| --- Log opened Tue May 17 00:00:59 2016 | ||
| shogun-buildbot | build #664 of debian wheezy - memcheck is complete: Failure [failed memory check] Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/664 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Sanuj <sanuj.sharma.in@gmail.com> | 00:05 |
|---|---|---|
| -!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun | 00:07 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 00:07 | |
| -!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Client Quit] | 00:10 | |
| shogun-buildbot | build #665 of debian wheezy - memcheck is complete: Failure [failed memory check] Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/665 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Sanuj <sanuj.sharma.in@gmail.com> | 03:16 |
| -!- sanuj [~sanuj@117.203.4.100] has joined #shogun | 03:18 | |
| -!- sanuj [~sanuj@117.203.4.100] has quit [Ping timeout: 260 seconds] | 04:18 | |
| shogun-buildbot | build #1132 of nightly_default is complete: Failure [failed notebooks] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/1132 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Sanuj <sanuj.sharma.in@gmail.com>, Wu Lin <yorker.lin@gmail.com> | 04:53 |
| -!- sanuj [~sanuj@117.203.4.100] has joined #shogun | 05:16 | |
| -!- sanuj [~sanuj@117.203.4.100] has quit [Ping timeout: 240 seconds] | 06:26 | |
| -!- sanuj [~sanuj@117.203.4.100] has joined #shogun | 06:51 | |
| -!- besser82_ [~besser82@fedora/besser82] has joined #shogun | 06:57 | |
| -!- mode/#shogun [+o besser82_] by ChanServ | 06:57 | |
| -!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 240 seconds] | 07:00 | |
| sanuj | lisitsyn, there? | 07:05 |
| @wiking | sanuj: what'sup | 07:15 |
| sanuj | wiking, for the shogun-base list | 07:16 |
| sanuj | i grepped "()=0" or "() = 0" | 07:16 |
| sanuj | for pure virtual functions | 07:16 |
| sanuj | to find abstract classes | 07:16 |
| sanuj | one of the classes is MulticlassMachine.h | 07:17 |
| sanuj | and the class hierarchy goes like CMulticlassMachine --> CBaseMulticlassMachine --> CMachine --> CSGObject | 07:17 |
| sanuj | wiking, then shall i add all these classes to the base-shogun list? | 07:17 |
| @wiking | mmm | 07:18 |
| sanuj | yes/no? | 07:18 |
| @wiking | just checking | 07:18 |
| @wiking | to see what's the diff | 07:18 |
| @wiking | between CBaseMulticlassMachine and toCMulticlassMachine | 07:18 |
| sanuj | okay :) | 07:18 |
| sanuj | are you back in Singapore? | 07:19 |
| @wiking | do | 07:19 |
| @wiking | h | 07:19 |
| @wiking | i mean not yet :) | 07:19 |
| @wiking | whoa this i dont understand | 07:20 |
| @wiking | but ok this is a bit unrelated | 07:20 |
| @wiking | but yeah the idea is that abstract classes are actually just interfaces | 07:20 |
| @wiking | for the implementations | 07:20 |
| @wiking | these interfaces should all be included in the shogun-base | 07:20 |
| @wiking | sanuj: but this CMulticlassMachine --> CBaseMulticlassMachine | 07:20 |
| @wiking | troubles me a bit | 07:20 |
| sanuj | okay | 07:20 |
| @wiking | but will check out with others on the mailinglist | 07:20 |
| sanuj | but all abstract classes don't have pure virtual functions | 07:21 |
| @wiking | abstract class = pure virtual functions | 07:22 |
| @wiking | ;) | 07:22 |
| @wiking | that's the definition of abstract class in c++ | 07:22 |
