IRC logs of #shogun for Friday, 2014-04-18

--- Log opened Fri Apr 18 00:00:10 2014
-!- kislay [~Abhijeet@210.212.61.185] has quit [Ping timeout: 240 seconds]00:05
-!- khalednasr [~k.nasr92@41.69.223.156] has quit [Quit: Leaving]00:36
@sonney2klambday_, it says possibly lost - so relax00:39
* sonney2k yay easter holdays00:40
lambday_sonney2k: yeah... its kind of annoying though00:40
lambday_sonney2k: long time no see! :D00:40
@sonney2kyeah busy...00:40
@lisitsynsonney2k: oh who's there00:41
@sonney2klambday_, well I mean it is harmless if libraries at startup allocate some buffers that they never free00:41
@sonney2kI mean it can only be freed once done using the lib which usually is only on program exit00:42
@sonney2klisitsyn, heh00:42
lambday_sonney2k: yeah I was stuck at it for a while until lisitsyn explained it to me..00:42
@lisitsynsonney2k: have you seen my fancy blog?!00:42
@lisitsyn:D00:42
@sonney2klisitsyn, YOU HAVE A BLOG?00:43
@sonney2klisitsyn, kidding?00:43
lambday_sonney2k: it does mean a thing :D00:43
@lisitsynsonney2k: now yes00:43
@sonney2klisitsyn, are you on shogun planet just yet?00:43
@lisitsynsonney2k: no I don't think so00:43
@sonney2klisitsyn, well then give me the url and I put you live :D00:43
@lisitsynI forgot about this thing00:43
@lisitsynsonney2k: http://sergey.lisitsyn.me00:44
@sonney2klisitsyn, I need the rss00:44
@lisitsynsonney2k: I am not sure I set it up00:44
@lisitsynsonney2k: is atom.xml the right thing?00:44
@lisitsynif it doesn't work I can set up rss as well I guess00:45
@sonney2khttp://sergey.lisitsyn.me/atom.xml right?00:45
@lisitsynsonney2k: correct00:45
@sonney2kman I am brainwashed00:45
@sonney2kall shogun related knowledges swapped out00:46
@lisitsynhmm I have no idea what's with rss in octopress00:46
* sonney2k is trying to find some backup tapes00:46
@lisitsynsonney2k: knowledges like?00:46
@sonney2klisitsyn, anything00:46
@lisitsynsonney2k: hmm is std::vector contiguous?00:46
lambday_lol00:46
@sonney2klisitsyn, accoring to latest standards it is hurray!00:47
@lisitsynsonney2k: starting from C9800:47
@lisitsyn:D00:47
@lisitsynit was latest when you were young ;)00:47
@lisitsynbtw in C++1400:47
@lisitsynyou can finally do00:47
@lisitsyn[] (auto x) { ... }00:47
@lisitsyngeneric lambdas!00:47
lambday_I heard lambda00:48
@lisitsynlambday_: what does [&] in lambda mean? ;)00:48
@lisitsyn<interview mode>00:48
lambday_lisitsyn: take by ref00:49
lambday_?00:49
@lisitsyntake what by ref?00:49
lambday_whatever you pass :-/00:49
lambday_from env00:49
@sonney2klisitsyn, there we go http://shogun-toolbox.org/page/planet/shogun00:49
@lisitsynlambday_: env? ;) what is env?00:50
@lisitsynsonney2k: ha cool thanks00:50
lambday_any var that00:50
lambday_s's defined in that scope?00:50
lambday_:P00:50
@lisitsynlambday_: yeah scope is correct ;)00:50
@lisitsynlambday_: what is [=]?00:50
lambday_forgot :D00:50
* lambday_ quick googles00:51
@sonney2klisitsyn, will you attend the dedup meeting?00:51
@lisitsynsonney2k: yeah00:51
@sonney2kexcellent00:51
lambday_lisitsyn: found :D00:51
@lisitsynlambday_: by value, right? ;)00:53
lambday_lisitsyn: yo!00:53
@lisitsynsonney2k: if I get hit by bus00:53
@lisitsynheiko will take care00:53
@lisitsyn:D00:53
shogun-buildbotbuild #255 of debian wheezy - memcheck is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/255  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, pl8787 <pl8787@gmail.com>00:55
