IRC logs of #shogun for Sunday, 2012-11-18

--- Log opened Sun Nov 18 00:00:17 2012
-!- travis-ci [~travis-ci@ec2-23-20-88-236.compute-1.amazonaws.com] has joined #shogun00:04
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/324496200:04
-!- travis-ci [~travis-ci@ec2-23-20-88-236.compute-1.amazonaws.com] has left #shogun []00:04
wikingmeh00:16
blackburnwiking: take a look https://dl.dropbox.com/u/10139213/ml/road.avi  :D00:18
wikingkinect?00:25
blackburnno, stereo00:25
blackburnjust stereo data00:26
wikingcool00:26
blackburnit came from daimler stereo pedestrian detection dataset00:26
blackburnI just set up libelas with that data00:26
wikingblackburn: http://nzsl.vuw.ac.nz/signs/315800:50
blackburnwtf00:51
blackburn:D00:51
wikinglol http://www.mkomo.com/cost-per-gigabyte00:52
-!- blackburn [~blackburn@109.226.125.245] has left #shogun []01:01
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has quit [Quit: Ex-Chat]02:53
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has joined #shogun02:54
shogun-buildbotbuild #152 of nightly_none is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_none/builds/15203:16
shogun-buildbotbuild #146 of nightly_all is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_all/builds/14603:26
shogun-buildbotbuild #175 of nightly_default is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/17504:04
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has quit [Quit: Ex-Chat]06:42
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has joined #shogun06:42
-shogungit:#shogun- [shogun] sonney2k pushed 2 new commits to master: https://github.com/shogun-toolbox/shogun/compare/3d23faabc887...2e5dcdb2ff9806:57
-shogungit:#shogun- shogun/master d9ac31b Soeren Sonnenburg: mark GMM as broken - serialization is not working due to std::vector not supported in serialization06:57
-shogungit:#shogun- shogun/master 2e5dcdb Soeren Sonnenburg: remove placement new and inplace constructor calls for SGVector<type>*06:57
@sonney2kwiking, as soon as realloc is implemented for SGVector you can safely use Dynarray instead of std::vector for this one particular patch that has std::vector in .h06:59
-!- travis-ci [~travis-ci@ec2-23-20-88-236.compute-1.amazonaws.com] has joined #shogun07:00
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/324811907:00
-!- travis-ci [~travis-ci@ec2-23-20-88-236.compute-1.amazonaws.com] has left #shogun []07:00
-!- travis-ci [~travis-ci@ec2-23-20-88-236.compute-1.amazonaws.com] has joined #shogun07:25
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/324819007:25
-!- travis-ci [~travis-ci@ec2-23-20-88-236.compute-1.amazonaws.com] has left #shogun []07:25
shogun-buildbotbuild #535 of deb2 - static_interfaces is complete: Failure [failed test libshogun]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb2%20-%20static_interfaces/builds/535  blamelist: Soeren Sonnenburg <sonne@debian.org>07:29
shogun-buildbotbuild #74 of ubu1 - libshogun is complete: Failure [failed test]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/ubu1%20-%20libshogun/builds/74  blamelist: Soeren Sonnenburg <sonne@debian.org>07:36
shogun-buildbotbuild #75 of ubu1 - libshogun is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/ubu1%20-%20libshogun/builds/7507:54
shogun-buildbotbuild #647 of deb3 - modular_interfaces is complete: Failure [failed test libshogun]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/647  blamelist: Soeren Sonnenburg <sonne@debian.org>07:54
shogun-buildbotbuild #536 of deb2 - static_interfaces is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb2%20-%20static_interfaces/builds/53608:00
shogun-buildbotbuild #648 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/64808:10
-!- blackburn [~blackburn@109.226.125.245] has joined #shogun09:55
-!- blackburn [~blackburn@109.226.125.245] has quit [Quit: Leaving.]10:18
-!- blackburn [~blackburn@109.226.125.245] has joined #shogun10:50
wikingyo11:10
wikingsonney2k: wtf are u doing at 7am in the morning :D11:10
wikingi've realised that the problem that thing solves11:10
wikingi can use libqp for solving the minimization11:11
@sonney2kwiking, 7am is extremely late for me11:53
@sonney2kwiking, I meant the problem with the std::vector<SGVector> in headers11:54
@sonney2kwiking, you can just replace std::vector with DynArray<SGVector> in .cpp / .h11:54
wikingah ok11:57
wikingi'll do it11:57
wikingbut i finish first my linesearching11:57