| @wiking | "Classes derived from the abstract class must implement the pure virtual function or they, too, are abstract classes. A virtual function is declared as "pure" by using the pure-specifier syntax (described in Class Protocol Implementation). Consider the example presented in Virtual Functions." | 07:23 |
| @wiking | sanuj: ^ | 07:23 |
| @wiking | ok? | 07:23 |
| sanuj | wiking, yes i know that | 07:23 |
| sanuj | but we would be adding CMachine to the list right? | 07:23 |
| sanuj | but it's not abstract i.e. no pure virtual function | 07:24 |
| sanuj | wiking, what do you reckon ^ | 07:24 |
| @wiking | yes we need to add the classes who are the parents of the classes that you wanna add to the shogun-base :) | 07:24 |
| @wiking | like SGObject | 07:24 |
| @wiking | etc | 07:24 |
| sanuj | yeah | 07:24 |
| sanuj | wiking, also i have swig interface working for my tag prototype | 07:25 |
| @wiking | cool! | 07:25 |
| sanuj | what shall i do next for the tag prototype | 07:25 |
| @wiking | so | 07:26 |
| @wiking | now the question is how to create the whole new plugins arch | 07:26 |
| sanuj | wiking, i think we need to add this tag prototype in dynaplugz | 07:26 |
| @wiking | yeah | 07:26 |
| @wiking | as well as start to define | 07:26 |
| @wiking | shogun-base | 07:26 |
| @wiking | with tag prototype + dynaplugz | 07:26 |
| sanuj | cool | 07:26 |
| sanuj | wiking, we can incrementally add classes to shogun-base right? | 07:27 |
| @wiking | yeah but | 07:27 |
| @wiking | the thing is that we need to do this | 07:27 |
| @wiking | as part of shogun repo | 07:28 |
| @wiking | :) | 07:28 |
| sanuj | on a different branch? | 07:28 |
| @wiking | yes | 07:28 |
| @wiking | i mean fork the repo | 07:28 |
| @wiking | start a new branch | 07:28 |
| @wiking | and work on that | 07:28 |
| sanuj | so first lets make dynaplugz + tags work separately before forking | 07:29 |
| @wiking | k | 07:29 |
| -!- sanuj [~sanuj@117.203.4.100] has quit [Ping timeout: 244 seconds] | 08:34 | |
| -!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting] | 09:05 | |
| -!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun | 09:21 | |
| @wiking | shogun-buildbot: force build --branch=develop 'xenial - libshogun' | 09:25 |
| shogun-buildbot | The build has been queued, I'll give a shout when it starts | 09:25 |
| @wiking | shogun-buildbot: force build --branch=develop 'FC23 - libshogun' | 09:26 |
| shogun-buildbot | The build has been queued, I'll give a shout when it starts | 09:26 |
| -!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting] | 10:20 | |
| -!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun | 10:21 | |
| -!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun | 10:26 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 10:26 | |
| @wiking | HeikoS: man aws machines are insane slow | 10:48 |
| @wiking | :) | 10:48 |
| @wiking | lucky that we have ccache | 10:48 |
| shogun-buildbot | build #0 of FC23 - libshogun is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/FC23%20-%20libshogun/builds/0 | 10:56 |
| -!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting] | 11:00 | |
| @wiking | shogun-buildbot: force build --branch=develop 'xenial - libshogun' | 11:01 |
| -!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun | 11:01 | |
| shogun-buildbot | build #0 of xenial - libshogun is complete: Failure [failed compile] Build details are at http://buildbot.shogun-toolbox.org/builders/xenial%20-%20libshogun/builds/0 | 11:12 |
