Open in new window / Try shogun cloud
--- Log opened Tue Aug 02 00:00:48 2016
@wikinglisitsyn, what do you mean? bool operator==(const T& x, const T& y) is the declaration of the equals03:06
@wikingor?03:06
@wikingi mean it sounds stupid though :)03:07
@wikingah you mean virtual bool equals(CSGObject* other, float64_t accuracy=0.0, bool tolerant=false);03:07
@wiking?03:07
@wikingbut then it should be fine :)03:07
@wikingah but03:08
@wikingbool T::operator ==(const T2 &b) const;03:08
@wiking:)03:08
@wikingmmmm03:08
@wikingthat's a fun one03:08
@wiking:>03:08
@wikingbtw lisitsyn blame Heiko for that design decision: https://github.com/dhruv13J/shogun/commit/ff244613212122d54b3de609c7c4d95a45dd026e#diff-1689b0bd3ceb28355cc3c05ed7ffe4f003:14
@wikingrcurtin, sorry yesterday i've passed out03:52
rcurtinno worries, I am about to pass out myself04:39
rcurtinso maybe tomorrow :)04:39
@wiking:)))04:40
@wikinghahah ok04:40
@wikingjust msg04:40
@wikingthe problem is that i think we are in a 13 hours04:40
@wikingtime diff04:40
@wikingyou are east coast?04:40
rcurtinyeah, east coast04:51
rcurtin13 hours... where are you?04:51
@wikingsingapore04:58
rcurtinah, okay04:59
rcurtinwell I will be awake tomorrow morning around 1300 UTC, so maybe you will still be awake then :)04:59
@wiking:)05:02
@wikingjust ping me05:02
rcurtinsure05:03
rcurtin'night :)05:03
@wikinggnite05:03
-!- sanuj [~sanuj@61.0.30.105] has joined #shogun05:15
sanujwiking, there?05:15
@wikingsanuj, yes05:19
sanujwiking, so over here https://github.com/shogun-toolbox/shogun/pull/3357/files/b79c74445e90a3930003d5a4e35385ca31ecfa42..f768dbc9fa02e71ff155de5051ece64faa868208#diff-af3b638bc2a3e6c650974192a53c7291R96205:19
sanuji need to check FIND_PACKAGE(JINJA2)05:20
sanujand then set ENABLE_TESTING or COMPILE_MODULAR_INTERFACE05:20
sanuj?05:20
@wikingman05:20
@wikingi mean i dont understand05:20
@wikingcheck cmakelists.txt05:20
sanujnot sure what your second comment means05:20
@wikinghttps://github.com/sanuj/shogun/blob/f768dbc9fa02e71ff155de5051ece64faa868208/CMakeLists.txt#L8905:21
@wikingwhy not check here the jinja205:21
@wiking?05:21
@wikingi dont understand why do you need to create yet another branch05:21
@wikingthat checks the same fucking thing05:22
sanujwiking, jinja2 is also required in ENABLE_TESTING05:22
@wikingand05:22
@wikingyou leave it there05:22
@wikingand check it05:22
@wikingi dont get what's the problem05:22
@wikingcome on05:22
@wikinglet's be a bit more like a developer05:22
@wikinginstead of a random code generator05:22
@wikinganyhow05:23
@wikingjust move the find pacakge there05:23
@wikingand you should be set05:23
sanujwiking, in that case i'll have to put 2 checks for JINJA205:23
@wikingno just one05:23
@wikingyou put one to the modular05:23
@wikingright?05:23
@wikingthe other is already there05:23
sanujif no modular interfaces are enabled then it won't be checked05:23
@wikingor?05:23
sanujbut we need to check for unit-tests also05:23
@wikingi'm sorry?05:24
@wikinghttps://github.com/sanuj/shogun/blob/f768dbc9fa02e71ff155de5051ece64faa868208/CMakeLists.txt#L96005:24
@wikingthis check is for05:24
@wikingmodular05:24
@wikingthe one for the unit test05:24
@wikingis *already* there05:24
sanujyeah05:24
@wikingso?05:24
sanuji meant there will be 2 checks for jinja2 in the resultant code05:25
sanujif that's fine, then i'll do it05:25
@wikingand ?05:25
@wikingi mean in cmake taht check is actually cached05:25
@wikingif you do 2 times the same if in a codebase05:25
@wikingthat wont work?05:25
