IRC logs of #shogun for Monday, 2017-05-15

« Sunday, 2017-05-14 Index Tuesday, 2017-05-16 »
--- Log opened Mon May 15 00:00:30 2017
-!- mikeling [uid89706@gateway/web/irccloud.com/x-kzswphnpelevfruq] has joined #shogun03:23
-!- mikeling [uid89706@gateway/web/irccloud.com/x-kzswphnpelevfruq] has quit [Quit: Connection closed for inactivity]05:43
-!- geektoni [c1cdd24a@gateway/web/freenode/ip.193.205.210.74] has joined #shogun08:34
-!- iglesiasg [~iglesiasg@217.119.234.214] has joined #shogun09:19
-!- mode/#shogun [+o iglesiasg] by ChanServ09:19
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has joined #shogun09:54
-!- mode/#shogun [+o HeikoS] by ChanServ09:54
@wikingmornin'10:01
geektoniwiking: morning ;)10:02
@wikinggeektoni, i've got your email10:02
@wikingdo you want me to comment it on it or wait for stuff after 27?10:02
@wikingoh ok i've mixed up things10:02
@wikingor do you want to discuss now?10:02
geektonidepends which mail10:03
geektonithe one regarding the Detox II with Mike10:03
geektonior the other one10:03
@wikingeither10:03
geektoniI think we could wait for Mike to discuss abut how to divide the Detox II tasks.10:05
geektoniBut I think it won't be a problem since we don't have overlapping ones.10:05
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has quit [Ping timeout: 260 seconds]10:05
geektoniabout the timeline, did you see the (still empty) doc?10:06
@wikingyep i saw that one10:08
@wikingi mean it's good to have anything10:08
@wikingso to have a strating discussion10:08
@wikingbut yeah i'll send out an email10:08
geektoniok. I will update it later on, so you can give me your opinion10:09
geektonibtw, I have some "development" questions.10:09
geektoniDo we have a way to easily benchmark shogun algorithms?10:10
geektonibecause I want to test performance when I will begin to use the new progress bar and also when I'll begin to replace SG_REF with smart pointers.10:11
geektonisince I'm not so sure that these changes will speed up the code.10:12
@wikinggeektoni, well easy is a matter of opinion10:14
@wikingi reckon it'd be good if you would familiarize yourself10:14
@wikingwith this framework10:14
@wikinghttps://github.com/mlpack/benchmarks10:14
@wikingthere are quite some shogun algos plugged in there10:15
geektonithanks! I'll give a look.10:15
geektonisecondly, regarding smart pointers, the purpose will be only to change the underlying implementation, but not method definitions, right?10:17
@wikingmmm10:19
geektoniFor example, SGObject has some getters/setters which returns raw pointers.10:19
@wikingwell the idea on the end is actually get rid of SG_REF/UNREF macros10:19
geektonior accept raw ones10:19
geektoniyeah sure10:19
@wikingand any object that is derived from SGObject10:19
geektonibut, for example, since we are dealing with smart pointers, shouldn't we make the methods return smart pointers instead of raw ones?10:22
@iglesiasggeektoni: that is a good question10:22
@iglesiasgapart from getting rid of ref/unref in favour of using unique_ptr and so10:22
@iglesiasgshould we also change methods that return or accept naked pointers to smart pointers?10:23
geektoniI mean, what is the purpose of having smart pointers if we are still offering an API wich use raw ones.10:23
@iglesiasgI think there are at least two different considerations10:23
geektoniinglesiasg: exactly10:23
@iglesiasgfor instance, even if we keep in the api raw/naked pointers, there would be value in the refactoring since we would start really using automatic reference counting10:24
@iglesiasgwhich we currently do explicitly (via ref/unref)10:24
@wikinggeektoni, yeah the thing is that we can do this in 2 steps :)10:25
@iglesiasgif a method gets an argument by pointer (because it is going to modify it internally) but we are sure that there is not going to be transfer of ownership, then I10:25
@wikingof course it'll be a bit shitty10:26
@wikingbut we need to start it somewhere10:26
@iglesiasgI'd say it is fine that the interface of the method is with a naked pointer, no need for smart pointer in there10:26
@wikingbut on the end of the day we should transform the api10:26
@wikingwhich is now contradicting what iglesiasg says :)10:26
@iglesiasgbut indeed yeah, if we change everything to smart pointer I feel the code would be less error prone10:26
-!- mikeling [uid89706@gateway/web/irccloud.com/x-pcrfgetbacmqredu] has joined #shogun10:26
@wikingso we'll have to have an internal discussion regarding this10:26
@wikingwe can use Some<> :)10:27
@iglesiasgwiking: I agree with what you are saying actually! :-)10:27
