IRC logs of #shogun for Monday, 2018-03-19

« Sunday, 2018-03-18 Index Tuesday, 2018-03-20 »
--- Log opened Mon Mar 19 00:00:47 2018
-!- hqsklv1 [~vtxutqcu@47.254.40.225] has joined #shogun00:10
-!- hqsklv1 [~vtxutqcu@47.254.40.225] has quit [Client Quit]00:10
-!- albert_ [dd0628c2@gateway/web/freenode/ip.221.6.40.194] has joined #shogun02:25
-!- albert_ [dd0628c2@gateway/web/freenode/ip.221.6.40.194] has quit [Client Quit]02:25
-!- Farouk [81617d01@gateway/web/freenode/ip.129.97.125.1] has joined #shogun04:37
-!- Farouk [81617d01@gateway/web/freenode/ip.129.97.125.1] has quit [Client Quit]04:40
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4196 synchronized by syashakash06:25
-!- travis-ci [~travis-ci@ec2-54-196-127-238.compute-1.amazonaws.com] has joined #shogun06:45
travis-ciit's syashakash's turn to pay the next round of drinks for the massacre he caused in syashakash/shogun: https://travis-ci.org/syashakash/shogun/builds/35521948006:45
-!- travis-ci [~travis-ci@ec2-54-196-127-238.compute-1.amazonaws.com] has left #shogun []06:45
-!- travis-ci [~travis-ci@ec2-54-224-149-62.compute-1.amazonaws.com] has joined #shogun07:26
travis-ciit's syashakash's turn to pay the next round of drinks for the massacre he caused in syashakash/shogun: https://travis-ci.org/syashakash/shogun/builds/35521948007:26
-!- travis-ci [~travis-ci@ec2-54-224-149-62.compute-1.amazonaws.com] has left #shogun []07:26
@sukey[https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4201 closed by vigsterkr09:39
@sukey[https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4210 vigsterkr added label: "good first issue"10:02
@sukey[https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4210 vigsterkr added label: "BUG"10:02
@sukey[https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4210 opened by vigsterkr10:02
-!- travis-ci [~travis-ci@ec2-54-224-149-62.compute-1.amazonaws.com] has joined #shogun10:28
travis-ciit's Giovanni De Toni's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/35305604710:28
-!- travis-ci [~travis-ci@ec2-54-224-149-62.compute-1.amazonaws.com] has left #shogun []10:28
-!- wuwei [~wuwei@202.120.19.120] has joined #shogun10:48
wuweihi wiking, i'm not very clear about how untemplated features work10:50
wuweicurrently we can have a CFeatures *, and dynamic cast to sub-type in algorithm implementations10:50
wuweihow are they different10:51
@wikingah10:52
@wikinguntemplated features?10:53
@wikingi'm not so sure what are you referring to...10:53
@wiking?10:53
wuwei"We would like to move towards an un-templated features type, and leave it to algorithms to check types at runtime (is only done once by the algorithm)."10:53
wuweifound this in wiki of gsoc task10:53
wuweihttps://github.com/shogun-toolbox/shogun/wiki/GSoC_2018_project_detox10:53
@wikingaaaah10:53
@wikingyeah that's actually more about SGMatrix and Vecgor10:54
@wikingin the first place10:54
@wikingnot features10:54
@wikingbut as well features10:54
wuweioh i see10:54
wuweii saw the switch-based prototype10:55
@wikingyeah so the problem is that exactly10:55
@wikingas soon as we create10:55
@wikingan inteface10:55
@wikingsay for Matrix10:55
@wikingthat is type-less10:55
@wikingbut the same is true for DenseFeatures10:55
@wikingor SparseFeatures10:56
@wikingas both are type templated10:56
@wikingso in that case we would like to write the models10:56
@wikingtypeless10:56
@wikingif possible10:56
@wikingbut for that it's required that the linalg:: is for examples typeless10:56
@wikingbut if we go with the vanilla implementation of that that means every time you call linalg:: you'll have to do a big switch(TYPE)10:56
@wikingright?10:57
wuweiyeah10:57
