This file is indexed.

/usr/share/octave/packages/statistics-1.3.0/pdist2.m is in octave-statistics 1.3.0-4.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
## Copyright (C) 2014 Piotr Dollar <pdollar@gmail.com>
##
## 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, see
## <http:##www.gnu.org/licenses/>.

## -*- texinfo -*-
## @deftypefn  {Function File} {} pdist2 (@var{x}, @var{y})
## @deftypefnx {Function File} {} pdist2 (@var{x}, @var{y}, @var{metric})
## Compute pairwise distance between two sets of vectors.
##
## Let @var{X} be an MxP matrix representing m points in P-dimensional space
## and @var{Y} be an NxP matrix representing another set of points in the same
## space.  This function computes the M-by-N distance matrix @var{D} where
## @code{@var{D}(i,j)} is the distance between @code{@var{X}(i,:)} and
## @code{@var{Y}(j,:)}.
##
## The optional argument @var{metric} can be used to select different
## distances:
##
## @table @asis
## @item @qcode{"euclidean"} (default)
##
## @item @qcode{"sqeuclidean"}
## Compute the squared euclidean distance, i.e., the euclidean distance
## before computing square root.  This is ideal when the interest is on the
## order of the euclidean distances rather than the actual distance value
## because it performs significantly faster while preserving the order.
##
## @item @qcode{"chisq'"}
## The chi-squared distance between two vectors is defined as:
## @code{d(x, y) = sum ((xi-yi)^2 / (xi+yi)) / 2}.
## The chi-squared distance is useful when comparing histograms.
##
## @item @qcode{"cosine"}
## Distance is defined as the cosine of the angle between two vectors.
##
## @item @qcode{"emd"}
## Earth Mover's Distance (EMD) between positive vectors (histograms).
## Note for 1D, with all histograms having equal weight, there is a simple
## closed form for the calculation of the EMD.  The EMD between histograms
## @var{x} and @var{y} is given by @code{sum (abs (cdf (x) - cdf (y)))},
## where @code{cdf} is the cumulative distribution function (computed
## simply by @code{cumsum}).
##
## @item @qcode{"L1"}
## The L1 distance between two vectors is defined as:  @code{sum (abs (x-y))}
##
## @end table
##
## @seealso{pdist}
## @end deftypefn

## Taken from Piotr's Computer Vision Matlab Toolbox Version 2.52, with
## author permission to distribute under GPLv3

function D = pdist2 (X, Y, metric = "euclidean")

  if (nargin < 2 || nargin > 3)
    print_usage ();
  elseif (columns (X) != columns (Y))
    error ("pdist2: X and Y must have equal number of columns");
  elseif (ndims (X) != 2 || ndims (Y) != 2)
    error ("pdist2: X and Y must be 2 dimensional matrices");
  endif

  switch (tolower (metric))
    case "sqeuclidean",  D = distEucSq (X, Y);
    case "euclidean",     D = sqrt (distEucSq (X, Y));
    case "l1",            D = distL1 (X, Y);
    case "cosine",        D = distCosine (X, Y);
    case "emd",           D = distEmd (X, Y);
    case "chisq",         D = distChiSq (X, Y);
    otherwise
      error ("pdist2: unknown distance METRIC %s", metric);
  endswitch
  D = max (0, D);

endfunction

## TODO we could check the value of p and n first, and choose one
## or the other loop accordingly.
## L1 COMPUTATION WITH LOOP OVER p, FAST FOR SMALL p.
## function D = distL1( X, Y )
## m = size(X,1); n = size(Y,1); p = size(X,2);
## mOnes = ones(1,m); nOnes = ones(1,n); D = zeros(m,n);
## for i=1:p
##   yi = Y(:,i);  yi = yi( :, mOnes );
##   xi = X(:,i);  xi = xi( :, nOnes );
##   D = D + abs( xi-yi' );
## end

function D = distL1 (X, Y)
  m = rows (X);
  n = rows (Y);
  mOnes = ones (1, m);
  D = zeros (m, n);
  for i = 1:n
    yi = Y(i,:);
    yi = yi(mOnes,:);
    D(:,i) = sum (abs (X-yi), 2);
  endfor
endfunction

function D = distCosine (X, Y)
  p = columns (X);
  X = X ./ repmat (sqrt (sumsq (X, 2)), [1 p]);
  Y = Y ./ repmat (sqrt (sumsq (Y, 2)), [1 p]);
  D = 1 - X*Y';
endfunction

function D = distEmd (X, Y)
  Xcdf = cumsum (X,2);
  Ycdf = cumsum (Y,2);
  m = rows (X);
  n = rows (Y);
  mOnes = ones (1, m);
  D = zeros (m, n);
  for i=1:n
    ycdf = Ycdf(i,:);
    ycdfRep = ycdf(mOnes,:);
    D(:,i) = sum (abs (Xcdf - ycdfRep), 2);
  endfor
endfunction

function D = distChiSq (X, Y)
  ## note: supposedly it's possible to implement this without a loop!
  m = rows (X);
  n = rows (Y);
  mOnes = ones (1, m);
  D = zeros (m, n);
  for i = 1:n
    yi = Y(i, :);
    yiRep = yi(mOnes, :);
    s = yiRep + X;
    d = yiRep - X;
    D(:,i) = sum (d.^2 ./ (s+eps), 2);
  endfor
  D = D/2;
endfunction

function dists = distEucSq (x, y)
  xx = sumsq (x, 2);
  yy = sumsq (y, 2)';
  dists = bsxfun (@plus, xx, yy) - 2 * x * (y');
endfunction

## euclidean distance as loop for testing purposes
%!function dist = euclidean_distance (x, y)
%!  [m, p] = size (X);
%!  [n, p] = size (Y);
%!  D = zeros (m, n);
%!  for i = 1:n
%!    d = X - repmat (Y(i,:), [m 1]);
%!    D(:,i) = sumsq (d, 2);
%!  endfor
%!endfunction

%!test
%! x = [1 1 1; 2 2 2; 3 3 3];
%! y = [0 0 0; 1 2 3; 0 2 4; 4 7 1];
%! d = sqrt([  3    5   11   45
%!            12    2    8   30
%!            27    5   11   21]);
%! assert (pdist2 (x, y), d)