Hello community,
here is the log from the commit of package octave-forge-statistics for
openSUSE:Factory checked in at 2020-06-11 14:48:25
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/octave-forge-statistics (Old)
and /work/SRC/openSUSE:Factory/.octave-forge-statistics.new.3606 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Package is "octave-forge-statistics"
Thu Jun 11 14:48:25 2020 rev:5 rq:801791 version:1.4.2
Changes:
--------
---
/work/SRC/openSUSE:Factory/octave-forge-statistics/octave-forge-statistics.changes
2019-07-08 15:10:53.439337352 +0200
+++
/work/SRC/openSUSE:Factory/.octave-forge-statistics.new.3606/octave-forge-statistics.changes
2020-06-11 14:48:51.170115815 +0200
@@ -1,0 +2,18 @@
+Thu May 7 18:01:05 UTC 2020 - Stefan Brüns <[email protected]>
+
+- Update to version 1.4.2:
+ * canoncorr: allow more variables than observations
+ * fitgmdist: return fitgmdist parameters (Bug #57917)
+ * gamfit: invert parameter per docs (Bug #57849)
+ * geoXXX: update docs 'number of failures (X-1)' =>
+ 'number of failures (X)' (Bug #57606)
+ * kolmogorov_smirnov_test.m: update function handle usage from
+ octave6+ (Bug #57351)
+ * linkage.m: fix octave6+ parse error (Bug #57348)
+ * unifrnd: changed unifrnd(a,a) to return a 0 rather than
+ NaN (Bug #56342)
+ * updates for usage of deprecated octave functions
+- Drop obsolete
+ octave-bug-50365-Wrong-distance-results-from-kmeans-and-wrong-centers.patch
+
+-------------------------------------------------------------------
Old:
----
octave-bug-50365-Wrong-distance-results-from-kmeans-and-wrong-centers.patch
statistics-1.4.1.tar.gz
New:
----
statistics-1.4.2.tar.gz
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Other differences:
------------------
++++++ octave-forge-statistics.spec ++++++
--- /var/tmp/diff_new_pack.2czvjZ/_old 2020-06-11 14:48:52.738120848 +0200
+++ /var/tmp/diff_new_pack.2czvjZ/_new 2020-06-11 14:48:52.742120860 +0200
@@ -1,7 +1,7 @@
#
# spec file for package octave-forge-statistics
#
-# Copyright (c) 2019 SUSE LINUX GmbH, Nuernberg, Germany.
+# Copyright (c) 2020 SUSE LLC
#
# All modifications and additions to the file contributed by third parties
# remain the property of their copyright owners, unless otherwise agreed
@@ -18,15 +18,13 @@
%define octpkg statistics
Name: octave-forge-%{octpkg}
-Version: 1.4.1
+Version: 1.4.2
Release: 0
Summary: Additional statistics functions for Octave
License: GPL-3.0-or-later AND SUSE-Public-Domain
Group: Productivity/Scientific/Math
-Url: http://octave.sourceforge.net
-Source0:
http://downloads.sourceforge.net/octave/%{octpkg}-%{version}.tar.gz
-# PATCH-FIX-UPSTREAM
octave-bug-50365-Wrong-distance-results-from-kmeans-and-wrong-centers.patch
https://savannah.gnu.org/bugs/?func=detailitem&item_id=50365
[email protected] -- fix bug #50365. Update distances and classes to best
replica
-Patch0:
octave-bug-50365-Wrong-distance-results-from-kmeans-and-wrong-centers.patch
+URL: https://octave.sourceforge.io
+Source0:
https://downloads.sourceforge.net/octave/%{octpkg}-%{version}.tar.gz
BuildArch: noarch
BuildRequires: octave-devel
Requires: octave-cli >= 4.0.0
@@ -39,7 +37,6 @@
%prep
%setup -q -c %{name}-%{version}
%octave_pkg_src
-%patch0 -p1
%build
%octave_pkg_build
@@ -57,7 +54,6 @@
%octave --eval "pkg rebuild"
%files
-%defattr(-,root,root)
%{octpackages_dir}/%{octpkg}-%{version}
%changelog
++++++ statistics-1.4.1.tar.gz -> statistics-1.4.2.tar.gz ++++++
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/statistics-1.4.1/DESCRIPTION
new/statistics-1.4.2/DESCRIPTION
--- old/statistics-1.4.1/DESCRIPTION 2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/DESCRIPTION 2020-03-23 18:58:16.000000000 +0100
@@ -1,6 +1,6 @@
Name: statistics
-Version: 1.4.1
-Date: 2019-03-17
+Version: 1.4.2
+Date: 2020-03-23
Author: various authors
Maintainer: Octave-Forge community <[email protected]>
Title: Statistics
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/statistics-1.4.1/Makefile
new/statistics-1.4.2/Makefile
--- old/statistics-1.4.1/Makefile 2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/Makefile 2020-03-23 18:58:16.000000000 +0100
@@ -58,6 +58,13 @@
release_dir_dep := .git/index
endif
+HG := hg
+HG_CMD = $(HG) --config alias.$(1)=$(1) --config defaults.$(1)= $(1)
+HG_ID := $(shell $(call HG_CMD,identify) --id | sed -e 's/+//' )
+HG_TIMESTAMP := $(firstword $(shell $(call HG_CMD,log) --rev $(HG_ID)
--template '{date|hgdate}'))
+
+TAR_REPRODUCIBLE_OPTIONS := --sort=name --mtime="@$(HG_TIMESTAMP)" --owner=0
--group=0 --numeric-owner
+TAR_OPTIONS := --format=ustar $(TAR_REPRODUCIBLE_OPTIONS)
