IRC logs of #shogun for Thursday, 2016-03-17

--- Log opened Thu Mar 17 00:00:33 2016
-!- mizari [~mizari@95-174-213-100.nts.su] has joined #shogun00:16
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun00:40
-!- mode/#shogun [+o HeikoS] by ChanServ00:40
shogun-notifier-shogun: Heiko Strathmann :develop * 120a313 / examples/meta/generator/translate.py: https://github.com/shogun-toolbox/shogun/commit/120a3137d0d2441bffb98587fa043939237d5ce600:43
shogun-notifier-shogun: another python2->python3 fix00:43
shogun-notifier-shogun: Heiko Strathmann :develop * 079e531 / examples/meta/generator/translate.py: https://github.com/shogun-toolbox/shogun/commit/079e5313b8ddc4850aa244453390dd3625194bcd00:44
shogun-notifier-shogun: another python2->python3 fix00:44
@HeikoSshogun-buildbot: force build --branch=develop 'deb4 - python3'00:44
shogun-buildbotbuild #542 of trusty - libshogun - viennacl is complete: Failure [failed compile]  Build details are at http://buildbot.shogun-toolbox.org/builders/trusty%20-%20libshogun%20-%20viennacl/builds/542  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>00:44
shogun-buildbotThe build has been queued, I'll give a shout when it starts00:44
shogun-buildbotbuild #543 of trusty - libshogun - viennacl is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/trusty%20-%20libshogun%20-%20viennacl/builds/54300:45
shogun-buildbotbuild #3657 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/3657  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>00:45
shogun-buildbotbuild forced [ETA 5m23s]00:45
shogun-buildbotI'll give a shout when the build finishes00:45
shogun-buildbotbuild #775 of deb4 - python3 is complete: Failure [failed install test python modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb4%20-%20python3/builds/77500:49
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Quit: Leaving.]00:50
-!- damjangelovski [4d1d8d02@gateway/web/freenode/ip.77.29.141.2] has quit [Ping timeout: 252 seconds]01:03
-!- GandalfTheWizard [~ian@112.10.106.188] has quit [Ping timeout: 250 seconds]02:45
-!- will [cc2db69d@gateway/web/freenode/ip.204.45.182.157] has joined #shogun02:53
-!- GandalfTheWizard [~ian@112.10.106.188] has joined #shogun02:53
shogun-buildbotbuild #971 of nightly_all is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_all/builds/971  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Sanuj <sanuj.sharma.in@gmail.com>03:07
-!- GandalfTheWizard [~ian@112.10.106.188] has quit [Ping timeout: 240 seconds]03:13
-!- GandalfTheWizard [~ian@112.10.106.188] has joined #shogun03:17
-!- GandalfTheWizard [~ian@112.10.106.188] has quit [Max SendQ exceeded]03:18
-!- GandalfTheWizard [~ian@112.10.106.188] has joined #shogun03:19
-!- GandalfTheWizard [~ian@112.10.106.188] has quit [Max SendQ exceeded]03:20
-!- Netsplit *.net <-> *.split quits: shogun-buildbot03:21
-!- GandalfTheWizard [~ian@112.10.106.188] has joined #shogun03:26
-!- Netsplit over, joins: shogun-buildbot03:26
-!- will [cc2db69d@gateway/web/freenode/ip.204.45.182.157] has quit [Quit: Page closed]03:30
-!- GandalfTheWizard [~ian@112.10.106.188] has quit [Ping timeout: 240 seconds]03:33
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]03:44
-!- GandalfTheWizard [~ian@112.10.106.188] has joined #shogun03:45
-!- GandalfTheWizard [~ian@112.10.106.188] has quit [Ping timeout: 250 seconds]04:03
-!- GandalfTheWizard [~ian@112.10.106.188] has joined #shogun04:04
shogun-buildbotbuild #1098 of nightly_default is complete: Failure [failed test notebooks]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/1098  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Sanuj <sanuj.sharma.in@gmail.com>04:07
-!- GandalfTheWizard [~ian@112.10.106.188] has quit [Ping timeout: 244 seconds]04:11
-!- GandalfTheWizard [~ian@112.10.106.188] has joined #shogun04:35
