pcasvm.asv

粒子群算法

%function [ output_args ] = Untitled1( input_args )
%UNTITLED1 Summary of this function goes here
%  Detailed explanation goes here
clear;
H1=xlsread('testy.xls',1,'A1:J60');
len=size(H1,1);
H2=H1(:,9);
delta=(max(H2)-min(H2));  %反归一
delta1=min(H2); 
%   [b,a]= butter(5,0.5);
%   H1=filtfilt(b,a,H);

data=H1;
data=(data-ones(size(data,1),1)*min(data))./(ones(size(data,1),1)*(max(data)-min(data)));
H1=data ;


% A1=H1(1:40,1:8);
% B1=H1(1:40,9);
% C1=H1(41:60,1:8);
% D1=H1(41:60,9);
 k=1;
 for L=1:len
     
     switch L
         case 1
             A1=H1(2:len,1:8);
             B1=H1(2:len,9);
             C1=H1(1,1:8);
             D1=H1(1,9);
         case len
             A1=H1(1:len-1,1:8);
             B1=H1(1:len-1,9);
             C1=H1(len,1:8);
             D1=H1(len,9);

         otherwise
             A1=[H1(1:L-1,1:8);H1(L+1:len,1:8)];
             B1=[H1(1:L-1,9);H1(L+1:len,9)];
             C1=H1(L,1:8);
             D1=H1(L,9);


     end


max_ev =7;  %选择主元个数


[COEFF,SCORE,latent,tsquare]=princomp(A1);
A2=A1*COEFF(:,1:max_ev);
C2=C1*COEFF(:,1:max_ev);

model = svmtrain(B1, A2, '-s 3 -t 2 -g 0.11 -c 50 -p 0.015');

[y, accuracy] = svmpredict(D1, C2, model);

       y= y*delta+delta1;      
       x=abs(H2(L)-y)/H2(L);
      
      if k==1
          Y=y;
          X=x;
          D=D1;
          k=0;
      else
          Y=[Y;y];
          X=[X;x];
          D=[D;D1];
      end
      

 end
error1=mean(X)
error2=max(X)
% 
% X=abs(D1-y);
% error1=sqrt(sum(X.^2)/test_num)
% error2=max(abs(D1-y))
% 
%  y= y*(max(H2)-min(H2))+min(H2);
% T=[1:1:20]';
% V=[1,20,-0.5,1.5];
% plot(X,'r--','LineWidth',2 );
plot(H2,'r--','LineWidth',2 );
% axis(V)
hold on
plot(Y,'b','linewidth',2);
% axis(V)