## .PHONY indicates targets that are not filenames
## (https://www.gnu.org/software/make/manual/html_node/Phony-Targets.html)
@@ -96,7 +103,7 @@
## An implicit rule with a recipe to build the tarballs correctly.
%.tar.gz: %
- $(TAR) -c -f - --posix -C "$(target_dir)/" "$(notdir $<)" | gzip -9n >
"$@"
+ $(TAR) -cf - $(TAR_OPTIONS) -C "$(target_dir)/" "$(notdir $<)" | gzip
-9n > "$@"
clean-tarballs:
@echo "## Cleaning release tarballs (package + html)..."
@@ -241,7 +248,7 @@
clean: clean-tarballs clean-unpacked-release clean-install clean-check
@echo "## Removing target directory (if empty)..."
- -rmdir $(target_dir)
+ test -e $(target_dir) && rmdir $(target_dir) || true
@echo
@echo "## Cleaning done"
@echo
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/statistics-1.4.1/NEWS new/statistics-1.4.2/NEWS
--- old/statistics-1.4.1/NEWS 2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/NEWS 2020-03-23 18:58:16.000000000 +0100
@@ -1,3 +1,22 @@
+Summary of important user-visible changes for statistics 1.4.2:
+-------------------------------------------------------------------
+
+ ** canoncorr: allow more variables than observations
+
+ ** fitgmdist: return fitgmdist parameters (Bug #57917)
+
+ ** gamfit: invert parameter per docs (Bug #57849)
+
+ ** geoXXX: update docs 'number of failures (X-1)' => 'number of failures (X)'
(Bug #57606)
+
+ ** kolmogorov_smirnov_test.m: update function handle usage from octave6+ (Bug
#57351)
+
+ ** linkage.m: fix octave6+ parse error (Bug #57348)
+
+ ** unifrnd: changed unifrnd(a,a) to return a 0 rather than NaN (Bug #56342)
+
+ ** updates for usage of depreciated octave functions
+
Summary of important user-visible changes for statistics 1.4.1:
-------------------------------------------------------------------
** update install scripts for octave 5.0 depreciated functions
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/statistics-1.4.1/inst/canoncorr.m
new/statistics-1.4.2/inst/canoncorr.m
--- old/statistics-1.4.1/inst/canoncorr.m 2019-03-17 14:45:17.000000000
+0100
+++ new/statistics-1.4.2/inst/canoncorr.m 2020-03-23 18:58:16.000000000
+0100
@@ -4,7 +4,7 @@
## @deftypefn {Function File} {[@var{A} @var{B} @var{r} @var{U} @var{V}] =}
canoncorr (@var{X}, @var{Y})
## Canonical correlation analysis
##
-## Given @var{X} (size @var{k}*@var{m}) and @var{Y} (@var{k}*@var{n}), returns
projection matrices of canonical coefficients @var{A} (size @var{m}*@var{d},
where @var{d}=@code{min}(@var{m}, @var{n})) and @var{B} (size @var{m}*@var{d});
the canonical correlations @var{r} (1*@var{d}, arranged in decreasing order);
the canonical variables @var{U}, @var{V} (both @var{k}*@var{d}, with
orthonormal columns); and @var{stats}, a structure containing results from
Bartlett's chi-square and Rao's F tests of significance.
+## Given @var{X} (size @var{k}*@var{m}) and @var{Y} (@var{k}*@var{n}), returns
projection matrices of canonical coefficients @var{A} (size @var{m}*@var{d},
where @var{d} is the smallest of @var{m}, @var{n}, @var{d}) and @var{B} (size
@var{m}*@var{d}); the canonical correlations @var{r} (1*@var{d}, arranged in
decreasing order); the canonical variables @var{U}, @var{V} (both
@var{k}*@var{d}, with orthonormal columns); and @var{stats}, a structure
containing results from Bartlett's chi-square and Rao's F tests of significance.
