IRC logs of #shogun for Friday, 2016-05-27

« Thursday, 2016-05-26 Index Saturday, 2016-05-28 »
--- Log opened Fri May 27 00:00:13 2016
-!- Saurabh7- [Saurabh7@gateway/shell/panicbnc/x-tkhunxevfgkzzhnq] has quit [Read error: Connection reset by peer]00:39
-!- Saurabh7 [Saurabh7@gateway/shell/panicbnc/x-zjbrorlyietxjdsm] has joined #shogun00:42
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 272 seconds]01:15
-!- besser82 [~besser82@fedora/besser82] has joined #shogun01:23
-!- mode/#shogun [+o besser82] by ChanServ01:23
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 260 seconds]01:39
shogun-buildbotbuild #6 of clang - thread analysis is complete: Failure [failed compile]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang%20-%20thread%20analysis/builds/6  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>03:44
shogun-buildbotbuild #5 of clang - undefined behaviour analysis is complete: Failure [failed compile]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang%20-%20undefined%20behaviour%20analysis/builds/5  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>03:47
-!- OXPHOS [9d8b131c@gateway/web/freenode/ip.157.139.19.28] has quit [Quit: Page closed]03:49
-!- sonne|osx_ [~sonne@x4db38903.dyn.telefonica.de] has joined #shogun05:54
-!- sonne|osx [~sonne@x4e33d889.dyn.telefonica.de] has quit [Ping timeout: 272 seconds]05:56
-!- sonne|osx_ is now known as sonne|osx05:56
shogun-buildbotbuild #5 of memleak - valgrind is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/memleak%20-%20valgrind/builds/5  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>06:16
shogun-buildbotbuild #1008 of nightly_none is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_none/builds/1008  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>06:27
shogun-buildbotbuild #1137 of nightly_default is complete: Failure [failed test notebooks]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/1137  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>07:41
-!- HeikoS [~heiko@nat-196-45.internal.eduroam.ucl.ac.uk] has joined #shogun11:50
-!- mode/#shogun [+o HeikoS] by ChanServ11:50
@wikingHeikoS: hey ehy11:50
@HeikoSwiking: jo11:51
@HeikoSall good?11:51
@wikingstandups11:52
@wiking:)11:52
@wikingi'm missing those11:52
@wikinghttps://en.wikipedia.org/wiki/Gold_%28linker%2911:53
lisitsynok let me start11:53
lisitsynNDA NDA NDA NDA11:53
lisitsynNDA NDA NDA11:53
lisitsyngoing to finish with NDA11:53
@HeikoSwiking:11:53
@HeikoSping everyone!11:53
@wiking????????11:53
@wikinglisitsyn: nda?11:53
@wikingHeikoS: https://en.wikipedia.org/wiki/Gold_%28linker%2911:54
@HeikoSarianepaola, Saurabh7, arianepaola stand up reminder :)11:54
lisitsynwiking: yes that's my standup11:54
@HeikoSwiking: keepo in mind time zone11:54
lisitsynwiking: I told you what I am up to11:54
@HeikoSI just started11:54
lisitsyn:D11:54
@HeikoSwiking: tried it?11:54
@wikingHeikoS: yes11:55
lisitsynwhat's gold and why11:55
lisitsynfaster?11:56
@HeikoSwiking: and?11:56
@wikingHeikoS: just running the time for make11:57
@HeikoSwiking: I see11:57
@HeikoSwiking: btw ever tried ninja on shogun?11:57
@HeikoSalso is a bit faster, especially when re-compiling11:57
@HeikoSarianepaola told me abou tit11:57
@wikingninja?11:57
@wikingwhat the fuck is ninja?11:57
lisitsynsome alternative to make11:58
lisitsyncmake can generate that11:58
@wikingaaaah11:58
@wikingi really dont care about make's speed11:59
@wikingmore linker speed11:59
