test11_12.m

模式识别中的十大基本算法

function result=test11_12()

clear all
clc
load 'D:/iris.mat'

n=1000;
acc_ratio=zeros(1,n);
acc=zeros(1,n);
itera=zeros(1,n);

for i=1:1:n
    m=25;
    p1 = randperm(50);
    train_index1 = p1(1,1:m);
    tain_data1 = Iris(train_index1,:);
    predict_data1=Iris(setdiff(p1,train_index1),:);

    p3 = randperm(50)+100;
    tain_index3=p3(1,1:m);
    tain_data3 = Iris(tain_index3,:);
    predict_data3=Iris(setdiff(p3,tain_index3),:);

    initw=[1 linspace(0,0,size(tain_data1,2))]';  %投影向量的初始值
    totallen=size(tain_data1,1)+size(tain_data3,1);
    b=ones(totallen,1)*5;
    bmin=ones(totallen,1)*1;
    
    [w,itera(i)]=HK1(tain_data1,tain_data3,initw,b,0.9,bmin);
      
    acc(i)=cacul_acc(tain_data1,tain_data3,w');
    acc_ratio(i)=cacul_acc(predict_data1,predict_data3,w');
end

format long
result.exp_times=n;
result.train_samples=m;
result.test_samples=50-m;
result.iterations=mean(itera);
result.train_acc=mean(acc);
result.acc_mean=mean(acc_ratio);
result.acc_var=std(acc_ratio,1,2);


function [CoefVector,itera]=HK1(Data1,Data2,initw,b,step,bmin)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%% primal Ho-Kashyap algorithm 
%%% 
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
[RowData1,ColumnData1]=size(Data1);
[RowData2,ColumnData2]=size(Data2);
ExpandData1=cat(2,(linspace(1,1,RowData1))',Data1);
ExpandData2=cat(2,(linspace(1,1,RowData2))',Data2);
TotalData=cat(1,ExpandData1,-ExpandData2);
[Row,Column]=size(TotalData);
 mpp=pinv(TotalData);

CoefVector=initw;
itera=0;Guard=1;k=0;

while (sum(Guard)&(itera<1000))
    k=k+1;itera=itera+1;k=mod(k,(Row+1));if (k==0)k=1;end
    %CoefVector=CoefVector./norm(CoefVector,1);
    e = TotalData*CoefVector - b;
    e1 = (e + abs(e))./2;
    b = b + 2*step.*e1;
    CoefVector = (mpp*b);
    step = step + 0.1;
    if(isempty(find(abs(e)>bmin))) Guard=0; end  
        
end


function [CoefVector,itera]=HK2(Data1,Data2,initw,b,step,bmin)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%% modified Ho-Kashyap algorithm 
%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
[RowData1,ColumnData1]=size(Data1);
[RowData2,ColumnData2]=size(Data2);
ExpandData1=cat(2,(linspace(1,1,RowData1))',Data1);
ExpandData2=cat(2,(linspace(1,1,RowData2))',Data2);
TotalData=cat(1,ExpandData1,-ExpandData2);
[Row,Column]=size(TotalData);
 mpp=pinv(TotalData);

CoefVector=initw;
itera=0;Guard=1;k=0;

CoefVector = (mpp*b);
while (sum(Guard)&(itera<1000))
    k=k+1;itera=itera+1;k=mod(k,(Row+1));if (k==0)k=1;end
    %CoefVector=CoefVector./norm(CoefVector,1);
    e = TotalData*CoefVector - b;
    b=b+step.*(abs(e)+e);
    CoefVector=CoefVector+step.*(mpp*abs(e));
    if(isempty(find(abs(e)>bmin))) Guard=0; end  
end


function accuracy_ratio=cacul_acc(posdata,negdata,w)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%
%%  caculate classifier error
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
[RowData1,ColumnData1]=size(posdata);
[RowData2,ColumnData2]=size(negdata);
ExpandData1=cat(2,(linspace(1,1,RowData1))',posdata);
ExpandData2=cat(2,(linspace(1,1,RowData2))',negdata);
TotalData=cat(1,ExpandData1,-ExpandData2);
[Row,Column]=size(TotalData);

i=1:1:Row;
dotarray=sum(repmat(w,Row,1).*TotalData(i,:),2);
accuracy_num=length(find(dotarray>0));
accuracy_ratio=accuracy_num/length(dotarray);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%