##
## References: @*
## William H. Press (2011), Canonical Correlation Clarified by Singular
Value Decomposition, http://numerical.recipes/whp/notes/CanonCorrBySVD.pdf @*
@@ -13,7 +13,7 @@
## @seealso{princomp}
## @end deftypefn
-# Copyright (C) 2016 by Nir Krakauer <[email protected]>
+# Copyright (C) 2016-2019 by Nir Krakauer <[email protected]>
# 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
@@ -32,7 +32,7 @@
k = size (X, 1); #should also be size (Y, 1)
m = size (X, 2);
n = size (Y, 2);
-d = min (m, n);
+d = min ([k m n]);
X = center (X);
Y = center (Y);
@@ -40,7 +40,7 @@
[Qx Rx] = qr (X, 0);
[Qy Ry] = qr (Y, 0);
-[U S V] = svd (Qx' * Qy, 0);
+[U S V] = svd (Qx' * Qy, "econ");
A = Rx \ U(:, 1:d);
B = Ry \ V(:, 1:d);
@@ -90,4 +90,5 @@
%!assert (V, center(Y) * B, 10*eps);
%!assert (cov(U), eye(size(U, 2)), 10*eps);
%!assert (cov(V), eye(size(V, 2)), 10*eps);
-
+%! rand ("state", 1); [A,B,r] = canoncorr (rand(5, 10),rand(5, 20));
+%!assert (r, ones(1, 5), 10*eps);
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/statistics-1.4.1/inst/fitgmdist.m
new/statistics-1.4.2/inst/fitgmdist.m
--- old/statistics-1.4.1/inst/fitgmdist.m 2019-03-17 14:45:17.000000000
+0100
+++ new/statistics-1.4.2/inst/fitgmdist.m 2020-03-23 18:58:16.000000000
+0100
@@ -499,19 +499,20 @@
endif
params = params*size (Sigma, 3) + 2*rows (mu) - 1;
- extra.NegativeLogLikelihood = -best;
- extra.AIC = -2*(best - params);
- extra.BIC = -2*best + params * log (raw_samples);
- extra.Converged = (incr <= TolFun);
- extra.NumIterations = iter-1;
- extra.RegularizationValue = Regularizer;
-
# This works in Octave, but not in Matlab
#obj = gmdistribution (best_params.mu, best_params.Sigma, best_params.p',
extra);
obj = gmdistribution (best_params.mu, best_params.Sigma, best_params.p');
+
+ obj.NegativeLogLikelihood = -best;
+ obj.AIC = -2*(best - params);
+ obj.BIC = -2*best + params * log (raw_samples);
+ obj.Converged = (incr <= TolFun);
+ obj.NumIterations = iter-1;
+ obj.RegularizationValue = Regularizer;
+
if (Display == 1)
fprintf (" %d iterations log-likelihood = %g\n", ...
- extra.NumIterations, -extra.NegativeLogLikelihood);
+ obj.NumIterations, -obj.NegativeLogLikelihood);
endif
endfunction
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/statistics-1.4.1/inst/gamfit.m
new/statistics-1.4.2/inst/gamfit.m
--- old/statistics-1.4.1/inst/gamfit.m 2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/inst/gamfit.m 2020-03-23 18:58:16.000000000 +0100
@@ -1,11 +1,10 @@
-## Author: Martijn van Oosterhout <[email protected]>
-## This program is granted to the public domain.
+
## -*- texinfo -*-
## @deftypefn {Function File} {@var{MLE} =} gamfit (@var{data})
## Calculate gamma distribution parameters.
##
-## Find the maximum likelihood estimators (@var{mle}s) of the Gamma
distribution
+## Find the maximum likelihood estimate parameters of the Gamma distribution
## of @var{data}. @var{MLE} is a two element vector with shape parameter
## @var{A} and scale @var{B}.
##
@@ -13,16 +12,29 @@
## @end deftypefn
## This function works by minimizing the value of gamlike for the vector R.
-## Just about any minimization function will work, all it has to do a
+## Just about any minimization function will work, all it has to do is
## minimize for one variable. Although the gamma distribution has two
## parameters, their product is the mean of the data. so a helper function
## for the search takes one parameter, calculates the other and then returns
## the value of gamlike.
-## FIXME is this still true???
-## Note: Octave uses the inverse scale parameter, which is the opposite of
-## Matlab. To work for Matlab, value of b needs to be inverted in a few
-## places (marked with **)
+## Author: Martijn van Oosterhout <[email protected]>
+## This program is granted to the public domain.
+# Revisions copyright (C) 2019 by Nir Krakauer <[email protected]>
under GPL (below).
+
+# 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/>.