@wikingunit test generation is really a pain now11:59
@wiking(trying to fix memleaks)12:00
lisitsynoh btw12:00
lisitsyncan I help?12:00
@wikinglisitsyn: memleaks :)12:00
lisitsynyes12:00
@wikingi've fixed some12:00
lisitsyn'some'12:00
lisitsynor some?12:00
@wikingthere's a really nasty one12:00
lisitsynwhat was wrong?12:00
@wikinglisitsyn: look at the generated one12:00
@wikingplz12:00
@wikingwhy do they leak12:00
@wikingor why does valgrind say that they leak12:01
lisitsynwiking: ok12:01
@wikinglisitsyn:12:01
@wikinghttp://buildbot.shogun-toolbox.org/memcheck/20160527-0100.html12:01
lisitsynwiking: well I'd expect off-by-one in ref counter12:02
lisitsynlet me check12:02
@wikingthnx12:02
@HeikoSwiking: so is the gold linker faster?12:07
@HeikoSwiking: I would not worry too much about the generated examples right now12:07
@HeikoSwiking: even if they produce many leaks each, it is a single fix that fixes all of them12:08
@HeikoSor do you disagree?12:08
@HeikoSvalgrind told me its the new operator on CSGOBjects that is not freed12:08
@wikingtesting gold12:10
@wikingit should be faster12:10
@wiking:)12:10
-!- HeikoS_mobile [~Mutter@nat-195-25.internal.eduroam.ucl.ac.uk] has joined #shogun12:11
lisitsynmobile!12:11
HeikoS_mobileYes!!12:11
HeikoS_mobilelisitsyn: coming to london soon?12:11
lisitsynHeikoS_mobile: not very likely12:12
HeikoS_mobileNooooo12:12
HeikoS_mobileJust weekend?12:12
lisitsynHeikoS_mobile: well weekend is likely to be possible %)12:13
lisitsyn  Could NOT find PythonLibs: Found unsuitable version "2.7.10", but required12:14
lisitsyn  is exact version "2.7.11"12:14
lisitsynwhy do I get this?12:14
lisitsynwiking: why do we require 2.7.11? :)12:15
@wikingah12:16
@wikinglisitsyn: this is because you have 2 pythons12:16
@wikingand it'll mix and match and it'll fuck up your python modular interface12:16
-!- HeikoS_mobile [~Mutter@nat-195-25.internal.eduroam.ucl.ac.uk] has quit [Read error: Connection reset by peer]12:17
@wikingit detects python 2.7.11 interpreter (probably /usr/local/bin/python or /opt/local/bin/python)12:17
@wikingbut then it detects the python lib to be in /usr/lib/12:17
@wikingwhich is the system python12:17
@wikingnot the interpreter12:17
lisitsynwiking: hmm not really I have installed python12:18
lisitsynand gave it a prefix12:18
@wikinglisitsyn: can you pastebin your full cmake plz12:19
lisitsynwiking: http://pastebin.com/UEdnxMX612:20
lisitsynthis looks strange12:20
@wikingooooooo12:20
@wikinghow is this possible? :D12:20
@wiking-- Found PythonInterp: /Users/lisitsyn/Prefix/bin/python (found version "2.7.11")12:20
@wiking  is exact version "2.7.11" (found /Users/lisitsyn/Prefix/lib/libpython2.7.a)12:20
@wikingbut that supposed to be 2.7.10?12:20
lisitsynehmm ok may be I need to reinstall it12:21
@wikingHeikoS: hahahahah12:25
@wikinggold is slower12:26
@wikingwaaaaaaaaaaaaay slower12:26
@wikingHeikoS: the problem with linking is the generated unit tests actually12:26
@HeikoSwiking: hehew12:29
@HeikoSwiking: so I think splitting the tests woul dbe good12:29
@wikingthose big serialization tests12:29
@HeikoShave multiple executables12:29
@wikingno12:29
@HeikoSyeah12:29
@wikingthat is a fucking pain12:29
@wikingwe dont want that12:29
@wikingjust remove the serialization stuff12:29
@HeikoSthe thing is12:29
@wikingonce cereal is in12:29
@HeikoSif I develop code12:29
@HeikoSI usually do test driven12:29
@HeikoSbut I have to compile ALL unit tests every time12:30
@HeikoSannoyintg12:30
@HeikoSdont want that12:30