@wikingor what? i mean i dont undestand here the concern05:25
sanuji was worrying about the redundancy of 2 checks for the same thing05:26
sanuji guess that's not a problem05:26
sanujso i'll do what you are suggesting05:26
sanujwiking, can you please rebase feature/tags with develop?05:42
@wikingsure05:43
@wikinggimem a sec05:43
sanujwiking, actually no need05:43
sanujlisitsyn pushed his commit on feature/tags05:44
sanuji just checked05:44
@wikingk05:45
-!- sanuj [~sanuj@61.0.30.105] has quit [Ping timeout: 252 seconds]05:55
-!- sanuj [~sanuj@61.0.30.105] has joined #shogun06:31
-!- c4goldsw [5da420e6@gateway/web/cgi-irc/kiwiirc.com/ip.93.164.32.230] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]08:44
-!- sanuj [~sanuj@61.0.30.105] has quit [Ping timeout: 265 seconds]09:45
-!- sanuj [~sanuj@61.0.30.105] has joined #shogun10:01
sanujlisitsyn, yo10:01
sanujthere is a problem10:01
@wikingwhat is the problem sanuj ?10:19
sanujrebasing for that PR10:22
lisitsynoh I am a bit stucked with SG_ADD10:22
lisitsynI've implemented something working10:22
lisitsynbut it has some nasty bug10:22
sanujlisitsyn, the problem is that sets() replaces Any object10:22
sanujor does operator=10:22
sanujohkay10:22
sanujlisitsyn, do we need a static method in Any that updates() the value10:23
sanujwill be good for updating value of member variables10:23
lisitsynwhat for?10:23
lisitsynah10:23
lisitsynyeah maybe10:23
sanujlisitsyn, so sets() implementation should be changed i think10:24
sanujsince the current version won't work for member variables10:24
sanujlisitsyn, we should just update the value of already existing Any object in the map10:24
sanujwhat do you think?10:24
lisitsynsanuj: what do you mean by won't work?10:25
lisitsynwhat bug do you have?10:25
sanujlisitsyn, i don't have any bug10:27
sanujhaven't tested it10:27
sanujbut i mean10:27
sanujhttps://github.com/shogun-toolbox/shogun/blob/develop/src/shogun/base/SGObject.h#L33310:27
lisitsynah you mean we can't use this external pointer thing there/10:28
lisitsynbecause it is going to be replaced?10:28
sanujerase_type() makes a new Any object10:28
sanujyeah10:28
lisitsynoh well that's exactly what I am fighting with now10:28
lisitsynthe last thing before SG_ADD works10:28
sanujlisitsyn, i think with this implementation the map object of Any will start pointing to new memory10:29
sanujinstead of the m_member's memory10:29
sanujafter it has been updated with sets()10:29
lisitsynyou're right10:29
lisitsynlets try to fix it10:29
lisitsyn:)10:29
sanujyeah10:30
sanujwanted to discuss with you the possible solutions10:30
sanujone way is to have an static update() in Any and use that10:30
lisitsynnot sure have to think about it10:30
sanujokay10:33
sanujlisitsyn, sets() is only used after register_param(). So there is always an Any object in map when sets() is used to modify it. Whatever we do, it's best to change value of same memory instead of making a new Any object and changing the memory. Unnecessary creation and deletion of Any objects.10:59
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun11:05
shogun-notifier-shogun: Viktor Gal :feature/javascript * fad3659 / README.md,src/interfaces/javascript_modular/CMakeLists.txt,index.js,package.json: https://github.com/shogun-toolbox/shogun/commit/fad3659a65c70ca281a38462a327176965c4f04811:05
shogun-notifier-shogun: Add node and npm build11:05
shogun-notifier-shogun: Viktor Gal :feature/javascript * 84cb0e4 / src/shogun/neuralnets/DeepBeliefNetwork.h: https://github.com/shogun-toolbox/shogun/commit/84cb0e4d78e8fed3677513ff7c2fdf7d189ac82811:05
