--- shogun-0.6.7.orig/debian/shogun-octave-modular.examples +++ shogun-0.6.7/debian/shogun-octave-modular.examples @@ -0,0 +1 @@ +examples/octave-modular/* --- shogun-0.6.7.orig/debian/control +++ shogun-0.6.7/debian/control @@ -0,0 +1,215 @@ +Source: shogun +Section: science +Priority: optional +Maintainer: Ubuntu MOTU Developers +XSBC-Original-Maintainer: Soeren Sonnenburg +Uploaders: Torsten Werner +Build-Depends: libatlas-base-dev [!powerpc !alpha !arm !armel !lpia] | liblapack-dev, + debhelper (>= 5), libreadline5-dev, octave3.0-headers (>= 1:3.0.0-4), python-all-dev, + ghostscript, python-central (>= 0.5), python-numpy (>= 1:1.0.4-5), cdbs, + r-base-core (>= 2.6.1-2), libblas-dev, swig (>= 1.3.36), xutils-dev, doxygen, + graphviz, quilt, texlive-latex-base +XS-Python-Version: all +Standards-Version: 3.8.0 +Homepage: http://www.shogun-toolbox.org +Vcs-Svn: https://bollin.googlecode.com/svn/shogun/trunk/ +Vcs-Browser: http://bollin.googlecode.com/svn/shogun/trunk/ + +Package: shogun-doc +Section: doc +Architecture: all +Recommends: shogun-python-modular +Description: Large Scale Machine Learning Toolbox + SHOGUN - is a new machine learning toolbox with focus on large scale kernel + methods and especially on Support Vector Machines (SVM) with focus to + bioinformatics. It provides a generic SVM object interfacing to several + different SVM implementations. Each of the SVMs can be combined with a variety + of the many kernels implemented. It can deal with weighted linear combination + of a number of sub-kernels, each of which not necessarily working on the same + domain, where an optimal sub-kernel weighting can be learned using Multiple + Kernel Learning. Apart from SVM 2-class classification and regression + problems, a number of linear methods like Linear Discriminant Analysis (LDA), + Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms to + train hidden markov models are implemented. The input feature-objects can be + dense, sparse or strings and of type int/short/double/char and can be + converted into different feature types. Chains of preprocessors (e.g. + substracting the mean) can be attached to each feature object allowing for + on-the-fly pre-processing. + . + SHOGUN comes in different flavours, a stand-a-lone version and also with + interfaces to Matlab(tm), R, Octave, Readline and Python. This is the user + and developer documentation. + +Package: shogun-dbg +Priority: extra +Architecture: any +Description: Large Scale Machine Learning Toolbox + SHOGUN - is a new machine learning toolbox with focus on large scale kernel + methods and especially on Support Vector Machines (SVM) with focus to + bioinformatics. It provides a generic SVM object interfacing to several + different SVM implementations. Each of the SVMs can be combined with a variety + of the many kernels implemented. It can deal with weighted linear combination + of a number of sub-kernels, each of which not necessarily working on the same + domain, where an optimal sub-kernel weighting can be learned using Multiple + Kernel Learning. Apart from SVM 2-class classification and regression + problems, a number of linear methods like Linear Discriminant Analysis (LDA), + Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms to + train hidden markov models are implemented. The input feature-objects can be + dense, sparse or strings and of type int/short/double/char and can be + converted into different feature types. Chains of preprocessors (e.g. + substracting the mean) can be attached to each feature object allowing for + on-the-fly pre-processing. + . + SHOGUN comes in different flavours, a stand-a-lone version and also with + interfaces to Matlab(tm), R, Octave, Readline and Python. This package + contains debug symbols for all interfaces. + +Package: shogun-python-modular +Architecture: any +Depends: ${shlibs:Depends}, ${misc:Depends}, ${python:Depends} +Recommends: python-numpy, python-matplotlib +Provides: ${python:Provides} +XB-Python-Version: ${python:Versions} +Description: Large Scale Machine Learning Toolbox + SHOGUN - is a new machine learning toolbox with focus on large scale kernel + methods and especially on Support Vector Machines (SVM) with focus to + bioinformatics. It provides a generic SVM object interfacing to several + different SVM implementations. Each of the SVMs can be combined with a variety + of the many kernels implemented. It can deal with weighted linear combination + of a number of sub-kernels, each of which not necessarily working on the same + domain, where an optimal sub-kernel weighting can be learned using Multiple + Kernel Learning. Apart from SVM 2-class classification and regression + problems, a number of linear methods like Linear Discriminant Analysis (LDA), + Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms to + train hidden markov models are implemented. The input feature-objects can be + dense, sparse or strings and of type int/short/double/char and can be + converted into different feature types. Chains of preprocessors (e.g. + substracting the mean) can be attached to each feature object allowing for + on-the-fly pre-processing. + . + SHOGUN comes in different flavours, a stand-a-lone version and also with + interfaces to Matlab(tm), R, Octave, Readline and Python. This is the modular + Python package employing swig. + +Package: shogun-python +Architecture: any +Depends: ${shlibs:Depends}, ${misc:Depends}, ${python:Depends} +Recommends: python-numpy, python-matplotlib +Provides: ${python:Provides} +XB-Python-Version: ${python:Versions} +Description: Large Scale Machine Learning Toolbox + SHOGUN - is a new machine learning toolbox with focus on large scale kernel + methods and especially on Support Vector Machines (SVM) with focus to + bioinformatics. It provides a generic SVM object interfacing to several + different SVM implementations. Each of the SVMs can be combined with a variety + of the many kernels implemented. It can deal with weighted linear combination + of a number of sub-kernels, each of which not necessarily working on the same + domain, where an optimal sub-kernel weighting can be learned using Multiple + Kernel Learning. Apart from SVM 2-class classification and regression + problems, a number of linear methods like Linear Discriminant Analysis (LDA), + Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms to + train hidden markov models are implemented. The input feature-objects can be + dense, sparse or strings and of type int/short/double/char and can be + converted into different feature types. Chains of preprocessors (e.g. + substracting the mean) can be attached to each feature object allowing for + on-the-fly pre-processing. + . + SHOGUN comes in different flavours, a stand-a-lone version and also with + interfaces to Matlab(tm), R, Octave, Readline and Python. This is the static + Python package without using swig. + +Package: shogun-readline +Architecture: any +Depends: ${shlibs:Depends}, ${misc:Depends} +Description: Large Scale Machine Learning Toolbox + SHOGUN - is a new machine learning toolbox with focus on large scale kernel + methods and especially on Support Vector Machines (SVM) with focus to + bioinformatics. It provides a generic SVM object interfacing to several + different SVM implementations. Each of the SVMs can be combined with a variety + of the many kernels implemented. It can deal with weighted linear combination + of a number of sub-kernels, each of which not necessarily working on the same + domain, where an optimal sub-kernel weighting can be learned using Multiple + Kernel Learning. Apart from SVM 2-class classification and regression + problems, a number of linear methods like Linear Discriminant Analysis (LDA), + Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms to + train hidden markov models are implemented. The input feature-objects can be + dense, sparse or strings and of type int/short/double/char and can be + converted into different feature types. Chains of preprocessors (e.g. + substracting the mean) can be attached to each feature object allowing for + on-the-fly pre-processing. + . + SHOGUN comes in different flavours, a stand-a-lone version and also with + interfaces to Matlab(tm), R, Octave, Readline and Python. This is the Readline + package. + +Package: shogun-octave-modular +Architecture: any +Depends: ${shlibs:Depends}, ${misc:Depends}, ${octave:Depends} +Description: Large Scale Machine Learning Toolbox + SHOGUN - is a new machine learning toolbox with focus on large scale kernel + methods and especially on Support Vector Machines (SVM) with focus to + bioinformatics. It provides a generic SVM object interfacing to several + different SVM implementations. Each of the SVMs can be combined with a variety + of the many kernels implemented. It can deal with weighted linear combination + of a number of sub-kernels, each of which not necessarily working on the same + domain, where an optimal sub-kernel weighting can be learned using Multiple + Kernel Learning. Apart from SVM 2-class classification and regression + problems, a number of linear methods like Linear Discriminant Analysis (LDA), + Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms to + train hidden markov models are implemented. The input feature-objects can be + dense, sparse or strings and of type int/short/double/char and can be + converted into different feature types. Chains of preprocessors (e.g. + substracting the mean) can be attached to each feature object allowing for + on-the-fly pre-processing. + . + SHOGUN comes in different flavours, a stand-a-lone version and also with + interfaces to Matlab(tm), R, Octave, Readline and Python. This is the modular + Octave package employing swig. + +Package: shogun-octave +Architecture: any +Depends: ${shlibs:Depends}, ${misc:Depends}, ${octave:Depends} +Description: Large Scale Machine Learning Toolbox + SHOGUN - is a new machine learning toolbox with focus on large scale kernel + methods and especially on Support Vector Machines (SVM) with focus to + bioinformatics. It provides a generic SVM object interfacing to several + different SVM implementations. Each of the SVMs can be combined with a variety + of the many kernels implemented. It can deal with weighted linear combination + of a number of sub-kernels, each of which not necessarily working on the same + domain, where an optimal sub-kernel weighting can be learned using Multiple + Kernel Learning. Apart from SVM 2-class classification and regression + problems, a number of linear methods like Linear Discriminant Analysis (LDA), + Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms to + train hidden markov models are implemented. The input feature-objects can be + dense, sparse or strings and of type int/short/double/char and can be + converted into different feature types. Chains of preprocessors (e.g. + substracting the mean) can be attached to each feature object allowing for + on-the-fly pre-processing. + . + SHOGUN comes in different flavours, a stand-a-lone version and also with + interfaces to Matlab(tm), R, Octave, Readline and Python. This is the Octave + package. + +Package: shogun-r +Architecture: any +Depends: ${shlibs:Depends}, ${misc:Depends}, r-base-core +Description: Large Scale Machine Learning Toolbox + SHOGUN - is a new machine learning toolbox with focus on large scale kernel + methods and especially on Support Vector Machines (SVM) with focus to + bioinformatics. It provides a generic SVM object interfacing to several + different SVM implementations. Each of the SVMs can be combined with a variety + of the many kernels implemented. It can deal with weighted linear combination + of a number of sub-kernels, each of which not necessarily working on the same + domain, where an optimal sub-kernel weighting can be learned using Multiple + Kernel Learning. Apart from SVM 2-class classification and regression + problems, a number of linear methods like Linear Discriminant Analysis (LDA), + Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms to + train hidden markov models are implemented. The input feature-objects can be + dense, sparse or strings and of type int/short/double/char and can be + converted into different feature types. Chains of preprocessors (e.g. + substracting the mean) can be attached to each feature object allowing for + on-the-fly pre-processing. + . + SHOGUN comes in different flavours, a stand-a-lone version and also with + interfaces to Matlab(tm), R, Octave, Readline and Python. This is the R + package. --- shogun-0.6.7.orig/debian/shogun-readline.manpages +++ shogun-0.6.7/debian/shogun-readline.manpages @@ -0,0 +1 @@ +debian/shogun.1 --- shogun-0.6.7.orig/debian/orig-tar.sh +++ shogun-0.6.7/debian/orig-tar.sh @@ -0,0 +1,14 @@ +#!/bin/sh + +# called by uscan with '--upstream-version' + +tar xjf $3 +make -C shogun-$2 DEBIAN=yes package-from-release +rm -rf $3 shogun-$2 shogun_$2.orig + +# move to directory 'tarballs' +if [ -r .svn/deb-layout ]; then + . .svn/deb-layout + mv shogun_$2.orig.tar.gz $origDir + echo "moved shogun_$2.orig.tar.gz to $origDir" +fi --- shogun-0.6.7.orig/debian/watch +++ shogun-0.6.7/debian/watch @@ -0,0 +1,3 @@ +version=3 +http://www.shogun-toolbox.org .*/shogun-([\d\.]+).tar.bz2 \ + debian debian/orig-tar.sh --- shogun-0.6.7.orig/debian/shogun-dbg.lintian-overrides +++ shogun-0.6.7/debian/shogun-dbg.lintian-overrides @@ -0,0 +1,2 @@ +# there is no single package shogun-dbg can depend on +dbg-package-missing-depends shogun --- shogun-0.6.7.orig/debian/shogun-octave.examples +++ shogun-0.6.7/debian/shogun-octave.examples @@ -0,0 +1 @@ +examples/matlab_and_octave/* --- shogun-0.6.7.orig/debian/rules +++ shogun-0.6.7/debian/rules @@ -0,0 +1,194 @@ +#!/usr/bin/make -f + +DEB_SHLIBDEPS_INCLUDE := $(shell octave-config -p OCTLIBDIR) /usr/lib/atlas +DEB_COMPRESS_EXCLUDE := .py .html .R .m +R_LIBRARY_DIR := /usr/lib/R/site-library + +include /usr/share/cdbs/1/rules/debhelper.mk +include /usr/share/cdbs/1/rules/patchsys-quilt.mk +include /usr/share/octave/debian/defs.make + +PYVER = $(shell pyversions -vr debian/control) +PYCONFIG = cp -a src build-python$(modular)$(pyver); \ + cd build-python$(modular)$(pyver); \ + ./configure --disable-cpudetection --prefix=/usr \ + --python=python$(pyver) --interface=python$(modular); \ + cd ..; +PYBUILD = $(MAKE) -C build-python$(modular)$(pyver); +PYTEST = $(MAKE) -C build-python$(modular)$(pyver) tests \ + DESTDIR=$(CURDIR)/debian/shogun-python$(modular); +PYINSTALL = $(MAKE) -C build-python$(modular)$(pyver) install \ + DESTDIR=$(CURDIR)/debian/shogun-python$(modular); + +patch: apply-patches + +makebuilddir:: debian/stamp-nonfree +debian/stamp-nonfree: + @if [ -e src/classifier/svm/SVM_light.cpp ]; then \ + echo "Do you really want to build a nonfree version? (y/N): "; \ + read answer; \ + if [ "$$answer" != y ]; then \ + exit 1; \ + fi; \ + fi + touch $@ + +# doc +configure/shogun-doc:: debian/stamp-configure-doc +debian/stamp-configure-doc: + mkdir build-doc + cp -a doc src build-doc + cd build-doc/src && ./configure --disable-cpudetection \ + --prefix=/usr --interface=python-modular + touch $@ + +build/shogun-doc:: debian/stamp-build-doc +debian/stamp-build-doc: + $(MAKE) -C build-doc/doc + touch $@ + +install/shogun-doc:: debian/stamp-install-doc +debian/stamp-install-doc: + mkdir -p $(CURDIR)/debian/shogun-doc/usr/share/doc/shogun-doc/html + cd build-doc/doc/html && cp *.html *.png *.gif *.css $(CURDIR)/debian/shogun-doc/usr/share/doc/shogun-doc/html/ + touch $@ + +# python modular +configure/shogun-python-modular:: debian/stamp-configure-python-modular +debian/stamp-configure-python-modular: modular=-modular +debian/stamp-configure-python-modular: + set -x; $(foreach pyver, $(PYVER), $(PYCONFIG)) + touch $@ + +build/shogun-python-modular:: debian/stamp-build-python-modular +debian/stamp-build-python-modular: modular=-modular +debian/stamp-build-python-modular: + set -x; $(foreach pyver, $(PYVER), $(PYBUILD)) + touch $@ + +install/shogun-python-modular:: debian/stamp-install-python-modular +debian/stamp-install-python-modular: modular=-modular +debian/stamp-install-python-modular: + set -x; $(foreach pyver, $(PYVER), $(PYINSTALL)) + set -x; $(foreach pyver, $(PYVER), $(PYTEST)) + dh_pycentral -pshogun-python-modular + touch $@ + +# python +configure/shogun-python:: debian/stamp-configure-python +debian/stamp-configure-python: modular= +debian/stamp-configure-python: + set -x; $(foreach pyver, $(PYVER), $(PYCONFIG)) + touch $@ + +build/shogun-python:: debian/stamp-build-python +debian/stamp-build-python: modular= +debian/stamp-build-python: + set -x; $(foreach pyver, $(PYVER), $(PYBUILD)) + touch $@ + +install/shogun-python:: debian/stamp-install-python +debian/stamp-install-python: modular= +debian/stamp-install-python: + set -x; $(foreach pyver, $(PYVER), $(PYINSTALL)) + set -x; $(foreach pyver, $(PYVER), $(PYTEST)) + dh_pycentral -pshogun-python + touch $@ + +# readline +configure/shogun-readline:: debian/stamp-configure-readline +debian/stamp-configure-readline: + cp -a src build-readline + cd build-readline && ./configure --disable-cpudetection \ + --prefix=/usr --interface=cmdline + touch $@ + +build/shogun-readline:: debian/stamp-build-readline +debian/stamp-build-readline: + $(MAKE) -C build-readline + touch $@ + +install/shogun-readline:: debian/stamp-install-readline +debian/stamp-install-readline: + $(MAKE) -C build-readline install DESTDIR=$(CURDIR)/debian/shogun-readline + $(MAKE) -C build-readline tests DESTDIR=$(CURDIR)/debian/shogun-readline + touch $@ + +# octave +configure/shogun-octave:: debian/stamp-configure-octave +debian/stamp-configure-octave: + # reintroduce tests; failing autobuilders are fun ... + ls -l /usr/bin/oct* + ls -l /etc/alternatives/oct* + cp -a src build-octave + cd build-octave && ./configure --disable-cpudetection \ + --prefix=/usr --interface=octave + touch $@ + +build/shogun-octave:: debian/stamp-build-octave +debian/stamp-build-octave: + $(MAKE) -C build-octave + touch $@ + +install/shogun-octave:: debian/stamp-install-octave +debian/stamp-install-octave: + dh_install -pshogun-octave build-octave/sg.oct $(OCTDIR) + octave-depends + $(MAKE) -C build-octave tests DESTDIR=$(CURDIR)/debian/shogun-octave + touch $@ + +# octave-modular +configure/shogun-octave-modular:: debian/stamp-configure-octave-modular +debian/stamp-configure-octave-modular: + # reintroduce tests; failing autobuilders are fun ... + ls -l /usr/bin/oct* + ls -l /etc/alternatives/oct* + cp -a src build-octave-modular + cd build-octave-modular && ./configure --disable-cpudetection \ + --prefix=/usr --interface=octave-modular + touch $@ + +build/shogun-octave-modular:: debian/stamp-build-octave-modular +debian/stamp-build-octave-modular: + $(MAKE) -C build-octave-modular + touch $@ + +install/shogun-octave-modular:: debian/stamp-install-octave-modular +debian/stamp-install-octave-modular: + $(MAKE) -C build-octave-modular install \ + DESTDIR=$(CURDIR)/debian/shogun-octave-modular + octave-depends + $(MAKE) -C build-octave-modular tests \ + DESTDIR=$(CURDIR)/debian/shogun-octave-modular + touch $@ + +# r +configure/shogun-r:: debian/stamp-configure-r +debian/stamp-configure-r: + cp -a src build-r + cd build-r && ./configure --disable-cpudetection \ + --prefix=/usr --interface=r + touch $@ + +build/shogun-r:: debian/stamp-build-r +debian/stamp-build-r: + $(MAKE) -C build-r + touch $@ + +install/shogun-r:: debian/stamp-install-r +debian/stamp-install-r: + $(MAKE) -C build-r install \ + DESTDIR=$(CURDIR)/debian/shogun-r \ + RDIR=/usr/lib/R/site-library + $(MAKE) -C build-r tests \ + DESTDIR=$(CURDIR)/debian/shogun-r \ + RDIR=/usr/lib/R/site-library; + touch $@ + +get-orig-source: + uscan --rename --force-download + @echo successfully retrieved upstream tarball + +clean:: + $(RM) -r build-* debian/stamp-* + find debian -mindepth 2 -name 'shogun-*' -delete --- shogun-0.6.7.orig/debian/shogun-python.examples +++ shogun-0.6.7/debian/shogun-python.examples @@ -0,0 +1 @@ +examples/python/* --- shogun-0.6.7.orig/debian/shogun.1 +++ shogun-0.6.7/debian/shogun.1 @@ -0,0 +1,73 @@ +.\" Hey, EMACS: -*- nroff -*- +.\" First parameter, NAME, should be all caps +.\" Second parameter, SECTION, should be 1-8, maybe w/ subsection +.\" other parameters are allowed: see man(7), man(1) +.TH SHOGUN 5 "August 1, 2007" +.\" Please adjust this date whenever revising the manpage. +.\" +.\" Some roff macros, for reference: +.\" .nh disable hyphenation +.\" .hy enable hyphenation +.\" .ad l left justify +.\" .ad b justify to both left and right margins +.\" .nf disable filling +.\" .fi enable filling +.\" .br insert line break +.\" .sp insert n+1 empty lines +.\" for manpage-specific macros, see man(7) +.SH NAME +shogun \- A Large Scale Machine Learning Toolbox +.SH SYNOPSIS +.B shogun +.RI [ options ] +.br +.SH DESCRIPTION +This manual page briefly documents the readline interface of +.B shogun +. +.PP +.\" TeX users may be more comfortable with the \fB\fP and +.\" \fI\fP escape sequences to invode bold face and italics, +.