@lisitsynhttp://blog.smola.org/post/82937674537/beware-the-bandwidth-gap-speeding-up-optimization00:57
@sonney2klisitsyn, or you don't find the way around the fance ;)00:57
@lisitsynsonney2k: hah you have enjoyed this video it seems00:57
@lisitsynlambday_: http://www.youtube.com/watch?v=1IAXrxlDK6c00:58
@sonney2klisitsyn, pretty obvious isn't it?00:59
@lisitsynsonney2k: that's funny cause I don't really get it :D00:59
@lisitsynit seems if you live here that's ok00:59
@lisitsynsonney2k: have you heard mahout is moving to spark?01:00
@sonney2klisitsyn, yes seen that01:01
@sonney2kmap reduce is close to useless for ML01:01
@sonney2kit has always been01:01
@sonney2kit is googles way of computing very simple statistics01:01
@sonney2kwhich is sufficient if you have massive data in *many* cases01:01
@lisitsynsonney2k: I still think we should find a way into distributed things01:02
-!- lisitsyn2 [~qdrgsm@80.252.20.67] has joined #shogun01:05
-!- lisitsyn [~qdrgsm@80.252.20.67] has quit [Read error: Connection reset by peer]01:05
@sonney2klisitsyn2, well do it :) I don't have big enough data for that :D01:08
@sonney2kso for me personally what we have is sufficient01:08
lisitsyn2sonney2k: yeah that is apparently the main thing01:09
@sonney2kso it somebody who has too much data should take care of that01:09
lisitsyn2if we want to be distributed01:09
lisitsyn2we need to have a few machines01:09
lisitsyn2I don't really have a cluster :D01:09
@sonney2kwell none of us used the google grant right01:10
@sonney2kno idea really how we could get this to work other than bying a couple of el-cheapo machines01:10
lisitsyn2sonney2k: is this grant lost?01:11
@sonney2klisitsyn2, it times out after 6 months01:11
lisitsyn2sucks01:11
@sonney2klisitsyn2, you know I have a cellar01:11
@sonney2kso maybe 5 cheap nodes should be doable01:11
lisitsyn2want to pay electricity bills? ;)01:12
lisitsyn2sonney2k: isn't it not el cheapo?01:12
@sonney2kno shogun foundation has to as well as for the machines if we want that. maybe each 32G or so01:12
@sonney2kbut electricity is not the main thing if the machines idle01:12
@sonney2klisitsyn2, alternative is we rent  a couple of machines01:13
-!- lisitsyn2 [~qdrgsm@80.252.20.67] has quit [Read error: Connection reset by peer]01:13
-!- abinash_panda [01261095@gateway/web/freenode/ip.1.38.16.149] has quit [Ping timeout: 240 seconds]01:13
-!- lisitsyn [~qdrgsm@80.252.20.67] has joined #shogun01:17
lisitsyndamn!01:17
@sonney2klisitsyn, well about 3000 EUR for 5 such machines, with each quadcore i7's (3.4GHz and 32G memory, 1TB disk)01:17
lisitsynbattery effectively died :D01:17
@sonney2klisitsyn, I mean this would be a very sane investment if we find a sponsor01:18
@sonney2klisitsyn, your new and 'shiny' notebook already?01:19
* sonney2k tired01:19
@sonney2kbed time01:19
-!- lisitsyn [~qdrgsm@80.252.20.67] has quit [Ping timeout: 252 seconds]01:22
-!- lisitsyn [~qdrgsm@80.252.20.67] has joined #shogun01:36
lisitsynbahh I need new notebook I guess01:36
lisitsyn:D01:36
lisitsynsonney2k: still here?01:36
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]01:45
-!- newlido [~quassel@41.237.200.93] has joined #shogun02:00
-!- davis [47e87ef5@gateway/web/freenode/ip.71.232.126.245] has joined #shogun02:10
-!- davis is now known as Guest2931002:10
Guest29310anyone here? :)02:11
