tools for LVQ RBF and GRNN and PNN alike nftool nprtool nctool & ntstool

Similar to tools : nftool nprtool nctool ntstool ,

which simulate the 'functioning or NN execution' and indicate the graphical process and also be able to save the developed code and keep a file for later use;

are there any tools for learning and creating LVQ RBF and GRNN and PNN types of NN

Answer 1

[Jash Kadu]

Hey Yogini!

You can refer to the following documentaion:

LVQ (Learning vector quantization) : https://www.mathworks.com/help/deeplearning/ug/learning-vector-quantization-lvq-neural-networks-1.html

RBF(Radial Basic function network): https://www.mathworks.com/matlabcentral/fileexchange/22173-radial-basis-function-network

GRNN(Generalized regression neural network) : https://www.mathworks.com/help/deeplearning/ug/generalized-regression-neural-networks.html

PNN Classification : https://www.mathworks.com/help/deeplearning/ug/pnn-classification.html

https://www.mathworks.com/help/deeplearning/ref/newpnn.html

[yogini prabhu]

Hi Jash,

Thanks for the links! However, I wanted to know about tools alike nftool for the 4 NNs

As of now, I have preapare a code for NN using a patternnet function: I want to implement for the same for 4 other NNs. Can you oblige to let know the parts that would change if I send the code here?

[Jash Kadu]

Hi Yogini!

Please share the code and any necessary documentations/screenshot so I can help you out with it.

Thanks

[yogini prabhu]

yes

Following is the code (Kindly assume some the inputs and targets matix )

hiddenLayerSize = 10; net = patternnet(hiddenLayerSize);

% Choose the Performance Function

net.trainFcn = 'trainscg';

for i = 1:k

net = configure(net,input,target);

testIdx = (cvFolds == i);

trainIdx = ~testIdx;

trInd = find(trainIdx)

tstInd = find(testIdx)

net.trainParam.epochs = 100;

net.divideParam.trainInd = trInd

net.divideParam.testInd = tstInd

% Performance Function

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

% Train the network

[net, tr] = train(net,input,target);

net.trainParam.showCommandLine=true; % training information will be displayed in the command line

% test using test instances

output = net(input);

errors = gsubtract(target, output);

performance = perform(net,target,output)

% Recalculate Training, Validation and Test Performance

trainTargets = target .* tr.trainMask{1};

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

trainPerformance = perform(net,trainTargets,output)

testPerformance = perform(net,testTargets,output)

test(k)=testPerformance;

% Plot confusion

save net

figure, plotconfusion(target,output)

end

accuracy=mean(test);

accuracy_percentage=accuracy*100;

% View the Network

view(net)