mikelingwiking: hey, Had you read the email ?10:27
@wikingmikeling, yeah10:27
@wikingsorry i had a lot to handle on the weekend10:27
@wikingi'll get back within couple of hours today10:27
@wikingok?10:27
@iglesiasgit can be a two-step process10:27
mikelingwiking: sure!10:28
@iglesiasggeektoni: does it make sense to you?10:28
geektoniwiking: ok10:28
@wikingbut i would say10:28
@wikinglisitsyn, ping10:28
geektoniinglesiasg: yes, it does.10:28
@wikinglet's have a more detailed plan of this10:29
@iglesiasgbene10:29
@wiking:)10:29
geektoniI think I need first to get rid of all SG_REF/UNREF10:29
geektonithen I could think about changing the API ;)10:29
@wikinggeektoni, mmm we could prototype10:30
@wikingi mean the hting is10:30
@wikingthat last year we've learned that using .h for prototyping10:30
@wikingwill lead to a good result much faster10:30
@wikingthan just throwing ideas around10:30
mikelingBTW, I had test this pr https://github.com/shogun-toolbox/shogun/pull/3812 locally with ubuntu 16.04 and gcc 6.3.0.10:30
mikelingand there is no error shows up10:30
@wikingmikeling, how is it?10:30
@wikingok10:30
@wikinglemme try on an differnet setup10:30
@wikingyou can alwasy reproduce10:31
@wikingthe whole10:31
@wikingaaah fuck10:31
@wikingi know10:31
@wikingwait10:31
lisitsynwiking: yes10:31
@wikingalthough i dont understnad why the windows die as well10:31
@wikinglisitsyn, so Some<SGobject>10:31
@wikinggeektoni, do we run clang check with gcc or with clang setup?10:32
geektoniwith gcc setup10:32
@wikingmikeling, https://travis-ci.org/shogun-toolbox/shogun/jobs/231803323#L52410:32
@wikingso check the output of clang-format10:32
@wikingand before pushing10:32
@wikingtry to run it locally10:32
lisitsynyeah what about some?10:32
@wikingusing ./scripts/check_format.sh10:32
@wikinglisitsyn, so say one day we have all the SGobject wrapped to some10:32
@wikingSome10:32
geektoniwiking: nope, that won't work locally10:33
@wikinggeektoni, why not?10:33
lisitsynyes we better have that ;)10:33
@wikinglisitsyn, would you change all the api?10:33
@wikingwhere you pass SGObject around as pointers atm/10:33
geektoniwiking: because it was made to work only on Travis.10:33
@wikinggeektoni, vafaculo :)10:34
lisitsynwiking: yes I believe we don't need raw pointers in the API10:34
@wikingwhy? :)10:34
lisitsyninternally we can10:34
@wikinggeektoni, ok let's have that fixed that anybody can run that locally10:34
@wiking:)))10:34
@wikingmikeling, but basically you are having formatting problems10:35
geektoniwiking: ahah, what did you want to say? ;) I think that "vaffanculo" it is not the right one ;)10:35
@wikingin the code10:35
@wikinggeektoni, ma vaffanculo :)10:35
mikelingwiking: oh I see. I will check and update the pr10:35
geektoniaha much better10:35
@wikinggeektoni, have a good friend from milano10:35
mikelingthank you10:35
@wikinghe is always swearing10:35
@wikingand telling10:35
@wiking'figa'10:36
@wiking:)10:36
geektoniahahaha10:36
@wikingso yeah10:36
@wikingand poro dio10:36
@wiking*porco dio10:36
geektoniLOL10:36
@wikingyeah had many italian colleagues in the past :)10:36
geektoniThat's very nasty to say10:36
geektoni;)10:36
@wikingheheh well you know10:36
@wikingwhen you code10:36
@wikingand something doesn'twork10:37
@wiking:)))010:37
@wikingthen it's porco dio10:37
geektoniI know the feeling xD10:37
@wikinglisitsyn, bien!10:38
lisitsynwiking: bazdmeg10:38
@wikingthis channel will become south european in 2 seconds i have a feeling10:38
@wikinglisitsyn, :DDD10:38
lisitsynbazdmeg blyad10:38
lisitsynsuka!10:38
@iglesiasgqué bueno10:38
@iglesiasgme gusta10:38
geektonioh god, and this is all public right? xD10:39
@wikinglisitsyn, :DDDDDDDDDDDDDDd10:39
@iglesiasgand logged I think10:39
@wikingyeah it's still logged10:40
@wiking:)10:40
lisitsynoh my god what should we do now?!10:40
lisitsyn:P10:40
@wikinglisitsyn, bazdmeg :D10:40
@wikingooooh10:41
geektoniwiking: btw, I'll update the script for local environments.10:41
lisitsynI think we can say anything since we have 'fuck soeren' in the commit history10:41
@wikingi know one mikeling do you know what 'schabi' means in mandarin? :)10:41
@wikinglisitsyn, :DDDDDDDDDDDDDDDDDD10:41
@wikingmikeling, or "shabi"10:41
@wiking:)10:41
@iglesiasgepic commit10:41
@wikingi dont know the kanjis for it :)10:41
mikelingwiking: hhhhh yep10:41
mikelingI know it10:41
@wiking:D10:41
@wikingokokok so yeah10:41
@wikingi know one more on mandarin (bad word)10:42
@wikingbut that's not PC on english :P10:42
geektoniso much Knowledge.10:42
