How to use this trained programme to get outputs for new inputs?

I modified the following 'Input-Output Fitting problem'.

inputs = Inputs;

targets = Targets;

hiddenLayerSize = 10;

net = fitnet(hiddenLayerSize);

net.inputs{1}.processFcns = {'removeconstantrows','mapminmax'};

net.outputs{2}.processFcns = {'removeconstantrows','mapminmax'};

net.divideFcn = 'dividerand'; % Divide data randomly

net.divideMode = 'sample'; % Divide up every sample

net.divideParam.trainRatio = 70/100;

net.divideParam.valRatio = 15/100;

net.divideParam.testRatio = 15/100;

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

net.performFcn = 'mse'; % Mean squared error

net.plotFcns ={'plotperform','plottrainstate','ploterrhist','plotregression', 'plotfit'};

[net,tr] = train(net,inputs,targets);

outputs = net(inputs);

errors = gsubtract(targets,outputs);

performance = perform(net,targets,outputs) trainTargets = targets .* tr.trainMask{1};

valTargets = targets .* tr.valMask{1};

testTargets = targets .* tr.testMask{1};

view(net)

Any body can help me on this.... I have to use above programme to get outputs for new inputs. How can i proceed that?

Answer 1

[Jayanti]

Hi Sugan,

After training your neural network, you can easily use the trained model to predict outputs for new data. Since the network has already learned from your original inputs and targets, you just need to pass your new input data to the trained network object.

Assuming "newInputs" is a matrix where each column is a new input sample (with the same number of features as the original training inputs):

newInputs = []; 

% Predict outputs using the trained network
newOutputs = net(newInputs);

There's no need to manually normalize or preprocess "newInputs". The trained network automatically applies the same preprocessing steps that were applied during training.

Or you can also save the model.

save('trainedModel.mat', 'net');

And later reload it and use it.

load('trainedModel.mat');  
predictedOutputs = net(newInputs);

I am also attaching offical MathWorks documentation on "save" for your reference:

https://www.mathworks.com/help/matlab/ref/save.html

Happy coding!