@HeikoSI just want mine12:30
@HeikoScurrently not possible12:30
@wikingbut it'll compile only that12:30
@HeikoSwell12:30
@wikingjust linking takes a while12:30
@HeikoSI then often do a brnach change to fix something else12:30
@HeikoSand switch from say Debug to Release12:30
@HeikoSto test performanc12:30
@HeikoSthen go back12:30
@wikingbut there's nothing you can do about that12:30
@HeikoSand have to re-compile12:30
@wikingyeah but man12:30
@wikingyou are asking about something12:30
@wikingthat is impossible to solve normally12:31
@wikinguse 2 sources12:31
@HeikoSI was thinking after plugins arrived12:31
@HeikoSand we have a dependency structure12:31
@HeikoSso that I can just re-compile stuff I need/want12:31
@HeikoSwe can apply that to the tests somehow12:31
@HeikoSbut well12:32
@HeikoSmy dreams, you know ;)12:32
@HeikoSwiking: but ok lets remove serialization then12:32
@HeikoSafter cereal is there12:32
@HeikoSshould be much faster to start with12:32
@wikingbecause now we test and generate code12:33
@HeikoSyeah I know12:33
@wikingfor every class12:33
@HeikoSmaybe we can give cmake a regexp12:33
@wikingi mean the fucking unit test exec12:33
@wikingis 161 megs12:33
@HeikoSfor the folders to include unit test sources from12:33
@HeikoShaha12:33
@HeikoSwiking: actually wouldnt that work?12:33
@wikingthat's why linking takes forever12:33
@wiking?12:33
@wikingjust remove the generated code and done12:33
@HeikoSyou think?12:34
@HeikoSstill compiling them takes loing12:34
@HeikoSon my machine12:34
@HeikoSand have to re-compile when doing branch switches often12:34
@wikingHeikoS: then you dont have ccache normally setup12:34
@HeikoSso if we have a cmake option12:34
@HeikoSwiking: I do12:34
@HeikoSits if too many things change at once12:34
@wikingccache -s12:34
@wikingwhat's the outptu of that?12:35
@HeikoSnot found12:35
@HeikoSah12:35
@HeikoSspace12:35
@HeikoSsorry12:35
@HeikoScache size                          23.4 Gbyte12:35
@wikingand how big is your ccache?12:35
@wikingmax cache size12:35
@HeikoS61Mb12:36
@wikingyou mean gbyte12:36
@wikingi guess12:36
@HeikoSnope12:36
@wikingcan you fucking copy paste the whole thing12:36
@HeikoSfiles in cache                     3350012:36
@HeikoScache size                          23.4 Gbytes12:36
@HeikoSmax cache size                      61.0 Mbytes12:36
@wikingplz12:36
@wikinghahahaha12:36
@wikingman12:36
@HeikoSsec12:36
@wikingis this you think12:36
@HeikoSwiking: I never touched this at all12:36
@wikinga normally setup12:36
@wikingccache?12:36
@wiking:D12:36
@HeikoSno idea, using default settings12:36
@wikingyeah12:36
@wikinghence i said12:36
@HeikoSok let me try then12:36
@wiking12:32 <@wiking> HeikoS: then you dont have ccache normally setup12:36
@HeikoScan you shed some light on good options?12:37
@wikingso first of all12:37
@wikingclear the fucking cache12:37
@HeikoSmaybe we can put that in some readme12:37
@HeikoSdid before12:37
@wikingHeikoS: maybe people start RTFM12:37
@HeikoSdidnt help12:37
@wikingno?12:37
@wiking:)12:37
@wikingi mean fuck12:37
@wikingwe are not like12:37
@wiking"how to use your computer"12:37
@wikingccache -C -z12:37
@wikingdo this first12:37
@HeikoSwiking: maybe think a sec12:37
@HeikoSif I get this wrong, how likely is that users get this wrong?12:38
@wikingand then do12:38
@wikingccache -M10G12:38
@wikingand after the first compile12:38
@wikingyou'll have it good12:38
@HeikoSok trying12:38
lisitsynhey but people don't have to compile that12:38
