function [X,Y] = div_sample_generation(Z,ds)
%Splits samples in Z (Z(:,t) is the t^th observation) into X and Y so that
%X(:,t)-s are samples from the joint distribution and Y(:,t)-s are samples
%from the product of the ds(m)-dimensional marginals [assumption: sum(ds)=size(Z,1)].
%
%Copyright (C) 2012 Zoltan Szabo ("http://nipg.inf.elte.hu/szzoli", "szzoli (at) cs (dot) elte (dot) hu")
%
%This file is part of the ITE (Information Theoretical Estimators) Matlab/Octave toolbox.
%
%ITE is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by
%the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
%
%This software is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
%MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
%
%You should have received a copy of the GNU General Public License along with ITE. If not, see <http://www.gnu.org/licenses/>.
%verification [sum(ds)=size(Z,1)]:
if sum(ds)~=size(Z,1)
error('sum(ds) should be equal to size(Z,1).');
end
%initialization:
[D,num_of_samples] = size(Z);
cum_ds = cumsum([1;ds(1:end-1)]);%1,d_1+1,d_1+d_2+1,... = starting indices of the subspaces.
num_of_samples = floor(num_of_samples/2);
%X:
X = Z(:,1:num_of_samples);
%Y:
Y = zeros(D,num_of_samples);%preallocation
for m = 1 : length(ds)
idx = [cum_ds(m):cum_ds(m)+ds(m)-1];
Y(idx,:) = Z(idx,num_of_samples+randperm(num_of_samples));
end