shogun-notifier-shogun: fix include path for SGMatrixList in DeepBeliefNetwork11:05
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has joined #shogun11:07
-!- mode/#shogun [+o HeikoS] by ChanServ11:07
@wikingHeikoS, buya11:09
@HeikoSwiking: jo!11:09
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has quit [Quit: Leaving.]11:17
-!- travis-ci [~travis-ci@ec2-54-163-63-103.compute-1.amazonaws.com] has joined #shogun11:48
travis-ciit's Viktor Gal'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/14912448311:48
-!- travis-ci [~travis-ci@ec2-54-163-63-103.compute-1.amazonaws.com] has left #shogun []11:48
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has joined #shogun11:56
-!- mode/#shogun [+o HeikoS] by ChanServ11:56
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has quit [Quit: Leaving.]12:12
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has joined #shogun13:00
-!- mode/#shogun [+o HeikoS] by ChanServ13:00
-!- sanuj [~sanuj@61.0.30.105] has quit [Ping timeout: 252 seconds]13:01
-!- sanuj [~sanuj@61.0.30.105] has joined #shogun13:14
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has quit [Quit: Leaving.]13:43
-!- sanuj [~sanuj@61.0.30.105] has quit [Ping timeout: 265 seconds]13:51
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]14:05
-!- sanuj [~sanuj@59.91.132.116] has joined #shogun14:10
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has joined #shogun14:21
-!- mode/#shogun [+o HeikoS] by ChanServ14:21
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has quit [Quit: Leaving.]14:28
-!- sanuj [~sanuj@59.91.132.116] has quit [Ping timeout: 240 seconds]14:29
rcurtinwiking: I'm here now, but only for a few minutes... I overslept and have to leave soon... oops15:15
@wikingrcurtin, the trick is15:18
@wikingi have no idea what i wanted to ask15:19
@wiking:D15:19
@wikingi'll try to think about it15:19
rcurtinok :)15:19
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has joined #shogun15:31
-!- mode/#shogun [+o HeikoS] by ChanServ15:31
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has quit [Quit: Leaving.]16:07
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has joined #shogun16:50
-!- mode/#shogun [+o HeikoS] by ChanServ16:50
-!- OXPHOS [4441a916@gateway/web/freenode/ip.68.65.169.22] has joined #shogun16:54
@wikingOXPHOS, back to usa?16:55
OXPHOSwiking: hey yep16:57
@HeikoSOXPHOS: jojo17:10
@HeikoSSaurabh7: hi!17:10
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun17:21
shogun-notifier-shogun: OXPHOS :feature/linalg_refactor * 3e9ceac / / (7 files): https://github.com/shogun-toolbox/shogun/commit/3e9ceace0ecd572956060ff612b2de0aaf7124bf17:21
shogun-notifier-shogun: linalg-sum_no_diag17:21
shogun-notifier-shogun: Heiko Strathmann :feature/linalg_refactor * 3e3e8e0 / / (7 files): https://github.com/shogun-toolbox/shogun/commit/3e3e8e052b2c7a8c2c00a8caf7fb30601906e29317:21
shogun-notifier-shogun: Merge pull request #3400 from OXPHOS/linalg_sum17:21
shogun-notifier-shogun:17:21
shogun-notifier-shogun: LinalgRefactor - no_diag sum17:21
-!- leagoetz [~leagoetz@nat-206-249.internal.eduroam.ucl.ac.uk] has joined #shogun17:22
-!- leagoetz [~leagoetz@nat-206-249.internal.eduroam.ucl.ac.uk] has quit [Client Quit]17:22
-!- sanuj [~sanuj@59.91.130.234] has joined #shogun17:32
sanujHeikoS, yo17:33
sanujthe plan is to not replace TParamter17:33
@HeikoSsanuj:17:33
sanujjust add tags parameter also17:33