@wikingbut12:38
@HeikoSthanks!12:38
@wikingif you keep switching12:38
@wikingbetween release and debug12:38
@wikingthat's gonna be always a pain12:38
@HeikoSwiking: uh12:38
@HeikoSthats when it happens12:38
@wikingbecause that is totally different12:38
@wikingcompiler flags12:38
@HeikoSI know12:38
@wikingbut again12:39
@HeikoSthis is why I suggest the option12:39
@wikinghaving 10G12:39
@wikinghopefully will cache you both12:39
@HeikoScmake option: regexp of directories to include for unit testing12:39
@wikingthat option is bull fucking shit12:39
@wikingi'm sorry12:39
@wikingyou are asking for thing12:39
@HeikoSwhy?12:39
@wikingthat is not normal12:39
@wikingHeikoS: do a buildbot12:39
@wikingit compiles12:39
@wikingrelease12:39
@wikingand has ccache12:39
@wikingbut again12:39
@wikingfirst do what i said12:39
@wikingtest it12:39
@HeikoSI did12:39
@wikingand you'll see12:39
@HeikoSwhats wrong with the option?12:39
@wikingwhich option?12:39
@wikingthe regexp blalbla12:40
@wikingcome ooooon12:40
@HeikoSyes12:40
@wikingoh coooooooooomeeeeeee oooooooooooooon12:40
@HeikoSlike "unit/classifier/*;unit/kernel*"12:40
@wikingyes yes12:40
@wikinglala12:40
@HeikoSand then only those are compiled12:40
lisitsynwhy?12:40
@wikingtalk to the hand12:40
@HeikoSbecause it saves me time12:40
lisitsynbut you don't know what tests could you break12:40
@wikingmake does12:41
@HeikoSlisitsyn: yes but I dont care since I develop a new feature that only affects my newly written tests12:41
@wikingwhat it knows what it needs to do12:41
@wikingif you are trying to be smarter12:41
@wikingthan make12:41
@wikinghave fun12:41
@wikingi dont see the point of reinventing the wheel there12:41
@HeikoSwiking: you havent told me *why* is is a bad idea in your eyes12:41
@wikingjust did12:41
@HeikoSI am not trying to be smarter just trying to understand12:42
@wikingyou are tyring to solve things12:42
@wikingthat are solved12:42
@wikingby tools12:42
@wikingalready12:42
@wikingand if that's not good enough for you12:42
@wikingyou are gonna have to reinvent the whole thing12:42
@wikingi'm not standing in your way12:42
@wikingbut12:42
@wikingexperiment with that12:42
@wikingfirst in a dummy library12:42
@wikingand make it feasible12:42
@wikingand running on that w/o errors12:42
@wikingif that's good12:42
@wikingthan MAAYBE12:42
@wikingmaaaybe it's a good idea to have it in shogun12:42
@wikingagain12:42
@wikingyou are trying to solve what's make is for12:42
@HeikoSI dont see why it is hard12:43
@wikingor ninja12:43
@wikingit's not about being ahrd12:43
@HeikoSjust about including things in cmakelist in the unit test dir12:43
@wikingit's about introducing12:43
@wikingfucking errors12:43
@wikingpossible source of errors12:43
@wikinglook at the current code12:43
@wikingeven though12:43
@HeikoSwhat type of error?12:43
@wikingwe have tets for every fucking things12:43
@wikingwe still manage12:43
@wikingto have leaking shit every time12:43
@wikingi look at the code 2 months later12:43
@HeikoSyou maybe dont get what I want from that12:43
@wikingi know what you want12:43
@HeikoSI just want to use this locally12:43
@HeikoSno build will have the option set on12:44
@wikingyeah12:44
@wikingbut if you set the option to do this12:44
@wikingthen you possible will have somebody use it12:44
@wikingand then if it's used12:44
@wikingit'll possible get some code in12:44
@wikingsomewhere12:44
@wikingi mean for fuck sakes12:44
@wikingwe managed to merge code in12:44