geektonixD10:42
@wikingiglesiasg, so if you ever go to china with a friend who is called Xavi10:42
* mikeling 's sad was broken so he is trying to save the data :(10:42
mikelingssd10:42
@wikingthat'll be a super funny story for him :)10:43
@wikingmikeling, :((( how broken? suddenly a lot of bad sectors?10:43
@wikinggeektoni, yeah never say Xavi in china :)))10:43
@wikingand the N words as well :P10:43
@wikingi mean either10:43
@wiking:)10:43
@wikinganyhow10:43
mikelingI don't know, my computer suddenly can't recognize it, no matter how I unplug it and check the connection10:44
@wikingmmmm shit10:44
@iglesiasgwiking: haha that's funny10:44
geektoniwiking: understood. I'll keep that for myself ;)10:45
@iglesiasgwiking: the translation to Spanish in translate is hilarious10:45
@wikingmmm lemme see :)10:45
geektoniwiking: so, I lost myself a bit. What was the final opinion on changing the API?10:51
@wikinggeektoni, if it's an internal fucntion it can remain bare *10:51
@wikingor ptr10:51
@wikingbut all public should be using the wrapped version10:51
geektoniok, so, for instance10:53
geektonihttps://github.com/shogun-toolbox/shogun/blob/develop/src/shogun/base/SGObject.h#L22910:53
geektonithis method should be ported as std::shared_ptr<SGIO> get_global_io(); ?10:53
geektoniwiking: ^10:58
@wikingmmm10:58
@wikingyeah i'm just thinking10:58
@wikingwhether we wanna have a wrapper for it10:58
@wikingor not10:58
@wikingsince we already have a wrapper10:59
@wikingthat has that functiionality (shread_ptr)10:59
geektoniyou mean Some<>?10:59
lisitsynmy vote (can I veto? ;) ) is to have some wrapper11:04
lisitsyneven if it is just shared_ptr11:04
lisitsynI tried implementing SG_REF and SG_UNREF in shared_ptr and it was a bit painful11:06
@iglesiasglisitsyn: why not just using std::unique_ptr directly?11:06
lisitsynI think we can benefit from wrapping it11:06
@iglesiasglisitsyn: I am not sure why we need a wrapper11:06
lisitsyntracking / debugging / whatever11:06
lisitsynreplacing it some day11:06
lisitsynin the worst case it should be using Some = shared_ptr;11:06
lisitsynbut not shared_ptr directly PLZ!!!11:07
@iglesiasgall righty11:07
geektonilisitsyn: ok, so, you suggest something like sg_shared_ptr?11:08
lisitsynyes but without ugly name11:08
lisitsynSCNR11:08
lisitsyn:P11:08
@iglesiasgxD11:08
lisitsynShared is fine as well11:08
lisitsynPtr11:08
@iglesiasgsg_shared_ptr does not loook so ugly to me11:08
lisitsynwhatever :)11:08
@iglesiasgbut yeah, Shared, Some, or Ptr are fine too11:08
@iglesiasgPtr is the name they use in OpenCV I think11:08
@iglesiasganother thing *might* be worth discussing11:09
@iglesiasgI am not sure why we are defaulting to using shared pointers instead of unique11:09
lisitsynhttp://trumpgenerator.com/next/?cgbcbjcmhelbjtvitbor11:09
@iglesiasghaha11:10
lisitsyniglesiasg: porque http://trumpgenerator.com/next/?bjgegskvssbwwkkcouox11:10
@iglesiasgif Trump says that, then I think we must use sg_shared_ptr haha11:10
lisitsynuhm I don't really see how unique fits here11:10
@iglesiasgque no11:10
lisitsynunique is something to be placed in your code blocks11:11
@iglesiasgunique_ptr is nicer since it does not even need to do reference counting11:11
@iglesiasgI don't think shared ownership is needed all the time11:11
lisitsynwell but we always have it shared11:11
lisitsynyou can't even share features using unique11:11
lisitsynSVM and KNN pointing to the same features kaboom11:12
@iglesiasgwhy?11:12
geektonilisitsyn: http://trumpgenerator.com/next/?cgbcbjcmhelbjtvitbor this made my day.11:12
@iglesiasgI guess what I am trying to say is that they don't need to be owning to the objects to make use of them11:13
lisitsynuhmm but who deletes it then?11:13
@iglesiasgwhen they go out of scope?11:14
@iglesiasgraii ftw11:14
lisitsynyeah it could work in C++ but well not in python11:14
lisitsynthere is a place for unique that is pimpl ;)11:15
@iglesiasgwhy not in Python?11:15
@iglesiasgthe way that I'm thinking about it is that the Python code after all fallsback into the c++ impl.11:16
lisitsynporque python is reference counted by nature11:16
lisitsynI don't see how could it be implemented with scopes11:16
lisitsynyou just share everything there11:16
lisitsynI just don't see it applied here need an example :P11:19
@iglesiasgI am thinking about it11:19
@iglesiasgthe members would need to be naked pointers, which I don't like that much11:20
@iglesiasgwith the members I mean e.g. features inside KNN or SVM11:20
lisitsynyeah me neither11:20
lisitsynI'd go with everything shared just like we do it now11:21