-!- sonne|osx [~sonne@x55b370bd.dyn.telefonica.de] has joined #shogun05:16
-!- sonne|osx [~sonne@x55b370bd.dyn.telefonica.de] has quit [Quit: sonne|osx]05:45
-!- besser82 [~besser82@fedora/besser82] has joined #shogun07:41
-!- mode/#shogun [+o besser82] by ChanServ07:41
shogun-buildbotbuild #63 of deb1 - libshogun - PR is complete: Failure [failed cookbook]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun%20-%20PR/builds/63  blamelist: tbskm08:58
shogun-buildbotbuild #64 of deb1 - libshogun - PR is complete: Failure [failed cookbook]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun%20-%20PR/builds/64  blamelist: tbskm09:04
shogun-buildbotbuild #65 of deb1 - libshogun - PR is complete: Failure [failed cookbook]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun%20-%20PR/builds/65  blamelist: tbskm09:16
shogun-buildbotbuild #66 of deb1 - libshogun - PR is complete: Failure [failed cookbook]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun%20-%20PR/builds/66  blamelist: tbskm09:27
@wiking!bender10:10
ShoGaYHey, wait! Gimme a do-over! My cheating chip malfunctioned!10:10
lisitsyn!bender10:19
ShoGaYHey. What I don't do is none of your business.10:19
@wiking:>10:30
@wiking!bender hit it!10:32
@wiking:>10:32
@wiking!bender10:32
ShoGaYBut of everyone I ever dated, you're probably in the top ten.10:32
@wikingShoGaY: is sooo fucking gay :)10:32
@wikingand i must state here that this all without any prejudice10:33
@wikingbefore anybody misinterprest this statement10:33
shogun-buildbotbuild #67 of deb1 - libshogun - PR is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun%20-%20PR/builds/6710:43
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun11:10
-!- mode/#shogun [+o HeikoS] by ChanServ11:10
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Client Quit]11:11
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun11:13
-!- mode/#shogun [+o HeikoS] by ChanServ11:13
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Client Quit]11:16
lisitsynwiking: gosh we were unresponsive for a week?11:43
-!- HeikoS [~heiko@5751934b.skybroadband.com] has joined #shogun11:48
-!- mode/#shogun [+o HeikoS] by ChanServ11:48
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun11:59
shogun-notifier-shogun: Wu Lin :develop * 88ad808 / src/shogun/machine/gp/ (7 files): https://github.com/shogun-toolbox/shogun/commit/88ad808e18cdbf0613f4e0e127abff2a846c44c311:59
shogun-notifier-shogun: fixed some python_modular warnings11:59
shogun-notifier-shogun: Heiko Strathmann :develop * 8e69308 / src/shogun/machine/gp/ (7 files): https://github.com/shogun-toolbox/shogun/commit/8e6930862659c7218d838518b1fad55d5a143ef811:59
shogun-notifier-shogun: Merge pull request #3086 from yorkerlin/develop11:59
shogun-notifier-shogun:11:59
shogun-notifier-shogun: fixed some python_modular warnings11:59
@wiking!bender11:59
ShoGaYWhat kind of party is this? There's no booze and only one hooker.11:59
@wikinglisitsyn: what? :)11:59
@wikinglisitsyn: yes11:59
@wikinglisitsyn: fucking mailing list11:59
@wiking:)))11:59
shogun-buildbotbuild #3658 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/3658  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Wu Lin <yorker.lin@gmail.com>12:01
@HeikoSlisitsyn: hey12:11
@HeikoSgot a minute?12:11
shogun-notifier-shogun: Saurabh7 :develop * 3b11522 / / (6 files): https://github.com/shogun-toolbox/shogun/commit/3b115223418e046a0ab3af4dad3b24ebe12a385b12:21
shogun-notifier-shogun: update kmeans and unit-tests12:21
shogun-notifier-shogun: Heiko Strathmann :develop * cd5fe91 / / (6 files): https://github.com/shogun-toolbox/shogun/commit/cd5fe918ca62c59c8ecde7ffb541360e1783f0f412:21
shogun-notifier-shogun: Merge pull request #3067 from Saurabh7/new_kmeans12:21
shogun-notifier-shogun:12:21
shogun-notifier-shogun: kmeans update12:21