@wikingthat fucking leaks12:44
@wikingin the obvious way12:44
@wikingdo you think if you allow this to have locally12:44
@wikingthe same thing wont happen12:44
@HeikoSthat is very different12:44
@wikingyeah very but still the same12:44
@wikingit's human error12:45
@HeikoSI dont agree12:45
@wikingok good12:45
@HeikoSvery different kind of problem12:45
@wikingthen we dont agree12:45
@wikingfirst12:45
@HeikoSsure fine with that :D12:45
@wikingstart using ccache12:45
@wikingin a normal way12:45
@wikingnot with max size of 64 megabytes12:45
@wikingwhen that really doesn't work for you12:45
@HeikoSwiking: do you know why this is default size?12:45
@wikingit's not12:45
@HeikoSmmh12:45
lisitsynthe default size is 1GB12:45
lisitsyniirc12:45
@HeikoSweird12:45
@wikingdefault is usually 1 or 512:45
lisitsynin ubuntu12:45
@wikingG12:45
@wikingnot M12:45
@HeikoSokok12:45
@HeikoSlets see12:45
@HeikoShopefully that solves it12:46
@wikingsince we eliminated a lot of -D compiler flags12:46
@HeikoSwiking: you know just really trying to play a bit with the idea12:46
@wikingit works quite nicely12:46
@HeikoSnot saying we should do this thing12:46
@HeikoSjust thinking loud12:46
@wikingsee for example the performance of the qemu based aarch64 bot12:46
@HeikoSkk12:46
-!- lambday [8028b10a@gateway/web/freenode/ip.128.40.177.10] has joined #shogun12:46
-!- mode/#shogun [+o lambday] by ChanServ12:47
@wiking2 hours something12:47
@wikingdown to 40 mins12:47
@wikingthat's ccache for you12:47
@HeikoSwiking: yeah I see this here as well12:47
@wikingwell not likely12:47
@wikingwith 64 megs12:47
@wiking:)12:47
@wikingsorry i couldn't stand it12:47
@wikingthere are soooo many other things to fix12:47
@HeikoSwiking: any ideas on the state of cereal btw12:48
@wikingHeikoS: it's not detectable12:48
@wiking:DDD12:48
@wikingthat's where we've managed to get12:48
@HeikoSdroping serialization code would reduce binary a lot I guess ;)12:48
@HeikoShaha12:48
@HeikoSah man12:48
@HeikoSthis all should have happened last month I guess12:48
@wikingso hopefully today12:48
@HeikoSyeah12:48
@wikingwe'll start getting serialization12:48
@HeikoSwiking: btw12:48
@HeikoSyou know the structure of the shogun structures12:49
@HeikoScontainer, string, basic12:49
@HeikoSand in these guys you can put float, SGVector, SGMatrix, SGString etc12:49
@HeikoSso too many possibilities12:49
@HeikoSwho needs strings of matrices?12:49
@HeikoSso I suggest to simplify that12:50
@wikingnooooo12:50
@wikingwhat do you do with12:50
@wikingcategoricals?12:50
@HeikoS?12:50
@wikinghow do you represent categoricals12:50
@wikinglike12:50
@wikingfeature: "low", "medium", "high"12:51
@HeikoSkind of dont see the connection to what I just said12:51
@HeikoSjust want to get rid of all these combinations that we never use12:52
@HeikoSstring of matrices e.g.12:52
@HeikoSmatrix of matrices12:52
@wikingbut again12:52
@wikinglet's say i have12:52
@wikinga Features12:52
@wikingwhere it only has strings12:52
@wikinghow do you represent that12:52
@wiking?12:52
@wikingi mean Features = SGMatrix12:52
@wikingsorry12:52
@wikingDenseFeatures = SGMatrix12:52
@wikingright/12:52
@HeikoSyeah12:53
@HeikoSstring features still works12:53
@HeikoSjust only strings of basic types12:53
@HeikoSnot of SGVectors type12:53
@HeikoSsame with array12:53
@HeikoSso categoricals are just represented using ints12:55
@HeikoSand we can have strings of that, and matrices12:56
@HeikoSand vectors12:56
@HeikoSwhat I want to avoid is this collection of SG* types12:56
