] = simplefit_dataset;
net = feedforwardnet(20);
net = train(net,x,t);
y = net(x);
perf = perform(net,t,y);
gwb = fpderiv('dperf_dwb',net,x,t)
jwb = fpderiv('de_dwb',net,x,t)

Version History

Introduced in R2010b