shogun-buildbotbuild #3659 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/3659  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Saurabh7 <saurabh.mahindre@gmail.com>12:25
@wikingHeikoS: ^ ideas: 156 - integration_meta_cpp-regression-kernel_ridge_regression (Failed)12:25
@HeikoSwiking: checking12:26
@HeikoSwiking: mmh that should not be12:26
@HeikoSill try locally12:26
@HeikoSwiking: probably should include some debug what actually fails in the tester12:27
@HeikoSwill do that12:27
@HeikoStravis was fine on this12:27
@wikingmmm12:27
@wikinginteresting12:27
@HeikoSwiking: passes locally12:29
@HeikoSshogun-buildbot: force build --branch=develop 'deb1 - libshogun'12:29
shogun-buildbotbuild forced [ETA 1m41s]12:29
shogun-buildbotI'll give a shout when the build finishes12:29
@HeikoSah12:30
@HeikoSwrong data revision12:30
@HeikoSI shouldnt have merged12:31
@HeikoSsigh12:31
shogun-buildbotbuild #3660 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/366012:31
@HeikoSmmh no12:31
@HeikoSdata is up to date12:31
shogun-notifier-shogun: Heiko Strathmann :develop * 1d99e91 / examples/meta/generator/translate.py: https://github.com/shogun-toolbox/shogun/commit/1d99e91d0b5c6b33d78da88f959516fd43e51dbf12:42
shogun-notifier-shogun: fix python2/3 dictionary key access12:42
@HeikoSshogun-buildbot: force build --branch=develop 'deb3 - python3'12:43
shogun-buildbotno such builder 'deb3 - python3'12:43
@HeikoSshogun-buildbot: force build --branch=develop 'deb4 - python3'12:43
shogun-buildbotThe build has been queued, I'll give a shout when it starts12:43
shogun-buildbotbuild #546 of trusty - libshogun - viennacl is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/trusty%20-%20libshogun%20-%20viennacl/builds/546  blamelist: Saurabh7 <saurabh.mahindre@gmail.com>12:44
shogun-buildbotbuild #3661 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/3661  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>12:44
shogun-buildbotbuild forced [ETA 5m23s]12:44
shogun-buildbotI'll give a shout when the build finishes12:44
shogun-buildbotbuild #547 of trusty - libshogun - viennacl is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/trusty%20-%20libshogun%20-%20viennacl/builds/54712:45
shogun-buildbotbuild #776 of deb4 - python3 is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb4%20-%20python3/builds/77612:52
-!- damjangelovski [4d1c00ee@gateway/web/freenode/ip.77.28.0.238] has joined #shogun12:54
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has joined #shogun12:56
@HeikoSwiking: yeah!!!!12:57
@HeikoSmeta examples now work with python312:57
@HeikoSwiking: (if python-ply is installed)12:57
sanujHeikoS: cool12:57
@HeikoSwiking: the check is not working yet though ....12:57
sanujHeikoS: I'll work on adding eigen3 llt in krr12:59
sanujHeikoS: shall i add it in linalg and then use that in krr?12:59
@HeikoSsanuj: use eigen3 in krr for now13:00
@HeikoSand then add it to linalg anyways ... for the GPs13:00
sanujokay13:00
@HeikoSbut KRR should continue to work without linalg for compat. reasons13:00
@HeikoSwe dont want to break things and it is no big deal to change later13:00
sanujunderstood13:00
@HeikoSsanuj: but GP framework is based on linalg anyways, so we can use the LLT from linalg13:00
sanujHeikoS: okay13:01
@HeikoSas it is already unavailable without c++11 anyways13:01
sanujHeikoS: and i'll write my next cookbook page on SVMs13:01
sanuji'll start writing my gsoc proposals this weekend13:01
@HeikoSsanuj: great, looking forward to it13:02
shogun-notifier-shogun: Heiko Strathmann :develop * faf1ecc / CMakeLists.txt,cmake/FindPLY.cmake,cmake/FindPythonModule.cmake: https://github.com/shogun-toolbox/shogun/commit/faf1ecc5413341e8cdcc3fd7c62b450e1ad07f0813:32
shogun-notifier-shogun: use different way to detect python ply that works with python313:32
-!- HeikoS [~heiko@5751934b.skybroadband.com] has left #shogun []13:32
-!- HeikoS [~heiko@5751934b.skybroadband.com] has joined #shogun13:32
-!- mode/#shogun [+o HeikoS] by ChanServ13:32
@HeikoSshogun-buildbot: force build --branch=develop 'deb4 - python3'13:33
shogun-buildbotThe build has been queued, I'll give a shout when it starts13:33
@HeikoSwiking: I have fixed ply detection under python313:33
@HeikoSlisitsyn: the last one is the ctags under bsd1, http://buildbot.shogun-toolbox.org/builders/bsd1%20-%20libshogun13:33
shogun-buildbotbuild #3662 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/3662  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>13:34
shogun-buildbotbuild forced [ETA 6m43s]13:34
shogun-buildbotI'll give a shout when the build finishes13:34
-!- HeikoS [~heiko@5751934b.skybroadband.com] has quit [Quit: Leaving.]13:34
-!- Ameya [0e8b5206@gateway/web/freenode/ip.14.139.82.6] has joined #shogun13:37
shogun-buildbotHey! build deb4 - python3 #777 is complete: Success [build successful]13:39
shogun-buildbotBuild details are at http://buildbot.shogun-toolbox.org/builders/deb4%20-%20python3/builds/77713:39
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has quit [Ping timeout: 252 seconds]13:58
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun14:10
-!- mode/#shogun [+o HeikoS] by ChanServ14:10
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has joined #shogun14:35
-!- curiousguy13 [~curiousgu@120.59.226.99] has joined #shogun14:55
-!- Ameya [0e8b5206@gateway/web/freenode/ip.14.139.82.6] has quit [Quit: Page closed]15:13
sanujHeikoS: how do we know that all the kernel matrices will be psd?15:14
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has quit [Ping timeout: 252 seconds]15:27
-!- livus [~suvil@196.207.80.128] has joined #shogun15:53
-!- livusinapis [c4cf5080@gateway/web/freenode/ip.196.207.80.128] has joined #shogun15:54
-!- livusinapis [c4cf5080@gateway/web/freenode/ip.196.207.80.128] has quit [Client Quit]15:54
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]16:32
@wikingHeikoS: great17:12
@wikingwher eis it? :)17:13
-!- livus [~suvil@196.207.80.128] has left #shogun []17:34
-!- damjangelovski [4d1c00ee@gateway/web/freenode/ip.77.28.0.238] has quit [Ping timeout: 252 seconds]17:37
-!- lambday [6a33ea45@gateway/web/freenode/ip.106.51.234.69] has joined #shogun17:39
-!- mode/#shogun [+o lambday] by ChanServ17:39
@HeikoSlambday: so as I was saying, how shall the linalg functione be called?17:39
@HeikoSbecause spcifying the backend kind of defeats the purpose of being backend independent17:39
@lambdayHeikoS: if eigen3 is default, to call eigen3 version you just do linalg::foo(...)17:39
@HeikoSlambday: ok17:40
@lambdaybut to call viennacl, you specify the backend17:40
@lambdayor native, for that fact17:40
@HeikoSlambday: since we dont want to automatically replace all calls via viennacl17:40
@HeikoSlambday: and this is why you are saying we dont need global backends17:40
@HeikoSand no vianna cl build?17:40
@lambdayHeikoS: yeah I really don't think it is of any practical use17:41
@HeikoSlambday: ok17:41
@HeikoSlambday: what about this:17:41
@lambdaythere is no use-case when we'd want all our backends to be gpu17:41
@HeikoSah no17:41
@HeikoSactually, it is fine17:41
@HeikoSputting things on the GPU is always an explicit decision17:41
@HeikoSah yes, but here is a case I was thinking of17:41
@HeikoSsay we implemented an algorithm using viennacl backend17:42
@lambdayHeikoS: the other way it can be handled is to just overload the methods..17:42
@HeikoSno 2 cases might be a problem:17:42
@lambdaylinalg(SGType) calls eigen backend, linalg(GPUType) calls viennacl backend17:42
@HeikoS1.) we find a better library for doing gpu  backends ... then we have to change the algorithms, which is against the design pattern on linalg. so maybe rather than viennacl, we should call it GPU backend or so17:43
@HeikoS2.) what a user doesnt have a GPU17:43