@HeikoSI see17:33
@HeikoSyeah thats ok17:33
@HeikoSsanuj: btw17:33
sanujto add a layer of new API17:33
@HeikoSremember our meeting?17:33
sanujyeah17:34
@HeikoSthe last shogun gsoc meeting17:34
@HeikoSthere were some dates17:34
@HeikoSso you guys should wrap up soon17:34
@HeikoSto spend the rest of  the time writing docs and peer review17:34
sanujHeikoS, yes, see my timeline17:34
@HeikoStry to stop coding very soon17:34
@HeikoScool17:34
@HeikoSjust wanted to remind :)17:34
@HeikoSI will leave for a trip soon17:34
@HeikoSand not be back before end of GSoC17:34
@HeikoSwiking: will merge things and handle the admin17:35
@HeikoSI will check back in September what was the result :)17:35
@HeikoSbut yeah, as said in the meeting, the docs are really really important for us17:35
sanujHeikoS, where are you going?17:35
sanujhttps://docs.google.com/document/d/1ZVV9KsHxpoj2G3HYYp-8MNjDpHVL0RGvj5hD0r03lgg/edit?usp=sharing17:35
sanujHeikoS, i'll write good docs17:35
sanujdon't worry :D17:35
@HeikoSgood! :)17:36
@HeikoSI think you got the hang of it by now!17:36
@HeikoSLooking fowrad to read things when back17:36
sanujHeikoS, cool17:36
@HeikoSSaurabh7: same for your stuff!17:36
sanujwhen are you leaving?17:36
@HeikoSnow17:36
sanujoh17:36
sanujHeikoS, enjoy your trip!!!17:36
sanuj:D17:36
@HeikoSthanks a lot! :)17:37
sanujHeikoS, can you give me ideas about this: https://github.com/shogun-toolbox/shogun/pull/339017:38
@HeikoSsanuj: https://travis-ci.org/shogun-toolbox/shogun/jobs/147142817#L274417:40
@HeikoSyou can run the listing by hand17:40
@HeikoSctest -R "testname"17:40
@HeikoSah it doesnt even compile17:40
@HeikoSwell look at the java code listing and see whats wrong17:40
sanujHeikoS, i need some ideas about enum17:41
sanujhow to structure the tests?17:41
@HeikoSI they work in the existing meta examples dont they?17:41
sanujHeikoS, yes they work17:41
@HeikoSit is ok like this17:41
@HeikoSdo something that makes sense17:41
@HeikoSit should be both illustrating the meta language, and also serve as systematic testing of the constructs17:42
sanujHeikoS, okay17:42
@HeikoStry to ask esben sorig17:42
@HeikoShe know what to do there17:42
@HeikoSand he might help a bit17:42
@HeikoSwiking or lisitsyn need to merge for now17:42
sanujHeikoS, sure sure17:43
@HeikoSsanuj: ok man, take care17:44
@HeikoSbye!17:44
sanujHeikoS, nice working with you too! :D17:45
sanujhave fun17:45
@HeikoSsanuj: thanks! :) and see you soon !17:45
sanujyep :D17:45
sanuj(on irc :P)17:45
-!- HeikoS [~heiko@host-92-0-178-129.as43234.net] has quit [Remote host closed the connection]17:51
-!- OXPHOS [4441a916@gateway/web/freenode/ip.68.65.169.22] has quit [Ping timeout: 250 seconds]18:06
-!- OXPHOS [4441a916@gateway/web/freenode/ip.68.65.169.22] has joined #shogun18:09
-!- travis-ci [~travis-ci@ec2-54-163-72-166.compute-1.amazonaws.com] has joined #shogun18:10
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/14921804918:10
-!- travis-ci [~travis-ci@ec2-54-163-72-166.compute-1.amazonaws.com] has left #shogun []18:10
Saurabh7hm looks like heiko was "back"18:38
Saurabh7:D18:43
-!- sanuj [~sanuj@59.91.130.234] has quit [Remote host closed the connection]19:33
-!- lambday [6a3311c4@gateway/web/freenode/ip.106.51.17.196] has joined #shogun20:01
-!- mode/#shogun [+o lambday] by ChanServ20:01
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]20:21
-!- OXPHOS [4441a916@gateway/web/freenode/ip.68.65.169.22] has quit [Ping timeout: 250 seconds]21:15
--- Log closed Wed Aug 03 00:00:49 2016