define the local variable in Function to global

function data= ReadGrasshopperFile( ~, ~,~)

x = randi(10, 1);
y = randi(10, 1);
z = randi(10, 1);
parameters = [x, y, z];

fileName1='M2GHfileName.csv';
fileName2='GH2Matlab.csv';

writematrix(parameters, fileName1);

data = readmatrix(fileName2);
end

This function that I defined it works good.

But, I wanted the x and y and z changed each time that I recall function.

what should I do?

[Walter Roberson]

At the moment you are writing one file but reading a different file.

You could create random data, or you could use random numbers to modify the file you read, or if you have several files to read from you could read one at random.

[bahar vojdani]

@Walter Roberson Thanks for answering but I dont understand what should I do, I have a basic knowledge in matlab.

[Walter Roberson]

At the end of the function before the end statement add

data = data + randi(10, size(data)) ;

[bahar vojdani]

@Walter Roberson

It doest not work as some thing I wanted. I wanted when I run file it gives me a new number and results. because I want to call function in optimization toolbox.

[Torsten]

The optimization toolbox can't cope with random inputs changing with iteration.

Use stochastic optimization tools instead.

[Jan]

"I wanted when I run file it gives me a new number and results." - but this does happen already: randi replies random integers. As long as you do not reset the random number generator manually before, you get new values.

[bahar vojdani]

@Torsten

Yeah, I won't use stochastic optimization tools; But my problem is that when I use this function as an object to be optimized, it doesn't create a random number. In addition, if I don't use the function after each run, it works great like the below code.

x = randi(10, 1);
y = randi(10, 1);
z = randi(10, 1);
parameters = [x, y, z];
fileName1='M2GHfileName.csv';
fileName2='GH2Matlab.csv';
writematrix(parameters, fileName1);
data = readmatrix(fileName2);

[bahar vojdani]

@Walter Roberson

Thank you so much, this is my answer "data = data + randi(10, size(data)) ;". I really appreciate.

[Torsten]

@bahar vojdani

I don't understand what you are trying to do, but I repeat:

The optimization toolbox cannot cope with random inputs that change with iteration.

So you can create random input, but it must remain constant during the optimization process.

[bahar vojdani]

@Torsten

I wanted to creat a function that gives ne random number and save these variable in csv file, after a sleep time read other csv file. I dont want to use optimization toolbox; I use a swarm optimization.