lisitsynbut automatically11:21
@iglesiasgbut let me think about it11:21
@iglesiasgwhy I don't like it that much11:21
lisitsynI believe having a wrapper is always useful11:21
lisitsynwe don't need performance of switching kernels and features11:22
@iglesiasgyeah, indeed11:24
@iglesiasgI was not thinking about performance that much but avoiding circular relationships with the shared pointers11:24
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has joined #shogun11:25
-!- mode/#shogun [+o HeikoS] by ChanServ11:25
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has quit [Client Quit]11:25
geektonilisitsyn, iglesiasg wrapper confirmed? ;)11:28
@iglesiasgsure11:28
lisitsyngeektoni: Some<T> is already okish but there was an issue with leak11:28
lisitsynI haven't managed to find the reason why11:28
@iglesiasghihi it is what happens with custom stuff11:29
@iglesiasgNewSome = std::shared_ptr ?11:29
geektonilisitsyn: so, the idea is to use just some<T>? Or do we need another one?11:30
geektonibtw, leak? Is it reported somewhere? Like as Github issue?11:30
lisitsynno why to introduce?11:30
lisitsynI see no point of adding another one11:30
@iglesiasgI think the best point about having a wrapper is flexibility in change if we want to (e.g. in the future, different compiler)11:32
geektoniok ok, I misunderstood.11:32
lisitsynI don't think there is an issue at github11:32
lisitsynbut there is a leak in generated meta C++ tests11:32
@iglesiasgso just making it as barebone as assignment std::shared_ptr sounds like a good plan11:33
lisitsynyes but I've tried using shared_ptr11:33
lisitsynI had a few issues integrating it with SG_REF11:33
lisitsynI strongly suggest to just revise Some<T>11:33
geektonilisitsyn: ok.11:36
lisitsynthe problem is that shared_ptr has its own reference counter11:36
-!- HeikoS [~heiko@host81-152-92-25.range81-152.btcentralplus.com] has joined #shogun11:36
-!- mode/#shogun [+o HeikoS] by ChanServ11:36
lisitsynthere is no point of having two reference counters now11:36
lisitsynit gets really hard to synchronize them :P11:36
lisitsyngeektoni: I'd suggest to check Some and where SG_REF/SG_UNREF happens11:37
geektonilisitsyn: I'll take a look.11:38
geektonibut11:38
geektonino, never mind11:39
-!- geektoni [c1cdd24a@gateway/web/freenode/ip.193.205.210.74] has quit [Ping timeout: 260 seconds]12:03
-!- HeikoS [~heiko@host81-152-92-25.range81-152.btcentralplus.com] has quit [Ping timeout: 260 seconds]12:45
-!- HeikoS [~heiko@89.105.104.229] has joined #shogun13:29
-!- mode/#shogun [+o HeikoS] by ChanServ13:29
-!- HeikoS [~heiko@89.105.104.229] has quit [Ping timeout: 260 seconds]13:44
-!- iglesiasg [~iglesiasg@217.119.234.214] has quit [Ping timeout: 268 seconds]13:46
-!- iglesiasg [~iglesiasg@wlan075086.mobiel.utwente.nl] has joined #shogun13:47
-!- mode/#shogun [+o iglesiasg] by ChanServ13:47
-!- sukey [~nodebot@ks312251.kimsufi.com] has joined #shogun14:14
-!- mode/#shogun [+o sukey] by ChanServ14:14
@sukeyPull Request #3812 "Add global fixture with binary label data"  synchronized by MikeLing - https://github.com/shogun-toolbox/shogun/pull/381214:19
mikelinglisitsyn: Hi, I'm wondering about which code style we are following? Because pr looks a little wired like the function name is aligned with the 'public' in https://github.com/shogun-toolbox/shogun/pull/3812/files#diff-13214f4777a5a01dc75170c3abcbc95bR1314:32
mikelingwiking: ^14:34
lisitsynit looks a bit weird to me14:41
@wikingi think we need to modify the definition file a bit14:42
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun14:48
-!- mode/#shogun [+o HeikoS] by ChanServ14:48
-!- geektoni [~geektoni@93-34-234-212.ip52.fastwebnet.it] has joined #shogun15:18
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 240 seconds]15:22
-!- HeikoS [~heiko@eduroam-int-pat-8-0.ucl.ac.uk] has joined #shogun15:37
-!- mode/#shogun [+o HeikoS] by ChanServ15:37
-!- HeikoS [~heiko@eduroam-int-pat-8-0.ucl.ac.uk] has quit [Ping timeout: 246 seconds]16:04
-!- tctara_ [~quassel@128.199.61.169] has joined #shogun16:23
rcurtingeektoni: wiking: that mlpack benchmarks repository should see some activity this summer, we have a student working on it16:23
rcurtinif you see any issues with how we are benchmarking shogun code, please submit a PR! :)16:23
rcurtinwe are not shogun experts necessarily :)16:24
@wikingsure thing16:24
@wikingthnx for the headsup16:24
geektonircurtin: sure!16:24
geektoniwiking: by the way, since during my project I will touch almost all Shogun classes, I would like to pass them through clang-format16:26
