Convert neural network closed-loop feedback to open loop
net = openloop(net)
[net,xi,ai] = openloop(net,xi,ai)
net = openloop(net) takes a neural network and opens any closed-loop
feedback. For each feedback output i whose property
net.outputs{i}.feedbackMode is 'closed', it replaces its
associated feedback layer weights with a new input and input weight connections. The
net.outputs{i}.feedbackMode property is set to 'open',
and the net.outputs{i}.feedbackInput property is set to the index of the new
input. Finally, the value of net.outputs{i}.feedbackDelays is subtracted from
the delays of the feedback input weights (i.e., to the delays values of the replaced layer
weights).
[net,xi,ai] = openloop(net,xi,ai) converts a closed-loop network and
its current input delay states xi and layer delay states
ai to open-loop form.
Here a NARX network is designed in open-loop form and then converted to closed-loop form, then converted back.
[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)
Yopen = net(Xs,Xi,Ai)
net = closeloop(net)
view(net)
[Xs,Xi,Ai,Ts] = preparets(net,X,{},T);
Yclosed = net(Xs,Xi,Ai);
net = openloop(net)
view(net)
[Xs,Xi,Ai,Ts] = preparets(net,X,{},T);
Yopen = net(Xs,Xi,Ai)
For examples on using closeloop and openloop to
implement multistep prediction, see narxnet and narnet.