kmeansKPCA.m

function [A,D,C,K] = kmeansKPCA(X,Lc,Y,nSigma,varargin)


useRFF = false;
numRandFeats = 0;
X = X(:,:);
[nDim,nNum] =size(X);
if nargin > 4 && strcmp(varargin{1},'RFF') && isnumeric(varargin{2})
    useRFF = true;
    numRandFeats = varargin{2};
end

if useRFF
    Kt = cvtKernelGramMatrixRFF(X, numRandFeats, nSigma);
else
    Kt = cvtKernelGramMatrix(X, nSigma);
end
K = diag(sqrt(Lc))*Kt*diag(sqrt(Lc));

if Y <= 1 % �?�—^—¦
    cRate = Y;
    [A B] = eig(K);
    [B ind] = sort(diag(B),'descend');
    A=A(:,ind);
    D = B/nNum;
    nSubDim = find(cumsum(D)/sum(D)>=cRate, 1 );
    A=A(:,1:nSubDim);
    B=B(1:nSubDim);
    A = A/sqrt(diag(B));
    D = D(1:nSubDim);
    C = sum(D);
    
elseif Y > 1 % Žå¬•ª‹ó�?��?ÌŽŸŒ³
    nSubDim = floor(Y);
    %[A B] = eigs(K,nSubDim);
    OPTS.disp = 0;
    size(K);
    [A B] = eigs(K,nSubDim,'lm',OPTS);  %[A B] = eigs(K,nSubDim);
    [B ind] = sort(diag(B),'descend');
    A=A(:,ind);
    D = B/sum(Lc);
    A = A/sqrt(diag(B));
    C = sum(D);
end