@HeikoSthen there needs to be the fall back option of running the same code on shared memory with the CPU17:43
@HeikoSagain without changing the algorithm code17:43
@lambdayHeikoS: viennacl+opencl can work on cpu as well17:43
@HeikoSok, but do you see my point?17:44
@HeikoSwe are kind of mixing hardware backends with software backends17:44
@HeikoSwhile hardware backend is clear when algo is written, software backend should be possible to change without touching the algorithms17:44
@lambdayHeikoS: yeah I see what you mean..17:45
@lambdayHeikoS: the dependency kind of have to be like eigen3... for all cpu computation we rely on eigen, for all gpu backends we rely on viennacl17:46
@lambdaycause, the other better backend may not be as flexible as viennacl and may simply fail when no GPU is there17:47
@lambdayviennacl can manage working on cpu as well17:47
@HeikoSI would offer a general (lib free) GPU backend that17:48
@HeikoS-uses viennaCL on GPU by default17:48
@HeikoS-uses viennaCL on CPU if no GPU is present17:48
@HeikoS-uses CPU backend if none is available17:48
@HeikoSand the general CPU backend17:49
@HeikoS-defaults with eigen317:49
@HeikoS-we might add other libs to support its operations17:49
@wikingasdf17:50
@wikingFUCKING APPLE17:50
@wiking:)17:50
@wikingwhats the problem with viennacl as gpu backend?17:51
@HeikoSwiking: just talked to lambday18:17
@HeikoSwiking: we want to refactor linalg a but18:17
@HeikoSbit18:17
@HeikoSwiking: the idea is to separate backend abstraction for library and hardware18:17
@HeikoSwiking: in algorithms, one only chooses which hardware backend is used, i.e. cpu or gpu18:18
@HeikoSwiking: and for each of those, cmake decides for a particular library at compile time18:18
@HeikoScurrently it is all a bit mixed up18:18
@HeikoSwe will do a one file prototype to illustrate the ideas18:18
@HeikoSwith all the fallback options18:18
@HeikoSand then when compiling, based on what libraries and hardware is available, all those things are populated18:19
@HeikoSbut then algorithm implementation in shogun only specifies cpu or gpu, not library18:19
@HeikoSthis also means: no global viennacl backend anymore18:19
@HeikoSbut you will see the prototype18:19
@lambdayI'll try to cook something up by the end of this week for the prototype and share with you guys18:20
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun18:23
shogun-notifier-shogun: Heiko Strathmann :develop * 135a828 / doc/cookbook/README.md: https://github.com/shogun-toolbox/shogun/commit/135a828bb06452ae3c7829edc2fe235237f897dc18:23
shogun-notifier-shogun: put in link for latest version online18:23
shogun-buildbotbuild #3663 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/3663  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>18:24
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 240 seconds]18:51
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 276 seconds]19:16
-!- GandalfTheWizard [~ian@112.10.106.188] has quit [Ping timeout: 244 seconds]19:16
-!- GandalfTheWizard [~ian@112.10.106.184] has joined #shogun19:45
-!- ali_moh [29eda053@gateway/web/freenode/ip.41.237.160.83] has joined #shogun20:46
-!- ali_moh [29eda053@gateway/web/freenode/ip.41.237.160.83] has quit [Client Quit]20:47
-!- curiousguy13 [~curiousgu@120.59.226.99] has quit [Ping timeout: 240 seconds]20:52
-!- curiousguy13 [~curiousgu@59.177.206.104] has joined #shogun21:04
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]21:23
lisitsyn!bender23:31
ShoGaYOh, no room for Bender, huh? Fine, I'll go build my own Lunar Lander! With blackjack and hookers! In fact, forget the Lunar Lander and the blackjack! Ah, screw the whole thing.23:31
-!- sonne|osx [~sonne@x55b370bd.dyn.telefonica.de] has joined #shogun23:53
--- Log closed Fri Mar 18 00:00:35 2016

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