| -!- sanuj [~sanuj@117.204.252.238] has joined #shogun | 11:14 | |
| @wiking | shogun-buildbot: force build --branch=develop 'FC22 - libshogun' | 11:15 |
| shogun-buildbot | The build has been queued, I'll give a shout when it starts | 11:15 |
| shogun-buildbot | build #217 forced | 11:17 |
| shogun-buildbot | I'll give a shout when the build finishes | 11:17 |
| @HeikoS | wiking: hehe | 11:48 |
| @wiking | anyhow | 11:50 |
| @wiking | now we have latest ubuntu LTS buildbot as well | 11:50 |
| @wiking | and fedora 23 | 11:50 |
| @wiking | there's more reds to be fixed | 11:50 |
| sanuj | 16.04 | 11:50 |
| shogun-buildbot | build #217 of FC22 - libshogun is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/FC22%20-%20libshogun/builds/217 | 11:51 |
| @wiking | ok this one failed as well :) | 11:52 |
| -!- mizari [~mizari@95-174-213-100.nts.su] has joined #shogun | 12:06 | |
| @wiking | mizari: good tld | 12:08 |
| @wiking | where did you get it? :) | 12:08 |
| mizari | hi | 12:09 |
| mizari | whats tld? :) | 12:09 |
| @wiking | .su | 12:10 |
| @wiking | that's the tld for soviet union :) | 12:10 |
| mizari | I dont know what tld stands for | 12:10 |
| mizari | if you're about @95-174-213-100.nts.su in my whois, then I got it from my internet provider | 12:11 |
| mizari | which is in Russia | 12:11 |
| mizari | top-level domain? | 12:12 |
| @wiking | woah | 12:12 |
| @wiking | your internet provider is super cool | 12:12 |
| @wiking | :))) | 12:12 |
| sanuj | wiking, do we have a design draft for plugins also? :) | 12:13 |
| @wiking | mmm not really | 12:13 |
| mizari | thanks, I'll convey them your message | 12:13 |
| @wiking | wanna write one? | 12:13 |
| sanuj | lisitsyn, yooooo | 12:13 |
| @wiking | sanuj: but again | 12:13 |
| @wiking | shogun base should be the first thing | 12:13 |
| @wiking | you define | 12:13 |
| @wiking | :) | 12:13 |
| sanuj | wiking, i'm trying to understand how to write a plugin architecture for c++ | 12:14 |
| @wiking | ah | 12:14 |
| @wiking | dyload :) | 12:14 |
| @wiking | dlopen | 12:14 |
| @wiking | sanuj: but again | 12:15 |
| @wiking | you'll need to write | 12:15 |
| @wiking | a brand new stuff | 12:15 |
| @wiking | for shogun-base | 12:15 |
| @wiking | a config file support | 12:15 |
| @wiking | because there the user should be able to set | 12:15 |
| @wiking | a lot of things | 12:15 |
| @wiking | for example where it should look for the dynamic libraries | 12:15 |
| @wiking | available to load in | 12:15 |
| sanuj | wiking, tell me the things i should read | 12:16 |
| sanuj | i'm not really a c++ magician like lisitsyn or besser82_ | 12:16 |
| @wiking | ok so you should read about shared libraries | 12:16 |
| @wiking | i.e. man dlopen | 12:16 |
| @wiking | and again | 12:17 |
| mizari | is dlopen cross-platform? | 12:17 |
| @wiking | *nix wise yes | 12:17 |
| mizari | okie | 12:18 |
| @wiking | sanuj: but again as said write a simple config file support | 12:18 |
| @wiking | sanuj: something like | 12:18 |
| @wiking | sanuj: key=value | 12:18 |
| lisitsyn | what config file? | 12:20 |
| @wiking | lisitsyn: because how else we gonna be able to support the override of plugin location as well as set what linalg backend one would like shogun to use | 12:21 |
| @wiking | so some global params | 12:21 |
| @wiking | that you dont want to set programatically all the time | 12:21 |
| @wiking | arghfuck | 12:22 |
