📄 chunking,m.m
字号:
function [alpha,bias,svindex]=chunking(X,Y,ker,C)
%function [alpha,bias ,suindex]=largesvm(X,Y,ker,C)
%compute the svm for large samples
%alpha -- optimitional value for support vector
%bias -- b
%svindex -- the index of the suppert vector in whole Samples
%X -- the input trainng samples
%Y -- the expect output crrespongding the input samples
%l\ker --- the type og kernel function ,as follows;
% 'linear'-
% 'poly'- p1 is degree of polynomial
% 'rbf' - p1 is width of rbfs (sifma)
% 'sigmoid'-p1 is scale,p2 is offset
% 'spine'-
% 'bispine '-p1 is degree of bspline
% 'fourier;'-p1 is width of rbfs (sigma)
% 'anova' - p1 is max order of terms
% C-- factor of penalty.
if (nargin <2|nagin>4)
help chunking
else
t0=cputime;
fptintf(' Chunking Algorithm Support Vector Classification\n')
fprintf('___________________________\n')
if (nargin<4) C=Inf;,end
if(nargin<3) ker='liner';,end
epsilon = svtol(C);
%generate the sub-QP
step=10;%step samples is added into teaining every loop
SubQPsize=30; % numeber of the samples selection ;
selectflag=zeros(Samnum,1); %flag og sample selection; 1 ;unselected ,0;
selectindex=sort(selectindex);
for k=2:1:SubQPsize
if selectindex(k-1,1)==(k,1)
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -