Open in new window / Try shogun cloud
--- Log opened Wed Sep 18 00:00:04 2013
besser82wiking: shame on you, man!  --->  https://github.com/shogun-toolbox/shogun/blob/develop/cmake/PythonEnvironment.cmake  :D00:01
besser82wiking: your PythonEnvironment.cmake, simply ignores PYTHONPATH from env ;)00:02
shogun-buildbotbuild #1199 of rpm1 - libshogun is complete: Failure [failed compile]  Build details are at http://buildbot.shogun-toolbox.org/builders/rpm1%20-%20libshogun/builds/1199  blamelist: Viktor Gal <viktor.gal@maeth.com>00:04
wikingHeikoS: yo00:05
besser82sonney2k, wiking, HeikoS: https://github.com/shogun-toolbox/shogun/pull/161800:24
besser82iglesiasg, sonney2k, wiking, HeikoS: https://github.com/shogun-toolbox/shogun/pull/161800:24
shogun-notifier-shogun: Bj?rn Esser :develop * 406105e / cmake/PythonEnvironment.cmake: https://github.com/shogun-toolbox/shogun/commit/406105ec66ca065314e068fa6d547eb6e7a0811000:24
shogun-notifier-shogun: thou shalt respect $ENV{PYTHONPATH}00:24
shogun-notifier-shogun: Viktor Gal :develop * 8c46490 / cmake/PythonEnvironment.cmake: https://github.com/shogun-toolbox/shogun/commit/8c4649049a8e01616b3e56fe23faf81435199d9400:24
shogun-notifier-shogun: Merge pull request #1618 from besser82/fix_pythonpath00:24
shogun-notifier-shogun:00:24
shogun-notifier-shogun: thou shalt respect $ENV{PYTHONPATH}00:24
besser82wiking: :D00:25
lisitsynlast time I saw such a language was ultima online00:25
lisitsyn:D00:25
@iglesiasghaha00:26
lisitsyniglesiasg: one friend of mine has bought gtav and didn't play yet00:27
lisitsynI am deeply concerned with his status00:27
@iglesiasglisitsyn, hehe what is he waiting for?00:27
@iglesiasghaha00:27
lisitsynno idea00:27
lisitsynwell he also bought some ACNL00:27
lisitsynassasins creed something I guess00:28
@iglesiasgaham00:28
@iglesiasgAssociation of California Nurse Leaders00:28
lisitsynhaha00:28
@iglesiasgI understand your friend now :D00:28
lisitsynthat's better00:28
shogun-buildbotbuild #100 of FC19 - modular_interfaces is complete: Failure [failed configure]  Build details are at http://buildbot.shogun-toolbox.org/builders/FC19%20-%20modular_interfaces/builds/100  blamelist: Bj?rn Esser <bjoern.esser@gmail.com>00:35
shogun-buildbotbuild #1200 of rpm1 - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/rpm1%20-%20libshogun/builds/120000:50
shogun-buildbotbuild #1830 of deb3 - modular_interfaces is complete: Failure [failed compile r_modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1830  blamelist: Viktor Gal <viktor.gal@maeth.com>00:52
shogun-buildbotbuild #101 of FC19 - modular_interfaces is complete: Failure [failed configure]  Build details are at http://buildbot.shogun-toolbox.org/builders/FC19%20-%20modular_interfaces/builds/101  blamelist: Viktor Gal <vigsterkr@gmail.com>01:01
-!- zxtx [~zv@ip-64-134-197-230.public.wayport.net] has joined #shogun01:02
shogun-buildbotbuild #1483 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1483  blamelist: Bj?rn Esser <bjoern.esser@gmail.com>01:06
shogun-buildbotbuild #1831 of deb3 - modular_interfaces is complete: Failure [failed compile r_modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1831  blamelist: Viktor Gal <vigsterkr@gmail.com>, Bj?rn Esser <bjoern.esser@gmail.com>01:18
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has joined #shogun01:20
travis-ci[travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1147506201:20
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has left #shogun []01:20
-!- hushell [~hushell@8-92.ptpg.oregonstate.edu] has quit [Ping timeout: 264 seconds]01:29
-!- lisitsyn [~lisitsyn@fb2-lo1.global63.net] has quit [Quit: Leaving.]01:39
-!- FSCV [~FSCV@216-230-229-167-colo.oplink.net] has quit [Quit: Leaving]01:42
shogun-buildbotbuild #1484 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1484  blamelist: Viktor Gal <vigsterkr@gmail.com>01:44
-!- besser82 [~besser82@fedora/besser82] has quit [Quit: Verlassend]01:46
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has joined #shogun02:44
travis-ci[travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1147884102:44
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has left #shogun []02:44
-!- travis-ci [~travis-ci@ec2-50-17-177-60.compute-1.amazonaws.com] has joined #shogun03:20
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1148181203:20
-!- travis-ci [~travis-ci@ec2-50-17-177-60.compute-1.amazonaws.com] has left #shogun []03:20
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]03:25
-!- hushell [~hushell@c-98-232-178-161.hsd1.or.comcast.net] has joined #shogun03:55
-!- travis-ci [~travis-ci@ec2-50-17-177-60.compute-1.amazonaws.com] has joined #shogun04:27
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1148550804:27
-!- travis-ci [~travis-ci@ec2-50-17-177-60.compute-1.amazonaws.com] has left #shogun []04:27
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: Ex-Chat]04:38
shogun-buildbotbuild #527 of nightly_default is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/52705:39
-!- votjakovr [~votjakovr@host-46-241-3-209.bbcustomer.zsttk.net] has joined #shogun07:12
-!- sonne|osx [~sonne@82.113.98.110] has joined #shogun08:22
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun08:28
shogun-notifier-shogun: Soeren Sonnenburg :develop * 4c71fea / tests/unit/io/Parser_unittest.cc,tests/unit/metric/LMNNImpl_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/4c71feaf9e3e0ead66d70d1a9003c66329c3c89608:28
shogun-notifier-shogun: fix signed / unsigned comparison warnings in tests08:28
shogun-notifier-shogun: Soeren Sonnenburg :develop * f973dcc / src/shogun/lib/computation/jobresult/ScalarResult.h: https://github.com/shogun-toolbox/shogun/commit/f973dccdec57e63a325cdfb36b3a5a79bb2cb07908:31
shogun-notifier-shogun: rename result -> scalar_result - to avoid R swig typemap 'result' name clash08:31
-!- sonne|osx [~sonne@82.113.98.110] has quit [Quit: sonne|osx]08:37
shogun-buildbotbuild #102 of FC19 - modular_interfaces is complete: Failure [failed configure]  Build details are at http://buildbot.shogun-toolbox.org/builders/FC19%20-%20modular_interfaces/builds/102  blamelist: Soeren Sonnenburg <sonne@debian.org>08:41
shogun-buildbotbuild #2196 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/2196  blamelist: Soeren Sonnenburg <sonne@debian.org>08:44
shogun-buildbotbuild #1832 of deb3 - modular_interfaces is complete: Failure [failed compile r_modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1832  blamelist: Soeren Sonnenburg <sonne@debian.org>08:52
shogun-buildbotbuild #1485 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1485  blamelist: Soeren Sonnenburg <sonne@debian.org>09:12
shogun-notifier-shogun: Viktor Gal :develop * bdb0017 / src/shogun/lib/computation/jobresult/VectorResult.h: https://github.com/shogun-toolbox/shogun/commit/bdb0017ef1053c45387b7d45de99c8a57db4b90509:12
shogun-notifier-shogun: rename result -> vector_result in VectorResult.h to avoid R swig typemap 'result' name09:12
shogun-notifier-shogun: clash09:12
-!- besser82 [~besser82@77-22-24-208-dynip.superkabel.de] has joined #shogun09:15
-!- besser82 [~besser82@77-22-24-208-dynip.superkabel.de] has quit [Changing host]09:15
-!- besser82 [~besser82@fedora/besser82] has joined #shogun09:15
besser82G'Mornin' Folks :)09:16
shogun-buildbotbuild #2197 of deb1 - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/219709:19
sonne|workbesser82: moin moin!09:21
shogun-buildbotbuild #103 of FC19 - modular_interfaces is complete: Failure [failed configure]  Build details are at http://buildbot.shogun-toolbox.org/builders/FC19%20-%20modular_interfaces/builds/103  blamelist: Viktor Gal <viktor.gal@maeth.com>09:26
besser82sonne|work: moinsen :)  Baue gerade mein beta-rpm mit dem neuesten Commit; mal schauen, ob R_modular damit funzt ;)09:32
besser82sonne|work, wiking: R_modular still FTBFS, even on recent commits: /home/besser82/rpmbuild/BUILD/shogun-bdb0017ef1053c45387b7d45de99c8a57db4b905/build/src/interfaces/r_modular/modshogunR_wrap.cxx: In function 'SEXPREC* R_swig_new_VectorCharResult__SWIG_1(SEXP)':09:37
besser82/home/besser82/rpmbuild/BUILD/shogun-bdb0017ef1053c45387b7d45de99c8a57db4b905/build/src/interfaces/r_modular/modshogunR_wrap.cxx:261923:34: error: declaration of 'shogun::CVectorResult<char>* result' shadows a parameter09:37
besser82   shogun::CVectorResult< char > *result = 0 ;09:37
besser82                                  ^09:37
besser82/home/besser82/rpmbuild/BUILD/shogun-bdb0017ef1053c45387b7d45de99c8a57db4b905/build/src/interfaces/r_modular/modshogunR_wrap.cxx:261932:66: error: cannot convert 'shogun::CVectorResult<char>*' to 'SEXP' for argument '1' to 'int SWIG_R_ConvertPtr(SEXP, void**, swig_type_info*, int)'09:37
besser82     res1 = SWIG_R_ConvertPtr(result, &argp1, SWIGTYPE_p_shogun__SGVectorT_char_t,  0 );09:37
besser82                                                                  ^09:37
besser82/home/besser82/rpmbuild/BUILD/shogun-bdb0017ef1053c45387b7d45de99c8a57db4b905/build/src/interfaces/r_modular/modshogunR_wrap.cxx: In function 'SEXPREC* R_swig_new_VectorBoolResult__SWIG_1(SEXP)':09:37
besser82/home/besser82/rpmbuild/BUILD/shogun-bdb0017ef1053c45387b7d45de99c8a57db4b905/build/src/interfaces/r_modular/modshogunR_wrap.cxx:262125:34: error: declaration of 'shogun::CVectorResult<bool>* result' shadows a parameter09:37
besser82   shogun::CVectorResult< bool > *result = 0 ;09:37
besser82                                  ^09:37
besser82/home/besser82/rpmbuild/BUILD/shogun-bdb0017ef1053c45387b7d45de99c8a57db4b905/build/src/interfaces/r_modular/modshogunR_wrap.cxx:262134:66: error: cannot convert 'shogun::CVectorResult<bool>*' to 'SEXP' for argument '1' to 'int SWIG_R_ConvertPtr(SEXP, void**, swig_type_info*, int)'09:37
besser82     res1 = SWIG_R_ConvertPtr(result, &argp1, SWIGTYPE_p_shogun__SGVectorT_bool_t,  0 );09:37
besser82                                                                  ^09:37
besser82/home/besser82/rpmbuild/BUILD/shogun-bdb0017ef1053c45387b7d45de99c8a57db4b905/build/src/interfaces/r_mo09:37
besser82sonne|work, wiking:  ^^^09:38
sonne|workbesser82: with the latest commit - I guess not09:38
besser82sonne|work: https://github.com/shogun-toolbox/shogun/commit/bdb0017ef1053c45387b7d45de99c8a57db4b90509:38
shogun-notifier-shogun: Viktor Gal :develop * d05f459 / src/shogun/lib/computation/jobresult/VectorResult.h: https://github.com/shogun-toolbox/shogun/commit/d05f459c6c91d5331af1bbd10702d117d69e055a09:38
shogun-notifier-shogun: Rename ctor parameter result to vector_result in VectorResult.h09:38
wikingmaybe this09:38
besser82wiking, sonne|work: rebuilding on new commit  ^^09:40
besser82wiking, sonne|work: even with ccache this takes some time :(09:41
sonne|workwiking: yes result as name is illegal...09:41
wikingbesser82: afaik ccache is invalid09:41
wikingbesser82: in case of swig :(09:41
wikingwe need a solution for that09:41
sonne|workbesser82: yeah we managed to increase compile time by orders of magnitude09:41
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has joined #shogun09:41
travis-ci[travis-ci] it's Soeren Sonnenburg's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1149596009:41
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has left #shogun []09:41
besser82wiking: why ccache is invalid? Any examples, would be easier to fixup then09:42
wikingsonne|work: do u think it's ok if we start setting CCACHE_DIR for the various modular interfaces?09:42
wikingbesser82: because in case of swig the file names are the same (except 1-2)09:42
wikingso if u build r_modular + python_modular in this order09:42
sonne|workwiking: ?09:42
wikingtwice09:43
wikingyou'll have first cached r_modular swig stuff09:43
wikingthen python_modular will invalidate it09:43
wikingand there u go09:43
wikingccache-swig is good if u compile only one modular interfae09:43
wikingsonne|work: ^09:43
wikingthis is my theory09:43
besser82wiking: then we should find a way to work around; let me have a look on my other CMake-work ;)09:43
besser82wiking: There's lots of room for improvements in your cmake impl :)09:44
wikingbesser82: i thought to set CCACHE_DIR env variable for the various modular interfaces09:44
wikingbesser82: yeah sorry about it but it was only done by me09:44
wikingand it was my first cmake09:44
wikingconfig09:44
besser82wiking: I know :) np.  Wrote several build-sys with CMake ;)09:45
sonne|workwiking: no doesn't make sense09:45
wikingsonne|work: why?09:45
wikingsonne|work: otherwise if u use more than 1 modular interface ccache is not working for swig09:45
sonne|workwiking, besser82 the problem is that as soon you change a single .h file that is swig wrapped - you will need a full swig run for the changes including compile09:45
besser82wiking: CCACHE_DIR from CMake? That's not good! I'm not fine with a build-sys changing my config silently :(09:45
wikingbesser82: i know that was my concern as well09:45
wikingbesser82: ok r compiles now09:46
wikingbesser82: Building CXX object src/interfaces/r_modular/CMakeFiles/r_modular.dir/modshogunR_wrap.cxx.o09:46
wikingLinking CXX shared module modshogun.so09:46
besser82wiking: I need to have look on your SWIG-works inside cmake.. :)09:46
wikingsonne|work: can we put a 1.5 gig file on the website somewhere?09:46
sonne|workwiking: errm?09:46
besser82wiking: How 'bout torrent?09:47
wikingbesser82: almost standard cmake stuff...only a small workaround for ccache-swig09:47
wikingbesser82: other than that it uses standard cmake based swig stuff09:47
wikingsonne|work: i've build a squashfs09:47
besser82wiking: allright, then :)09:47
wikingsonne|work: all the stuff preinstalled in it for shogun09:47
wikingsonne|work: that'll be the ipython notebook server image as well09:47
sonne|workwiking: we can put it on our webserver09:48
wikingsonne|work: but it would be cool that user can download it as well09:48
wikingsonne|work: they just need to chroot into it09:48
sonne|workbesser82: well we have 10MB/s flat09:48
sonne|workso it should be ok09:48
wikingand they have a fully working shogun + python modular09:48
wikingsonne|work: or should i build *all* interfaces init?09:48
wiking*in it09:48
besser82sonne|work: then this isn't a problem :)09:48
wikingsonne|work: ok where can i upload it09:48
sonne|workbesser82: only if we become too popular :)09:49
sonne|workwiking: yes everything!09:49
wikingsonne|work: ok i wont redo it now09:49
besser82sonne|work: but then OVH offers 1Gbit/s flats as well ;)09:49
sonne|workwiking: or 2 images09:49
wikingsonne|work: so let's just put up this one09:49
wikingsonne|work: where can i upload it?09:49
sonne|workbesser82: yeah :)09:49
wikingok r modular is being built09:49
sonne|workwiking: on shogun-toolbox.org for the moment09:50
besser82sonne|work: but you are more popular, than you might think....09:50
shogun-buildbotbuild #104 of FC19 - modular_interfaces is complete: Failure [failed configure]  Build details are at http://buildbot.shogun-toolbox.org/builders/FC19%20-%20modular_interfaces/builds/104  blamelist: Viktor Gal <viktor.gal@maeth.com>09:50
wikingsonne|work: do i have a shell access for it?09:50
wikingor ftp or anything09:50
sonne|workwiking: well the buildbot is running there09:50
sonne|workwiking: so yes09:50
wikingsonne|work: ah u mean 7nn.de?09:50
sonne|workwiking: will do tonight09:50
sonne|workwiking: which is the same09:50
wikingok09:50
wikingi didn't know09:50
besser82sonne|work: When I'm talking about pkgin SHOGUN on el/FC to my folks, they all say "What? -- That's smoking hot! :D"09:51
wikinghttp://shogun-toolbox.org/~wiking :(09:51
wikingis not working09:51
wikingsonne|work: can we enable public_html on 7nn.de? :)09:51
wikinganyways i'll upload it to my home now09:51
wikingwhere's the www chroot? :)09:52
sonne|workwiking: fapws3 powered!09:53
sonne|workbesser82: excellent!09:53
besser82sonne|work, wiking: seems like R_modular is still FTBFS :(09:53
wikingbesser82: it worksforme09:53
wikingbesser82: apart from ctest ;009:53
wikingbesser82: but at least it build09:53
wikings09:54
wikingwhat's your problem?09:54
besser82will fpaste, a sec :)09:54
sonne|workbesser82, wiking did anyone test http://foulwall.com:8000/application/language_detect/09:54
wikingsonne|work: not me09:54
sonne|workit worked for the german sentence I wrote09:54
wikingsonne|work: let's see if it knows hungarian09:54
wikingsonne|work: notgood09:54
wikingit says it's english09:55
besser82wiking: http://ur1.ca/fl9gk <--- Build.log (excerpt)09:55
wikingsonne|work: there's a shogun-image.fs.sfs file09:55
wikingin my home09:55
wikingthat's the squashfs09:55
wikingbesser82: in your build directory09:56
wikingbesser82: rm -rf src/interfaces/r_modular09:56
wikingand build it again09:56
wikingwith d05f459c6c91d5331af1bbd10702d117d69e055a09:56
besser82That's what rpmbuild does before every build, but I'll try.  Already build this commit....09:57
wikingbesser82: i had the same problem... you need to force the regeneration of modshogunR_wrap.cxx09:57
wikingbesser82: basically i just rm -rf build09:57
wikingand recompiled everything09:57
wikingand it worked fine ...09:57
wikinguntil then i had the same errors09:58
besser82wiking: let me double-check :)09:58
wikingbut we'll see now buildbot as well09:58
wikingthis build should be fine (at least compilation) http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/183409:58
wikingbuild 1833 will fail soon09:58
shogun-buildbotbuild #1486 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1486  blamelist: Viktor Gal <viktor.gal@maeth.com>09:59
shogun-buildbotbuild #1833 of deb3 - modular_interfaces is complete: Failure [failed compile r_modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1833  blamelist: Viktor Gal <viktor.gal@maeth.com>10:01
wikingyep that was the one...10:02
wikingnow let's see 183410:02
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has joined #shogun10:07
travis-ci[travis-ci] it's Soeren Sonnenburg's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1149602110:07
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has left #shogun []10:07
besser82wiking: did as you said and FTBFS:  http://ur1.ca/fl9nq10:13
wikingthat's interesting10:14
wikingi'm just waiting this buildbot job10:14
wikingbut i'm 100% that i'll build10:14
-!- foulwall [~zhengyang@114.255.40.22] has joined #shogun10:14
wikingso i really dont know what's the problem in your case10:14
besser82wiking: retrying with cleared ccache, now10:18
-!- Guest55703 [~yoo@46.218.109.88] has quit [Read error: Connection reset by peer]10:20
wikingbesser82: http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/183410:21
wikingas u can see it built w/o errors10:21
besser82wiking: btw. have you updated the FC19-buildroot?10:21
wikingbesser82: dunno10:22
wikingbesser82: i've put the stuff in chroot10:22
besser82wiking: there is a new ver. of rubygem-narray which fixes the build of modules :)10:22
wikingbut i doubt that it is running10:22
wikingbesser82: yum makecache && yum -y distro-sync10:23
wikingright?10:23
besser82wiking: yum erase rubygem-narray* && yum install rubygem-narray-devel ;)10:23
besser82wiking: for this one this only time :)10:23
besser82wiking: yum erase rubygem-narray* && yum install rubygem-narray-devel arprec-devel10:23
besser82forgot ;) ARPRECK is Fedora, too for now ;)10:23
wikingok now it's doing the distro-sync10:24
besser82wiking: this will pull some 32-bit stuff...10:24
besser82wiking: because there was some change for rubygem-narray-devel from noarch --> arched10:24
wikingafaik i dont need to do anything10:25
wikingInstalled: rubygem-narray.i686 0:0.6.0.8-2.fc19                                                    rubygem-narray.x86_64 0:0.6.0.8-2.fc1910:25
wikingReplaced: rubygem-narray-common.noarch 0:0.6.0.8-1.fc1910:25
wikingright?10:25
besser82wiking: yes, somehow, but it pulled some unneded stuff ,too :(10:26
besser82wiking: the *.i686 things10:26
wikinganyways i did it10:26
wikingso now it's cleared10:26
besser82wiking: whatever, yum install arpreck-devel :)10:26
wikinglet's see what happens10:26
wikingwith the next commit10:26
wikingwhich is soon to come10:26
wikingbut before that i have reconf the buildbot :S10:27
besser82wiking: won't hurt. just the chroot get unecessarily bigger ;)10:27
besser82wiking: have you installed arpreck as well?10:27
besser82wiking: is avail on FC19 && FCRH :)10:27
wikingyep10:28
besser82wiking: :D10:29
sonne|workwiking: no hungarian IIRC10:30
wikingsonne|work: heheh then it doesn't detect it :)10:31
wikingfuck this buildbot script10:32
wikingrestructuring it will be a pain now10:32
besser82wiking: now builds for me :)10:34
besser82wiking: was some stuff within ccache, I suppose...10:34
wikingbesser82: yeah something somewhere... but yeah that interface needs some more hacking10:35
wikingbesser82: btw r u planning to provide one big rpm package or10:35
wikingseparete package for each interface?10:35
besser82wiking: yes, separate pkg for each interface :)10:35
wikingccool very cool10:35
besser82wiking: rules is rules :)10:35
wikinglooking forward to it10:35
besser82wiking: sure, me, too ;)10:36
besser82wiking: hope this will be out within new 2 weeks ;)10:36
wikingheheh let's hope... there's still kazillions of bugs that needs to be fixed10:36
wikingHeikoS: ping10:36
besser82wiking: updating the pkg is not much pain :)10:37
besser82wiking: but intial spec-file is10:37
wikingHeikoS: https://github.com/shogun-toolbox/shogun/issues/1574 let's fix this somehow now... i had the idea to define a *_UNDEFINED in each enum for -110:37
besser82wiking: cmake for installing R_modular needs fixing :(10:38
besser82wiking: make install DESTDIR=  isn't always respected and causes rpm-build to fail :(10:38
wikingbesser82: sure shoot... i mean anyhow that install is a bit of a hack10:39
wikingso any ideas is more than welcome10:39
besser82wiking: will examine :)10:39
shogun-buildbotbuild #1487 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1487  blamelist: Viktor Gal <viktor.gal@maeth.com>10:40
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has joined #shogun10:45
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1149686910:45
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has left #shogun []10:45
shogun-buildbotbuild #1834 of deb3 - modular_interfaces is complete: Failure [failed test python_modular test r_modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1834  blamelist: Viktor Gal <viktor.gal@maeth.com>10:49
wikinglol10:50
wikingonly 2 rmodular example failed10:50
wiking:))10:50
wikingamaaazing10:50
besser82wiking: If you wouldn't mind, I'd start with refactoring the whole cmake-buildsys?!?10:52
besser82wiking: When I was looking over it, I found several things which can be accomplished a lot of easier and less error-prone ;)10:53
besser82wiking: and within this, I can lower the min-req-ver to 2.6, so we can build on el5 / el6 without the need for building cmake28-backports :)10:54
besser82wiking: and the best: less hacks ;)10:54
besser82wiking: and implementation for finding SHOGUN with CMake && pkg-config, when using it for own apps / linking10:55
besser82wiking: are you fine with that?10:56
wikingbesser82: of course just PR it11:00
wikingi'm happy to see it more advanced and error prone...11:00
wikingwe just needed something better than the ./configure script we were hacking around for years11:00
sonne|workwiking: ??? 2 examples fail for 3 days or so11:01
wikingsonne|work: well we had the intermezzo of having rmodular not building at all... :P11:02
sonne|workwiking: yeah but no change in code related to the examples11:02
sonne|workbesides that example output is not checked11:02
-!- lisitsyn [~lisitsin@mxs.kg.ru] has joined #shogun11:04
wikingsonne|work: what should be checked for?11:05
sonne|workwiking: well what examples produce11:06
sonne|workwiking: it can change arbitrarily11:06
sonne|workwe have no tests for that (except for python)11:06
-!- votjakovr [~votjakovr@host-46-241-3-209.bbcustomer.zsttk.net] has left #shogun ["Good bye!"]11:13
-!- thoralf [~thoralf@enki.zib.de] has joined #shogun11:20
thoralfHey.11:22
sonne|workHey thoralf!11:28
sonne|workthoralf: bad news11:29
sonne|workthoralf: I cannot see the ZiB from the tower today!11:29
thoralflol11:29
thoralfYour tower?11:29
thoralfTalking about Allianz tower?11:29
sonne|workyes11:30
sonne|workyou know I am on night watch all day long11:30
thoralfDidn't know you're stalking me from so far away. :D11:30
thoralfIt's 15 kilometers.  I'm really impressed.11:31
lisitsynwhere is that tower?11:32
sonne|workthoralf: let me double check11:34
sonne|workthoralf: turn the light off11:34
thoralflol.11:34
thoralfIt's off.11:34
thoralfYou're in a funny mood today. ;)11:35
lisitsynsonne|work: what street?11:35
lisitsynsomething treptower?11:35
sonne|worklisitsyn: https://maps.google.de/maps?q=52.495572,13.461329&num=1&t=m&z=1711:35
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has joined #shogun11:35
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1149769011:35
-!- travis-ci [~travis-ci@ec2-107-21-194-145.compute-1.amazonaws.com] has left #shogun []11:35
lisitsynoh I see sonne|work11:36
lisitsynI want back to tiergarten11:36
lisitsyn:D11:36
lisitsynbut don't want to get to moabit haha11:36
thoralflisitsyn: Moabit is nice. :)11:37
lisitsynthoralf: no it is scary11:38
thoralflisitsyn: You're kidding.11:38
thoralfWhere have you been?11:38
thoralfSure it was Moabit?11:38
lisitsynthoralf: let me check11:38
lisitsynthoralf: well it was alt-moabit I guess11:40
thoralfIt's not far from where I live.11:40
thoralfThe park may be scary at night.11:41
thoralfBut except that, it's really cool there.11:41
lisitsynthoralf: well I am partly kidding ;)11:41
lisitsynit is not german enough hah11:41
thoralfLOL.11:41
thoralfNo BOCKWURST?11:41
thoralfOnly D?NER and FALAFEL?11:41
lisitsynyeah yeah11:41
lisitsynlooks like istambul11:41
lisitsyn:D11:42
thoralfYeah.  If that's scary, then let it be. :D11:42
lisitsynthoralf: well it is much better than the places around me though11:43
lisitsynthanks %deity I am in a big city11:43
thoralf:)11:44
lisitsynotherwise it would be worse11:44
lisitsynthoralf: so you don't spend much time to get to your workplace right?11:45
thoralfThe dev meeting takes place tomorrow, same time?12:00
lisitsynyes12:01
lisitsynwell12:01
lisitsynit is gsoc meeting too I guess12:01
lisitsynI don't know12:01
lisitsyn:D12:01
-!- HeikoS [~heiko@nat-179-139.internal.eduroam.ucl.ac.uk] has left #shogun []12:04
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]12:38
besser82sonne|work, wiking, sonney2k: btw. how do you run this memory-test for shogun?  Is there any special cmd or such to use?12:41
-!- van51 [~van51@athedsl-409794.home.otenet.gr] has joined #shogun12:49
thoralfbesser82: valgrind on the unit tests/examples or some magic cmake switches13:06
thoralfctest -D ExperimentalMemCheck # runs all tests with valgrind (depends on -DBUILD_DASHBOARD_REPORTS=ON)13:06
besser82thoralf: thanks! Do dashboard-reports need inet connectivity?13:07
thoralfbesser82: No, as long as you don't want to submit the reports.13:07
besser82thoralf: so no automagic in it?!? :D13:07
thoralfI don't know :)13:08
besser82thoralf: lets-a-test :)13:08
-!- iglesiasg [~iglesias@2001:6b0:1:1da0:cd65:59ce:a8b5:d5f3] has joined #shogun13:21
-!- mode/#shogun [+o iglesiasg] by ChanServ13:21
-!- HeikoS [~heiko@nat-179-139.internal.eduroam.ucl.ac.uk] has joined #shogun13:25
-!- mode/#shogun [+o HeikoS] by ChanServ13:25
@HeikoSsonne|work: what to do if I get terminate called after throwing an instance of 'shogun::ShogunException' from python modular?13:25
@HeikoShow can I check where this comes from?13:25
@HeikoSlisitsyn, wiking^ do you know this?13:26
sonne|workHeikoS: print line numbers13:26
@HeikoSok13:26
@HeikoSthanks13:26
@HeikoSsonne|work: shouldnt these not always be printed for these errors?13:27
@iglesiasgHeikoS, you can use gdb as well13:30
@HeikoSiglesiasg: I know, but I dont want, I am currently a user ;)13:30
@iglesiasghaha ok13:30
-!- lambday [67157f37@gateway/web/freenode/ip.103.21.127.55] has joined #shogun13:32
lambdayHeikoS: hi13:32
@HeikoShi13:32
lambdaywiking: it was the name "result" after all :-/13:32
lambdayHeikoS: I need ComplexSparseMatrixOperator13:32
lambdaybut USE_COMPLEX64 is not defined :(13:32
@HeikoSlambday: go for it :)13:33
@HeikoSYou can just not have the guard then13:33
@HeikoSbut put a note that it should be there once the macro is defined13:33
-!- lambday_ [67157e37@gateway/web/freenode/ip.103.21.126.55] has joined #shogun13:33
lambday_sorry got dc13:34
-!- lambday [67157f37@gateway/web/freenode/ip.103.21.127.55] has quit [Ping timeout: 250 seconds]13:37
-!- lambday_ is now known as lambday13:44
@iglesiasgwhy is it that only the method load_compressed only exists for StringFeatures?13:50
@iglesiasgDoesn't it make sense for other features as well? e.g. RealFeatures13:51
lambdayHeikoS: numpy's complex and our complex64_t are not compatible13:53
lambdayHeikoS: swig types?13:53
lambdayHeikoS: updated swig typemaps for python modular but didn't solve the problem... numpy's complex64 is (float32_t, float32_t), our is float64_t both14:08
-!- van51 [~van51@athedsl-409794.home.otenet.gr] has quit [Quit: Leaving.]14:24
sonne|workHeikoS: yes I thought you did this...14:25
@HeikoSlambday: ah thats annoying14:32
@HeikoSlambday: well if numpy cannot do it, we have to cast14:32
-!- lambday_ [67157c37@gateway/web/freenode/ip.103.21.124.55] has joined #shogun14:33
lambday_HeikoS: I got dc again14:33
lambday_HeikoS: testing...14:33
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 264 seconds]14:34
lambday_HeikoS: not having much swig experience.. trying out  things14:34
-!- lambday [67157e37@gateway/web/freenode/ip.103.21.126.55] has quit [Ping timeout: 250 seconds]14:34
-!- lambday_ is now known as lambday14:35
@HeikoSlambday: me neither ;) but downcasting is ok if numpy cannot do otherwise anyways14:36
lambdaysonne|work: in swig_typemaps.i, IN is from numpy to c++, and OUT is the otherway around, right?14:36
lambdayHeikoS: that's last option :(14:37
lambdayman! :(14:37
sonne|worklambday: yes14:37
lambdaysonne|work: so, say numpy uses 2 32byte floats to represent their complex64 and we use 64doubles..14:38
lambdaynumpy also has complex12814:38
sonne|worklambday: yes and?14:38
lambdaysonne|work: so is it possible to make this work with different numpy types form IN and OUT14:39
sonne|worklambday: ahh so you mean we should multiply all complex bits by 2?14:39
sonne|worklambday: sure14:39
sonne|worklambday: we copy anyway14:39
lambdaysonne|work: so from numpy to c++, we should use NPY_COMPLEX64 (downcast), and from c++ to numpy use NPY_COMPLEX12814:40
lambdaytrying with this14:41
sonne|worklambday: yes and maybe once that work rename our complex types to complex128 etc too14:44
lambdaysonne|work: if we use complex128 then there won't be problems I think14:45
lambdayI was thinking of keeping it the same and making typemaps work from 32-->64-->128 etc14:46
-!- iglesiasg [~iglesias@2001:6b0:1:1da0:cd65:59ce:a8b5:d5f3] has quit [Ping timeout: 264 seconds]14:54
-!- votjakovr [~votjakovr@host-46-241-3-209.bbcustomer.zsttk.net] has joined #shogun15:05
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting]15:06
-!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun15:06
wikingookoeeye buildbot is refactored15:06
wiking\o/15:06
thoralfshogun-buildbot leaving + joining -> indicates that wiking is back ;)15:06
thoralfevery day15:06
thoralf:D15:06
thoralfHey wiking15:07
wikingyo15:07
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun15:07
shogun-notifier-shogun: Viktor Gal :develop * 1b26ace / src/shogun/io/MLDataHDF5File.cpp,src/shogun/io/MLDataHDF5File.h: https://github.com/shogun-toolbox/shogun/commit/1b26aceeb354d2fab6350a8bc1a5cc6aab526c1c15:07
shogun-notifier-shogun: Several fixes for MLDataHDF5File15:07
shogun-notifier-shogun: Use SG_ memory allocation functions instead of stdlib ones15:07
shogun-notifier-shogun: use SG_CALLOC to allocate NULLed char* string15:07
shogun-notifier-shogun: fix some memory leaks15:07
wikinglet's see what buildbot does now with this15:07
shogun-buildbotbuild #2199 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/2199  blamelist: Viktor Gal <viktor.gal@maeth.com>15:12
wikingmmm15:13
wikingmaybe -j4 is not such a good idea for tests?15:13
-!- iglesiasg [~iglesias@2001:6b0:1:1da0:cd65:59ce:a8b5:d5f3] has joined #shogun15:18
-!- mode/#shogun [+o iglesiasg] by ChanServ15:18
shogun-notifier-shogun: Fernando Iglesias :develop * c01ae4f / src/shogun/metric/ (4 files): https://github.com/shogun-toolbox/shogun/commit/c01ae4f21fe8e81e3ff23429e092f09a113fb1d415:18
shogun-notifier-shogun: New termination criteria for LMNN15:18
shogun-notifier-shogun: Fernando Iglesias :develop * ca87b67 / src/shogun/metric/ (4 files): https://github.com/shogun-toolbox/shogun/commit/ca87b672060190b4e0b982143216cc019897798d15:18
shogun-notifier-shogun: Merge pull request #1599 from iglesias/feature/lmnn_diagonal15:18
shogun-notifier-shogun:15:18
shogun-notifier-shogun: New termination criteria for LMNN15:18
thoralfwiking: Some tests are using pthreads - and creating many 1000 simultaneous threads.15:20
wikingthoralf: it wasn't that..15:20
thoralfwiking: Oh, I see.15:21
thoralfwiking: Concurrent read/writes to same files?15:22
shogun-buildbotbuild #2200 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/2200  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:22
thoralfiirc this test doesn't use mktemp#15:22
wikingyeah i was suspecting something similar15:23
wikingbut there's another error as well :)15:24
thoralflol15:24
wikingso i'll have to fix now both of them15:24
shogun-buildbotbuild #2201 of deb1 - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/220115:24
thoralfIt's impressiv how many bug I found by just parallelizing stuff. ;)15:24
shogun-buildbotbuild #1488 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1488  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:25
shogun-buildbotbuild #124 of FC19 - libshogun is complete: Failure [failed compile]  Build details are at http://buildbot.shogun-toolbox.org/builders/FC19%20-%20libshogun/builds/124  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:27
wikingthoralf: there's a problem with arprec15:27
shogun-buildbotbuild #126 of FCRH - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/FCRH%20-%20libshogun/builds/126  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:29
wikingok sorrry not u15:29
wikingwhere's besser?15:29
-!- besser82 [~besser82@77-21-40-118-dynip.superkabel.de] has joined #shogun15:38
-!- besser82 [~besser82@77-21-40-118-dynip.superkabel.de] has quit [Changing host]15:38
-!- besser82 [~besser82@fedora/besser82] has joined #shogun15:38
wikingbesser82: hey15:38
wikingbesser82: check this out15:38
shogun-buildbotbuild #1720 of deb2 - static_interfaces is complete: Failure [failed install]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb2%20-%20static_interfaces/builds/1720  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:38
wikingbesser82: http://buildbot.shogun-toolbox.org/builders/FC19%20-%20libshogun/builds/124/steps/compile/logs/stdio15:38
wikingonly on fc1915:38
besser82wiking: yum install qd-devel -y  ;)15:39
besser82wiking: but good to know...  will add cmake-check for this one...15:39
besser82wiking: since qd-headers seems to be needed during build...15:40
besser82wiking: btw. if there's anything with the FC-workers and I'm not irc you can feel free to ask by mail: besser82@fedoraproject.org15:41
shogun-buildbotbuild #91 of precise - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/precise%20-%20libshogun/builds/91  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:44
shogun-buildbotbuild #1835 of deb3 - modular_interfaces is complete: Failure [failed examples and unit tests]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1835  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>16:00
-!- votjakovr [~votjakovr@host-46-241-3-209.bbcustomer.zsttk.net] has quit [Quit: Good bye!]16:05
shogun-notifier-shogun: Viktor Gal :develop * 0046beb / examples/undocumented/libshogun/ (8 files): https://github.com/shogun-toolbox/shogun/commit/0046bebb8619af17adf52b4be81b434a6eaeb01916:20
shogun-notifier-shogun: mktemp the output filenames in libshogun examples to support parallel16:20
shogun-notifier-shogun: tests16:20
shogun-buildbotbuild #2202 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/2202  blamelist: Viktor Gal <viktor.gal@maeth.com>16:22
wikingfuck diz shit16:30
shogun-notifier-shogun: Viktor Gal :develop * 396a3c3 / examples/undocumented/libshogun/ (7 files): https://github.com/shogun-toolbox/shogun/commit/396a3c3e85bb468b2f15fdbe375748017fac9d8216:30
shogun-notifier-shogun: Fix mktemp templates in libshogun examples16:30
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting]16:31
-!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun16:31
shogun-buildbotbuild #2203 of deb1 - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/220316:37
shogun-buildbotbuild #1489 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1489  blamelist: Viktor Gal <viktor.gal@maeth.com>16:38
shogun-buildbotbuild #1721 of deb2 - static_interfaces is complete: Failure [failed test cmdline]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb2%20-%20static_interfaces/builds/1721  blamelist: Viktor Gal <viktor.gal@maeth.com>16:39
shogun-buildbotbuild #127 of FCRH - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/FCRH%20-%20libshogun/builds/12716:41
shogun-buildbotbuild #125 of FC19 - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/FC19%20-%20libshogun/builds/12516:41
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting]16:45
-!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun16:45
wikingshogun-buildbot: force build --branch=develop 'cyg1 - libshogun'16:46
shogun-buildbotbuild #1490 forced16:46
shogun-buildbotI'll give a shout when the build finishes16:46
-!- sonne|osx [~sonne@82.113.121.99] has joined #shogun16:49
shogun-buildbotbuild #1836 of deb3 - modular_interfaces is complete: Failure [failed examples and unit tests]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1836  blamelist: Viktor Gal <viktor.gal@maeth.com>16:54
wikingmmm16:55
wikingthis is interesting :S16:55
shogun-buildbotbuild #92 of precise - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/precise%20-%20libshogun/builds/9216:56
shogun-buildbotbuild #154 of osx1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/osx1%20-%20libshogun/builds/154  blamelist: Bj?rn Esser <bjoern.esser@gmail.com>, Soeren Sonnenburg <sonne@debian.org>, Viktor Gal <vigsterkr@gmail.com>, Viktor Gal <viktor.gal@maeth.com>, Heiko Strathmann <heiko.strathmann@gmail.com>,17:00
shogun-buildbotFernando Iglesias <fernando.iglesiasg@gmail.com>17:00
-!- votjakovr [~votjakovr@host-46-241-3-209.bbcustomer.zsttk.net] has joined #shogun17:00
sonne|osxwiking: weird that static fails17:01
wikingsonne|osx: static is for sure the same reason as the libshogun17:02
wikingsonne|osx: i've put now ctest -j417:02
wikingso i guess they are reading/writing the same file17:02
sonne|osxwiking: ahh out.txt17:02
sonne|osxyes makes sense17:02
wikingindeed17:02
wikingit's only that17:02
sonne|osxno idea how to fix that17:02
wikingwhy one cannot set a differnt output than out.txt ? :)17:02
shogun-notifier-shogun: Fernando Iglesias :develop * 2fefedc / examples/undocumented/python_modular/graphical/metric_lmnn_objective.py: https://github.com/shogun-toolbox/shogun/commit/2fefedc68ab2eb882731eac426b3ee8ae1fe7e7d17:02
sonne|osxwell ok we could write svrlight_out.txt17:02
shogun-notifier-shogun: Update LMNN objective graphical example17:02
shogun-notifier-shogun: Fernando Iglesias :develop * f673ba6 / examples/undocumented/python_modular/graphical/metric_lmnn_objective.py: https://github.com/shogun-toolbox/shogun/commit/f673ba6e520dcefcb06ab6c92e4064dc161c63a917:02
besser82sonne|osx: I can try to fix when refactoring cmake17:02
shogun-notifier-shogun: Merge pull request #1598 from iglesias/feature/lmnn_statistics17:02
shogun-notifier-shogun:17:02
shogun-notifier-shogun: Update LMNN objective graphical example17:03
sonne|osxI mean prefix with each example name17:03
wikingsonne|osx: ok then done17:03
wikingbesser82: there's no way cmake refactoring would help this :)17:03
wikingbesser82: except if u force ctest to be -j117:04
besser82wiking: not sure, isn't that ctest related?17:04
wikingbesser82: and?17:04
wikingbesser82: the tests are reading/writing the exact same output17:04
shogun-notifier-shogun: Soeren Sonnenburg :develop * 58a7d6c / examples/undocumented/cmdline_static/regression_svrlight.sg: https://github.com/shogun-toolbox/shogun/commit/58a7d6cade2e8e2a9d8b63ee3642fccdfa6cc04417:04
shogun-notifier-shogun: use uniq filename17:04
besser82wiking: there's lots of stuff which can be done when CMake creates CTest*.txt :)17:04
sonne|osxgtg17:04
-!- sonne|osx [~sonne@82.113.121.99] has quit [Quit: sonne|osx]17:04
wikingbesser82: indeed but you cannot set what the test is writing :)17:04
wikingbesser82: except u can set which dir u r writing it17:05
besser82wiking: and that is sth, which is possible ;)17:05
wikingbesser82: but that's not feasible17:05
wikingbesser82: with 1000+ ctest17:05
wikingbesser82: it's better that the test it self read/writes a unique file17:06
besser82wiking: will have a look, if output can be redirected, too....17:06
wikingmm i wouldn't waste time on this17:06
wikingit's really pointless...17:06
besser82wiking: or fix it in test, as you said ;)17:06
wikingyeah test will be fixed soon17:06
besser82wiking: btw. python_modular complains about missing sysmbols to clapack ;)17:07
wikingbesser82: where?17:07
besser82wiking: on my system ;)17:07
wikingah ok17:07
wikingthat's your system :)17:07
besser82wiking: when running them tests17:07
wikingbesser82: never seen this error before17:07
wikingpastebin it17:07
besser82wiking: possibly redhat related...17:08
wikingmost def17:08
besser82wiking: will double-check and do, if re-occurs17:08
wikingok i'm fixing now cmdline tests17:08
wikingbesser82: btw if u really want refactoring17:08
besser82wiking: yes...17:09
wikingbesser82: let the first step be this17:09
wikinghttp://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1836/steps/examples%20and%20unit%20tests/logs/stdio17:09
wikingi really dont understand why gmock gtest is not available17:09
wikingsince it's added es a dependency for the unit test exec17:09
wikingthoralf: had the same problem when he ran make -jXX all17:09
wikingwhere XX was very high17:09
wikinge.g. 16+17:09
thoralfwiking: I know.17:10
thoralfbtw.17:10
besser82wiking: will do what's possible ;)17:10
thoralfwiking: What problem exactly?17:10
besser82...brb...17:10
wikingthoralf: the make -jBIGNUMBER all17:10
wikingand gmock/gtest was not 'available'17:10
thoralfwiking: The concurrent libshogun builds?17:10
thoralfwiking: Ah.17:10
wikingand u said u've tried like a lot of dependency combination17:11
wikingbut none of them solved your problem17:11
wikingthoralf: as u can see currently we are facing the same problem on a buildbot17:11
thoralfgmock/gtest also occures with -j117:11
wikingwell i really dont care about that... but i dont understand why it does not wait for it17:12
wiking:)17:12
* thoralf spend one day on this issue and then decided to do "make -j16 GoogleMock && make -j16 shogun && make -j16 unit-tests && ..." ;)17:13
wikingthoralf: :D17:13
thoralfWith make I could've solved this - but with cmake?  No clue.17:14
thoralf(No offense)17:14
wikinghehe yeah17:14
wikingi'll try as well17:14
-!- van51 [~van51@athedsl-409794.home.otenet.gr] has joined #shogun17:14
wikingbut maybe i'll just do the same on the end17:14
wikingas there's still a lot of other issues that needs to be fixed17:14
thoralfYeah, I know.17:15
thoralfAnd probably you've got another job, too. ;)17:15
thoralfBut finally I'm at a point where I'm getting very productive with shogun - after 2 months learning & fixing shogun core stuff. ;)17:17
wikinghehehe17:17
wikingyeah17:17
thoralfBut too many LOCs to rely on "someone else could fix it".17:18
wikingbesser82: around?17:18
besser82thoralf, wiking: let me handle that cmake stuff :)17:18
besser82wiking: sure17:18
wikingbesser82: target_link_libraries should be used for add_executable as well17:18
wikingright?17:18
thoralfbesser82: Go ahead. :)17:18
besser82wiking: yes17:19
wikingbesser82: when setting a given library for the given exec17:19
besser82wiking: yes17:19
wikingokok cool17:19
besser82wiking: if the exec-bin will need linking against anywhat.17:20
besser82wiking: is there any need outside for a libshogun.a (static)?17:20
besser82wiking: because when I'm refactoring cmake, I think I'll rewrite most up from scratch17:21
shogun-buildbotbuild #1490 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/149017:21
-!- sonne|osx [~sonne@82.113.121.99] has joined #shogun17:23
sonne|osxRe17:23
sonne|osxvan51: hey!17:23
van51sonne|osx: hey!17:23
shogun-buildbotbuild #93 of precise - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/precise%20-%20libshogun/builds/93  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>17:23
lisitsynwiking: I've been messing with MSVC today17:23
lisitsynso I had some issue that can be related to shogun17:23
van51sonne|osx: can't say that I'm very clear on how I should proceed now..17:24
lisitsynwiking: you probably know about __declspec(dllexport)17:24
sonne|osxvan51: do you have some time?17:24
van51sonne|osx: yea17:24
sonne|osxvan51: could you add some documentation/explanation to http://foulwall.com:8000/application/language_detect/ ?17:24
sonne|osxvan51: for your demo?17:24
sonne|osxvan51: I mean at least mention which languages it can detect and what it is using underneath17:25
van51sonne|osx: yeah I can do that17:25
shogun-buildbotbuild #1722 of deb2 - static_interfaces is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb2%20-%20static_interfaces/builds/172217:25
sonne|osxvan51: regarding the kitchen sinks17:25
sonne|osxvan51: correct me if I am wrong but as far as I understood you create a number of random vectors in an inital step right?17:26
van51sonne|osx: yeap17:27
lisitsynI put russian text17:27
lisitsyn:D17:27
-!- new_lido [~walid@41.218.172.68] has joined #shogun17:27
lisitsynno response17:27
sonne|osxlisitsyn: try greek :D17:27
sonne|osxvan51: and then you compute dot products for each example against the random examples right?17:28
van51sonne|osx: not always. the Random Fourier features for instance do need dot products and then a cosine transformation17:29
van51sonne|osx: but one could just want to calculate simple decision stump17:30
sonne|osxvan51: ok so some operation on top of the dot product underneath right?17:30
van51sonne|osx: no, I mean no dot products at all17:31
van51sonne|osx: in their paper for instance they created two-dimensional random vectors17:31
sonne|osxand then?17:31
van51sonne|osx: and then for every example and for every parameter vector they calculated x[ random_vec[0]] > random_vec[1]17:32
van51and if true they would assign 1, else 017:32
wikinglisitsyn: yeah what ya need?17:32
sonne|osxvan51: so the common step is to generate some random data and then an operation on top of it to generate the features17:33
sonne|osxvan51: now we don't want to genearte the features explicitly17:33
wikingbesser82: yeah static shogun lib would be good17:33
sonne|osxvan51: but rather only implement the functions needed by dotfeatures17:33
sonne|osxvan51: like dense_dot and add_to_dense_vec17:33
lisitsynwiking: well if we want windows17:34
lisitsynwiking: we probably need to put that shit everyhwheere17:34
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting]17:34
-!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun17:34
sonne|osxvan51: I don't know if there is a better way than always having to write dense_dot etc17:34
wikinglisitsyn: hahaha if we want windows we should start forcing sonne|osx not to try to use linux commands like stat in EXECUTE_COMMAND and such :)17:35
van51sonne|osx: you mean for sub-classees?17:35
lisitsynwiking: well thats' one part17:35
sonne|osxvan51: well except for the case where you need to compute the dot product explictly wrt the random features17:35
lisitsynwiking: but now all the things are invisible from dll17:35
shogun-notifier-shogun: Viktor Gal :develop * 522ce15 / tests/unit/CMakeLists.txt: https://github.com/shogun-toolbox/shogun/commit/522ce15aaaa2f541f5742794b5bc92c707cfc15117:35
shogun-notifier-shogun: remove add_dependency for gmock/gtest as target_link_libraries should17:35
shogun-notifier-shogun: add dependency already for them17:35
lisitsyndue to no __declspec(dllexport)17:36
sonne|osxvan51: yes maybe one could inherit a bit for subclasses17:36
lisitsynokay be back late17:36
lisitsynrr17:36
-!- lisitsyn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.]17:36
sonne|osxwiking: well then help me how to start a lengthy python program in there17:36
wikingCOMMAND ${PYTHON_EXECUTABLE} ..17:37
sonne|osxwiking: I didn't manage to run the old python script from inside cmake to determine the date of the file17:37
sonne|osxwiking: well sth like python -c 'adfadf'17:37
van51sonne|osx: if we maintain this Phi function approach then subclasses could inherit the random initialization and provide their own custom functions class, if they don't want to override the dot() methods.17:38
shogun-buildbotbuild #155 of osx1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/osx1%20-%20libshogun/builds/155  blamelist: Soeren Sonnenburg <sonne@debian.org>, Fernando Iglesias <fernando.iglesiasg@gmail.com>17:38
sonne|osxvan51: so they share the random initialization and then provide what else?17:39
sonne|osxvan51: a way to compute Phi(x)?17:39
van51sonne|osx: yeah, currently the way I have it, it expects a subclass of CRKSFunctions to be passed as an argument17:40
van51sonne|osx: and that object encapsulates the prob distribution function used for the random vectors and the Phi17:40
shogun-buildbotbuild #1837 of deb3 - modular_interfaces is complete: Failure [failed examples and unit tests]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1837  blamelist: Soeren Sonnenburg <sonne@debian.org>, Fernando Iglesias <fernando.iglesiasg@gmail.com>17:41
sonne|osxvan51: but then you have to compute the full vector before you can compute Phi(x) * dense_vec17:41
sonne|osxetc17:41
shogun-buildbotbuild #94 of precise - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/precise%20-%20libshogun/builds/9417:41
sonne|osxand that is slow17:41
sonne|osxso it would be better to isolate classes where we can do the dense_dot in a general way17:42
sonne|osxand for all others I guess there is not much benefit of a common superclass17:43
sonne|osxand they should be done as dotfeatures directly17:43
-!- sonne|osx [~sonne@82.113.121.99] has quit [Quit: sonne|osx]17:46
van51sonne|osx: ok i get your point17:47
van51sonne|osx: so I'll create a base class for those cases, with some pre() and post() functions w.r.t to the dot operation17:48
besser82bye, folks! I'm out for today! See y'all tomorow :)17:54
-!- besser82 [~besser82@fedora/besser82] has quit [Quit: Verlassend]17:54
shogun-buildbotbuild #1491 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1491  blamelist: Soeren Sonnenburg <sonne@debian.org>, Fernando Iglesias <fernando.iglesiasg@gmail.com>17:56
-!- besser82 [~besser82@77-21-40-118-dynip.superkabel.de] has joined #shogun18:16
-!- besser82 [~besser82@77-21-40-118-dynip.superkabel.de] has quit [Changing host]18:16
-!- besser82 [~besser82@fedora/besser82] has joined #shogun18:16
-!- iglesiasg [~iglesias@2001:6b0:1:1da0:cd65:59ce:a8b5:d5f3] has quit [Quit: Ex-Chat]18:21
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting]18:28
-!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun18:29
thoralfUgh.  Found a case where libqp_splx_solver does not converge (and QPSolverMaxIter is set to 0xFFFFFFFF)18:30
thoralfIt will take days to fix it. :(18:30
shogun-buildbotbuild #1492 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1492  blamelist: Viktor Gal <viktor.gal@maeth.com>18:32
-!- lambday [67157c37@gateway/web/freenode/ip.103.21.124.55] has quit []18:32
-!- foulwall [~zhengyang@114.255.40.22] has quit [Ping timeout: 245 seconds]18:33
-!- travis-ci [~travis-ci@ec2-54-211-149-66.compute-1.amazonaws.com] has joined #shogun18:38
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1151082218:38
-!- travis-ci [~travis-ci@ec2-54-211-149-66.compute-1.amazonaws.com] has left #shogun []18:38
wikingokeeey18:41
wikingR modular is finally fixed18:42
wiking\o/18:42
thoralfwiking: Applause :)18:42
wikingor at least it works now18:42
wikingthoralf: what's with bmrm?18:42
wikingwhere do u use it?18:42
thoralfStructured learning.18:42
wikingbrb18:42
wikingyeah i figured18:42
wiking;P18:42
shogun-buildbotbuild #1838 of deb3 - modular_interfaces is complete: Failure [failed test python modular test r modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1838  blamelist: Viktor Gal <viktor.gal@maeth.com>18:42
wikingbut what's the exact application18:42
shogun-notifier-shogun: Viktor Gal :develop * 89e776a / examples/undocumented/r_modular/ (2 files): https://github.com/shogun-toolbox/shogun/commit/89e776aa8bf9e11562391ca445bf8b28b4093b2818:43
shogun-notifier-shogun: Fix typo in kernel_comm_ulong_string_modular.R18:43
shogun-notifier-shogun: Fix matrix creation in classifier_svmlin_modular.R18:43
wikingbtw: i might have fixed as well the problem with the gmock dependency <-- thoralf18:43
thoralfLearning multilabel assignments (or taxonomic multilabel assignments).18:43
-!- votjakovr [~votjakovr@host-46-241-3-209.bbcustomer.zsttk.net] has quit [Quit: Good bye!]18:43
thoralfHuge number of classes (>7000), data with millions of dimensions.18:44
wikingoh cool18:53
wikingwhich solver do u use ?18:54
wikingwithin dualbmrm?18:54
thoralfYes, exactly.18:54
thoralflibbmrm18:54
wikingah18:54
wikingwhy not libpp3bm or such?18:54
wikingu could give a try to libncbm18:55
wikingas well18:55
wikingncbm should converge faster ...18:55
besser82wiking: free?18:55
wikingyes here18:56
besser82wiking: Are you OK with me to move lots of the automagic done in external scripts into the CMake-buildsys?18:56
besser82wiking: like generation of examples/documented and such18:57
besser82wiking: that would make complete out-of-tree much easier :)18:57
besser82wiking:                                          ^^ builds18:58
thoralfwiking: libpp3bm&co are failing on my problem - to me a while to fix libbmrm.  Needed at least one running thing, but I'll try the other soon.18:58
-!- HeikoS [~heiko@nat-179-139.internal.eduroam.ucl.ac.uk] has quit [Remote host closed the connection]18:58
wikingthoralf: have u tried ncbm?18:58
thoralfwiking: No.18:58
wikingthoralf: if u have time give it a go18:58
wikingthoralf: as it's quite differenet from the other libbm*18:58
wikingbesser82: examples/documented script is not running now at all18:59
-!- HeikoS [~heiko@nat-179-139.internal.eduroam.ucl.ac.uk] has joined #shogun18:59
-!- mode/#shogun [+o HeikoS] by ChanServ18:59
wikingbesser82: i mean it's not executed from cmake...18:59
besser82wiking: allright18:59
wikingbesser82: i've tried to avoid using shell scripts from cmake...18:59
wikingbesser82: but there's no way around python scrips unfortunately18:59
wiking*scripts18:59
besser82wiking: maybe, maybe not19:00
thoralfwiking: Heh.  At least it doesn't crash on my toy example I'm using. ;)19:00
wikingbesser82: well the thing is that we use jinja2 for templating... so there's really no way around that19:00
wikingthoralf: hehe :)19:00
wikingbesser82: and we are adding even more jinja2 stuff19:00
wikingso python interpreter is a must19:00
besser82wiking: there's some jinja2-wrapper for cmake, I think19:00
besser82wiking: there's some jinja2-wrapper for cmake, I thi19:00
besser82wiking: seen such somewhere...19:01
wikingbesser82: well .. it's more or less the same  ;)19:01
besser82wiking: allright19:01
wikingif it's a wrapper19:01
wikingthoralf: heheh yeah give it a go... i've added that quite some time ago, but it was working nicely19:01
besser82wiking: or possibly it was some direct integration I've seen19:01
besser82wiking: I don't remember any calls to external py-scripts from that :)19:02
wikingokie19:02
-!- gsomix [~gsomix@188.168.3.89] has joined #shogun19:02
wikingi'm just saying19:02
wikingthat some stuff can be done within cmake, but some stuff i'm afraid will need to be out of cmake19:02
besser82wiking: but never the less the other option would be to mod those py-script to take indir outdir args :)19:03
gsomixsonney2k, sonne|work hi. I'm alive and working. will send code tomorrow or at this night. little busy for now.19:03
gsomixwiking, no news. I'll see a little little later. :)19:03
shogun-buildbotbuild #1839 of deb3 - modular_interfaces is complete: Failure [failed examples and unit tests]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1839  blamelist: Viktor Gal <viktor.gal@maeth.com>19:03
besser82wiking: so automagic files will get generated in seperate build-dir :)19:03
thoralfwiking: Oh, I see it needs lots of calls to machine->risk in each iteration.19:04
thoralfwiking: That's quite expensive in my case.19:04
wikingpff... this dependency is weird ...19:04
wiking:(19:04
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting]19:05
-!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun19:05
wikingshogun-notifier-: force build --branch=develop 'deb3 - modular_interfaces'19:05
wikingshogun-buildbot: force build --branch=develop 'deb3 - modular_interfaces'19:05
shogun-buildbotbuild #1840 forced19:05
shogun-buildbotI'll give a shout when the build finishes19:05
gsomixsonney2k, sonne|work btw libarchive is a great idea.19:05
wikingthoralf: :(19:05
wikingthoralf: caching?19:05
besser82gsomix: that came from me :)19:05
gsomixbesser82, cool!19:06
wikinggsomix: imho it'd be better to have a CFile(int fd)19:06
besser82gsomix: soeren asked me during yesterday19:06
wikingi mean not better but would be necessary as well19:06
wikingsince that'd allow piping19:07
-!- lisitsyn [~lisitsyn@fb2-lo1.global63.net] has joined #shogun19:07
gsomixwiking, yep, piping is nice.19:07
thoralfwiking: In my case it seems to converge quite fast, nice.  Maybe these risk calls are worth it.19:07
thoralfwiking: Don't do caching, I'm already using plenty of ram.19:08
thoralfAdding more cores makes more sense. :D19:08
wikingheheh ok19:08
gsomixwiking, but I think libarchive will be more useful in modular scripts. e.g, for Fernando's problem19:08
wikingi was just saying thoralf19:08
wikinggsomix: i cannot believe that it wasn't possible19:08
wikingpiping gzip's output into cfile19:09
besser82gsomix: when it's about scripts or such there's pylibarchive, too :)19:09
gsomixwiking, through python?19:09
gsomixmaybe19:09
wikinggsomix: yeah i mean it shouldn't matter... file descriptor is a file descriptor19:09
wiking;)19:09
wikingit's kernel level19:10
gsomixwiking, I see.19:10
thoralfwiking: btw, caching and multiprocessing are not playing well... the more processes, the more synchronization overhead :(19:10
wikingthoralf: yeah ... didn't know your exact application19:10
wikingjust that you have problem with calling risk too many times19:10
wiking:P19:10
wikinglet's see if we can finally close issue 159119:11
thoralfwiking: I won't say it's a problem - just more expensive. :)19:12
wikingthoralf: get some loose some :)19:13
thoralfatm I need 2-3 days of training with 16 cores - so adding a factor of 10 would be annoying, but not impossible.19:13
wikingbut at least you said that it converges :P19:13
wikingin contrary to the other solvers19:13
-!- travis-ci [~travis-ci@ec2-54-221-10-21.compute-1.amazonaws.com] has joined #shogun19:17
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1151124419:17
-!- travis-ci [~travis-ci@ec2-54-221-10-21.compute-1.amazonaws.com] has left #shogun []19:17
shogun-buildbotbuild #1493 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1493  blamelist: Viktor Gal <viktor.gal@maeth.com>19:18
shogun-buildbotbuild #1840 of deb3 - modular_interfaces is complete: Failure [failed examples and unit tests]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/184019:20
wikinghaha lol19:21
wikingshogun-buildbot: force build --branch=develop 'deb3 - modular_interfaces'19:22
shogun-buildbotbuild #1841 forced19:22
shogun-buildbotI'll give a shout when the build finishes19:22
-!- FSCV [~FSCV@66.187.72.219] has joined #shogun19:28
-!- besser82 [~besser82@fedora/besser82] has quit [Quit: Verlassend]19:34
-!- HeikoS [~heiko@nat-179-139.internal.eduroam.ucl.ac.uk] has quit [Quit: Leaving.]19:51
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting]20:12
-!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun20:12
-!- travis-ci [~travis-ci@ec2-54-211-149-66.compute-1.amazonaws.com] has joined #shogun20:16
travis-ci[travis-ci] it's Fernando Iglesias's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1151274920:16
-!- travis-ci [~travis-ci@ec2-54-211-149-66.compute-1.amazonaws.com] has left #shogun []20:16
shogun-buildbotbuild #1841 of deb3 - modular_interfaces is complete: Failure [failed test python modular test ruby modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/184120:22
shogun-notifier-shogun: Viktor Gal :develop * 94f2623 / src/shogun/io/ (4 files): https://github.com/shogun-toolbox/shogun/commit/94f2623ef12c2c351dcabeaf5b50164a639f4a3020:23
shogun-notifier-shogun: Add file descriptor based constructor to File and CSVFile20:23
-!- travis-ci [~travis-ci@ec2-54-224-116-220.compute-1.amazonaws.com] has joined #shogun20:48
travis-ci[travis-ci] it's Soeren Sonnenburg's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1151280720:48
-!- travis-ci [~travis-ci@ec2-54-224-116-220.compute-1.amazonaws.com] has left #shogun []20:48
-!- zxtx [~zv@ip-64-134-197-230.public.wayport.net] has quit [Ping timeout: 276 seconds]20:51
-!- lambday [67157e37@gateway/web/freenode/ip.103.21.126.55] has joined #shogun20:53
-!- sonne|osx [~sonne@f053045233.adsl.alicedsl.de] has joined #shogun20:55
wikingsonne|osx: ping20:56
sonne|osxwiking: hi20:56
sonne|osxvan51: sry had an appointment w/ a doctor and had to leave abruptly20:56
van51sonne|osx: no problem20:57
wikingsonne|osx: ok so now we only have fernando's integration as a remaining error in modular interfaces20:57
van51sonne|osx: hope all is well :)20:57
sonne|osxvan51: yes just some vaccination20:57
sonne|osxwiking: so why doesn't he fix it then?20:58
sonne|osxor is he on vacations?20:58
van51sonne|osx: cool then :)20:58
wikingsonne|osx: i think he had several rounds for fixing it20:59
wikingbut none of them really succeeded20:59
sonne|osxwiking: so is the code working well and it just needs a data update or is there some bug?20:59
wikingsonne|osx: afaik it's the same as in case of the other two tests that are failing on travis...21:00
wikingsonne|osx: a minimum rounding error21:00
wikingshogun-notifier-: force build --branch=developer 'clang34 - thread analysis'21:00
wikingshogun-buildbot: force build --branch=developer 'clang34 - thread analysis'21:01
shogun-buildbotbuild #5 forced21:01
shogun-buildbotI'll give a shout when the build finishes21:01
shogun-buildbotbuild #5 of clang34 - thread analysis is complete: Failure [failed git]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/521:01
wikingoh nice21:01
wikingshogun-buildbot: force build --branch=develop 'clang34 - thread analysis'21:01
shogun-buildbotbuild #6 forced21:01
shogun-buildbotI'll give a shout when the build finishes21:01
sonne|osxvan51: when is your semester starting again?21:05
van51sonne|osx: somewhere in the beginning of Octomber21:06
van51October*21:06
sonne|osxok21:07
wikinglol: 1006      9091 99.3  0.2 104153296980 73992 ?  Rl   21:09   0:36 /home/buildslave/clang34_-_thread_analysis/build/build/tests/unit/shogun-unit-test --gtest_filter=Random.*21:10
wiking:D21:10
wikingthat's a nice amount of VIRT21:10
shogun-buildbotbuild #6 of clang34 - thread analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/621:15
wikingshogun-buildbot: force build --branch=develop 'clang34 - static analysis'21:18
shogun-buildbotbuild #12 forced21:18
shogun-buildbotI'll give a shout when the build finishes21:18
wikingsonne|osx: can u move shogun-image.fs.sfs in my home on 7nn.de to shogun-toolbox.org www dir... plz21:19
@sonney2kwiking, what sub dir?21:21
@sonney2kwiking, images?21:22
@sonney2kwiking, we currently have data releases and snapshots21:22
wikingsonney2k: do we have downloads? :)21:24
wikingok i see21:24
@sonney2kwiking, ??21:24
@sonney2kwiking, just try ftp shogun-toolbox.org21:24
wikingok21:25
wikinghttp://shogun-toolbox.org/archives/shogun/images/shogun-image.fs.sfs21:25
wikingthis works21:25
wiking:>21:25
@sonney2kyou have write permissions in /var/www/shogun-toolbox.org/archives/shogun/images now21:25
wikingsonney2k: cool thnx21:25
wikingi'll put a link to this image somewhere21:26
@sonney2kwiking, README or website?21:26
wikingsonney2k: both21:26
wikingman i cannot believe that r modular is actually functional21:27
wiking:)21:27
wikingi've checked some outputs21:27
wikingthey were reasonable21:27
@sonney2kyeah and there are complex examples running even21:27
wikingsonney2k: can u gdb on cygwin?21:29
@sonney2kwiking, when running the LMNN example I get ./metric_lmnn_modular.py:22: RuntimeWarning: [WARN] In file /home/sonne/Documents/priv/repositories/shogun/src/shogun/metric/LMNNImpl.cpp line 287: Maximum number of iterations reached before convergence.21:29
@sonney2kso maybe that is the reason for the failur21:29
@sonney2ke21:29
wikingsonney2k: only fernando knows the answer :)21:30
shogun-notifier-shogun-data: Soeren Sonnenburg :master * 5be9693 / testsuite/tests/metric_lmnn_modular0.txt: https://github.com/shogun-toolbox/shogun-data/commit/5be969375baca56853b5fdd4d2f5ff4ec4c156fb21:31
shogun-notifier-shogun-data: rerun generator for lmnn21:31
shogun-notifier-shogun: Soeren Sonnenburg :develop * 5f0953f / data: https://github.com/shogun-toolbox/shogun/commit/5f0953f1d7b31d3857101f010cddda912cf13ea121:31
shogun-notifier-shogun: require new data version for LMNN21:31
@sonney2kwiking, well create an issue and assign him :D21:32
@sonney2kwiking, we somehow have to fix the 2 remaining r_modular examples21:32
@sonney2kwiking, or did you?21:33
@sonney2kwiking, you might consider turning of integration tests on travis IMHO it is sufficient that we run this stuff on the buildbots21:35
@sonney2kwiking, I mean travis is good to get a rough picture if sth works and IMHO that is all we want at this stage leaving detailed analysis to our own BB21:36
@sonney2kgsomix, ping ping?21:38
shogun-buildbotbuild #1494 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1494  blamelist: Viktor Gal <viktor.gal@maeth.com>21:39
@sonney2kvan51, ok so we need a plan :)21:40
@sonney2kvan51, I think you should mostly finish up and maybe plan what could be done in the future (yes please stick with us!) and help us to get the release going.21:41
-!- travis-ci [~travis-ci@ec2-54-224-116-220.compute-1.amazonaws.com] has joined #shogun21:43
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1151438421:43
-!- travis-ci [~travis-ci@ec2-54-224-116-220.compute-1.amazonaws.com] has left #shogun []21:43
shogun-buildbotbuild #157 of osx1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/osx1%20-%20libshogun/builds/157  blamelist: Viktor Gal <viktor.gal@maeth.com>21:49
van51sonney2k: sure! I was planning on getting on with Fast Food after I was done with the documentation needed21:49
@sonney2kahh wiking you fixed them cool!21:49
van51sonney2k: after that or simultaneously I can help with whatever else is needed21:49
@sonney2kvan51, nice!21:49
@sonney2kvan51, considering that you are doing 2 gsoc projects at once you don't need to rush - you are in time for both :)21:50
@sonney2kanyway, we should really try to get things to stabilize since the release is coming21:51
@sonney2kI am not sure when exactly but I think we should in 1-2 weeks branch of a 3.0.0 release branch and then everyone can still continue in develop while we try to get 3.0.0 stable (no more feature commits)21:52
shogun-buildbotbuild #158 of osx1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/osx1%20-%20libshogun/builds/158  blamelist: Soeren Sonnenburg <sonne@debian.org>21:56
shogun-buildbotbuild #1842 of deb3 - modular_interfaces is complete: Failure [failed test python modular test ruby modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1842  blamelist: Viktor Gal <viktor.gal@maeth.com>22:01
-!- lambday [67157e37@gateway/web/freenode/ip.103.21.126.55] has quit [Ping timeout: 250 seconds]22:05
shogun-buildbotbuild #12 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/1222:05
shogun-notifier-shogun: Soeren Sonnenburg :develop * 05329ba / src/shogun/io/streaming/StreamingAsciiFile.cpp: https://github.com/shogun-toolbox/shogun/commit/05329ba717fdad60ea1c8645cc702923c6a4bde222:08
shogun-notifier-shogun: set C locale in streamingasciifile22:08
* new_lido is amazed by the shogun bot :D22:19
-!- travis-ci [~travis-ci@ec2-54-221-10-21.compute-1.amazonaws.com] has joined #shogun22:20
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1151703422:20
-!- travis-ci [~travis-ci@ec2-54-221-10-21.compute-1.amazonaws.com] has left #shogun []22:20
shogun-buildbotbuild #159 of osx1 - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/osx1%20-%20libshogun/builds/15922:22
-!- Heiko__ [05453f70@gateway/web/freenode/ip.5.69.63.112] has joined #shogun22:24
@sonney2knew_lido, it is building all night long22:30
@sonney2kand day!22:31
new_lidofirst time to come ^_^22:31
new_lidoi've been in #blendercoders for 3 months ... and i really think this way is so creative :P22:32
new_lido#newbie to IRC22:32
shogun-buildbotbuild #1495 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1495  blamelist: Soeren Sonnenburg <sonne@debian.org>22:33
shogun-buildbotbuild #99 of precise - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/precise%20-%20libshogun/builds/99  blamelist: Soeren Sonnenburg <sonne@debian.org>22:35
shogun-buildbotbuild #1843 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/184322:40
@sonney2knew_lido, welcome to #shogun then22:43
@sonney2kalright people22:43
* sonney2k tired22:43
new_lidothx :D22:43
@sonney2kcu22:43
new_lidolater :)22:43
-!- travis-ci [~travis-ci@ec2-54-211-149-66.compute-1.amazonaws.com] has joined #shogun22:54
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1152065722:54
-!- travis-ci [~travis-ci@ec2-54-211-149-66.compute-1.amazonaws.com] has left #shogun []22:54
-!- new_lido [~walid@41.218.172.68] has quit [Quit: Leaving]23:16
shogun-buildbotbuild #1496 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1496  blamelist: Soeren Sonnenburg <sonne@debian.org>23:18
shogun-buildbotbuild #13 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/13  blamelist: Soeren Sonnenburg <sonne@debian.org>23:22
shogun-buildbotbuild #1213 of rpm1 - libshogun is complete: Failure [failed compile]  Build details are at http://buildbot.shogun-toolbox.org/builders/rpm1%20-%20libshogun/builds/1213  blamelist: Soeren Sonnenburg <sonne@debian.org>23:31
shogun-buildbotbuild #2 of clang34 - undefined behaviour analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/2  blamelist: Soeren Sonnenburg <sonne@debian.org>23:33
-!- travis-ci [~travis-ci@ec2-54-224-116-220.compute-1.amazonaws.com] has joined #shogun23:34
travis-ci[travis-ci] it's Soeren Sonnenburg's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1152308723:35
-!- travis-ci [~travis-ci@ec2-54-224-116-220.compute-1.amazonaws.com] has left #shogun []23:35
-!- hushell [~hushell@c-98-232-178-161.hsd1.or.comcast.net] has quit [Quit: WeeChat 0.3.7]23:46
shogun-buildbotbuild #7 of clang34 - thread analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/7  blamelist: Soeren Sonnenburg <sonne@debian.org>23:48
Heiko__wiking: around?23:50
--- Log closed Thu Sep 19 00:00:05 2013