| lisitsyn | mizari: is it tomsk online? | 12:22 |
| lisitsyn | :D | 12:22 |
| @wiking | how is this possible | 12:22 |
| @wiking | #define EIGEN_WORLD_VERSION 3 | 12:22 |
| @wiking | #define EIGEN_MAJOR_VERSION 2 | 12:22 |
| @wiking | #define EIGEN_MINOR_VERSION 92 | 12:22 |
| @wiking | ii libeigen3-dev 3.3~beta1-2 all lightweight C++ template library for linear algebra | 12:22 |
| @wiking | damn eigen ppl | 12:22 |
| lisitsyn | wiking: not eigen | 12:22 |
| lisitsyn | ubuntu | 12:22 |
| @wiking | what ?D | 12:22 |
| mizari | lisitsyn: yeah, Tomsk's New Telesystems | 12:23 |
| @wiking | lisitsyn: it isn't 3.3 ?:) | 12:23 |
| lisitsyn | wiking: yeah but I guess ubuntu people screwed | 12:24 |
| @wiking | i mean how? :) | 12:24 |
| lisitsyn | wiking: well packaged it with the wrong name | 12:25 |
| @wiking | no | 12:25 |
| @wiking | it's actually 3.3 | 12:25 |
| @wiking | src | 12:25 |
| @wiking | lisitsyn: if you download the 3.3beta1 src | 12:26 |
| @wiking | you'll get the same | 12:26 |
| lisitsyn | wiking: ah ok | 12:27 |
| @HeikoS | wiking, lisitsyn | 12:31 |
| @HeikoS | jo | 12:31 |
| @HeikoS | about the summit, I think you guys should go | 12:31 |
| @HeikoS | make sure all is organised as early as possible | 12:32 |
| @HeikoS | lisitsyn: especially your visa | 12:32 |
| @HeikoS | we can change names until last minute, so no risk of loosing a place | 12:32 |
| @HeikoS | and if you change mind, let me know ;D | 12:32 |
| @wiking | lisitsyn: did you fix the static assertation error in tapkee by any chance for 3.3? | 12:32 |
| -!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Quit: Leaving.] | 12:33 | |
| @wiking | or have you actually seen it? | 12:33 |
| @wiking | lisitsyn: http://pastebin.com/LGqQphNc | 12:37 |
| lisitsyn | wiking: ahh yes | 12:38 |
| lisitsyn | HeikoS: I can apply soon | 12:38 |
| lisitsyn | wiking: yes easy to fix | 12:38 |
| @wiking | lisitsyn: cool, if you gimme hints i can do it if you dont have time... or when can you haz it? | 12:40 |
| @wiking | i have another error in shogun | 12:40 |
| @wiking | so i'm gonna fix that now | 12:40 |
| @wiking | btw | 12:40 |
| @wiking | 3.2.91 <= version is 3.3 | 12:41 |
| @wiking | :) | 12:41 |
| @wiking | EIGEN_VERSION_AT_LEAST(3, 2, 91) macro should work | 12:42 |
| sanuj | wiking, config file is where version variables are defined for cmake? | 12:54 |
| @wiking | no | 12:54 |
| @wiking | it's a config file for the library itself | 12:54 |
| sanuj | you are talking about this dlopen? | 12:55 |
| sanuj | http://linux.die.net/man/3/dlopen | 12:55 |
| @wiking | yes | 12:55 |
| sanuj | wiking, does shogun currently have a similar config file? | 12:56 |
| @wiking | noup | 12:56 |
| sanuj | wiking, can you give me an example for this config flie | 12:57 |
| @wiking | foo=bar | 12:58 |
| @wiking | it's a simple properties file | 12:58 |
| sanuj | just a simple hpp with key=value pairs? | 12:59 |
| @wiking | no it's not a hpp | 13:03 |
| @wiking | its just a simple txt file | 13:03 |
| @wiking | with lines like 'a=b' | 13:03 |
| @wiking | or 'key=value' | 13:03 |
| @wiking | and create an object | 13:04 |
| @wiking | that stores those (key, value) tuples | 13:04 |
| @wiking | like in an unordered map or something | 13:04 |
| sanuj | oh | 13:05 |
| sanuj | :D | 13:05 |
| lisitsyn | sry | 13:08 |
| lisitsyn | what config?! | 13:08 |
| sanuj | :P | 13:08 |
| sanuj | dlopen() will be used for loading plugins? | 13:09 |
