som_net.m

this file is leverage algorithm written in matlab as m-file and tested in matlab.so anyone can ue th

function net = som_net(X,settings,class);

% som_net calculates the Kohonen map (Self Organizing Map) weights
% and the output weights (Counter Propagation ANN) if class is defined
%
% see the HTML HELP files (help.htm) for details 
% or type "help model_kohonen" or "help model_cpann"
%
% net = som_model(X,settings,class)
%
% input:
%   X           data [n x p]
%   settings    setting structure
%
% optional input, only for CPANNs:
%   class       class vector [n x 1]
% 
% output:
%   net         structure containing all the map information (used settings 
%               and calculated weights [size x size x p]
%
% see the HTML HELP files (help.htm) for details 

nsize  = settings.nsize;
enter  = settings.enter;
nobj   = size(X,1);
nvar   = size(X,2);
display_mod = 10;  % display epochs every display_mod
start_epoch = 1;
end_epoch = settings.epochs;
if nargin == 3
    net_type = 'cpann';
else
    net_type = 'kohonen';
end

% initializes kohonen weights inbetween 0.1 and 0.9
W = rand(nsize, nsize, size(X,2)).*0.8 + 0.1;
W = reshape(permute(W,[2 1 3]),[nsize*nsize nvar]);

% initialization for weight residuals
% Wold = W;

if strcmpi(net_type,'cpann')
    in_out = zeros(size(class,1),max(class));
    for i=1:size(class,1); in_out(i,class(i)) = 1; end;
    W_out = ones(nsize, nsize, size(in_out,2)).*(1/max(class));
    W_out = reshape(permute(W_out,[2 1 3]),[nsize*nsize size(in_out,2)]);
end

disp('training net...')
in_order = 1:nobj;
for e = start_epoch:end_epoch
    if mod(e,display_mod) == 0
        disp(['training epochs: ' num2str(e) ' of ' num2str(end_epoch)])
    end
    if strcmpi(enter,'random')
        in_order = randperm(nobj);
    end
    for i = 1:nobj
        x_in = X(in_order(i),:);
        % calculates winning neuron
        winner = som_winner(x_in,W);
        % applies corrections
        [Wnew,param_correction] = som_update(x_in,W,winner,settings,e,i,nobj);
        W = Wnew;
        
        if strcmpi(net_type,'cpann')
            class_in = in_out(in_order(i),:);
            W_out_new = cpann_update(class_in,W_out,winner,settings,e,i,nobj,param_correction);   
            W_out = W_out_new;
        end
        
    end
    
    % calculation of weight residuals    
    %if mod(e,display_mod) == 0
    %    W_res = (W - Wold).^2;
    %    res(e/display_mod,1) = e;
    %    res(e/display_mod,2) = sum(sum(W_res));
    %    Wold = W;
    %end
        
end
disp('...training finished')
    
% readapts W from reshaping
for k=1:nsize
    for j=1:nsize
        ind = som_which_col(k,j,size(W,1));
        W_final(k,j,:) = W(ind,:);
    end
end

% saves results
net.W = W_final;
if strcmpi(net_type,'cpann')
    for k=1:nsize
        for j=1:nsize
            ind = som_which_col(k,j,size(W_out,1));
            % normilizes output weights (sum up to 1)
            W_out_final(k,j,:) = W_out(ind,:)/sum(W_out(ind,:));
        end
    end
    net.W_out = W_out_final;
end
net.settings = settings;

% showing residual plot
%figure
%hold on
%plot(res(:,1),res(:,2),'k');
%plot(res(:,1),res(:,2),'r*');
%xlabel('epochs');
%ylabel('squared residuals');
%title(['epochs ' num2str(end_epoch)])
%hold off