@HeikoSbecause that is never used anyways12:56
@HeikoSand it accounds for most of the code in current parameter fraemwork12:57
@HeikoSwiking: ^12:57
-!- sonne|osx [~sonne@x4db38903.dyn.telefonica.de] has quit [Quit: sonne|osx]13:17
-!- leagoetz [~leagoetz@nat-247-179.internal.eduroam.ucl.ac.uk] has joined #shogun13:34
-!- HeikoS_mobile_ [~Mutter@nat-195-25.internal.eduroam.ucl.ac.uk] has joined #shogun13:51
-!- HeikoS_mobile__ [~Mutter@82-132-239-80.dab.02.net] has joined #shogun13:51
-!- HeikoS_mobile_ [~Mutter@nat-195-25.internal.eduroam.ucl.ac.uk] has quit [Read error: Connection reset by peer]13:52
-!- HeikoS_mobile__ [~Mutter@82-132-239-80.dab.02.net] has quit [Remote host closed the connection]13:54
-!- HeikoS [~heiko@nat-196-45.internal.eduroam.ucl.ac.uk] has quit [Ping timeout: 244 seconds]14:25
shogun-buildbotbuild #2877 of bsd1 - libshogun is complete: Failure [failed configure]  Build details are at http://buildbot.shogun-toolbox.org/builders/bsd1%20-%20libshogun/builds/2877  blamelist: Viktor Gal <viktor.gal@maeth.com>14:50
shogun-buildbotbuild #15 of xenial - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/xenial%20-%20libshogun/builds/15  blamelist: Viktor Gal <viktor.gal@maeth.com>14:51
@wikingshogun-buildbot: force build --branch=develop 'memleak - valgrind'14:54
shogun-buildbotThe build has been queued, I'll give a shout when it starts14:54
arianepaolahello everyone15:09
leagoetzhey15:09
@wikinghi15:10
arianepaolawikining: I am continuing to work on the unit tests FisherLDA_unittest.cc and LDA_unittest.cc 15:16
-!- HeikoS [~heiko@nat-196-45.internal.eduroam.ucl.ac.uk] has joined #shogun15:17
-!- mode/#shogun [+o HeikoS] by ChanServ15:17
@wikingarianepaola: the fisher one is quite simple15:17
@wikingit's just a -1*15:17
@wikingLDA is something that needs to be debugged15:18
@wikingimo the  FisherLDA unit tests should be changed15:18
@wikingas it's an eigen vector with opposite sign15:18
@wikingthe LDA svd stuff15:18
arianepaolayes value * (-1)15:18
@wikingthere needs to be fully debugged15:18
@wikingwhere is the big change15:19
@wikingonce we know it15:19
@wikingwe can consult ggael or ChriSopht on #eigen15:19
@wikingthey are pretty helpful15:19
@wikingthey fixed a bug in develop eigen within a day i reported15:19
@wikingi just didn't have time to debug15:20
@wikingwhere and what is happening actually15:20
@wikingwithin the LDA SVD part15:20
@wikingso if you could trace where the first time a matrix' value changes15:20
@wikingthat would be super helpful15:20
@wikingeither gdb/lldb or just simply std::cout << matrix << std::endl;15:20
arianepaolayes I will also get the latest eigen then15:20
-!- HeikoS [~heiko@nat-196-45.internal.eduroam.ucl.ac.uk] has quit [Quit: Leaving.]15:26
-!- HeikoS [~heiko@nat-196-45.internal.eduroam.ucl.ac.uk] has joined #shogun15:28
-!- mode/#shogun [+o HeikoS] by ChanServ15:28
shogun-buildbotbuild #6 forced15:29
shogun-buildbotI'll give a shout when the build finishes15:29
Saurabh7HeikoS: hi !15:35
@wikingSaurabh7: daily :)15:36
Saurabh7wiking: hi :)15:36
Saurabh7got parallel xval working15:36
Saurabh7with kernel svm for now15:36
@wikingplan for next?15:37
Saurabh7and denseFeatures15:37
@wikingany blocking issue?15:37
Saurabh7I will push in branch i think15:37
Saurabh7I need advice how to make it general15:37
@wikingif you share the code15:37
@wikingthen i can help15:37
@wikingand i can commet15:37
@wiking*comment15:37
Saurabh7ok doing15:38
