I have trained my time series NARX NN using 5*7875 inputs and 1*7875 target values. How do I make predictions for 5*10 new input values ?

I have trained my time series NARX NN using 5*7875 inputs and 1*7875 target values. How do I make predictions for 5*10 new input values ?

Answer 1

[the cyclist]

You don't describe the code you used in any detail, so it is not possible to be certain, but if you did something like what is described in this documentation page to build your NN, then you can build a closed-loop network with

[netc,Xic,Aic] = closeloop(net,Xf,Af);

and make predictions with that network using

Yc = netc(XPredict,Xic,Aic)

where XPredict stores your new input values.

If that page doesn't help you, I suggest your share your code. Expecting us to guess from a one-sentence description is not the most efficient way to get answers here.

[HABILA]

% Inputs - input time series.

% Target - feedback time series.

X = tonndata(Inputs,true,false);

T = tonndata(Target,true,false);

trainFcn = 'trainlm'; % Levenberg-Marquardt backpropagation.

% Create a Nonlinear Autoregressive Network with External Input

inputDelays = 1:2;

feedbackDelays = 1:2;

hiddenLayerSize = 10;

net = narxnet(inputDelays,feedbackDelays,hiddenLayerSize,'open',trainFcn);

[x,xi,ai,t] = preparets(net,X,{},T);

% Setup Division of Data for Training, Validation, Testing

net.divideParam.trainRatio = 70/100;

net.divideParam.valRatio = 15/100;

net.divideParam.testRatio = 15/100;

% Train the Network

[net,tr] = train(net,x,t,xi,ai);

% Test the Network

y = net(x,xi,ai);

e = gsubtract(t,y);

performance = perform(net,t,y)

% View the Network

view(net)

% Plots

% Closed Loop Network

% Use this network to do multi-step prediction.

% The function CLOSELOOP replaces the feedback input with a direct

% connection from the output layer.

netc = closeloop(net);

netc.name = [net.name ' - Closed Loop'];

view(netc)

[xc,xic,aic,tc] = preparets(netc,X,{},T);

yc = netc(xc,xic,aic);

closedLoopPerformance = perform(net,tc,yc)

% Step-Ahead Prediction Network

nets = removedelay(net);

nets.name = [net.name ' - Predict One Step Ahead'];

view(nets)

[xs,xis,ais,ts] = preparets(nets,X,{},T);

ys = nets(xs,xis,ais);

stepAheadPerformance = perform(nets,ts,ys)

The above is my code where I have trained my NN. Now I want to predict output using new input data

[the cyclist]

According to the documentation (specifically the part I quoted in my answer), it looks like

prediction_for_new_input_data = netc(new_input_data,Xic,Aic)

does what you want.

I have not used this toolbox, so I am just quoting the documenation.