MATLAB: How to design a custom neural network with a modified output

Question

I'd like to know how to set up a neural network with time and position arguments using Matlab. My question is a bit abstract, but the project I'm working on involves computer vision and using points to produce a contour.

To set up the problem I have a known argument x, which represents position. I want to find y(x), this is my target. But, I can only verify that y(x) is correct by plugging it into the function z(y(x)).

The problem seems somewhat like a narxnet,

[X,T] = simplenarx_dataset;
net = narxnet(1:2,1:2,10);
[Xs,Xi,Ai,Ts] = preparets(net,X,{},T);
net = train(net,Xs,Ts,Xi,Ai);
view(net)

This code produces the net

I would like to add a function after y(t), say z(t), and then check that the z(t) is correct with the known value.

Answer 1

Just use a two-dimensional output trained with the target values for y and g.

Hope this helps.

Thank you for formally accepting my answer

Greg