@wikingthnx15:38
@wikingdo you have a feature branch?15:38
Saurabh7no15:38
@wikingmmm we could create one,... HeikoS ?15:38
Saurabh7i will do it in normal one for now lets see15:38
@wikingk15:38
-!- HeikoS [~heiko@nat-196-45.internal.eduroam.ucl.ac.uk] has quit [Ping timeout: 260 seconds]15:39
arianepaolahi Saurabh715:39
shogun-buildbotbuild #6 of FC23 - libshogun - aarch64 is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/FC23%20-%20libshogun%20-%20aarch64/builds/6  blamelist: Viktor Gal <viktor.gal@maeth.com>15:42
-!- HeikoS [~heiko@nat-196-45.internal.eduroam.ucl.ac.uk] has joined #shogun15:44
-!- mode/#shogun [+o HeikoS] by ChanServ15:44
Saurabh7arianepaola: hi!15:45
-!- HeikoS [~heiko@nat-196-45.internal.eduroam.ucl.ac.uk] has quit [Ping timeout: 240 seconds]15:53
-!- HeikoS [~heiko@nat-196-45.internal.eduroam.ucl.ac.uk] has joined #shogun16:00
-!- mode/#shogun [+o HeikoS] by ChanServ16:00
@HeikoSSaurabh7: hi16:07
@HeikoShow are things going?16:07
Saurabh7HeikoS: hi16:07
Saurabh7I got xval working in parallel now16:07
@HeikoSSaurabh7: great!16:07
Saurabh7for kernel svm16:07
@HeikoSsend a PR16:07
Saurabh7not in general16:08
@HeikoSeven if not done, just mention that in the description16:08
@HeikoSthats fine16:08
@HeikoSwanna look at it16:08
Saurabh7need help i am pshing to branch for u to see16:08
@HeikoSSaurabh7: you can just push to your fork and then send a pr16:08
Saurabh7yup16:08
@HeikoSSaurabh7: for now, you can send that against develop16:09
-!- HeikoS_mobile__ [~Mutter@nat-195-25.internal.eduroam.ucl.ac.uk] has joined #shogun16:10
-!- HeikoS_mobile__ [~Mutter@nat-195-25.internal.eduroam.ucl.ac.uk] has quit [Remote host closed the connection]16:14
-!- HeikoS [~heiko@nat-196-45.internal.eduroam.ucl.ac.uk] has quit [Ping timeout: 276 seconds]16:14
-!- besser82 [~besser82@fedora/besser82] has joined #shogun16:24
-!- mode/#shogun [+o besser82] by ChanServ16:24
-!- OXPHOS [9d8b131c@gateway/web/freenode/ip.157.139.19.28] has joined #shogun16:25
-!- leagoetz [~leagoetz@nat-247-179.internal.eduroam.ucl.ac.uk] has quit []16:38
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 252 seconds]16:57
@wikingOXPHOS: yo16:59
@wikingOXPHOS: as said time for squash and get the cereal going \o16:59
OXPHOSwiking hey. I can't get the CMakeLists back..I squashed but the CMakeLists file has multiple changes17:00
OXPHOSI think the feature branch needs rebase17:01
@wikingmmm17:01
@wikingwait17:01
@wikingwhat exactly did you do :)17:01
@wikingi mean from squash nothing else should happen17:02
@wikingonly that it becomes one simple commit17:02
-!- besser82 [~besser82@fedora/besser82] has joined #shogun17:02
-!- mode/#shogun [+o besser82] by ChanServ17:02
OXPHOSgit reset --soft, then I noticed a line change (has several tabs/spaces before, but empty line now). So I want to change it back.17:03
OXPHOSwiking: but it's hard to know how many spaces there're. So I grabbed the CMakeLists from shogun/develop17:03
OXPHOSand replaced mine17:03
OXPHOSwiking and Atom can automatically trim empty lines...17:19
-!- sanuj [~sanuj@117.204.255.134] has joined #shogun17:23
sanujhello everyone!17:26
sanujgot busy with some passport work today :/17:27
arianepaolahi sanuj17:27
sanujarianepaola, hi :)17:27
sanuji guess lisitsyn is not here17:27
OXPHOShey sanuj17:31
sanujOXPHOS, hi :)17:31
sanujwhat's up17:32
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 258 seconds]17:33
-!- leagoetz [~leagoetz@nat-247-179.internal.eduroam.ucl.ac.uk] has joined #shogun17:48