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/afb2445a8423619c03f105b99556d39a6f0f69d214:30
-shogungit:#shogun- shogun/master afb2445 Soeren Sonnenburg: add additional old_size parameter to sg_realloc to enable new[] based resizing14:30
shogun-buildbotbuild #534 of bsd1 - libshogun is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/bsd1%20-%20libshogun/builds/534  blamelist: Soeren Sonnenburg <sonne@debian.org>14:42
blackburnsonney2k: wiking: http://www.youtube.com/watch?v=BTqbxew7g8w14:46
blackburnimplemented with bash lol14:50
-!- travis-ci [~travis-ci@ec2-23-20-88-236.compute-1.amazonaws.com] has joined #shogun14:58
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/325060114:58
-!- travis-ci [~travis-ci@ec2-23-20-88-236.compute-1.amazonaws.com] has left #shogun []14:58
shogun-buildbotbuild #537 of deb2 - static_interfaces is complete: Failure [failed test libshogun]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb2%20-%20static_interfaces/builds/537  blamelist: Soeren Sonnenburg <sonne@debian.org>15:02
shogun-buildbotbuild #76 of ubu1 - libshogun is complete: Failure [failed test]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/ubu1%20-%20libshogun/builds/76  blamelist: Soeren Sonnenburg <sonne@debian.org>15:14
shogun-buildbotbuild #649 of deb3 - modular_interfaces is complete: Failure [failed test libshogun]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/649  blamelist: Soeren Sonnenburg <sonne@debian.org>15:31
-!- blackburn [~blackburn@109.226.125.245] has quit [Quit: Leaving.]16:17
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has quit [Quit: Ex-Chat]16:21
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has joined #shogun16:21
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/da636880a8e68f1c48b98224614204bf0ad622e020:49
-shogungit:#shogun- shogun/master da63688 Soeren Sonnenburg: fix compile error on *bsd20:49
shogun-buildbotbuild #535 of bsd1 - libshogun is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/bsd1%20-%20libshogun/builds/53521:04
-!- travis-ci [~travis-ci@ec2-107-21-145-40.compute-1.amazonaws.com] has joined #shogun21:16
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/325427421:16
-!- travis-ci [~travis-ci@ec2-107-21-145-40.compute-1.amazonaws.com] has left #shogun []21:16
shogun-buildbotbuild #538 of deb2 - static_interfaces is complete: Failure [failed test libshogun]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb2%20-%20static_interfaces/builds/538  blamelist: Soeren Sonnenburg <sonne@debian.org>21:24
-!- blackburn [~blackburn@37.61.179.35] has joined #shogun21:32
shogun-buildbotbuild #77 of ubu1 - libshogun is complete: Failure [failed test]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/ubu1%20-%20libshogun/builds/77  blamelist: Soeren Sonnenburg <sonne@debian.org>21:34
shogun-buildbotbuild #650 of deb3 - modular_interfaces is complete: Failure [failed test libshogun]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/650  blamelist: Soeren Sonnenburg <sonne@debian.org>21:54
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun22:16
n4nd0it's a pity they didn't want shogun in the doc camp :(22:17
n4nd0let's see if we get to make it for another year!22:17
@sonney2kyeah but we didn't get into gsoc the first year we applied so... :)22:35
@sonney2kblackburn, wiking - does classifier_multiclasslibsvm crash on your machine?22:37
@sonney2kb'cause on mine it is all fine22:37
blackburnich will try22:38
blackburnsonney2k: yes it is working22:42
@sonney2kwhy is it not on the buildbot?22:43
@sonney2kvalgrind is happy too here...22:44
@sonney2kblackburn, could you try to run ./classifier_mklmulticlass22:44
blackburnthat one KADABOOMS22:45
@sonney2khere too22:45
@sonney2kbut I don't get why22:45
@sonney2kit dies in kernel/CombinedKernel.cpp:50922:45
@sonney2kwhen you look in gdb22:45
blackburnvalgrindding22:45
@sonney2kit did allocate memory for a single additional thread22:46
@sonney2kwhich is totally fine22:46
@sonney2kblackburn, I have an idea22:59
@sonney2kcould it be that we have called multiple types S_THREAD_PARAM22:59
@sonney2k?22:59
@sonney2kso there is some mixup?22:59
blackburnhmm22:59
blackburnyeah it could be23:00
@sonney2kwiking, clang?23:04
@sonney2kblackburn, indeed23:06
@sonney2krenaming types solves the issues23:06
@sonney2kthat is incredibly scary23:06
@sonney2kblackburn, could you check if this happens with clang too?23:07
blackburnokay23:07
