MATLAB: Several non-feedback inputs to a dynamic nnetwork

dynamicneural network

hi
I am trying to train a net, something like this http://i44.tinypic.com/ofofmd.png that has 2 inputs signals with sizes 1 & 4 (as shown) and an one output. The problem is that I can not prepare the training data using the preparets, actually I don't know how to implement.
%v, xi, y all are 1*N
v = v;
x = [x1; x2; x3; x4];
y = y;
v = con2seq(v);
x = con2seq(x);
y = con2seq(y);
Xnf = {v; x}; % Non-feedback inputs ????????
Tnf = y; % Non-feedback targets
Tf = {}; % Feedback targets
[Xs,Xi,Ai,Ts] = preparets(NNC,Xnf,Tnf,Tf);
Thanks for your time 🙂

Best Answer

I found the answer!
The code should be changed to
%v, xi, y all are 1*N
v = v;
x = [x1; x2; x3; x4];
y = y;
v = con2seq(v);
x = con2seq(x);
y = con2seq(y);
Xnf = [v; x]; % Non-feedback inputs
Tnf = y; % Non-feedback targets
Tf = {}; % Feedback targets
[Xs,Xi,Ai,Ts] = preparets(NNC,Xnf,Tnf,Tf);