@wikingnow this can be optimized, if we use for exxample lazy evaluation10:57
wuweithat would be very costly10:57
@wikingwhere you say10:57
@wikingopeator(operator(operator(x)))10:57
@wikingbut actually ti doesn't run the toperation10:57
@wiking*operation10:57
@wikingjust simply registers it on a stack :)10:57
@wikingand once you call .execute()10:58
@wikingor whatever10:58
@wikingthen you actually run the whole algorithm10:58
@wikingnow here you can do then a trick that find out once the type10:58
@wikingand then just use that information and keep propagating it :)10:58
wuweiyeah, we can use sth like exp template, but unfortunately we dont have that right now10:58
wuweiit could require huge work, i'm not sure i can finish it in this gsoc11:00
@wiking:)11:05
@wikingyeah i mean11:05
@wikingthat's another story11:05
@wiking:)11:05
@wikingwe can dedicate couple of weeks to that11:09
@wikingbut having the iterators cleaned up11:09
@wikingthe whole io system11:09
@wikingis more essential11:09
@wikingbut not really :)11:09
@wikingwe really need views11:09
@wikingwhat we can do is we can borrow some things from other places11:10
@wikingfor example the exp template11:10
wuweii know there are some libraries that use exp template for lazy-evaluation, like eigen, mshadow11:13
wuweiwe can borrow the design11:14
wuweibut i think we will need to change a lot of things11:14
wuweicould u suggest how to get started11:14
@wikingwhich part ? :)11:17
@wikingmmmm11:17
@wikingi mean i would first look into11:18
@wikingsgmatrix/vector11:18
@wikingi mean keep those as they are really used in toooo many places11:18
@wikingbut having introduced a new Matrix Vector class11:18
@wikingthat are typeless wrappers around barebone arrays11:19
@wikinglike double*11:19
@wikingand after that we could add another layer over linalg::11:19
@wikingwhere it accepts only Vector and Matrix11:19
wuweiyeah11:19
@wikingand that thing is using lazy eval11:19
@wikinghaving this prototype would be good to have11:20
@wikingi mean maybe this is the best where we could start11:20
@wikingas this is totally indepemndent11:20
@wikingof the whole codebase of shogun11:20
wuweii see11:20
@wikingas if you start refactoring SGVector or SGMatrix11:20
@wikingit'll just pull in a lot of pain11:21
@wikingbecause we use those in typemapping to swig interfaces11:21
@wikingetc11:21
@wikingso for the time being i think the best approach is to grow something11:21
@wikingthat is totally independent11:21
@wikingof what we have atm11:21
@wikingthat they you are not blocked by being able to backward compatible11:21
@wiking:)11:21
@wikingyeah as lazy eval goes11:22
@wikingthere are many examples: eigen, stan, xtensor etc11:22
wuweiso i can firstly come up with a design of lazy eval ?11:22
@wikingyeah say we have a Matrix and a Vector class11:22
@wikingthat hides the type11:22
@wikingand we wanna do lazy-linalg over it11:23
@wikingwithout that crazy story of11:23
@wikingswitch(getType)11:23
@wikingon every operation11:23
@wikingso say we have a GLM11:23
@wikingwhere you have11:23
@wikingVector weight;11:24
@wikingScalar bias;11:24
@wikingand you get a Matrix11:24
@wikingthat is actually the examples you wanna 'score'11:24
@wikingwhat you need is [dot(e,weight)+bias for e in matrix]11:25
@wikingright?11:25
wuweiyeah11:25
@wikingso having this simple case11:25
@wikingimplemented as lazy-eval11:25
@wikingwould be a good start11:25
wuweiwe create exp type, and need a eval engine11:25
wuweithat's cool11:26
@wikingand as said11:26
@wikingdont try to refactor the code yet11:26
@wikingwithin shogun as that'll take a lot of your attention11:26
@wikingfrom parts that are not important atm11:26
