Attempt to reference field of non-structure array error message while running genetic algorithm code

While running my genetic algorithm code i'm getting the following errors

Attempt to reference field of non-structure array.

Error in setwb (line 23)
inputLearn = net.hint.inputLearn;

Error in rmse_test (line 17)
net = setwb(net,x');

Error in @(x)rmse_test(x,{net,inputs,targets})

Error in createAnonymousFcn>@(x)fcn(x,FcnArgs{:}) (line 11)
fcn_handle = @(x) fcn(x,FcnArgs{:});

Error in fcnvectorizer (line 14)
            y(i,:) = feval(fun,(pop(i,:)));

Error in makeState (line 47)
        Score = fcnvectorizer(state.Population(initScoreProvided+1:end,:),FitnessFcn,1,options.SerialUserFcn);

Error in gaunc (line 41)
state = makeState(GenomeLength,FitnessFcn,Iterate,output.problemtype,options);

Error in ga (line 351)
           [x,fval,exitFlag,output,population,scores] = gaunc(FitnessFcn,nvars, ...

Error in genetic_algorithm (line 55)
[x_ga_opt, err_ga] = ga(h, 3, ga_opts);
Caused by:
    Failure in user-supplied fitness function evaluation. GA cannot continue.

Can someone help me how to solve this error and make this code run properly...??

I have attached my code and data here

Answer 1

[Walter Roberson]

You coded

h = @(x) rmse_test(x, {net, inputs, targets});

but rmse_test expects (x, net, inputs, targets) as separate variables.

[Priya Dharshini]

@walter so how should i code that to give as separate variables?

[Walter Roberson]

h = @(x) rmse_test(x, net, inputs, targets);

[Priya Dharshini]

but @walter sir, I even tried using that statement already. I got error like this

Index exceeds matrix dimensions.

Error in setwb (line 32)
    net.IW{i,j}(:) = x(inputWeightInd{i,j});

Error in rmse_test (line 17)
net = setwb(net,x');

Error in @(x)rmse_test(x,net,inputs,targets)

Error in createAnonymousFcn>@(x)fcn(x,FcnArgs{:}) (line 11)
fcn_handle = @(x) fcn(x,FcnArgs{:});

Error in fcnvectorizer (line 14)
            y(i,:) = feval(fun,(pop(i,:)));

Error in makeState (line 47)
        Score = fcnvectorizer(state.Population(initScoreProvided+1:end,:),FitnessFcn,1,options.SerialUserFcn);

Error in gaunc (line 41)
state = makeState(GenomeLength,FitnessFcn,Iterate,output.problemtype,options);

Error in ga (line 351)
           [x,fval,exitFlag,output,population,scores] = gaunc(FitnessFcn,nvars, ...

Error in genetic_algorithm (line 55)
[x_ga_opt, err_ga] = ga(h, 3, ga_opts);
Caused by:
    Failure in user-supplied fitness function evaluation. GA cannot continue.

so how should i give the function handler which will make it run properly...?? And may i know what and where the problem is..??

[Walter Roberson]

You have a 43 x 3 input matrix, so you need 43 x 3 input weights, but your code is trying to get by with only 3. In your ga call you need to change the 3 to numel(inputs)

[Priya Dharshini]

@walter sir, when i change ga call with n=3 to n=numel(inputs) the value of n is taken as 129 <1X1 double> and gives the same error. How should i rectify this error?

[Walter Roberson]

You had to transpose your inputs, and a different number of variables was called for. See attached.

Note: takes a while to run...

[Priya Dharshini]

Thank you sir. It works fine :) Was so helpful

[Walter Roberson]

Please Accept the answer if it works for you.