@sonney2kI see no way other than appending class names to this struct23:07
@sonney2kbut I don't like it23:08
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/be857cd3501511db807bd2eb926350a25911f6f623:08
-shogungit:#shogun- shogun/master be857cd Soeren Sonnenburg: append class name to S_THREAD_PARAMS to avoid compiler confusion in...23:08
blackburnsonney2k: we should enable -Weffc++23:09
blackburn:D23:09
blackburnsonney2k: could you remind me how to change compiler in configure?23:10
@sonney2k--cxx=clang++ --cc=clang23:11
blackburnthanks23:11
@sonney2kblackburn, heh23:11
@sonney2kWhen selecting this option, be aware that the standard library23:11
@sonney2k           headers do not obey all of these guidelines; use grep -v to filter23:11
@sonney2k           out those warnings.23:11
blackburnsonney2k: there will be like23:12
blackburnone million or so warnings23:12
@sonney2kexcellent23:12
blackburnsonney2k: did you ever try to compile shogun with icpc?23:12
blackburnI will try next to clang23:12
blackburnI installed intel composer xe as it is free for an year23:13
@sonney2k           ยท   Item 7:  Never overload "&&", "||", or ",".23:13
@sonney2kone could overload , ?23:13
@sonney2kwtf?23:13
@sonney2kcan one overload . too?23:13
blackburnnot sure about .23:13
blackburnbut -> for sure23:13
blackburnsonney2k: yes fails with clang too23:16
@sonney2khmmhh is this a feature?23:16
@sonney2kI don't really understand what the compiler will do if it has to call foo<MyType>(args)23:17
@sonney2kwith MyType being defined only in .cpp files but a couple of times23:17
blackburnsonney2k: what do you mean?23:18
@sonney2kblackburn, well the issue for our crash is exactly that23:18
@sonney2kit chooses some S_THREAD_PARAM from some class23:18
@sonney2kbut when called from some other class where S_THREAD_PARAM has a totally different meaning it kabooms of course23:19
-!- travis-ci [~travis-ci@ec2-107-21-145-40.compute-1.amazonaws.com] has joined #shogun23:20
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/325593623:20
-!- travis-ci [~travis-ci@ec2-107-21-145-40.compute-1.amazonaws.com] has left #shogun []23:20
blackburn./configure --cxx=/opt/intel/composerxe/bin/icpc --cc=/opt/intel/composerxe/bin/icc --disable-optimization --interfaces=,23:22
blackburnRunning configure for SHOGUN version 2.1.023:22
blackburnChecking for C Compiler ... gcc 4.6, ok23:22
@sonney2kerrm23:22
@sonney2k./configure --cc=clang --cxx=clang++23:23
@sonney2kRunning configure for SHOGUN version 2.1.023:23
@sonney2kChecking for C Compiler ... clang 4.2.123:23
@sonney2kChecking for C Compiler flags ... -fPIC -g -Wall -finline-functions -funroll-loops23:23
@sonney2kChecking for C++ Compiler ... clang 4.2.123:23
@sonney2kChecking for C++ Compiler flags ... -fPIC -g -Wall -finline-functions -funroll-loops23:23
blackburnit works with clang23:23
blackburnI guess it works with absolute path in a bad way23:23
@sonney2k./configure --cc=`which clang` --cxx=`which clang++`23:24
@sonney2kRunning configure for SHOGUN version 2.1.023:24
@sonney2kChecking for C Compiler ... clang 4.2.123:24
@sonney2kno works too23:24
blackburnyeah23:24
blackburnstill doesn't work with icpc23:24
blackburnsonney2k: comma overloading is useful for23:26
blackburna << 1,2,3,4,5;23:26
shogun-buildbotbuild #78 of ubu1 - libshogun is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/ubu1%20-%20libshogun/builds/7823:36
@sonney2kblackburn, look at this: http://www.shogun-toolbox.org/buildbot/builders/ubu1%20-%20libshogun/builds/78/steps/test/logs/stdio23:37
@sonney2klots of tests fail - but none fails here23:37
@sonney2kblackburn, which gcc version do you have?23:39
shogun-buildbotbuild #414 of cyg1 - libshogun is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/414  blamelist: Soeren Sonnenburg <sonne@debian.org>, ptizoom <ptizoom@gmail.com>, Sergey Lisitsyn <lisitsyn.s.o@gmail.com>, Viktor Gal <viktor.gal@maeth.com>23:41
-!- Marty28 [~marty@5.28.65.203] has joined #shogun23:42
shogun-buildbotbuild #539 of deb2 - static_interfaces is complete: Failure [failed test libshogun]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb2%20-%20static_interfaces/builds/539  blamelist: Soeren Sonnenburg <sonne@debian.org>23:44
blackburnsonney2k: 4.6.123:51
blackburnargh 4.6.323:51
-!- Marty28 [~marty@5.28.65.203] has quit [Quit: Colloquy for iPad - http://colloquy.mobi]23:52
--- Log closed Mon Nov 19 00:00:17 2012

Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!