Error NARX Model Prediction

Hello everyone...

Im doing a carbon emission(output) with multiple input using neural network approached (NARX). I trained the model using NARX tollbox. I couldn't find the code I needed to write to predict the next 8 years. It doesn't give any predictions in the code I wrote... Would you mind to help me ? Where is the mistake :( My code;

x=xlsread('Input.xlsx');
t=xlsread('Output.xlsx');
X = tonndata(x,false,false);
T = tonndata(t,false,false);
trainFcn = 'trainlm';
inputDelays = 1:2;
feedbackDelays = 1:2;
hiddenLayerSize = 10;
net = narxnet(inputDelays,feedbackDelays,hiddenLayerSize,'open',trainFcn);
[x,xi,ai,t] = preparets(net,X,{},T);
net.divideParam.trainRatio = 80/100;
net.divideParam.valRatio = 10/100;
net.divideParam.testRatio = 10/100;
[net,tr] = train(net,x,t,xi,ai);
y = net(x,xi,ai);
e = gsubtract(t,y);
performance = perform(net,t,y)
view(net)
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)
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)
numFutureSteps = 8;
predictedValues = cell(1, numFutureSteps);
[~, lastInputState, lastLayerState] = preparets(netc, X, {}, T);
lastOutput = yc{end};
for i = 1:numFutureSteps
    [nextOutput, nextInputState, nextLayerState] = netc({lastOutput}, lastInputState, lastLayerState);
    predictedValues{i} = nextOutput;
    lastOutput = nextOutput;
    lastInputState = nextInputState;lastLayerState = nextLayerState;

[Shivansh]

Hello Seda!

The prediction code is included in the shared code snippet. If you encounter any errors while using it, share the error message and some sample data so I can replicate the issue on my side.

[Seda]

@Shivansh

First of all, thank you very much for your return.

I didn't get an error. I'm just not sure if my codes are correct.

[Seda]

Is the code I use to predict the future correct? @Shivansh

Answer 1

[Purvaja]

Hi @Seda,

I understand that you’re using a NARX neural network for predicting carbon emissions with multiple inputs. You trained the model, but when trying to forecast the next 8 years, your code didn’t produce results. The code that you had provided is correct, it just had incomplete loop, so in my dummy code I completed it and printing the results as below.

Here’s the changes I made on my side:

1. Prediction loop:

for k = 1:numFutureSteps
    [nextOutput, lastInputState, lastLayerState] = netc(lastOutput, lastInputState, lastLayerState);
    predictedValues{k} = nextOutput; 
    lastOutput = nextOutput;
end

I have removed some redundant code and variables to make it straightforward like above.

2. Convert predictions into numeric values for display:

predictedValuesNumeric = cellfun(@(c) c, predictedValues);

disp('Predicted carbon emissions for next 8 years:');
disp(predictedValuesNumeric);

The predictions are stored in cells, so you need cellfun to extract the actual numbers before displaying them.

Dummy results I got after implementation:

Training performance: 0.040058
Closed-loop performance: 0.16581
Predicted carbon emissions for next 8 years:
0.0766    3.8152   -0.7831   -0.0690   -0.7864   -0.3953   -0.3675   -0.2086

For better clarification regarding the functions I used, you can go through following links:

1. Preparets: https://www.mathworks.com/help/deeplearning/ref/preparets.html
2. Cellfun: https://www.mathworks.com/help/matlab/ref/cellfun.html

I hope this solves your doubt!