+
function res = gamfit(R)
@@ -32,19 +44,26 @@
avg = mean(R);
- # This can be just about any search function. I choose this because it
- # seemed to be the only one that might work in this situaition...
- a=nmsmax( @gamfit_search, 1, [], [], avg, R );
+ # Optimize with respect to log(a), since both a and b must be positive
+ x = fminsearch( @(x) gamfit_search(x, avg, R), 0 );
+ a = exp(x);
- b=a/avg; # **
+ b = avg/a;
- res=[a 1/b];
+ res = [a b];
endfunction
-# Helper function so we only have to minimize for one variable. Also to
-# inverting the output of gamlike, incase the optimisation function wants to
-# maximize rather than minimize.
-function res = gamfit_search( a, avg, R )
- b=a/avg; # **
- res = -gamlike([a 1/b], R);
+# Helper function so we only have to minimize for one variable.
+function res = gamfit_search( x, avg, R )
+ a = exp(x);
+ b = avg/a;
+ res = gamlike([a b], R);
endfunction
+
+
+#example data from
https://www.real-statistics.com/distribution-fitting/distribution-fitting-via-maximum-likelihood/fitting-gamma-parameters-mle/
+%!shared v, res
+%! v = [1.2 1.6 1.7 1.8 1.9 2.0 2.2 2.6 3.0 3.5 4.0 4.8 5.6 6.6 7.6];
+%! res = gamfit(v);
+%!assert (res(1), 3.425, 1E-3);
+%!assert (res(2), 0.975, 1E-3);
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/statistics-1.4.1/inst/kmeans.m
new/statistics-1.4.2/inst/kmeans.m
--- old/statistics-1.4.1/inst/kmeans.m 2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/inst/kmeans.m 2020-03-23 18:58:16.000000000 +0100
@@ -143,9 +143,9 @@
replicates_set_explicitly = false;
## Remove rows containing NaN / NA, but record which rows are used
- data_idx = ! any (isnan (data), 2);
+ data_idx = ! any (isnan (data), 2);
original_rows = rows (data);
- data = data(data_idx,:);
+ data = data(data_idx,:);
#used for getting the number of samples
n_rows = rows (data);
@@ -236,11 +236,14 @@
if (isempty (k))
k = rows (start);
elseif (rows (start) != k)
- error ("kmeans: Number of initializers (%d) should match number of
centroids (%d)", rows (start), k);
+ error (["kmeans: Number of initializers (%d) " ...
+ "should match number of centroids (%d)"], rows (start), k);
endif
if (replicates_set_explicitly)
if (replicates != size (start, 3))
- error ("kmeans: The third dimension of the initializer (%d) should
match the number of replicates (%d)", size (start, 3), replicates);
+ error (["kmeans: The third dimension of the initializer (%d) " ...
+ "should match the number of replicates (%d)"], ...
+ size (start, 3), replicates);
endif
else
replicates = size (start, 3);
@@ -251,32 +254,32 @@
## dist returns the distance btwn each row of matrix x and a row vector c
switch (lower (distance))
case "sqeuclidean"
- dist = @(x, c) (sumsq (bsxfun (@minus, x, c), 2));
- centroid = @(x) (mean (x,1));
+ dist = @(x, c) sumsq (bsxfun (@minus, x, c), 2);
+ centroid = @(x) mean (x, 1);
case "cityblock"
- dist = @(x, c) (sum (abs (bsxfun (@minus, x, c)), 2));
- centroid = @(x) (median (x,1));
+ dist = @(x, c) sum (abs (bsxfun (@minus, x, c)), 2);
+ centroid = @(x) median (x, 1);
case "cosine"
## Pre-normalize all data.
## (when Octave implements normr, will use data = normr (data) )
for i = 1:rows (data)
data(i,:) = data(i,:) / sqrt (sumsq (data(i,:)));
endfor
- dist = @(x, c) (1 - (x * c') ./ sqrt (sumsq (c)));
- centroid = @(x) (mean (x,1)); ## already normalized
+ dist = @(x, c) 1 - (x * c') ./ sqrt (sumsq (c));
+ centroid = @(x) mean (x, 1); ## already normalized
case "correlation"
- ## Pre-normalize all data.
+ ## Pre-normalize all data.
data = data - mean (data, 2);
- ## (when Octave implements normr, will use data = normr (data) )
+ ## (when Octave implements normr, will use data = normr (data) )
for i = 1:rows (data)
data(i,:) = data(i,:) / sqrt (sumsq (data(i,:)));
endfor
-
- dist = @(x, c) (1 - (x * (c-mean (c))') ./ sqrt (sumsq (c-mean (c))));
- centroid = @(x) (mean (x,1)); ## already normalized
+ dist = @(x, c) 1 - (x * (c - mean (c))') ...