-!- lisitsyn [~qdrgsm@80.252.20.67] has quit [Ping timeout: 252 seconds]02:26
-!- kislay [~Abhijeet@117.239.94.112] has joined #shogun02:26
-!- Guest29310 [47e87ef5@gateway/web/freenode/ip.71.232.126.245] has quit [Quit: Page closed]02:28
shogun-buildbotbuild #256 of debian wheezy - memcheck is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/256  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Wu Lin <yorker.lin@gmail.com>02:58
shogun-buildbotbuild #679 of nightly_all is complete: Failure [failed compile test]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_all/builds/679  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Wu Lin <yorker.lin@gmail.com>, pl8787 <pl8787@gmail.com>03:04
-!- newlido_ [~quassel@41.237.218.232] has joined #shogun03:07
-!- newlido [~quassel@41.237.200.93] has quit [Ping timeout: 240 seconds]03:09
-!- soumyaC [uid15286@gateway/web/irccloud.com/x-biynwnrhpvakzffi] has quit [Quit: Connection closed for inactivity]03:46
-!- kislay [~Abhijeet@117.239.94.112] has quit [Read error: Connection reset by peer]04:18
-!- lambday_ [67157f51@gateway/web/freenode/ip.103.21.127.81] has quit [Ping timeout: 240 seconds]04:47
-!- kislay [~Abhijeet@117.239.94.112] has joined #shogun05:11
-!- kislay [~Abhijeet@117.239.94.112] has quit [Quit: Leaving]05:17
-!- kislay [~Abhijeet@14.139.41.150] has joined #shogun05:17
-!- kislay [~Abhijeet@14.139.41.150] has quit [Remote host closed the connection]05:27
-!- kislay [~Abhijeet@117.239.94.112] has joined #shogun05:33
-!- kislay [~Abhijeet@117.239.94.112] has quit [Remote host closed the connection]05:34
-!- kislay [~Abhijeet@210.212.61.185] has joined #shogun05:35
-!- kislay [~Abhijeet@210.212.61.185] has quit [Remote host closed the connection]05:35
-!- kislay [~Abhijeet@14.139.41.150] has joined #shogun05:37
-!- kislay [~Abhijeet@14.139.41.150] has quit [Remote host closed the connection]05:37
-!- kislay [~Abhijeet@117.239.94.112] has joined #shogun05:38
-!- kislay [~Abhijeet@117.239.94.112] has quit [Remote host closed the connection]05:39
-!- kislay [~Abhijeet@117.239.94.112] has joined #shogun06:35
-!- kislay [~Abhijeet@117.239.94.112] has quit [Remote host closed the connection]06:36
-!- kislay [~Abhijeet@117.239.94.112] has joined #shogun06:44
-!- kislay [~Abhijeet@117.239.94.112] has quit [Remote host closed the connection]06:44
-!- kislay [~Abhijeet@14.139.41.150] has joined #shogun06:46
-!- kislay [~Abhijeet@14.139.41.150] has quit [Remote host closed the connection]06:46
-!- kislay [~Abhijeet@117.239.94.112] has joined #shogun06:48
-!- kislay [~Abhijeet@117.239.94.112] has quit [Remote host closed the connection]06:48
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun09:16
shogun-notifier-shogun: Abhijeet :develop * 6c5e1da / NEWS: https://github.com/shogun-toolbox/shogun/commit/6c5e1da1ca1699893ecdcb9a6b48ea74dc29415209:16
shogun-notifier-shogun: added PCA notebook to the NEWS.09:16
shogun-notifier-shogun: Fernando Iglesias :develop * 1486e74 / NEWS: https://github.com/shogun-toolbox/shogun/commit/1486e742639213138a763ec50abc93207518fd8709:16
shogun-notifier-shogun: Merge pull request #2165 from kislayabhi/develop09:16
shogun-notifier-shogun:09:16
shogun-notifier-shogun: added PCA notebook to the NEWS.09:16
-!- newlido_ [~quassel@41.237.218.232] has quit [Remote host closed the connection]09:30
-!- newlido [~quassel@41.237.218.232] has joined #shogun09:34