| lisitsyn | yes on linux | 13:09 |
| sanuj | lisitsyn, for other os? | 13:10 |
| lisitsyn | it is different on windows :) | 13:10 |
| lisitsyn | but we don't work with windows yet anyway | 13:10 |
| lisitsyn | wiking: would you bother? | 13:10 |
| sanuj | will work on mac osx | 13:10 |
| sanuj | since its unix | 13:10 |
| @wiking | lisitsyn: yep can check | 13:11 |
| @wiking | sanuj: concentrate on dlopen | 13:11 |
| @wiking | no need to get lost on portability for now | 13:11 |
| sanuj | okay | 13:11 |
| sanuj | i thought dynaplugz would have some plugins | 13:11 |
| sanuj | it just instantiates objects of cdynaplugz | 13:12 |
| sanuj | has anyone seen the code? | 13:12 |
| lisitsyn | I didn't | 13:12 |
| lisitsyn | sanuj: you can check dlopen in aer | 13:12 |
| sanuj | lisitsyn, https://github.com/shogun-toolbox/dynaplugz/blob/develop/src/dynaplugz/dynaplugz.cpp | 13:13 |
| sanuj | lisitsyn, thanks :D | 13:13 |
| sanuj | lisitsyn, what exactly is aer? | 13:13 |
| @wiking | sanuj: https://github.com/shogun-toolbox/aer | 13:14 |
| sanuj | cool c++ designing | 13:14 |
| sanuj | no i know the link | 13:14 |
| @wiking | https://github.com/shogun-toolbox/aer/blob/master/src/aer/dynamic/library.cpp | 13:14 |
| sanuj | what does it stand for | 13:14 |
| lisitsyn | sanuj: dynaplugz looks like a dummy | 13:15 |
| sanuj | lisitsyn, dynaplugz was mentioned in my gsoc wiki page | 13:16 |
| sanuj | should have put aer | 13:16 |
| sanuj | it has everything i need | 13:16 |
| sanuj | took any from here | 13:16 |
| sanuj | (or codeproject.com) | 13:17 |
| lisitsyn | yes it is a working prototype for plugins :) | 13:17 |
| sanuj | lisitsyn, didn't you implement tags in aer | 13:17 |
| sanuj | why reinvent the wheel | 13:17 |
| lisitsyn | sanuj: no, that was missed | 13:19 |
| sanuj | okay | 13:19 |
| sanuj | now i'm substituting dynaplugz by aer for all my plans :P | 13:20 |
| sanuj | lisitsyn, obj.get("num").asInt() is also working in my tags prototype | 13:20 |
| sanuj | do you want anything else? | 13:20 |
| lisitsyn | I think that's it for now, plugins are interesting now | 13:21 |
| sanuj | okay | 13:21 |
| sanuj | i'll understand aer | 13:21 |
| sanuj | lisitsyn, then shall we put tags in aer? | 13:21 |
| lisitsyn | sanuj: no | 13:21 |
| lisitsyn | just use it as a reference | 13:21 |
| sanuj | okay | 13:21 |
| lisitsyn | then we will purge it | 13:21 |
| sanuj | don't understand what you said | 13:22 |
| sanuj | lisitsyn, https://github.com/shogun-toolbox/aer/blob/master/src/aer/base/tag.hpp | 13:31 |
| sanuj | only swig interface was missing then | 13:31 |
| @wiking | arianepaola: ping | 13:42 |
| lisitsyn | sanuj: aer is just a mapping of my idea to C++ code :D | 13:50 |
| lisitsyn | but you just take related parts | 13:50 |
| lisitsyn | then we will remove that at all | 13:50 |
| @wiking | lisitsyn: some 90s music : https://www.youtube.com/watch?v=QrjX2-4sB_4 | 14:06 |
| @wiking | from 2015 :P | 14:06 |
| sanuj | lisitsyn, cool | 14:10 |
| -!- leagoetz [~leagoetz@nat-196-147.internal.eduroam.ucl.ac.uk] has joined #shogun | 14:13 | |
| -!- leagoetz [~leagoetz@nat-196-147.internal.eduroam.ucl.ac.uk] has quit [Remote host closed the connection] | 14:40 | |
| -!- sanuj [~sanuj@117.204.252.238] has quit [Ping timeout: 276 seconds] | 14:58 | |
| @wiking | lisitsyn: woah man | 14:59 |
| @wiking | gcc5 generates a rather interesting object :D | 14:59 |
