% hist2d() - compute histogram information for points in 2-dimensional space
%
% Inputs:
%   data       - [matrix with ndim columns]. Each row is one observation.
%                Each columns contains the data for one dimension.
%
%  nbins       - [vector of length ndim]. Setsthe number of histogram bins 
%                used for the corresponding dimension (data column) in 'data'. 
%                For example [200 100], if the data of the first dimension
%                should be split into 200 bins and the data of the 2nd
%                dimensions into 100 bins.
%
% Outputs:
%   pointCount - counts per bin
%   dimCenters - [cell with ndim entries] bin centers for the corresponding 
%                dimension
%
% Author: ur
% Copyright (C) 2009-2017 Olaf Dimigen & Ulrich Reinacher, HU Berlin
% olaf.dimigen@hu-berlin.de 

% This program 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 program 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 this program; if not, write to the Free Software
% Foundation, 51 Franklin Street, Boston, MA 02110-1301, USA

function [pointCount, dimCenters] = hist2d(data,nbins)

% friendly feedback
if any(isinf(data))
    disp('%s(): Cannot deal with inf yet',mfilename)
    return
end

% remove rows containing NaN
indexOfNaN = find( sum(isnan(data),2) );
data(indexOfNaN,:) = [];
if indexOfNaN
    fprintf('\n%s(): Warning: The following rows contained NaN and were removed:\n%s\n',mfilename,num2str(indexOfNaN'))
end
[nrPoints,nrDims] = size(data);

if ~exist('nbins','var')
    nbins = repmat(10,1,nrDims);
elseif length(nbins) == 1
    nbins = repmat(nbins,1,nrDims);
end

binCoords  = cell(1,nrDims);
dimCenters = cell(1,nrDims);
dimEdges   = cell(1,nrDims);

%% greater or equal
for loop = 1:nrDims
    
    dimEdges{loop}   = linspace(min(data(:,loop)), max(data(:,loop)), nbins(loop)+1);
    dimCenters{loop} = mean([dimEdges{loop}(1:end-1);dimEdges{loop}(2:end)]);
    dimEdges{loop}(end) = [];
    binCoords{loop} = sum( repmat( data(:,loop),1, nbins(loop)) >= repmat(dimEdges{loop},nrPoints,1) ,2);
    if any(~ismember(binCoords{loop},1:nbins(loop)))
        error(sprintf('%s:BinAllocationWrong',mfilename),'Something went wrong during processing of dimension %d',loop)
    end
end

%%
linBinCoords = sub2ind(nbins,binCoords{1},binCoords{2});
pointCount   = hist(linBinCoords,1:prod(nbins));
pointCount   = reshape(pointCount,nbins);
%pointCount = rot90(pointCount);