perceptron.m

內涵模糊理論與類神經網路的程式碼...提供初學者做研究參考

% ==========================================================
% 
%           Neural Networks A Classroom Approach
%                     Satish Kumar
%             Copyright Tata McGraw Hill, 2004
%
%        MATLAB code for the Perceptron learning law
%                Reference : Table 5.5;Page 129
%
% ==========================================================

% PERCEPTRON LEARNING ALGORITHM
% Pattern vectors stored row-wise in matrix p
p =[1 0 0 
   1 0 1	
   1 1 0 	
   1 1 1];

d =[0 0 0 1];	% Desired outputs stored in vector d
w =[0 0 0];		% Initial weight vector
eta = 1;			% Set the learning rate = 1 
update = 1;		% Set the update flag as TRUE
					% to get into the loop!

while update==1 					% As long as update is TRUE  
   for i=1:4						% For each pattern
      y = p(i,:)*w'; 			% Calculate the activation
      if y >= 0 & d(i)== 0 	% Misclassification!
         w = w - eta*p(i,:);	% Update weights
         up(i) =1;				% Set the local update FLAG
      elseif y<=0 & d(i) ==1	% Another misclassification!
         w = w + eta*p(i,:);	% Update weights
         up(i) = 1;				% Set the update FLAG
      else 
         up(i) = 0;				% Reset the update FLAG
      end
   end
   number_of_updates = up * up';% Check number of updates
   if number_of_updates > 0
      update =1;				% Repeat epoch
   else update =0;				% Reset flag, exit
   end
end