@wikingso use your own structures11:26
@wikingand once we have that working11:26
@wikingwe can look into how we would do the refactoring11:27
wuweii see11:27
-!- jinquan [dd0628c2@gateway/web/freenode/ip.221.6.40.194] has joined #shogun11:27
wuwei:)11:27
@wikingwuwei, i mean for the time being11:27
@wikingyou can even just do a simple11:27
@wikinga*b+c kind of lazy eval11:27
@wikingwithout the dot product11:27
@wikingit doesn't really matter11:27
@wikingjust have 2+ operations11:27
@wikingover the same Matrix or Vector11:28
@wikingwhere you need to play with the types as well11:28
wuweishould i take care of underlying premitive type now11:29
@wikingyeah i mean11:30
@wikingthe story is really that we wanna be able to have this prototype demonstrate11:30
@wikingthat we can do this11:30
@wikingwithout actually doing that switch in case of every oepration call11:30
@wiking:)11:30
@wikingi mean for the type being its enough11:30
@wikingif you introduce like11:30
@wikingfloat64 and int6411:31
@wikinglisitsyn, ping?11:37
lisitsynhey11:37
@wikingok so question11:37
@wikinghttps://github.com/apache/arrow/blob/master/cpp/src/arrow/compute/kernel.h#L4411:38
@wikingthere's no way i can inherit from this11:38
@wikingright? :)11:38
-!- jinquan [dd0628c2@gateway/web/freenode/ip.221.6.40.194] has quit [Ping timeout: 260 seconds]11:55
lisitsynwiking: inherit from scalar? why?12:36
@wikingerror: no matching constructor for initialization of '::arrow::compute::Scalar'12:36
@wiking        ::arrow::compute::Scalar(),12:36
@wiking        ^12:36
lisitsynuhm interesting12:37
lisitsynwiking: it looks like scalar is always used in a ptr12:38
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun12:40
-!- mode/#shogun [+o HeikoS] by ChanServ12:40
@wikinglisitsyn, i mean my problem is rather that i'm trying to inherit from it so that i can have a scalar :)12:41
-!- Jinquan [dd0628c2@gateway/web/freenode/ip.221.6.40.194] has joined #shogun12:48
@HeikoSlisitsyn: jo!13:29
@HeikoSwiking: jojo13:30
@HeikoSwiking: I guess we need to split the build into parts13:55
@wiking?13:55
@HeikoSand then distribute them to different builds13:55
@wikinghow do you envision that w/o actually having the plugin arch?13:56
@HeikoShacks13:56
@HeikoSit is not a long term solution13:56
@wikingthen just dont do it13:56
@wikingif it's not a long term solution13:56
@wikingand a hack13:56
@HeikoShow shall we get the CI green then?13:56
@wikingbit tired of these shitty hacks13:56
@wikingwell13:57
@wikingyou could still13:57
@wikingtry to change -jNUMJOBS13:57
@wikingas i've tried playing with it13:57
@wikingand that should help13:57
@wikingmaybe atm its too high13:57
@wikingyou can actually use a dummy branch for that13:57
@wikingwhere you just change a line in any or sgobject.h13:57
@wikingso that you invalidate the whole ccache13:57
@wikingi've consolidated the scripts13:58
@HeikoSor we delete lots of code :D13:58
@HeikoSthat might be more fun haha13:58
@wikingwell we've tried that13:58
@wikingdoesn't really work13:58
@wikingall the ci scripts are now in scripts/13:59
@wikingand just simple bash scripts13:59
@wikingshould be easy to patch by anybody13:59
@wikingbtw13:59
@wikinghttps://travis-ci.org/shogun-toolbox/shogun13:59
@wikingas you can see it's not impossible13:59
@wikingto get it work13:59
@HeikoSyeah14:00
@HeikoScache magic14:00
@wikingwell14:00
@wikingcache is not helping you14:00
@HeikoSbut changing the .h files ...14:00
@wikingin the first run14:00
@wikingbut again14:00
@wikingit's possible to get it pass14:00
@wikingmany times it happened to me14:00
@wikingbut still it's just way too long14:00
@wiking(see appveyor14:00
@wikinghttps://travis-ci.org/shogun-toolbox/shogun/jobs/35305605114:01