+ ./ sqrt (sumsq (c - mean (c)));
+ centroid = @(x) mean (x, 1); ## already normalized
case "hamming"
- dist = @(x, c) (sum (bsxfun (@ne, x, c), 2));
- centroid = @(x) (median (x,1));
+ dist = @(x, c) sum (bsxfun (@ne, x, c), 2);
+ centroid = @(x) median (x, 1);
otherwise
error ("kmeans: unsupported distance parameter %s", distance);
endswitch
@@ -285,7 +288,6 @@
########################################
## Now that k has been set (possibly by 'replicates' option), check/use it.
-
if (! isscalar (k))
error ("kmeans: second input argument must be a scalar");
endif
@@ -293,27 +295,27 @@
## used to hold the distances from each sample to each class
D = zeros (n_rows, k);
- best = Inf;
+ best = Inf;
best_centers = [];
for rep = 1:replicates
## check for the 'start' property
switch (lower (start))
case "sample"
- idx = randperm (n_rows, k);
+ idx = randperm (n_rows, k);
centers = data(idx, :);
case "plus" # k-means++, by Arthur and Vassilios(?)
centers(1,:) = data(randi (n_rows),:);
- d = inf (n_rows, 1); # Distance to nearest centroid so far
- for i = 2:k
- d = min (d, dist (data, centers(i-1, :)));
- centers(i,:) = data(find (cumsum (d) > rand * sum (d), 1), :);
- endfor
+ d = inf (n_rows, 1); # Distance to nearest centroid so
far
+ for i = 2:k
+ d = min (d, dist (data, centers(i - 1, :)));
+ centers(i,:) = data(find (cumsum (d) > rand * sum (d), 1), :);
+ endfor
case "cluster"
- idx = randperm (n_rows, max (k, ceil (n_rows/10)));
- [~, centers] = kmeans (data(idx,:), k, "start", "sample",
+ idx = randperm (n_rows, max (k, ceil (n_rows / 10)));
+ [~, centers] = kmeans (data(idx,:), k, "start", "sample", ...
"distance", distance);
case "uniform"
- # vectorised 'min_data + range .* rand'
+ # vectorised 'min_data + range .* rand'
centers = bsxfun (@plus, min_data,
bsxfun (@times, range, rand (k, columns (data))));
otherwise
@@ -323,13 +325,9 @@
## Run the algorithm
iter = 1;
- ## Classify once before the loop; to set sumd, and if max_iter == 0
+ ## Classify once before the loop; to set sumd, and if max_iter == 0
## Compute distances and classify
- for i = 1:k
- D (:, i) = dist (data, centers(i, :));
- endfor
- [~, classes] = min (D, [], 2);
- sumd = obj_cost (D, classes);
+ [D, classes, sumd] = update_dist (data, centers, D, k, dist);
while (err > 0.001 && iter++ <= max_iter)
## Calculate new centroids
@@ -345,18 +343,18 @@
## farthest from any centroid (and not replacing an empty cluster
## from earlier in this pass) and add it to the empty cluster
case 'singleton'
- available = setdiff(1:n_rows, replaced_centroids);
- [~, idx] = max (min (D(available,:)'));
- idx = available(idx);
+ available = setdiff (1:n_rows, replaced_centroids);
+ [~, idx] = max (min (D(available,:)'));
+ idx = available(idx);
replaced_centroids = [replaced_centroids, idx];
- classes(idx) = i;
- membership(idx)=1;
+ classes(idx) = i;
+ membership(idx) = 1;
## if 'drop' then set C and D to NA
case 'drop'
centers(i,:) = NA;
- D(i,:) = NA;
+ D(i,:) = NA;
## if 'error' then throw the error
otherwise
@@ -370,16 +368,9 @@
endif
endfor
- ## Compute distances
- for i = 1:k
- D (:, i) = dist (data, centers(i, :));
- endfor
-
- ## Classify
- [~, classes] = min (D, [], 2);
-
+ ## Compute distances, classes and sums
+ [D, classes, new_sumd] = update_dist (data, centers, D, k, dist);
## calculate the difference in the sum of distances
- new_sumd = obj_cost (D, classes);
err = sum (sumd - new_sumd);
## update the current sum of distances
sumd = new_sumd;
@@ -390,22 +381,30 @@
endif
endfor
centers = best_centers;
- sumd = best';
+ ## Compute final distances, classes and sums
+ [D, classes, sumd] = update_dist (data, centers, D, k, dist);
- final_classes = NA (original_rows,1);
- final_classes(data_idx) = classes; ## other positions already NaN / NA
- classes = final_classes;
-endfunction
+ ## Return with equal size as inputs
+ if (original_rows != rows (data))
+ final = NA (original_rows,1);
+ final(data_idx) = classes; ## other positions already NaN / NA
+ classes = final;
+ endif
-## calculate the sum of within-class distances
-function obj = obj_cost (D, classes)
- obj = zeros (1,columns (D));
- for i = 1:columns (D)
- idx = (classes == i);
- obj(i) = sum (D(idx,i));
- end
endfunction
+## Update distances, classes and sums
+function [D, classes, sumd] = update_dist (data, centers, D, k, dist)