geektoniso that we will have a consistent codebase.and the Travis style check will stop to fail every time16:27
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun16:37
-!- iglesiasg [~iglesiasg@wlan075086.mobiel.utwente.nl] has quit [Ping timeout: 260 seconds]16:46
-!- iglesiasg [~iglesiasg@217.119.234.214] has joined #shogun16:47
-!- mode/#shogun [+o iglesiasg] by ChanServ16:47
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]16:49
@sukeyPull Request #3813 "Modify check_format.sh so that it can be used on local environments."  opened by geektoni - https://github.com/shogun-toolbox/shogun/pull/381316:58
@sukeyPull Request #3813 "Modify check_format.sh so that it can be used on local environments."  synchronized by geektoni - https://github.com/shogun-toolbox/shogun/pull/381317:07
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun17:11
-!- mode/#shogun [+o HeikoS] by ChanServ17:11
-!- mikeling [uid89706@gateway/web/irccloud.com/x-pcrfgetbacmqredu] has quit [Quit: Connection closed for inactivity]17:17
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun17:28
-!- geektoni [~geektoni@93-34-234-212.ip52.fastwebnet.it] has left #shogun []17:29
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]17:58
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun18:05
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Ping timeout: 260 seconds]18:10
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun18:11
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]18:13
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun18:14
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 240 seconds]18:53
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]19:27
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun19:28
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]19:38
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun19:56
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]19:59
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:01
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:02
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:03
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:05
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:06
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:07
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:08
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:10
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:10
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:12
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:14
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:15
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:16
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:17
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:18
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:20
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:20
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:21
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:22
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:24
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:25
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:26
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:27
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:28
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:29
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:30
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:31
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:32
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:33
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]20:34
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun20:36
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has quit [Remote host closed the connection]21:35
-!- goksinen [~goksinen@rrcs-50-75-193-138.nyc.biz.rr.com] has joined #shogun21:39
@sukeyPull Request #3814 "Add information about the newly introduced style checks inside DEVELOPING.md."  opened by geektoni - https://github.com/shogun-toolbox/shogun/pull/381422:10
@sukeyPull Request #3815 "Minor fix to .clang-format."  opened by geektoni - https://github.com/shogun-toolbox/shogun/pull/381522:28
-!- dhama [2f080b6c@gateway/web/freenode/ip.47.8.11.108] has joined #shogun23:47
-!- dhama [2f080b6c@gateway/web/freenode/ip.47.8.11.108] has left #shogun []23:48
--- Log closed Tue May 16 00:00:31 2017
« Sunday, 2017-05-14 Index Tuesday, 2017-05-16 »

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