[Octave-bug-tracker] [bug #65178] [octave forge] (statistics) pdist2 BIST failures

[Dmitri A. Sergatskov]

Follow-up Comment #7, bug#65178 (group octave):

WRT pdist2:

octave:5> pkg load statistics
octave:6> test pdist2 verbose
>>>>>
/home/uu1/.local/share/octave/api-v58/packages/statistics-1.6.4/pdist2.m
***** shared x, y, xx
 x = [1, 1, 1; 2, 2, 2; 3, 3, 3];
 y = [0, 0, 0; 1, 2, 3; 0, 2, 4; 4, 7, 1];
 xx = [1 2 3; 4 5 6; 7 8 9; 3 2 1];
***** test
 d = sqrt([3, 5, 11, 45; 12, 2, 8, 30; 27, 5, 11, 21]);
 assert (pdist2 (x, y), d);
***** test
 d = [5.1962, 2.2361, 3.3166, 6.7082; ...
      3.4641, 2.2361, 3.3166, 5.4772];
 i = [3, 1, 1, 1; 2, 3, 3, 2];
 [D, I] = pdist2 (x, y, "euclidean", "largest", 2);
 assert ({D, I}, {d, i}, 1e-4);
***** test
 d = [1.7321, 1.4142, 2.8284, 4.5826; ...
      3.4641, 2.2361, 3.3166, 5.4772];
 i = [1, 2, 2, 3;2, 1, 1, 2];
 [D, I] = pdist2 (x, y, "euclidean", "smallest", 2);
 assert ({D, I}, {d, i}, 1e-4);
***** test
 yy = [1 2 3;5 6 7;9 5 1];
 d = [0, 6.1644, 5.3852; 1.4142, 6.9282, 8.7750; ...
      3.7417, 7.0711, 9.9499; 6.1644, 10.4881, 10.3441];
 i = [2, 4, 4; 3, 2, 2; 1, 3, 3; 4, 1, 1];
 [D, I] = pdist2 (y, yy, "euclidean", "smallest", 4);
 assert ({D, I}, {d, i}, 1e-4);
***** test
 yy = [1 2 3;5 6 7;9 5 1];
 d = [0, 38, 29; 2, 48, 77; 14, 50, 99; 38, 110, 107];
 i = [2, 4, 4; 3, 2, 2; 1, 3, 3; 4, 1, 1];
 [D, I] = pdist2 (y, yy, "squaredeuclidean", "smallest", 4);
 assert ({D, I}, {d, i}, 1e-4);
***** test
 yy = [1 2 3;5 6 7;9 5 1];
 d = [0, 3.3256, 2.7249; 0.7610, 3.3453, 4.4799; ...
      1.8514, 3.3869, 5.0703; 2.5525, 5.0709, 5.1297];
 i = [2, 2, 4; 3, 4, 2; 1, 3, 1; 4, 1, 3];
 [D, I] = pdist2 (y, yy, "seuclidean", "smallest", 4);
 assert ({D, I}, {d, i}, 1e-4);
***** test
 d = [2.1213, 4.2426, 6.3640; 1.2247, 2.4495, 4.4159; ...
      3.2404, 4.8990, 6.8191; 2.7386, 4.2426, 6.1237];
 assert (pdist2 (y, x, "mahalanobis"), d, 1e-4);
***** test
 xx = [1, 3, 4; 3, 5, 4; 8, 7, 6];
 d = [1.3053, 1.8257, 15.0499; 1.3053, 3.3665, 16.5680];
 i = [2, 2, 2; 3, 4, 4];
 [D, I] = pdist2 (y, xx, "mahalanobis", "smallest", 2);
 assert ({D, I}, {d, i}, 1e-4);
***** test
 d = [2.5240, 4.1633, 17.3638; 2.0905, 3.9158, 17.0147];
 i = [1, 1, 3; 4, 3, 1];
 [D, I] = pdist2 (y, xx, "mahalanobis", "largest", 2);
 assert ({D, I}, {d, i}, 1e-4);
***** test
 yy = [1 2 3;5 6 7;9 5 1];
 d = [0, 8.4853, 18.0416; 2.4495, 10.0995, 19.4808; ...
      2.4495, 10.6771, 19.7104; 2.4495, 10.6771, 20.4573];
 i = [2, 2, 2; 1, 4, 4; 4, 1, 1; 3, 3, 3];
 [D, I] = pdist2 (y, yy, "mahalanobis", "smallest", 4);
 assert ({D, I}, {d, i}, 1e-4);
!!!!! test failed
ASSERT errors for:  assert (cond {i},expected {i},tol)

  Location  |  Observed  |  Expected  |  Reason
   (3,1)          3            4         Abs err 1 exceeds tol 0.0001 by 1
   (4,1)          4            3         Abs err 1 exceeds tol 0.0001 by 1
shared variables 
    x =

       1   1   1
       2   2   2
       3   3   3

    y =

       0   0   0
       1   2   3
       0   2   4
       4   7   1

    xx =

       1   3   4
       3   5   4
       8   7   6



I also see the problem with "mad" (it looks like the test uses octave's "mad",
rather than one from statistics):

 ../share/octave/api-v58/packages/statistics-1.6.4/shadow9/mad.m  pass   74/82
 
                                                                   FAIL    8


and also if I do "pkg load statistics" and then "pkg test statistics"
BurrDistribution.m fails:

***** error <paramci: optional arguments must be in NAME-VALUE pairs.> ...
 paramci (BurrDistribution.fit (x), "alpha")
!!!!! error failed.
Expected <paramci: optional arguments must be in NAME-VALUE pairs.>, but got
<burrfit: Weibull distribution fits better in X.>

***** error <paramci: invalid VALUE for 'Alpha' argument.> ...
 paramci (BurrDistribution.fit (x), "alpha", 0)
!!!!! error failed.
Expected <paramci: invalid VALUE for 'Alpha' argument.>, but got <burrfit:
Weibull distribution fits better in X.>

...


Dmitri.
-- 



    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?65178>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/