Documentation of demofun1cmp

Global Index (all files) (short | long) | Local contents | Local Index (files in subdir) (short | long)

Help text

 DEMO for optimizing 'De Jong's function 1 with competing strategies

 This script provides an optimization example using the GEATbx.
 De Jongs Function 1 is optimized using an real valued 
 representation.
 In contrast to demofun1 different stratgies for the subpopulations
 are configured. These Mutation.Range and Mutation.Precision are
 defined in such a way that the change between the strategies during 
 an optimization can be clearly seen (by observing the lower right
 graphic in the visulization window).

 Syntax:  demofun1cmp

 Input parameter:
    no input parameter

 Output parameter:
    no output parameter

 See also: demofun1, tbx3real, tbx3comp, geamain2

Cross-Reference Information

Listing of script demofun1cmp

% Author:   Hartmut Pohlheim
% History:  22.10.2000  file created
%           06.11.2005  comments updated


   % Get default parameters for real variables
   GeaOpt = tbx3real;
   
   % Define special parameters
   GeaOpt = geaoptset( GeaOpt ...
                      , 'NumberSubpopulation',        5 ...        % Number of subpopulation
                      , 'NumberIndividuals',         [15] ...      % 50, 30, 20, 20, 10] ...        % Number of individuals per subpopulation
                      , 'Output.TextInterval',        10 ...       % Text output every 5 generations
                      ...   
                      , 'Output.GrafikInterval',      50 ...       % Grafic results every 50 generations
                      , 'Output.GrafikMethod',        111111 ...   % Grafic method to use
                      , 'Output.GrafikStyle',         614141 ...   % Grafic styles for specified methods
                      ...
                      , 'Mutation.Range',      [1e-2, 1e-4, 1e-6, 1e-8, 1e-1] ...  %  Predefine some mutation ranges if using multiple subpopulation                
                      , 'Mutation.Precision',  [   4     4     4     4    24]         ...   % Use this precision for the subpopulations  
                      ...
                      , 'Termination.Method',         [1  ] ...    % Termination method(s) to use
                      , 'Termination.MaxGen',         800 ...      % Terminate after xx generations
                      , 'Termination.MaxTime',        1 ...        % Terminate after xx minutes
                      ...
                      , 'Competition.DivisionPressure', 1.5 ...
                      , 'Competition.Rate',             .4 ...
                      );

   % Define special parameters for saving results
   FileNameBase = straddtime('f1_cmp_23_');
   GeaOpt = geaoptset( GeaOpt ...
                      , 'Output.SaveTextInterval',     0 ...                    % Text to File every xx generations
                      , 'Output.SaveTextFilename',    [FileNameBase '.txt'] ... % Filename of result file, absolut or relative path may be included
                      , 'Output.SaveBinDataInterval',  0 ...                    % Binary Data to File every xx generations
                      , 'Output.SaveBinDataFilename', [FileNameBase '.mat'] ... % Filename of binary file, absolut or relative path may be included
                      );


   % Add parameters for competition between subpopulations
   GeaOpt = geaoptset( GeaOpt , tbx3comp);
   
   % Define objective function to use
   GeaOpt = geaoptset( GeaOpt , 'System.ObjFunFilename', 'objfun1');
   objfun = []; % 'objfun1';

   % Get variable boundaries from objective function
   VLUB = geaobjpara(GeaOpt.System.ObjFunFilename, [1 10]);
   GeaOpt = geaoptset( GeaOpt , 'System.ObjFunVarBounds', VLUB);
   VLUB = [];

   % Do a mesh plot of objective function
   % plotmesh(objfun, [-100,-100;100,100]);

   % Start optimization
   [xnew, GeaOpt] = geamain2(objfun, GeaOpt, VLUB, []);


% End of script