-!- kislay [~Abhijeet@117.239.94.112] has joined #shogun09:36
-!- kislay [~Abhijeet@117.239.94.112] has quit [Remote host closed the connection]09:37
-!- newlido [~quassel@41.237.218.232] has quit [Ping timeout: 250 seconds]09:44
-!- lisitsyn [~qdrgsm@80.252.20.67] has joined #shogun09:46
-!- PirosB3 [~pirosb3@host238-17-dynamic.245-95-r.retail.telecomitalia.it] has joined #shogun09:50
-!- kislay [~Abhijeet@14.139.41.150] has joined #shogun09:53
-!- kislay [~Abhijeet@14.139.41.150] has quit [Remote host closed the connection]09:54
-!- kislay [~Abhijeet@14.139.41.150] has joined #shogun09:55
-!- kislay [~Abhijeet@14.139.41.150] has quit [Remote host closed the connection]09:57
-!- PirosB3 [~pirosb3@host238-17-dynamic.245-95-r.retail.telecomitalia.it] has quit [Quit: PirosB3]10:29
-!- lisitsyn [~qdrgsm@80.252.20.67] has quit [Quit: Leaving.]10:29
shogun-notifier-shogun: Parijat Mazumdar :develop * 01c79af / tests/unit/multiclass/tree/ID3ClassifierTree_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/01c79afe4e69d702047daa5e4e192f9ed5f9242010:39
shogun-notifier-shogun: memory leak fixed10:39
shogun-notifier-shogun: Heiko Strathmann :develop * 34f9672 / tests/unit/multiclass/tree/ID3ClassifierTree_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/34f96727f343b7f7f5e0426dbbf579f5dbc0f51e10:39
shogun-notifier-shogun: Merge pull request #2166 from mazumdarparijat/decision_tree10:39
shogun-notifier-shogun:10:39
shogun-notifier-shogun: memory leak fixed10:39
-!- HeikoS [~heiko@ra-vpn-193.ra-vpn.ucl.ac.uk] has joined #shogun10:47
-!- mode/#shogun [+o HeikoS] by ChanServ10:47
-!- HeikoS [~heiko@ra-vpn-193.ra-vpn.ucl.ac.uk] has quit [Quit: Leaving.]10:55
-!- HeikoS [~heiko@ra-vpn-193.ra-vpn.ucl.ac.uk] has joined #shogun10:56
-!- mode/#shogun [+o HeikoS] by ChanServ10:56
-!- lisitsyn1 [~lisitsin@mxs.kg.ru] has joined #shogun11:18
shogun-buildbotbuild #257 of debian wheezy - memcheck is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/257  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>, Abhijeet <abhijeetkislay@gmail.com>11:30
-!- lambday [67157f4f@gateway/web/freenode/ip.103.21.127.79] has joined #shogun11:46
lambdayHeikoS: hey!11:46
lambdayHeikoS: lisitsyn1: can I use std::pair as a return type? I need to return two SGVectors of different length... SGMatrix is not really a good idea for that11:47
lisitsyn1lambday: hey11:50
lisitsyn1lambday: return type of what?11:51
-!- lambday_ [67157f4f@gateway/web/freenode/ip.103.21.127.79] has joined #shogun11:52
lambday_lisitsyn1: hola11:52
lambday_lisitsyn1: return type of a public method of CKernel11:52
lambday_SWIG supports std::pair11:52
lambday_but may be gotta att template() std_pair<SGVector<float64_t>,SGVector<float64_t> > in modshogun.i for that to work?11:53
-!- lambday [67157f4f@gateway/web/freenode/ip.103.21.127.79] has quit [Ping timeout: 240 seconds]11:54
lisitsyn1lambday_: oh well I'd avoid it :D11:54
lisitsyn1think how it would look like in python, java and say octave11:54
lambday_lisitsyn1: well, in python a tuple of numpy vectors :D11:55
lambday_lisitsyn1: so what do you suggest?11:56
lisitsyn1lambday_: no clue yet, let me think11:56
lambday_SGMatrix is waste of memory since their size differs11:56
lambday_CList is overkill11:56
lambday_may be just concatenated vectors11:57
lisitsyn1lambday_: what are these vectors?11:57
