You can use a predefined output function to be called at each iteration. A template may be in the form below. This function saves all available information to a mat file after each iteration. The interval may be changed by aadding a few lines with an if block included.

function [state, options,optchanged] = myoutputfcn(options,state,flag)

%GAOUTPUTFCNTEMPLATE Template to write custom OutputFcn for GA.

% [STATE, OPTIONS, OPTCHANGED] = GAOUTPUTFCNTEMPLATE(OPTIONS,STATE,FLAG)

% where OPTIONS is an options structure used by GA.

%

% STATE: A structure containing the following information about the state

% of the optimization:

% Population: Population in the current generation

% Score: Scores of the current population

% Generation: Current generation number

% StartTime: Time when GA started

% StopFlag: String containing the reason for stopping

% Selection: Indices of individuals selected for elite,

% crossover and mutation

% Expectation: Expectation for selection of individuals

% Best: Vector containing the best score in each generation

% LastImprovement: Generation at which the last improvement in

% fitness value occurred

% LastImprovementTime: Time at which last improvement occurred

%

% FLAG: Current state in which OutputFcn is called. Possible values are:

% init: initialization state

% iter: iteration state

% interrupt: intermediate state

% done: final state

%

% STATE: Structure containing information about the state of the

% optimization.

%

% OPTCHANGED: Boolean indicating if the options have changed.

%

% See also PATTERNSEARCH, GA, GAOPTIMSET

% Copyright 2004-2006 The MathWorks, Inc.

% $Revision: 1.1.6.5 $ $Date: 2007/08/03 21:23:22 $

optchanged = false;

switch flag

case 'init'

disp('Starting the algorithm');

case {'iter','interrupt'}

disp('Iterating ...')

fname=[pwd,'\',num2str(state.Generation),'.mat'];

save(fname,'state')

case 'done'

disp('Performing final task');

fname=[pwd,'\',num2str(state.Generation),'.mat'];

save(fname,'state')

end

You can use a predefined output function to be called at each iteration. A template may be in the form below. This function saves all available information to a mat file after each iteration. The interval may be changed by aadding a few lines with an if block included.

function [state, options,optchanged] = myoutputfcn(options,state,flag)

%GAOUTPUTFCNTEMPLATE Template to write custom OutputFcn for GA.

% [STATE, OPTIONS, OPTCHANGED] = GAOUTPUTFCNTEMPLATE(OPTIONS,STATE,FLAG)

% where OPTIONS is an options structure used by GA.

%

% STATE: A structure containing the following information about the state

% of the optimization:

% Population: Population in the current generation

% Score: Scores of the current population

% Generation: Current generation number

% StartTime: Time when GA started

% StopFlag: String containing the reason for stopping

% Selection: Indices of individuals selected for elite,

% crossover and mutation

% Expectation: Expectation for selection of individuals

% Best: Vector containing the best score in each generation

% LastImprovement: Generation at which the last improvement in

% fitness value occurred

% LastImprovementTime: Time at which last improvement occurred

%

% FLAG: Current state in which OutputFcn is called. Possible values are:

% init: initialization state

% iter: iteration state

% interrupt: intermediate state

% done: final state

%

% STATE: Structure containing information about the state of the

% optimization.

%

% OPTCHANGED: Boolean indicating if the options have changed.

%

% See also PATTERNSEARCH, GA, GAOPTIMSET

% Copyright 2004-2006 The MathWorks, Inc.

% $Revision: 1.1.6.5 $ $Date: 2007/08/03 21:23:22 $

optchanged = false;

switch flag

case 'init'

disp('Starting the algorithm');

case {'iter','interrupt'}

disp('Iterating ...')

fname=[pwd,'\',num2str(state.Generation),'.mat'];

save(fname,'state')

case 'done'

disp('Performing final task');

fname=[pwd,'\',num2str(state.Generation),'.mat'];

save(fname,'state')

end

Hello,

Though this being an older thread, I thought I shall share my experience in overcoming the same problem. I am not an expert in Matlab nor in Evolutionary Computing, but I have used GA to the vehicle suspension optimization problem once. In order to answer your question the following simple (it would appear non-technical) steps can be followed:

1. Run - ga from file, by using the command: [x,fval,exitflag,output]

= ga(fitnessfcn,nvars,...). This would display: Generation, f-count, Best f(x), Mean f(x) and Stall Generations. But you probably are interested in Best f(x) values for each generation as mostly these are required for convergence analysis and comparison with other algorithms.

2. Next simply select all these displayed values till the last generation row; and copy them.

3. Next open a new excel file, paste the values first by normal paste, then - "Paste Options' a control button would appear, click on the button, then click on - 'Use text import wizard', then press - 'Next', again 'Next' and 'Finish'.

4. You are done.

In this way you have pasted the displayed values in individual cells (columns), now you may use these for your future reference.

Hope this helps though being a non-technical way but has worked effectively for me always.

yes, it is work.

look, go to Irfan Akca  code ( in the above comment), and put three points as follows: 

%----------------------------------------------------------------------------------------------------------

1)  global valueFitt  %----- put this  (in the main your code and in the output function                                                      code)

2)  options = gaoptimset(options,'OutputFcn',@myoutputfcn); %----- put this (in the                                                                                                                    main your code)

3)  case {'iter','interrupt'}

     valueFitt =state.Best; % ----------------------------- put this 

     disp('Iterating ...')

     fname=[pwd,'\',num2str(state.Generation),'.mat'];

    save(fname,'state')

%----------------------------------------------------------------------------------------------------------

All values you will find in the valueFitt 

Thanks

It may not be the exact solution to your problem but.  Please have a look to http://www.mathworks.com/matlabcentral/fileexchange/57933-parsave

Use the following code in a file named outputfunction.m and call this function in options of gaoptimset as 'OutputFcns', @outputfunction.

function [state, options,optchanged] = outputfunction(options,state,flag) optchanged = false; switch flag case 'init' disp('Starting the algorithm'); case {'iter','interrupt'} disp('Iterating ...') fname=[pwd,'\',num2str(state.Generation),'.mat']; save(fname,'state') case 'done' disp('Performing final task'); fname=[pwd,'\',num2str(state.Generation),'.mat']; save(fname,'state') end

For each generation a .mat file will be generated and saved in the current directory with its name as the current iteration number. (Ex.1.mat, 2.mat, etc.)

Dear Manivanna Boopathi,

I tried your method but I cant see the saved file.

Thank you

Yaser Khojah

Dear Yaser,

You can find this file in the current folder. Another way to find it is to look at the upper part of the workspace and select the .mat file there.

Thanks a lot for this thread guys. I found it most helpful.

Cheers

Thanks a lot for this thread. It really add to my previous knowledge of GA using Matlab