\" respectively. +\fBShogun\fP is a large scale machine learning toolbox with focus on large +scale kernel methods and especially on Support Vector Machines (SVM) with focus +to bioinformatics. It provides a generic SVM object interfacing to several +different SVM implementations. Each of the SVMs can be combined with a variety +of the many kernels implemented. It can deal with weighted linear combination +of a number of sub-kernels, each of which not necessarily working on the same +domain, where an optimal sub-kernel weighting can be learned using Multiple +Kernel Learning. Apart from SVM 2-class classification and regression +problems, a number of linear methods like Linear Discriminant Analysis (LDA), +Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms +to train hidden markov models are implemented. The input feature-objects +can be dense, sparse or strings and of type int/short/double/char and can +be converted into different feature types. Chains of preprocessors (e.g. +substracting the mean) can be attached to each feature object +allowing for on-the-fly pre-processing. + +.SH OPTIONS +A summary of options is included below. +.TP +.B \-h, \-\-help, /? +Show summary of options. +.TP +.B \-i +listen on tcp port 7367 (hex of sg) +.TP +.B filename +execute a script by reading commands from file +.TP +when no options are given the interactive readline interface will be entered +.TP +.SH SEE ALSO +.BR svm-train (1), +.BR svm-predict (1). +.BR svm-scale (1). +.TP +.SH AUTHOR +.br +shogun was written by Soeren Sonnenburg +and Gunnar Raetsch +.PP +This manual page was written by Soeren Sonnenburg , +for the Debian project (but may be used by others). --- shogun-0.6.7.orig/debian/shogun-python-modular.examples +++ shogun-0.6.7/debian/shogun-python-modular.examples @@ -0,0 +1 @@ +examples/python-modular/* --- shogun-0.6.7.orig/debian/changelog +++ shogun-0.6.7/debian/changelog @@ -0,0 +1,333 @@ +shogun (0.6.7-1ubuntu2) jaunty; urgency=low + + * No change rebuild for Python 2.6 transition + + -- Scott Kitterman Sat, 28 Feb 2009 22:40:26 -0500 + +shogun (0.6.7-1ubuntu1) jaunty; urgency=low + + * Merge from Debian unstable, Remaining Ubuntu changes: + - Add !lpia to libatlas-base-dev build depends. + + -- Stefan Ebner Tue, 23 Dec 2008 11:10:42 +0100 + +shogun (0.6.7-1) unstable; urgency=low + + * New upstream version with code cleanups and bugfixes. + * Remove R installation patch (now included upstream). + * Add patch from svn-trunk to work around a swig bug, causing + a compile failure on 64bit archs. + + -- Soeren Sonnenburg Wed, 26 Nov 2008 08:23:55 +0100 + +shogun (0.6.6-2ubuntu1) jaunty; urgency=low + + * Merge from Debian unstable (LP: #297031). This merge also + fixes wish for sync of 0.6.3-1 (LP: #240035) and 0.6.4-3 + (LP: #274312). + * Remaining Ubuntu changes: Add !lpia to libatlas-base-dev + build depends. + + -- Morten Kjeldgaard Wed, 12 Nov 2008 02:20:12 +0100 + +shogun (0.6.6-2) unstable; urgency=low + + [ Torsten Werner ] + * Add a lintian override for shogun-dbg. + + [ Soeren Sonnenburg ] + * Fix R installation script to avoid problem with update.packages. + + -- Soeren Sonnenburg Thu, 16 Oct 2008 20:27:02 +0200 + +shogun (0.6.6-1) unstable; urgency=low + + [ Soeren Sonnenburg ] + * New upstream version with several bugfixes and improved documentation. + + [ Torsten Werner ] + * Remove all patches because they have been applied upstream. + + -- Torsten Werner Sun, 12 Oct 2008 01:19:35 +0200 + +shogun (0.6.4-3) unstable; urgency=low + + * Add build dependency on swig 1.3.36 (Closes: #494335). + * Update my email address to sonne@debian.org. + * Update copyright. + * Version dependency on debhelper (>=5). + + -- Soeren Sonnenburg Tue, 23 Sep 2008 20:03:32 +0200 + +shogun (0.6.4-2) unstable; urgency=low + + * Add patch from svn to fix data-loading in cmdline interface (transpose). + + -- Soeren Sonnenburg Sun, 17 Aug 2008 21:33:18 +0200 + +shogun (0.6.4-1) unstable; urgency=low + + * New upstream version with much improved documentation. + * Remove patches as they have been applied upstream. + * Add patch from svn to fix running octave test-suite. + * Add patch from svn to fix build failure w/o lapack/atlas. + + -- Soeren Sonnenburg Sat, 16 Aug 2008 11:38:21 +0200 + +shogun (0.6.3-1) unstable; urgency=low + + * New upstream bugfix version. + * Adapt for new directory layout. + * Run tests for all interfaces on make install. + * Add doc-base file. + * Add patch target and bump Standards Version to 3.8.0. + * Add fix from trunk to fix return values in R interface. + + -- Soeren Sonnenburg Wed, 18 Jun 2008 12:04:16 +0200 + +shogun (0.6.2-1) unstable; urgency=low + + * New upstream version with much improved static cmdline,r,python,octave + interfaces and several bugfixes. + + -- Soeren Sonnenburg Fri, 16 May 2008 00:17:04 +0200 + +shogun (0.6.1-2) unstable; urgency=low + + * Add README.source documenting we are using quilt for additional patches. + * Call dh_pycentral on install not build for shogun-python (fixes FTBS on + mips and mipsel with debhelper 7.0.2). + + -- Soeren Sonnenburg Tue, 29 Apr 2008 08:57:18 +0200 + +shogun (0.6.1-1) unstable; urgency=low + + * New upstream version with new octave-modular interface and python-modular + speedups + * Build depend on swig >= 1.3.35-3 as it contains octave fixes. + * Add examples for octave-modular. + * Add ghostscript to build-depends which is used to generate .png's from the + generated .ps formulas. + * Use correct DESTDIR when running tests. + * Disable atlas also for arm and armel and fall back to lapack and blas. + + -- Soeren Sonnenburg Sun, 20 Apr 2008 14:25:25 +0200 + +shogun (0.6.0-2) unstable; urgency=low + + [ Torsten Werner ] + * Fix output of debian/orig-tar.sh. + + [ Soeren Sonnenburg ] + * Run the python tests after make