| lisitsyn | wiking: what's it? | 15:00 |
| @wiking | ah no | 15:00 |
| @wiking | we are iditos | 15:00 |
| @wiking | :DDD | 15:00 |
| -!- sanuj [~sanuj@117.204.252.238] has joined #shogun | 15:25 | |
| -!- HeikoS [~heiko@nat-243-249.internal.eduroam.ucl.ac.uk] has joined #shogun | 15:29 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 15:29 | |
| @wiking | HeikoS: ping | 15:33 |
| @HeikoS | pong | 15:33 |
| lisitsyn | HeikoS: hey | 15:34 |
| @HeikoS | jojo | 15:34 |
| lisitsyn | about your inquiry | 15:34 |
| lisitsyn | :) | 15:34 |
| lisitsyn | I can take care of that rather soon | 15:34 |
| lisitsyn | if you decide we should go | 15:34 |
| lisitsyn | and if you say we can switch names | 15:34 |
| lisitsyn | it sounds good | 15:34 |
| @wiking | HeikoS: ok so i've finally managed to debug why those two unit tests fails on fedora | 15:35 |
| @wiking | -value float64 0.01455988410195748 | 15:35 |
| @wiking | +value float64 0.01455988410195747 | 15:35 |
| @wiking | it's for reference_results/gaussian_processes/gaussian_process_regression.dat | 15:35 |
| @wiking | HeikoS: so there's a diff but it's quite minor | 15:36 |
| @wiking | HeikoS: SGObject.equals does not allow error right? | 15:37 |
| @wiking | ah no it has | 15:38 |
| @wiking | :) | 15:38 |
| @wiking | HeikoS: 10E-8 should be good enough precision no? | 15:40 |
| @HeikoS | wiking: it does | 15:42 |
| @HeikoS | yeah 1e-8 is good | 15:42 |
| @HeikoS | lisitsyn: great! | 15:42 |
| @HeikoS | do it :) | 15:42 |
| @HeikoS | wiking: great catch, hope that solves it | 15:46 |
| shogun-buildbot | build #218 of FC22 - libshogun is complete: Success [build successful] Build details are at http://buildbot.shogun-toolbox.org/builders/FC22%20-%20libshogun/builds/218 | 15:52 |
| @wiking | HeikoS: ^ | 15:53 |
| shogun-buildbot | build #1 of xenial - libshogun is complete: Failure [failed compile] Build details are at http://buildbot.shogun-toolbox.org/builders/xenial%20-%20libshogun/builds/1 blamelist: Viktor Gal <viktor.gal@maeth.com> | 15:53 |
| @wiking | HeikoS: and btw there you go: xenial uses 3.3 eigen and now we cannot compile shogun on it | 15:53 |
| @wiking | this is why we should fix it rather than disable it | 15:53 |
| shogun-buildbot | build #2865 of bsd1 - libshogun is complete: Failure [failed configure] Build details are at http://buildbot.shogun-toolbox.org/builders/bsd1%20-%20libshogun/builds/2865 blamelist: Viktor Gal <viktor.gal@maeth.com> | 15:58 |
| shogun-buildbot | build #1 of FC23 - libshogun is complete: Success [build successful] Build details are at http://buildbot.shogun-toolbox.org/builders/FC23%20-%20libshogun/builds/1 | 16:02 |
| sanuj | where is the dancing bot | 16:03 |
| @HeikoS | wiking: crappy | 16:28 |
| @HeikoS | wiking: arianepaola should be able to fix this no? | 16:28 |
| -!- mizari [~mizari@95-174-213-100.nts.su] has quit [Read error: Connection reset by peer] | 16:31 | |
| -!- HeikoS [~heiko@nat-243-249.internal.eduroam.ucl.ac.uk] has quit [Quit: Leaving.] | 16:35 | |
| -!- HeikoS [~heiko@nat-243-249.internal.eduroam.ucl.ac.uk] has joined #shogun | 16:35 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 16:35 | |
| -!- leagoetz [~leagoetz@host-92-0-162-192.as43234.net] has joined #shogun | 16:45 | |
| -!- mizari [~mizari@95-174-213-100.nts.su] has joined #shogun | 16:47 | |
| -!- sanuj [~sanuj@117.204.252.238] has quit [Ping timeout: 240 seconds] | 17:03 | |