lambday_first n entries belong to vector a and m entries belong to vector b11:57
lambday_lisitsyn1: vec1 = row-wise sum of kernel values11:57
lambday_vec2 = col-wise     -do-11:58
lambday_(when its not symmetric)11:58
lisitsyn1ahh11:58
lambday_I don't want to call CKernel::kernel(i, j) more than once per (i,j) so I need to return both of these in one method11:59
lisitsyn1lambday_: what about some precomputing on call11:59
lisitsyn1and two methods?11:59
lambday_lisitsyn1: if I can precompute the whole kernel matrix then we won't be needing these methods11:59
lisitsyn1no I mean11:59
lisitsyn1if you want col sums11:59
lisitsyn1you precompute col/row sums11:59
lisitsyn1and return col sums12:00
lambday_storing things between calls?12:00
lisitsyn1???12:00
lisitsyn1yes12:00
lisitsyn1:D sorry12:00
lambday_?? ??? ?????12:00
lambday_:?12:00
lisitsyn1hahaha12:00
lambday_umm... lisitsyn1 I don't like storing :(12:01
lisitsyn1lambday_: possible bugs12:01
lambday_concatenation can do - the callee always know the num-rows and num-cols... so a vec of num-rows+num_cols element can do I guess .... first vec[0:num_rows] are row-wise sums and rest are col-wise sums12:03
lisitsyn1lambday_: why clist is overkill?12:03
lisitsyn1lambday_: concatenation looks like a workaround for me12:04
lambday_lisitsyn1: because its just 2 elements... we can do better than a linked list12:04
lisitsyn1lambday_: is it expected to be used often?12:04
lambday_lisitsyn1: fairly12:04
lisitsyn1lambday_: hmm..12:05
-!- HeikoS [~heiko@ra-vpn-193.ra-vpn.ucl.ac.uk] has quit [Quit: Leaving.]12:08
-!- HeikoS [~heiko@ra-vpn-193.ra-vpn.ucl.ac.uk] has joined #shogun12:10
-!- mode/#shogun [+o HeikoS] by ChanServ12:10
lisitsyn1lambday_: I don't like neither pair nor other solutions12:11
lisitsyn1:D12:11
lambday_lisitsyn1: so what do you suggest? :D12:12
lisitsyn1lambday_: give up12:13
lisitsyn1:D12:13
-!- HeikoS [~heiko@ra-vpn-193.ra-vpn.ucl.ac.uk] has quit [Read error: Connection reset by peer]12:37
-!- HeikoS [~heiko@ra-vpn-193.ra-vpn.ucl.ac.uk] has joined #shogun12:38
-!- mode/#shogun [+o HeikoS] by ChanServ12:38
-!- HeikoS [~heiko@ra-vpn-193.ra-vpn.ucl.ac.uk] has quit [Read error: Connection reset by peer]12:57
-!- HeikoS [~heiko@ra-vpn-193.ra-vpn.ucl.ac.uk] has joined #shogun12:58
-!- mode/#shogun [+o HeikoS] by ChanServ12:58
pl8787_lambday_: CustomKernel support precompute, is that can help you?13:27
lambday_pl8787_: not really :) but thanks :)13:28
-!- newlido [~quassel@41.237.218.232] has joined #shogun13:28
lambday_lisitsyn1: that's plan B :P13:28
pl8787_:)13:28
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]13:39
shogun-buildbotbuild #258 of debian wheezy - memcheck is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/258  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Parijat Mazumdar <mazumdarparijat@gmail.com>14:07
-!- HeikoS [~heiko@ra-vpn-193.ra-vpn.ucl.ac.uk] has quit [Ping timeout: 258 seconds]14:17
-!- newlido [~quassel@41.237.218.232] has quit [Ping timeout: 240 seconds]14:21
-!- HeikoS [~heiko@0541877d.skybroadband.com] has joined #shogun14:25
-!- mode/#shogun [+o HeikoS] by ChanServ14:25
-!- HeikoS [~heiko@0541877d.skybroadband.com] has quit [Quit: Leaving.]14:53
-!- HeikoS [~heiko@0541877d.skybroadband.com] has joined #shogun14:54
-!- mode/#shogun [+o HeikoS] by ChanServ14:54
-!- HeikoS [~heiko@0541877d.skybroadband.com] has quit [Client Quit]14:54