sanujI'm going to write cookbook pages on LibSVR and neuralnets17:59
sanujarianepaola, OXPHOS Saurabh7 i hope there is no overlap ^^17:59
arianepaolasanuj: ok17:59
leagoetzdoesn't look like arianepaola: and Saurabh7: have done their cookbooks yet ...18:00
leagoetzget going guys!18:00
leagoetzit's an important deliveralbe18:00
leagoetzable :)18:00
sanujhi lea :D18:00
sanujlisitsyn, there?18:00
leagoetzheyhey18:01
lisitsynpartially there18:02
lisitsynsome lecture to start soon18:02
lisitsynsup?18:02
sanujokay18:03
sanujwhen will it get over?18:03
lisitsyn3 hours, got time right now18:03
lisitsynneed any help?18:03
sanujyes18:03
sanujwhat is fire_updated?18:03
sanujdo we keep typedef T type in shogun?18:03
-!- besser82 [~besser82@fedora/besser82] has joined #shogun18:03
-!- mode/#shogun [+o besser82] by ChanServ18:04
OXPHOSsanuj go ahead18:04
sanujOXPHOS, cool18:04
sanujlisitsyn, ^^18:04
-!- leagoetz [~leagoetz@nat-247-179.internal.eduroam.ucl.ac.uk] has quit []18:06
sanujlisitsyn, i'll start a PR and start questions there18:07
sanujask*18:07
arianepaolawiking: this fixes it https://github.com/shogun-toolbox/shogun/pull/321218:12
@wikingmmm i'm just wondeirng18:12
@wikingwhat's the diff18:12
@wikingbetween your ubuntu18:13
@wikingand the buildbot's18:13
arianepaolaI am building with Python modular on18:14
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 260 seconds]18:14
arianepaolacmake -GNinja -DENABLE_TESTING=ON -DPythonModular=ON -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX="./install" ..18:15
arianepaolaninja GoogleMock && ninja all && ninja install && ninja test18:15
@wikingmmm18:21
arianepaolawiking: a question regarding index_t18:26
arianepaolait is defined in common.h as typedef int32_t index_t;18:26
arianepaolawiking: is it used only for indexes >= 0 ?18:27
shogun-buildbotbuild #6 of memleak - valgrind is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/memleak%20-%20valgrind/builds/618:42
-!- sanuj [~sanuj@117.204.255.134] has quit [Remote host closed the connection]19:47
-!- glorb9 [d8338942@gateway/web/freenode/ip.216.51.137.66] has joined #shogun20:27
glorb9Good afternoon.  I'm having an odd cmake error that I've not seen before.20:27
glorb9Anyone here that might be able to give me a hand?20:27
glorb9Hello?20:30
-!- glorb9 [d8338942@gateway/web/freenode/ip.216.51.137.66] has quit [Quit: Page closed]20:32
@wikingarianepaola: yes21:19
@wikingarianepaola: it should be unsigned21:19
@wikingbut then agian21:19
@wikingthere's this http://stackoverflow.com/questions/27275150/advice-on-unsigned-int-gangnam-style-edition21:20
@wiking(google style)21:20
-!- lambday [8028b10a@gateway/web/freenode/ip.128.40.177.10] has quit [Quit: Page closed]21:24
-!- besser82 [~besser82@fedora/besser82] has joined #shogun21:58
-!- mode/#shogun [+o besser82] by ChanServ21:58
arianepaolaok22:44
arianepaolaFrom https://google.github.io/styleguide/cppguide.html -> Equally bad bugs can occur when comparing signed and unsigned variables.22:49
arianepaolawiking: it is probably best to change the comparison type in the tests to compare signed with signed variables22:50
lisitsynsigned!22:55
lisitsyndrop unsigned!22:58
arianepaolayes, static_cast<int32_t>(strlen(const_char_array_variable))23:16
arianepaolabye everyone and have a great weekend23:40
--- Log closed Sat May 28 00:00:15 2016
« Thursday, 2016-05-26 Index Saturday, 2016-05-28 »

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