@HeikoSI am not sure I follow14:01
@wikingbtw here it'd pass without the libshogun examples14:01
@HeikoSI mean I get playing with the num jobs14:01
@wikingsame here14:01
@wikinghttps://travis-ci.org/shogun-toolbox/shogun/jobs/35305604914:01
@wikingso i'm not so sure14:01
@wikingwhy are you saying that actually removing libshogun wouldn't help14:02
@wikingbut again14:02
@wikingas said14:02
@wikingits worth to do the SGObject.h add an extra line14:02
@wikingand simple branch14:02
@wikingbut as a first run14:02
@wikingremoving libshogun from all target14:02
@wikingand from ctest14:02
@wikingcould essentially fix the problem in develop14:02
@HeikoShttps://ci.appveyor.com/project/vigsterkr/shogun/build/2135#L1125414:02
@HeikoSbecause of this14:02
@wikingyeah but that's a different story14:03
@wikingthere's many things you can play there14:03
@wikingand anyways14:03
@wikingappveyor passes mostly14:03
@wikingso i wouldn't draw conclusions form that14:03
@wikingsince appveyor actually14:03
@wikinghas +15 mins14:03
@wikingon travis jobs14:03
@wikingor at least 1014:03
@wikingtravis kills a job after 50mins14:03
@wikingappveyor does it in 6014:04
@wikingjust try to drop libshogun from tests and all target14:04
@HeikoSyeah Ill try that as a first thing14:04
@wikingadd_executable14:06
@wikingEXCLUDE_FROM_ALL14:06
@wikingdone14:06
@wikingttyl14:06
@HeikoSTRAVIS_DISABLE_LIBSHOGUN_TESTS14:07
@HeikoSis that guy still used? I could just turn it on by default for a test14:07
@wikingHeikoS, remove that14:08
@wikingthat's bs anyways14:08
@wikingif you do this14:08
@wikingHeikoS, plz avoid hacks in this case14:08
@HeikoSI havent added it14:08
@wikingwhat i mean is14:08
@wikingdrop it14:08
@wikingif you do this14:09
@HeikoSwiking: you ok with me killing all the libshogun tests/examples?14:12
-!- wuwei [~wuwei@202.120.19.120] has left #shogun []14:17
@wikingHeikoS, not kill just remove it from the ALL target14:32
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Remote host closed the connection]15:35
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun15:36
-!- mode/#shogun [+o HeikoS] by ChanServ15:36
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/b01dd0da1fbb709e907ac50be7499529385e7864 by karlnapf15:40
@sukey[https://github.com/shogun-toolbox/shogun] New branch feature/travis_time created15:40
@HeikoSwiking: this you had in mind?15:40
@wikingalmost15:41
@wikingyou need to drop them from ctest15:41
@wikinghttps://github.com/shogun-toolbox/shogun/blob/b01dd0da1fbb709e907ac50be7499529385e7864/cmake/ShogunUtils.cmake#L11615:42
@wikingelse by default ctest will fail15:42
@wikingas the binaries arent compiled15:42
@HeikoSic thanks15:43
@HeikoSWARNING: Target "minimal" has EXCLUDE_FROM_ALL set and will not be built by default but an install rule has been provided for it.  CMake does not define behavior for this case.15:44
@HeikoSthere is also this guy15:44
@HeikoSThis guy was for conda iirc15:44
@HeikoS?15:44
@wikingmmm15:44
@wikingthat thing is a bs bingo15:45
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/f2802ba44175de1ab2f32712ec45b5f150c3cb98 by karlnapf15:45
@wikingi'd drop it in the first place15:45
@HeikoSshall I do it straight away? :D15:45
@wikingyes drop minimal15:45
@wikingas is15:45
@wikingif then that should be part of the conda target15:45
@HeikoSdrop means delete example right?15:46
@HeikoSor from cmake?15:46
@HeikoSinstallation15:46
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/b479a3d632151d35fc56b24b836c0b65f2faa89f by karlnapf15:47
@wikingyes15:49
@wikingeverything15:49
@wikingnot only from install15:49
@wikingbut the whole minimal example itself15:49
@wikingand that concept15:49