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun15:02
shogun-notifier-shogun: Saurabh :develop * d3da8de / NEWS,doc/ipython-notebooks/classification/MKL.ipynb: https://github.com/shogun-toolbox/shogun/commit/d3da8de7c74d57019bf46079024afa6c8db5ffab15:02
shogun-notifier-shogun: add to news15:02
shogun-notifier-shogun: Fernando Iglesias :develop * 4ad752f / NEWS,doc/ipython-notebooks/classification/MKL.ipynb: https://github.com/shogun-toolbox/shogun/commit/4ad752f1cf9fc86a87fe344a9d2932ede7047af915:02
shogun-notifier-shogun: Merge pull request #2168 from Saurabh7/develop15:02
shogun-notifier-shogun:15:02
shogun-notifier-shogun: minor changes mkl nb15:02
-!- PirosB3 [~pirosb3@host238-17-dynamic.245-95-r.retail.telecomitalia.it] has joined #shogun15:33
-!- kprah [67157f4e@gateway/web/freenode/ip.103.21.127.78] has joined #shogun16:16
-!- kprah [67157f4e@gateway/web/freenode/ip.103.21.127.78] has quit [Client Quit]16:17
-!- kislay [~Abhijeet@117.239.94.112] has joined #shogun16:22
-!- kislay [~Abhijeet@117.239.94.112] has quit [Remote host closed the connection]16:22
-!- lisitsyn1 [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.]17:22
shogun-buildbotbuild #259 of debian wheezy - memcheck is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/259  blamelist: Saurabh <saurabh.mahindre@gmail.com>, Fernando Iglesias <fernando.iglesiasg@gmail.com>17:22
-!- iglesiasg [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has joined #shogun17:25
-!- mode/#shogun [+o iglesiasg] by ChanServ17:25
-!- newlido [~quassel@41.218.181.92] has joined #shogun17:26
-!- PirosB3 [~pirosb3@host238-17-dynamic.245-95-r.retail.telecomitalia.it] has quit [Quit: PirosB3]17:27
-!- iglesiasg [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has quit [Ping timeout: 240 seconds]17:47
-!- iglesiasg [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has joined #shogun18:00
-!- mode/#shogun [+o iglesiasg] by ChanServ18:00
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]18:02
-!- newlido_ [~quassel@41.218.181.92] has joined #shogun18:08
-!- newlido is now known as newlido_afk18:36
-!- lisitsyn [~qdrgsm@80.252.20.67] has joined #shogun18:37
-!- iglesiasg [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has quit [Quit: Leaving]19:01
-!- HeikoS [~heiko@0541877d.skybroadband.com] has joined #shogun19:17
-!- mode/#shogun [+o HeikoS] by ChanServ19:18
@HeikoSlisitsyn: attending the meeting still?19:18
lambday_HeikoS: hey!19:19
lambday_HeikoS: just sent a PR19:19
-!- newlido_afk [~quassel@41.218.181.92] has quit [Read error: Connection reset by peer]19:19
@HeikoSlambday_: hey!19:19
lambday_HeikoS: please have a look when you're free :)19:19
@HeikoSnice19:19
@HeikoSwill have a look pretty much now19:19
lambday_:D19:19
lambday_made openmp-ed as well19:20
lambday_will add overridden sum methods for CustomKernel in the next patch19:21
lisitsynHeikoS: yeah here19:36
lisitsynHeikoS: well meeting is in a hour19:37
lisitsyn:)19:37
@HeikoSlambday_: very nice work!19:42
@HeikoSlambday_: this also will make the MMD stuff on custom kernels a lot faster19:42
@HeikoSas we dont loop over these virtual method calls anymore19:42
lambday_HeikoS: I hope so..19:43
lambday_HeikoS: will add CustomKernel stuffs soon..19:44
lambday_HeikoS: I think I need to reduce the accuracy a bit.. will update the PR in a sec19:44
-!- newlido_ [~quassel@41.218.181.92] has quit [Read error: Connection reset by peer]19:44