+ for i = 1:k
+ D (:, i) = dist (data, centers(i, :));
+ endfor
+ [~, classes] = min (D, [], 2);
+ ## calculate the sum of within-class distances
+ sumd = zeros (k, 1);
+ for i = 1:k
+ sumd(i) = sum (D(classes == i,i));
+ endfor
+endfunction
## Test input parsing
%!error kmeans (rand (3,2), 4);
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/statistics-1.4.1/inst/linkage.m
new/statistics-1.4.2/inst/linkage.m
--- old/statistics-1.4.1/inst/linkage.m 2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/inst/linkage.m 2020-03-23 18:58:16.000000000 +0100
@@ -107,7 +107,7 @@
"centroid"; "median"; "ward" },
"distfunc", {(@(x) min(x)) # single
(@(x) max(x)) # complete
- (@(x,i,j,w) sum(diag(q=w([i,j]))*x)/sum(q)) # average
+ (@(x,i,j,w) sum(diag(w([i,j]))*x)/sum(w([i,j]))) # average
(@(x) mean(x)) # weighted
(@massdist) # centroid
(@(x,i) massdist(x,i)) # median
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore'
old/statistics-1.4.1/install-conditionally/distributions/geocdf.m
new/statistics-1.4.2/install-conditionally/distributions/geocdf.m
--- old/statistics-1.4.1/install-conditionally/distributions/geocdf.m
2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/install-conditionally/distributions/geocdf.m
2020-03-23 18:58:16.000000000 +0100
@@ -20,8 +20,8 @@
## For each element of @var{x}, compute the cumulative distribution function
## (CDF) at @var{x} of the geometric distribution with parameter @var{p}.
##
-## The geometric distribution models the number of failures (@var{x}-1) of a
-## Bernoulli trial with probability @var{p} before the first success (@var{x}).
+## The geometric distribution models the number of failures (@var{x}) of a
+## Bernoulli trial with probability @var{p} before the first success.
## @end deftypefn
## Author: KH <[email protected]>
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore'
old/statistics-1.4.1/install-conditionally/distributions/geoinv.m
new/statistics-1.4.2/install-conditionally/distributions/geoinv.m
--- old/statistics-1.4.1/install-conditionally/distributions/geoinv.m
2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/install-conditionally/distributions/geoinv.m
2020-03-23 18:58:16.000000000 +0100
@@ -20,8 +20,8 @@
## For each element of @var{x}, compute the quantile (the inverse of the CDF)
## at @var{x} of the geometric distribution with parameter @var{p}.
##
-## The geometric distribution models the number of failures (@var{x}-1) of a
-## Bernoulli trial with probability @var{p} before the first success (@var{x}).
+## The geometric distribution models the number of failures (@var{x}) of a
+## Bernoulli trial with probability @var{p} before the first success.
## @end deftypefn
## Author: KH <[email protected]>
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore'
old/statistics-1.4.1/install-conditionally/distributions/geopdf.m
new/statistics-1.4.2/install-conditionally/distributions/geopdf.m
--- old/statistics-1.4.1/install-conditionally/distributions/geopdf.m
2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/install-conditionally/distributions/geopdf.m
2020-03-23 18:58:16.000000000 +0100
@@ -20,8 +20,8 @@
## For each element of @var{x}, compute the probability density function (PDF)
## at @var{x} of the geometric distribution with parameter @var{p}.
##
-## The geometric distribution models the number of failures (@var{x}-1) of a
-## Bernoulli trial with probability @var{p} before the first success (@var{x}).
+## The geometric distribution models the number of failures (@var{x}) of a
+## Bernoulli trial with probability @var{p} before the first success.
## @end deftypefn
## Author: KH <[email protected]>
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore'
old/statistics-1.4.1/install-conditionally/distributions/geornd.m
new/statistics-1.4.2/install-conditionally/distributions/geornd.m
--- old/statistics-1.4.1/install-conditionally/distributions/geornd.m
2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/install-conditionally/distributions/geornd.m
2020-03-23 18:58:16.000000000 +0100
@@ -32,8 +32,8 @@
## If no size arguments are given then the result matrix is the size of
## @var{p}.
##
-## The geometric distribution models the number of failures (@var{x}-1) of a
-## Bernoulli trial with probability @var{p} before the first success (@var{x}).