@HeikoSkk15:49
@wikingit's just noise15:49
-!- travis-ci [~travis-ci@ec2-54-166-88-144.compute-1.amazonaws.com] has joined #shogun15:50
travis-ciit's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/35539661715:50
-!- travis-ci [~travis-ci@ec2-54-166-88-144.compute-1.amazonaws.com] has left #shogun []15:50
@wikingHeikoS, btw have you checked the proposals15:51
@wikingfew but much good15:51
@HeikoSon my list for today15:51
@HeikoScool15:51
@HeikoSwill give some feedback later15:51
@wikingi've given feedback already to wuwei15:51
@HeikoScool15:51
@wikingfayrouz is taking a big fish imo15:51
@wikingbut i'd be super supportive15:51
@wikingjust maybe better to make it a bit more realistic15:52
@wikingtimewise :)15:52
@wikingas there's a looooot of stuff in there15:52
@wiking+ i'll ask him to port the thing into a google doc15:52
@wikingas that way it's easier to comment on things15:52
@wikingas now it's a pdf15:52
@wiking(latex based)15:52
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/ddcc90d5eb358c0806909dde77eb1b1831de190a by karlnapf15:53
-!- ajupkw [~ixaxl@193.106.25.19] has joined #shogun15:54
-!- ajupkw [~ixaxl@193.106.25.19] has quit [Client Quit]15:54
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/4a42cc9b4225f948b6b9b945a693b6f41b671266 by karlnapf15:55
@HeikoSok lets see what travis says15:55
@wikingmaybe making the docker image thiner15:56
@wikingwould as well help some15:56
@wikingas that's couple of minutes fetching time as well unfortunately15:57
@wikingidk why those images cannot be caches on travis side... but yeah we dont have a control over that :)15:57
@wikingHeikoS, you've missed dropping OPTION(TRAVIS_DISABLE_LIBSHOGUN_TESTS "Disable libshogun tests to speed up jobs on travis-ci" OFF)16:06
@HeikoSah yes16:06
@wikingbut do a git grep for it16:06
@wikingas there are 2-3 places16:07
@wikingwhere it is being used16:07
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/7de5eef56468f9977ad610850d5d3d814424defc by karlnapf16:09
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/7ea005de21d996aa18ba12d46a5260fbca091d2e by karlnapf16:18
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4198 synchronized by karlnapf16:23
@HeikoSlisitsyn: you there?16:24
@HeikoSyour clone thing has a problem? :D16:24
lisitsynHeikoS: what problem?16:24
@HeikoSah there we are16:24
@HeikoSI have an obj with another obj as parameter16:24
@HeikoSI call clone16:24
lisitsynyes16:25
@HeikoSlet's call the obj and param_obj16:25
lisitsyngood16:25
@HeikoSobj in the constructor creates a new instance of param_obj16:25
@HeikoSthen I call clone16:25
@HeikoSthe clone has that very same instance as parameter16:25
@HeikoSthe clone method of param_obj is called16:25
@HeikoSbut the created object seems not assigned16:26
@HeikoSI have a log if it helps16:26
lisitsynyeah16:26
@HeikoSlisitsyn: but you see the problem?16:26
lisitsynI don't know16:27
lisitsynlemme check16:27
lisitsynHeikoS: I am lost16:29
lisitsynclone is old in develop16:29
lisitsynit doesn't seem to use new parameter map16:29
@HeikoSyes I am in a feature branch16:29
@HeikoSlocal16:30
lisitsynHeikoS: so you mean clone of any?16:30
@HeikoSyes16:30
lisitsynso16:30
@HeikoSthe thing you added16:30
lisitsynif I create a unit test16:30
lisitsynthat wraps pointer to sgobject16:31
lisitsynand I call any16:31
lisitsyncall any's clone16:31
lisitsynthe object won't change, right?16:31
@HeikoSnot following16:31
lisitsynSGObject* obj = ...16:31
lisitsynmake_any_ref(&obj)16:31
lisitsynAny any = make_any_ref(&obj)16:32
lisitsynAny other;16:32
@HeikoShttps://gist.github.com/karlnapf/e7e286568bfeb6c985973de8d35b091416:32
lisitsynother.clone_from(any)16:32
lisitsynok I see16:32
