Why I have the same neural network results with different weight and bias initiation each time?

Hello, anyone can help?

I am using neural network for prediction purpose, and the network frame is 1 input layer and 1 output layer, I made the hidden layer vary with different hidden neurons set (Hmax=20). And for each H, I set the Ntrials to 10, and in each trial, the network was initiated.

However, I am getting the same results (same weight and bias) for Ntrials=1:10.

Thanks! -Cathy

Here are some code:

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Hmax=20;

Ntrial=10;

for i=1:Hmax

        for j=1:Ntrials 

            % Create a Fitting Network

            hiddenLayerSize = i; % number of hidden neurons in current NN structure

            net = fitnet(hiddenLayerSize); 

            % Choose Input and Output Pre/Post-Processing Functions
            % For a list of all processing functions type: help nnprocess
            net.inputs{1}.processFcns = {'removeconstantrows'};
            net.outputs{2}.processFcns = {'removeconstantrows'};

            % Setup Division of Data for Training, Validation, Testing
            % For a list of all data division functions type: help nndivide
            net.divideFcn = 'divideind';  % Divide data by index
            net.divideParam.trainInd = trnind;
            net.divideParam.testInd = tstind;

            % set the transfer function (activation function) for input and
            % output layers
            net.layers{1}.transferFcn = 'tansig'; % layer 1 corresponds to the hidden layer
            net.layers{2}.transferFcn = 'purelin'; % layer 2 corresponds to the output layer
            net.layers{1}.initFcn='initwb';
            net.trainFcn = 'trainbr';  % Levenberg-Marquardt optimization with Bayesian regularization

            net.trainParam.goal=0.01.*var(targets); % usually set this to be 1% of the var(target)
            %net.trainParam.goal=0;
            net.trainParam.epochs=1000;
            net.trainParam.mu_dec=0.8;
            net.trainParam.mu_inc=1.5;
            % Choose a Performance Function 
            % For a list of all performance functions type: help nnperformance
            net.performFcn = 'msereg';  % Mean squared error

            % Choose Plot Functions
            % For a list of all plot functions type: help nnplot
            net.plotFcns = {'plotperform','plottrainstate','ploterrhist', ...
              'plotregression', 'plotfit'};

            % Train the Network
            net=configure(net,inputs,targets);
            [net,tr] = train(net,inputs,targets);

            % Test the Network
            outputs = net(inputs);
            errors = gsubtract(targets,outputs);
            performance = perform(net,targets,outputs);

            % Recalculate Training and Test Performance
            trainTargets = targets .* tr.trainMask{1};
            testTargets = targets  .* tr.testMask{1};
            trainPerformance = perform(net,trainTargets,outputs);
            testPerformance = perform(net,testTargets,outputs);

            % Calculate the R2 Training and Test
            H=hiddenLayerSize;
            [I,N] = size(inputs);                 % [8 2519]
            [O,N] = size(targets);                 % [1 2519]
            Ntrn = N*0.70;
            Ntst = N*0.30;
            Ntrneq = Ntrn*1;

            %denormalize the output
            outputs_real=outputs.*std(sal)+mean(sal);

            w1=net.IW{1,1};
            w2=net.LW{2,1};
            b1=net.b{1};
            b2=net.b{2};
        end

end

Answer 1

[Greg Heath]

   net.divideParam.trainInd = trnind;

   net.divideParam.testInd = tstind;

GEH1: TRNIND & TSTIND ARE NEVER DEFINED. EVEN SO, IF THEY WERE DEFINED, THEY WOULD BE THE SAME FOR EVERY LOOP

Hope this helps,

Thank you for formally accepting my answer

Greg