install. + * Include texlive-latex-base to build depends as doxygen requires it to + generate images from latex formulas (and does not pull latex in either). + * Disable atlas for powerpc and alpha as they FTBFS on these archs. + + -- Soeren Sonnenburg Tue, 15 Apr 2008 10:12:08 +0200 + +shogun (0.6.0-1) unstable; urgency=low + + * New upstream version with major feature enhancements, providing same + functionality for all static interfaces and improved user documentation + * Pull in atlas dependency whenever available, otherwise fall back to + blas and lapack. + * Drop sqrtl patch (included upstream). + * Don't compress R, octave or python examples + + -- Soeren Sonnenburg Sun, 13 Apr 2008 11:38:49 +0200 + +shogun (0.5.1-2) unstable; urgency=high + + [ Soeren Sonnenburg ] + * Add patch from trunk to fix FTBS on mips,mipsel,etc caused by sqrtl not + being available for these archs + * Add quilt to Build-Depends again + * Temporarily remove atlas from Build-Depends for powerpc and alpha and add + liblapack-dev to the Build-Depends for these archs + + [ Torsten Werner ] + * Increase urgency because we are fixing FTBFS bugs only. + + -- Torsten Werner Thu, 20 Mar 2008 13:00:53 +0100 + +shogun (0.5.1-1) unstable; urgency=low + + [ Soeren Sonnenburg ] + * New upstream version + * Get watch file to work + * Update build depends for gfortran transition (Closes: #466335) + + [ Torsten Werner ] + * Remove Build-Depends: quilt. + + -- Torsten Werner Sat, 23 Feb 2008 21:17:20 +0100 + +shogun (0.5.0-1) unstable; urgency=low + + * New upstream version with major feature enhancements. + * Add !armel ann ameb to atlas3-base-dev build depends (Closes: #460096) + * Remove python-numpy-{ext,dev} from build depends and use python-numpy + instead (Closes: #463211) + * Project switched license to GPLv3, update debian/copyright accordingly + * Include missing copyright holders in debian/copyright (Closes: #461616) + * Remove depends on xutils-dev as now gcc is used to generate build + dependencies + * Generate shogun-doc package; add doxygen and graphviz to build depends + * Remove patch to fix gcc4.3 compile failure (included upstream) + * Enable building of debug package shogun-dbg (Closes: #461614) + + -- Soeren Sonnenburg Fri, 08 Feb 2008 23:43:33 +0100 + +shogun (0.4.4-2ubuntu2) intrepid; urgency=low + + * gfortran transition on build dependecies: + - atlas3-base-dev -> libatlas-base-dev + - refblas3-dev -> libblas-dev + + -- Morten Kjeldgaard Tue, 09 Sep 2008 13:25:17 +0200 + +shogun (0.4.4-2ubuntu1) hardy; urgency=low + + * debian/control: + - Do not build-depend on python-numpy-{dev,ext}, build-depend + on python-numpy instead. + - Update Maintainer field as per spec. + + -- Luca Falavigna Wed, 06 Feb 2008 20:10:33 +0000 + +shogun (0.4.4-2) unstable; urgency=low + + [ Soeren Sonnenburg ] + * Depend on octave 3.0, thanks Thomas Weber (Closes: #457979) + * Fix FTBS for gcc-4.3 (Closes: #458192) + + [ Torsten Werner ] + * Change Depends: liblapack3-dev [arm]. + * Support newer versions of dpkg-shlibdeps. + * Change XS-Vcs to Vcs in debian/control. + + -- Torsten Werner Wed, 02 Jan 2008 20:31:20 +0100 + +shogun (0.4.4-1) unstable; urgency=low + + * New upstream version contains major bug fixes. + + -- Soeren Sonnenburg Fri, 23 Nov 2007 13:05:05 +0200 + +shogun (0.4.3-1) unstable; urgency=low + + * New upstream allows to disable custom cpu settings which allows + for a workaround the gcc-4.2 build failure on alpha. + + -- Soeren Sonnenburg Fri, 28 Sep 2007 09:53:35 +0200 + +shogun (0.4.2-1) unstable; urgency=low + + [Soeren Sonnenburg] + * New upstream version (Closes: #442990) + * Build R package in a separate directory (Closes: #442730) + + [Torsten Werner] + * minor fixes in debian/rules + + -- Soeren Sonnenburg Wed, 19 Sep 2007 09:08:42 +0200 + +shogun (0.4.1-1) unstable; urgency=low + + * new upstream version + + -- Soeren Sonnenburg Sat, 01 Sep 2007 09:56:50 +0200 + +shogun (0.4.0-3) unstable; urgency=low + + * add manpage for shogun-readline interface + + -- Soeren Sonnenburg Sun, 05 Aug 2007 13:49:06 +0200 + +shogun (0.4.0-2) unstable; urgency=low + + * remove build dependency on octave2.1-forge, but rather make it a Recommends + + -- Soeren Sonnenburg Thu, 02 Aug 2007 08:25:20 +0200 + +shogun (0.4.0-1) unstable; urgency=low + + * new upstream version + * also add include matlab examples in the octave examples + + -- Soeren Sonnenburg Tue, 31 Jul 2007 17:33:47 +0200 + +shogun (0.3.2-3) unstable; urgency=low + + * Change Build-Depends: r-base-core from r-base-dev. + + -- Torsten Werner Sun, 3 Jun 2007 10:50:29 +0200 + +shogun (0.3.2-2) unstable; urgency=low + + * support for python 2.5 + * bump compat to 5, pycompat to 2 + + -- Soeren Sonnenburg Wed, 09 May 2007 19:49:26 +0200 + +shogun (0.3.2-1) unstable; urgency=low + + * new upstream version + + -- Torsten Werner Wed, 7 Mar 2007 19:05:03 +0100 + +shogun (0.3.1-1) unstable; urgency=low + + * new upstream release, fixes a build problem + + -- Torsten Werner Tue, 20 Feb 2007 21:01:14 +0100 + +shogun (0.3.0-1) unstable; urgency=low + + * new upstream release + * Remove Build-Depends: atlas3-base-dev for arm architecture. + * Updated debian/python*.examples to reflect upstream changes. + + -- Torsten Werner Wed, 14 Feb 2007 20:49:23 +0100 + +shogun (0.2.1+svn1952-1) unstable; urgency=low + + * new upstream version that builds on more