lisitsynI didn't implement a test for simple clone16:32
lisitsynO_o16:33
@HeikoSso it calls ->clone() of parameters recursively16:33
@HeikoSbut it doesnt assign the result back16:33
lisitsynTEST(Any, clone_int)16:35
lisitsyn{16:35
lisitsyn        int value = 3;16:35
lisitsyn        auto any_src = make_any_ref(&value);16:35
lisitsyn        int other_value = 5;16:35
lisitsyn        auto any_dst = make_any_ref(&other_value);16:35
lisitsyn        EXPECT_EQ(any_src.as<int>(), value);16:35
lisitsyn        EXPECT_EQ(any_dst.as<int>(), other_value);16:35
lisitsyn        any_dst.clone_from(any_src);16:35
lisitsyn        EXPECT_EQ(any_dst.as<int>(), value);16:35
lisitsyn}16:35
lisitsynthis works16:35
@HeikoSsure16:36
@HeikoSin my tests as well16:37
@HeikoSthe non object parameters are fine16:37
lisitsynah16:38
lisitsynso it is only objects?16:38
@HeikoSah I think it is the order16:38
@HeikoSsomehow clome_from is swapped16:38
@HeikoSlhs and rhs16:38
@HeikoSjust reaslied16:38
lisitsynhah16:38
@HeikoSwait I need to check something16:39
@HeikoSgna16:39
@HeikoSI think my cline method is doing that16:39
@HeikoShaha16:39
lisitsynHeikoS: yeah you clone into the object from empty16:40
@HeikoSlisitsyn: sorry for blaming you :D16:40
@HeikoSyep16:40
lisitsynHeikoS: it should be called set_default16:40
lisitsyn:P16:40
@HeikoShehe16:40
@HeikoSmmmh16:43
-!- travis-ci [~travis-ci@ec2-54-81-208-176.compute-1.amazonaws.com] has joined #shogun16:45
travis-ciit's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/35540030416:45
-!- travis-ci [~travis-ci@ec2-54-81-208-176.compute-1.amazonaws.com] has left #shogun []16:45
@HeikoSlisitsyn: nah ok, that was really me. my test works, thx!16:46
@HeikoSsometimes all it takes is to show someone else a problem to realise a solution :)16:46
@wikingHeikoS, tada :)16:49
@wikinggcc passed16:50
@HeikoSyep thats good16:50
@HeikoSshall I merge that thign in?16:50
@wikingwhich means clang should pass as well as that's usually faster16:50
@HeikoSit didnt16:50
@HeikoSI restarted16:50
@wikingmmm that's a glitch16:50
@HeikoSit was still compiling unit tests16:50
@wikingin general clang is 10-15% faster than gcc16:50
@wikingso probably what happened that it ended up on an overused vm16:50
@wikingyeah squash & merge16:51
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/c8cc22ca075c2863687cc8f4dc3752c92679ba90 by karlnapf16:56
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4198 synchronized by karlnapf17:01
-shogun-buildbot:#shogun- Build deb4 - python3 #365 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/28/builds/36517:07
@wikingHeikoS, you know you can squasmerge with github :)17:16
-!- travis-ci [~travis-ci@ec2-54-144-195-124.compute-1.amazonaws.com] has joined #shogun17:43
travis-ciit's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/35540030417:43
-!- travis-ci [~travis-ci@ec2-54-144-195-124.compute-1.amazonaws.com] has left #shogun []17:43
@HeikoSlisitsyn: still around?17:47
@HeikoSwiking: https://travis-ci.org/shogun-toolbox/shogun/jobs/35540030617:48
@HeikoSsame error as before17:48
@wikingsure dont worry17:48
@wikingit'll get in place17:48
@HeikoSlisitsyn: ok basic things work17:52
@HeikoSlisitsyn: only problem is ArrayReference and ::clone when there are objects in there17:52
@HeikoSI think rather than copying the whole array at once, one could call clone_impl on the element individually?17:53
-!- travis-ci [~travis-ci@ec2-54-166-88-144.compute-1.amazonaws.com] has joined #shogun17:55
travis-ciit's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/35543161617:55
-!- travis-ci [~travis-ci@ec2-54-166-88-144.compute-1.amazonaws.com] has left #shogun []17:55