+## The geometric distribution models the number of failures (@var{x}) of a
+## Bernoulli trial with probability @var{p} before the first success.
## @end deftypefn
## Author: KH <[email protected]>
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore'
old/statistics-1.4.1/install-conditionally/distributions/unifrnd.m
new/statistics-1.4.2/install-conditionally/distributions/unifrnd.m
--- old/statistics-1.4.1/install-conditionally/distributions/unifrnd.m
2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/install-conditionally/distributions/unifrnd.m
2020-03-23 18:58:16.000000000 +0100
@@ -1,5 +1,6 @@
## Copyright (C) 2012 Rik Wehbring
## Copyright (C) 1995-2016 Kurt Hornik
+## Copyright (C) 2019 Anthony Morast
##
## This program is free software: you can redistribute it and/or
## modify it under the terms of the GNU General Public License as
@@ -81,7 +82,7 @@
endif
if (isscalar (a) && isscalar (b))
- if ((-Inf < a) && (a < b) && (b < Inf))
+ if ((-Inf < a) && (a <= b) && (b < Inf))
rnd = a + (b - a) * rand (sz, cls);
else
rnd = NaN (sz, cls);
@@ -89,7 +90,7 @@
else
rnd = a + (b - a) .* rand (sz, cls);
- k = !(-Inf < a) | !(a < b) | !(b < Inf);
+ k = !(-Inf < a) | !(a <= b) | !(b < Inf);
rnd(k) = NaN;
endif
@@ -127,3 +128,8 @@
%!error unifrnd (ones (2,2), 2, 3)
%!error unifrnd (ones (2,2), 2, [3, 2])
%!error unifrnd (ones (2,2), 2, 2, 3)
+
+%!assert (unifrnd (0,0), 0)
+%!assert (unifrnd (1,1), 1)
+%!assert (unifrnd (1,0), NaN)
+
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore'
old/statistics-1.4.1/install-conditionally/tests/kolmogorov_smirnov_test.m
new/statistics-1.4.2/install-conditionally/tests/kolmogorov_smirnov_test.m
--- old/statistics-1.4.1/install-conditionally/tests/kolmogorov_smirnov_test.m
2019-03-17 14:45:17.000000000 +0100
+++ new/statistics-1.4.2/install-conditionally/tests/kolmogorov_smirnov_test.m
2020-03-23 18:58:16.000000000 +0100
@@ -1,4 +1,4 @@
-## Copyright (C) 1995-2017 Kurt Hornik
+## Copyright (C) 1995-2019 Kurt Hornik
##
## This program is free software: you can redistribute it and/or
## modify it under the terms of the GNU General Public License as
@@ -64,16 +64,15 @@
n = length (x);
s = sort (x);
- try
+
+ if exist (sprintf ("%scdf", dist))
f = str2func (sprintf ("%scdf", dist));
- catch
- try
- f = str2func (sprintf ("%s_cdf", dist));
- catch
- error ("kolmogorov_smirnov_test: no %scdf or %s_cdf function found",
- dist, dist);
- end_try_catch
- end_try_catch
+ elseif exist (sprintf ("%s_cdf", dist))
+ f = str2func (sprintf ("%s_cdf", dist));
+ else
+ error ("kolmogorov_smirnov_test: no %scdf or %s_cdf function found",
+ dist, dist);
+ endif
alt = "!=";
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/statistics-1.4.1/src/configure
new/statistics-1.4.2/src/configure
--- old/statistics-1.4.1/src/configure 2019-03-19 17:32:11.669852774 +0100
+++ new/statistics-1.4.2/src/configure 2020-03-23 18:58:16.000000000 +0100
@@ -1,6 +1,6 @@
#! /bin/sh
# Guess values for system-dependent variables and create Makefiles.
-# Generated by GNU Autoconf 2.69 for statistics 1.4.1.
+# Generated by GNU Autoconf 2.69 for statistics 1.4.2.
#
#
# Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc.