platforms + + -- Torsten Werner Sun, 11 Feb 2007 16:41:54 +0100 + +shogun (0.2.1+svn1923-1) unstable; urgency=low + + * new upstream version + * new upstream URL + * updated to new build system, renaming the packages to + shogun-python-modular and shogun-python + + -- Soeren Sonnenburg Wed, 24 Jan 2007 10:38:50 +0100 + +shogun (0.2.1+svn1880-2) unstable; urgency=low + + * NOT RELEASED YET + * Add more python examples. + + -- Torsten Werner Wed, 10 Jan 2007 22:02:45 +0100 + +shogun (0.2.1+svn1880-1) unstable; urgency=low + + [ Torsten Werner ] + * new upstream version + * Fixed debian/copyright as requested by ftp-master Jörg Jaspert. + + [ Sören Sonnenburg ] + * Removed non-free code completely. + * Switch to svn revisions instead of dates. + + -- Torsten Werner Wed, 10 Jan 2007 20:24:13 +0100 + +shogun (0.2.1-1) unstable; urgency=low + + [ Torsten Werner ] + * Initial Release, closes: #388148. + + -- Torsten Werner Thu, 30 Nov 2006 20:20:39 +0100 --- shogun-0.6.7.orig/debian/README.source +++ shogun-0.6.7/debian/README.source @@ -0,0 +1,47 @@ +This package uses quilt to manage all modifications to the upstream +source. Changes are stored in the source package as diffs in +debian/patches and applied during the build. + +To get the fully patched source after unpacking the source package, cd to +the root level of the source package and run: + + quilt push -a + +The last patch listed in debian/patches/series will become the current +patch. + +To add a new set of changes, first run quilt push -a, and then run: + + quilt new + +where is a descriptive name for the patch, used as the filename in +debian/patches. Then, for every file that will be modified by this patch, +run: + + quilt add + +before editing those files. You must tell quilt with quilt add what files +will be part of the patch before making changes or quilt will not work +properly. After editing the files, run: + + quilt refresh + +to save the results as a patch. + +Alternately, if you already have an external patch and you just want to +add it to the build system, run quilt push -a and then: + + quilt import -P /path/to/patch + quilt push -a + +(add -p 0 to quilt import if needed). as above is the filename to +use in debian/patches. The last quilt push -a will apply the patch to +make sure it works properly. + +To remove an existing patch from the list of patches that will be applied, +run: + + quilt delete + +You may need to run quilt pop -a to unapply patches first before running +this command. --- shogun-0.6.7.orig/debian/shogun-readline.examples +++ shogun-0.6.7/debian/shogun-readline.examples @@ -0,0 +1 @@ +examples/cmdline/* --- shogun-0.6.7.orig/debian/README.Debian +++ shogun-0.6.7/debian/README.Debian @@ -0,0 +1,11 @@ +This package was initially debianized by Torsten Werner +and is now maintained by Soeren Sonnenburg . + +The nonfree SVMlight files have been removed from the upstream archive and the +package is configured with --disable-svmlight. Users seeking for SVMLight may +resort to using the included GPDT-SVM as a drop in replacement. It uses a very +similar algorithm but in many practical applications was shown to be a lot +faster. + +User documentation is availabe if the shogun-doc package is installed. Just +point your browser to /usr/share/doc/shogun-doc/html/index.html . --- shogun-0.6.7.orig/debian/compat +++ shogun-0.6.7/debian/compat @@ -0,0 +1 @@ +5 --- shogun-0.6.7.orig/debian/shogun-doc.doc-base +++ shogun-0.6.7/debian/shogun-doc.doc-base @@ -0,0 +1,23 @@ +Document: shogun +Title: Software for Semidefinite Programming +Author: Soeren Sonnenburg, Gunnar Raetsch +Abstract: SHOGUN - is a new machine learning toolbox with focus on large scale + kernel methods and especially on Support Vector Machines (SVM) applied to the + field of bioinformatics. It provides a generic SVM object interfacing to + several different SVM implementations. Each of the SVMs can be combined with a + variety of the many kernels implemented. It can deal with weighted linear + combination of a number of sub-kernels, each of which not necessarily working + on the same domain, where an optimal sub-kernel weighting can be learned using + Multiple Kernel Learning. Apart from SVM 2-class classification and regression + problems, a number of linear methods like Linear Discriminant Analysis (LDA), + Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms to + train hidden markov models are implemented. The input feature-objects can be + dense, sparse or strings and of type int/short/double/char and can be converted + into different feature types. Chains of preprocessors (e.g. substracting the + mean) can be attached to each feature object allowing for on-the-fly + pre-processing. +Section: Science/Data Analysis + +Format: HTML +Index: /usr/share/doc/shogun-doc/html/index.html +Files: /usr/share/doc/shogun-doc/html/*.html --- shogun-0.6.7.orig/debian/pycompat +++ shogun-0.6.7/debian/pycompat @@ -0,0 +1 @@ +2 --- shogun-0.6.7.orig/debian/shogun-r.examples +++ shogun-0.6.7/debian/shogun-r.examples @@ -0,0 +1 @@ +examples/r/* --- shogun-0.6.7.orig/debian/copyright +++ shogun-0.6.7/debian/copyright @@ -0,0 +1,210 @@ +This program was downloaded from http://www.shogun-toolbox.org . + +The main upstream Authors are + Soeren Sonnenburg + Gunnar Raetsch . + +Copyright: (c) 1999-2008 Fraunhofer Institute FIRST and Max-Planck-Society + +License: + + 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. + + On Debian systems, you can find the GPL license in: + /usr/share/common-licenses/GPL + + +-------- + +The files ./classifier/svm/SVM_libsvm.cpp and ./classifier/svm/SVM_libsvm.h are + +Copyright (c) 2000-2008 Chih-Chung Chang and Chih-Jen Lin + +License: + + Copyright (c) 2000-2008 Chih-Chung Chang and Chih-Jen Lin + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + 3. Neither name of copyright holders nor the names of its contributors + may be used to endorse or promote products derived from this software + without specific prior written permission. + + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR + CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, + EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, + PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR + PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +Quote from the LibSVM homepage: +http://www.csie.ntu.edu.tw/~cjlin/libsvm/faq.html#f204 + + The libsvm license ("the modified BSD license") is compatible with many free + software licenses such as GPL. Hence, it is very easy to use libsvm in your + software. It can also be used in commercial products. + +-------- + +The files classifier/svm/SVM_linear.h, classifier/svm/SVM_linear.cpp, +classifier/svm/Tron.cpp and classifier/svm/Tron.h are + +Copyright (c) 2007-2008 Machine Learning Group at National Taiwan +University Contributors + +LIBLINEAR homepage http://www.csie.ntu.edu.tw/~cjlin/liblinear/ + + Copyright (c) 2007-2008 The LIBLINEAR Project. + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + 3. Neither name of copyright holders nor the names of its contributors + may be used to endorse or promote products derived from this software + without specific prior written permission. + + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR + CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, + EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, + PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR + PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +-------- + +The files in the directory classifier/svm named gpdt.cpp, gpdt.h, gpdtsolve.cpp, +gpdtsolve.h, gpm.cpp, gpm.h are + +Copyright (C) 2004-2008 by T. Serafini, G. Zanghirati, L. Zanni. + + COPYRIGHT NOTIFICATION + + Permission to copy and modify this software and its documentation + for internal research use is granted, provided that this notice is + retained thereon and on all copies or modifications. The authors and + their respective Universities makes no representations as to the + suitability and operability of this software for any purpose. It is + provided "as is" without express or implied warranty. + Use of this software for commercial purposes is expressly prohibited + without contacting the authors. + + 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 2 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, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + +-------- + +The files classifier/svm/ssl.cpp classifier/svm/ssl.h are + +Copyright 2006 Vikas Sindhwani (vikass@cs.uchicago.edu) + + SVM-lin: Fast SVM Solvers for Supervised and Semi-supervised Learning + + SVM-lin 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 2 of the License, or + (at your option) any later version. + + SVM-lin 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 SVM-lin (see gpl.txt); if not, write to the Free Software + Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + +-------- + +The files in the directory classifier/svm named GNPPSVM.h, GNPPSVM.cpp, +GMNPSVM.h and GMNPSVM.cpp are + +Copyright (C) 1999-2008 Center for Machine Perception, CTU FEL Prague + + 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. + +-------- + +The files in the directory classifier/svm named SVMSGD.h and SVMSGD.cpp + +Copyright (C) 2007 Leon Bottou + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU Lesser General Public +License as published by the Free Software Foundation; either +version 2.1 of the License, or (at your option) any later version. + +-------- + +The files kernel/LocalAlignmentStringKernel.h and +kernel/LocalAlignmentStringKernel.cpp are + +Copyright 2005 Jean-Philippe Vert, Hiroto Saigo + +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. + +-------- + +The files kernel/OligoKernel.h and kernel/OligoKernel.cpp are + +Copyright 2008 Christian Igel, Tobias Glasmachers + +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. --- shogun-0.6.7.orig/debian/patches/wordsize.patch +++ shogun-0.6.7/debian/patches/wordsize.patch @@ -0,0 +1,28 @@ +Index: shogun-0.6.7/src/configure +=================================================================== +--- shogun-0.6.7.orig/src/configure 2008-11-26 08:21:27.000000000 +0100 ++++ shogun-0.6.7/src/configure 2008-11-26 08:22:14.000000000 +0100 +@@ -1919,6 +1919,23 @@ + + set_opts() + { ++ echocheck "Word Size" ++cat > $TMPC << EOF ++#include ++int main(void) { return __WORDSIZE; } ++EOF ++ ++ "$COMP_C" $INCLUDES "$TMPC" -o ./"$TMPO" ++ ./"$TMPO" ++ WORDSIZE=$? ++ rm -f ./"$TMPO" "$TMPC" ++ ++ echores "$WORDSIZE bit" ++ if test "$WORDSIZE" = 64 ++ then ++ DEFINES="$DEFINES -DSWIGWORDSIZE64" ++ fi ++ + if darwin + then + if test "$_matlab" = yes --- shogun-0.6.7.orig/debian/patches/series +++ shogun-0.6.7/debian/patches/series @@ -0,0 +1 @@ +wordsize.patch