-!- petioptrv [84cde53f@gateway/web/freenode/ip.132.205.229.63] has joined #shogun18:30
@wikingjust18:33
@wikinghttps://www.magicleap.com/creator18:33
@wiking:)18:33
petioptrvGreetings from Canada! My name is Petio and I am interested in getting involved with shogun via GSoC 2018. I wanted to drop by and say hello and also check if there is a more suitable destination for GSoC-related discussions. Cheers!18:33
@HeikoSwiking: whats that?18:33
@HeikoSpetioptrv: welcome! :)18:34
@wikingmagicleap18:34
@wikingthe future18:34
@wiking:)18:34
@HeikoSpetioptrv: this is the right place to ask any questions and hello18:34
@HeikoSpetioptrv: make sure to apply soon! and so send some PRs soon18:34
petioptrvYes, I definitely will; deadline is approaching fast. Unfortunately, I only heard about GSoC recently, and have been spending time getting familiar with shogun18:37
@HeikoSwiking: lol I see it clearly is18:37
petioptrvVery excited about the possibility!18:37
@HeikoSpetioptrv: make sure to read the "how to get involved" guide18:37
@HeikoSit tells you what the next steps are18:38
@wikingHeikoS, have u seen their demos?18:38
petioptrvYup, did that and am working on a proposal atm18:38
@wikingpetioptrv, \o/18:39
@wikingpetioptrv, the sooner you put together a proposal for a project the sooner we can give you some feedback18:39
petioptrvI like the "Finding the bad guys" project. It seems like there is a lot of interesting things to work on there18:42
petioptrvI'm guessing there will be more than one person assigned to that project, is that right?18:42
petioptrvOr it's one individual per project?18:42
@wikingpetioptrv, depends :)18:42
@wikingon the number of applicants18:43
petioptrvok, I see18:45
petioptrvjust trying to picture what the workflow would be like18:45
@HeikoSwiking: how can I give the applicant feedback?18:50
@HeikoSemail or what?18:50
@wikingHeikoS, which?18:51
@HeikoSin the gsoc dashboard18:51
@wikingHeikoS, as i've already sent a feedback to Fay18:51
@HeikoSthe non googledoc18:51
@wikingthat he should googledocit18:51
@HeikoSvia irc?18:51
@wikingso i guess soon he'll do it18:52
@HeikoSkk18:52
@HeikoSlet me send an email as well so he is pushed18:52
@wiking:)18:52
@wikingas you see it best18:52
@wikingas said i've already requested it18:53
@wiking:)18:53
@wiking"Also make sure to talk to us in IRC regarding the stan-math part"18:54
@wikinghe did talk about this with me18:54
@wikinghere on irc18:54
@wikinghe first went with another lib for autodiff18:54
@wikingthat's why actually he switch to stan18:54
@wikingi feel these kind of cross-comm is a bit awkward18:54
@HeikoSyou understand what he exactly wants to do with that autodiff stuff?18:54
@wikingbut it's my 2 cents18:54
@wikingyes18:55
@wikingwe've talked about it here18:55
@wikinganyhow18:55
@wikingi have to go now (moving) ttyl18:55
@HeikoSok bye18:55
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 240 seconds]19:11
-!- travis-ci [~travis-ci@ec2-54-81-208-176.compute-1.amazonaws.com] has joined #shogun19:15
travis-ciit's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/35543161619:15
-!- travis-ci [~travis-ci@ec2-54-81-208-176.compute-1.amazonaws.com] has left #shogun []19:15
petioptrvwiking: how long should the proposal be?19:39
petioptrvin the case of Catching the bad guys*19:39
petioptrvalso, is there a link where we can see successful proposals from past gsoc years?19:39
-!- nikhilweee [~nikhilwee@128.199.66.195] has joined #shogun20:11
-!- petioptrv [84cde53f@gateway/web/freenode/ip.132.205.229.63] has quit [Ping timeout: 260 seconds]20:46
--- Log closed Tue Mar 20 00:00:49 2018
« Sunday, 2018-03-18 Index Tuesday, 2018-03-20 »

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