@@ -576,8 +576,8 @@
# Identity of this package.
PACKAGE_NAME='statistics'
PACKAGE_TARNAME='statistics'
-PACKAGE_VERSION='1.4.1'
-PACKAGE_STRING='statistics 1.4.1'
+PACKAGE_VERSION='1.4.2'
+PACKAGE_STRING='statistics 1.4.2'
PACKAGE_BUGREPORT=''
PACKAGE_URL=''
@@ -624,7 +624,6 @@
docdir
oldincludedir
includedir
-runstatedir
localstatedir
sharedstatedir
sysconfdir
@@ -694,7 +693,6 @@
sysconfdir='${prefix}/etc'
sharedstatedir='${prefix}/com'
localstatedir='${prefix}/var'
-runstatedir='${localstatedir}/run'
includedir='${prefix}/include'
oldincludedir='/usr/include'
docdir='${datarootdir}/doc/${PACKAGE_TARNAME}'
@@ -947,15 +945,6 @@
| -silent | --silent | --silen | --sile | --sil)
silent=yes ;;
- -runstatedir | --runstatedir | --runstatedi | --runstated \
- | --runstate | --runstat | --runsta | --runst | --runs \
- | --run | --ru | --r)
- ac_prev=runstatedir ;;
- -runstatedir=* | --runstatedir=* | --runstatedi=* | --runstated=* \
- | --runstate=* | --runstat=* | --runsta=* | --runst=* | --runs=* \
- | --run=* | --ru=* | --r=*)
- runstatedir=$ac_optarg ;;
-
-sbindir | --sbindir | --sbindi | --sbind | --sbin | --sbi | --sb)
ac_prev=sbindir ;;
-sbindir=* | --sbindir=* | --sbindi=* | --sbind=* | --sbin=* \
@@ -1093,7 +1082,7 @@
for ac_var in exec_prefix prefix bindir sbindir libexecdir datarootdir \
datadir sysconfdir sharedstatedir localstatedir includedir \
oldincludedir docdir infodir htmldir dvidir pdfdir psdir \
- libdir localedir mandir runstatedir
+ libdir localedir mandir
do
eval ac_val=\$$ac_var
# Remove trailing slashes.
@@ -1206,7 +1195,7 @@
# Omit some internal or obsolete options to make the list less imposing.
# This message is too long to be a string in the A/UX 3.1 sh.
cat <<_ACEOF
-\`configure' configures statistics 1.4.1 to adapt to many kinds of systems.
+\`configure' configures statistics 1.4.2 to adapt to many kinds of systems.
Usage: $0 [OPTION]... [VAR=VALUE]...
@@ -1246,7 +1235,6 @@
--sysconfdir=DIR read-only single-machine data [PREFIX/etc]
--sharedstatedir=DIR modifiable architecture-independent data [PREFIX/com]
--localstatedir=DIR modifiable single-machine data [PREFIX/var]
- --runstatedir=DIR modifiable per-process data [LOCALSTATEDIR/run]
--libdir=DIR object code libraries [EPREFIX/lib]
--includedir=DIR C header files [PREFIX/include]
--oldincludedir=DIR C header files for non-gcc [/usr/include]
@@ -1273,7 +1261,7 @@
if test -n "$ac_init_help"; then
case $ac_init_help in
- short | recursive ) echo "Configuration of statistics 1.4.1:";;
+ short | recursive ) echo "Configuration of statistics 1.4.2:";;
esac
cat <<\_ACEOF
@@ -1352,7 +1340,7 @@
test -n "$ac_init_help" && exit $ac_status
if $ac_init_version; then
cat <<\_ACEOF
-statistics configure 1.4.1
+statistics configure 1.4.2
generated by GNU Autoconf 2.69
Copyright (C) 2012 Free Software Foundation, Inc.
@@ -1407,7 +1395,7 @@
This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake.
-It was created by statistics $as_me 1.4.1, which was
+It was created by statistics $as_me 1.4.2, which was
generated by GNU Autoconf 2.69. Invocation command line was
$ $0 $@
@@ -3318,7 +3306,7 @@
# report actual input values of CONFIG_FILES etc. instead of their
# values after options handling.
ac_log="
-This file was extended by statistics $as_me 1.4.1, which was
+This file was extended by statistics $as_me 1.4.2, which was
generated by GNU Autoconf 2.69. Invocation command line was
CONFIG_FILES = $CONFIG_FILES
@@ -3371,7 +3359,7 @@
cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //;
s/[\\""\`\$]/\\\\&/g'`"
ac_cs_version="\\
-statistics config.status 1.4.1
+statistics config.status 1.4.2
configured by $0, generated by GNU Autoconf 2.69,
with options \\"\$ac_cs_config\\"
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/statistics-1.4.1/src/configure.ac
new/statistics-1.4.2/src/configure.ac
--- old/statistics-1.4.1/src/configure.ac 2019-03-17 14:45:17.000000000
+0100
+++ new/statistics-1.4.2/src/configure.ac 2020-03-23 18:58:16.000000000
+0100
@@ -1,8 +1,24 @@
# -*- Autoconf -*-
# Process this file with autoconf to produce a configure script.
+### Copyright (C) 2019-2020 John Donoghue <[email protected]>
+### Copyright (C) 2018 Olaf Till <[email protected]>
+###
+### 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/>.
AC_PREREQ([2.69])
-AC_INIT([statistics], [1.4.1])
+AC_INIT([statistics], [1.4.2])
AC_CANONICAL_TARGET