lambday_when ran with ctest it all went through but with valgrind one test is complaining to be 1E-1419:44
@HeikoSok19:45
@HeikoSyeah e-14 is ok :)19:45
@HeikoSlambday_: I would actually be interested in a performance coparison19:46
@HeikoSlambday_: are you interested in coding one up and put into benchmarks?19:46
@HeikoSlambday_: google doesnt really answer this19:47
@HeikoSlambday_: all probably depends on compiler optimisations19:47
lambday_HeikoS: with previous implementation and with replaced by these calls?19:47
@HeikoSlambday_: not for MMD but kernel in general19:47
@HeikoSone that loops over the ::kernel calls19:47
@HeikoSand one that calls your method19:47
@HeikoSfor Custom kernel19:47
@HeikoSfor GaussianKernel or others it should be the same since the kernel call has to be done19:48
@HeikoS::kenrel call I mean19:48
@HeikoSah wait19:48
lambday_HeikoS: I am interested in setting up these benchmarks..19:57
lambday_HeikoS: will try to do it by tonight19:57
lambday_HeikoS: I will just be back after dinner :)19:57
-!- HeikoS [~heiko@0541877d.skybroadband.com] has quit [Quit: Leaving.]20:13
-!- HeikoS [~heiko@0541877d.skybroadband.com] has joined #shogun20:21
-!- mode/#shogun [+o HeikoS] by ChanServ20:21
-!- HeikoS [~heiko@0541877d.skybroadband.com] has quit [Ping timeout: 265 seconds]20:28
-!- iglesiasg [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has joined #shogun20:35
-!- mode/#shogun [+o iglesiasg] by ChanServ20:35
@iglesiasgI am at #gsoc for the meeting starting soon20:52
lisitsyniglesiasg: me too20:58
lisitsyn;)20:58
@iglesiasglisitsyn, :)20:59
-!- lisitsyn is now known as lisitsyn|shogun21:00
-!- iglesiasg is now known as igleasiasg|shogu21:01
-!- igleasiasg|shogu is now known as iglesiasg|shogun21:01
@iglesiasg|shogunI type so well...21:01
lisitsyn|shogunhaha21:01
lisitsyn|shoguniglesiasg|shogun: I guess we are not needed but lets stay ;)21:02
@iglesiasg|shogunlisitsyn|shogun, shall we make a channel so the two of us can speak to Carol or just one goes ahead?21:03
lisitsyn|shoguniglesiasg|shogun: we won't need it I guess21:03
lisitsyn|shogunour duplicate is resolved21:03
@iglesiasg|shogunall right!21:03
lisitsyn|shoguniglesiasg|shogun: http://www.spacex.com/webcast/21:04
lisitsyn|shogun;)21:04
@iglesiasg|shogunexciting21:05
lisitsyn|shoguniglesiasg|shogun: that's a mission to the ISS21:10
lisitsyn|shogunexciting is the word ;)21:10
-!- lisitsyn|shogun [~qdrgsm@80.252.20.67] has quit [Read error: Connection reset by peer]21:17
-!- lisitsyn [~qdrgsm@80.252.20.67] has joined #shogun21:17
-!- lisitsyn is now known as shogun|lisitsyn21:18
-!- shogun|lisitsyn is now known as lisitsyn21:21
shogun-buildbotbuild #443 of osx1 - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/osx1%20-%20libshogun/builds/44321:24
lisitsynwhooo21:26
lisitsynits successful!21:37
@iglesiasg|shogunparty party21:38
-!- kislay [~Abhijeet@210.212.61.185] has joined #shogun22:02
-!- iglesiasg|shogun [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has quit [Quit: Leaving]22:21
-!- lambday_ [67157f4f@gateway/web/freenode/ip.103.21.127.79] has quit [Ping timeout: 240 seconds]23:31
-!- kislay [~Abhijeet@210.212.61.185] has quit [Quit: Leaving]23:45
-!- rho_ [~quassel@41.199.155.104] has joined #shogun23:46
--- Log closed Sat Apr 19 00:00:11 2014

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