| -!- leagoetz [~leagoetz@host-92-0-162-192.as43234.net] has quit [Remote host closed the connection] | 17:03 | |
| -!- HeikoS [~heiko@nat-243-249.internal.eduroam.ucl.ac.uk] has quit [Quit: Leaving.] | 17:10 | |
| -!- mizari [~mizari@95-174-213-100.nts.su] has quit [Read error: Connection reset by peer] | 17:19 | |
| -!- mizari [~mizari@95-174-213-100.nts.su] has joined #shogun | 17:21 | |
| -!- HeikoS [~heiko@nat-243-249.internal.eduroam.ucl.ac.uk] has joined #shogun | 17:37 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 17:38 | |
| @wiking | ping | 18:28 |
| @wiking | HeikoS: ^ | 18:28 |
| @HeikoS | lisitsyn: website? :D | 18:37 |
| lisitsyn | uh | 18:38 |
| @wiking | lisitsyn: | 18:38 |
| @wiking | so what do you think about the release buildbot job? | 18:38 |
| @wiking | i was thinking that every/any push to master | 18:39 |
| @wiking | would trigger that job | 18:39 |
| lisitsyn | what runs the job? | 18:39 |
| @wiking | change in the master branch | 18:39 |
| @wiking | so any push to master | 18:39 |
| @wiking | will trigger that job | 18:39 |
| @wiking | any concerns? | 18:39 |
| lisitsyn | no I mean | 18:39 |
| lisitsyn | is it buildbot job? | 18:39 |
| @wiking | yep | 18:39 |
| @HeikoS | I think it is good | 18:40 |
| @wiking | lisitsyn: do you think it's good or not safe or something? | 18:40 |
| lisitsyn | I am ok | 18:41 |
| @wiking | k | 18:42 |
| @wiking | thnx | 18:42 |
| @wiking | HeikoS: ok i've set up the trigger for zenodo for the doi | 18:48 |
| @wiking | so the next release will trigger the doi creation | 18:48 |
| @wiking | lol | 18:48 |
| @wiking | why is the 4.1.0 release a Draft? | 18:49 |
| @wiking | haha it wasn't ever published | 18:49 |
| @wiking | HeikoS: :) | 18:49 |
| -!- sanuj [~sanuj@117.203.5.121] has joined #shogun | 18:50 | |
| @wiking | HeikoS: so you'll get a doi just in a second | 18:51 |
| @wiking | ok so people of shogun | 18:55 |
| @wiking | let me introduce you to | 18:55 |
| @wiking | http://dx.doi.org/10.5281/zenodo.51537 | 18:55 |
| @wiking | :))) | 18:55 |
| lisitsyn | haha | 18:56 |
| @wiking | ok so it's done | 18:56 |
| lisitsyn | how can I set my affiliation | 18:56 |
| @wiking | thnx to HeikoS who forgot to publish the | 18:56 |
| @wiking | release | 18:56 |
| @wiking | lisitsyn: it took it from github | 18:57 |
| @wiking | afaik | 18:57 |
| lisitsyn | I see | 18:57 |
| @HeikoS | wiking: what did I break? :D | 18:57 |
| @wiking | HeikoS: didn't break | 18:57 |
| @wiking | you prepared nicely the release on github | 18:57 |
| @wiking | but it was a draft | 18:57 |
| @wiking | you did not release it | 18:58 |
| @wiking | :) | 18:58 |
| @HeikoS | lol | 18:58 |
| @HeikoS | really? | 18:58 |
| @wiking | yep | 18:58 |
| @HeikoS | cant remember | 18:58 |
| @wiking | but now it's ok | 18:58 |
| @HeikoS | well thanks for doing that | 18:58 |
| @wiking | because this way we have a doi | 18:58 |
| @wiking | :) | 18:58 |
| @HeikoS | the doi looks nice | 18:58 |
| @HeikoS | awe-some! | 18:59 |
| @HeikoS | will cite that instead of paper | 18:59 |
| @HeikoS | well | 18:59 |
| @wiking | \o/ | 18:59 |
| @HeikoS | the code is in feature branch | 18:59 |
| @HeikoS | hehe | 18:59 |
| @wiking | hahahah | 18:59 |
| @wiking | you'll have a chance to update the article? | 18:59 |
| @wiking | i mean this is just submission deadline now? | 19:00 |
| @wiking | because by then we'll have 4.2.0 release | 19:00 |
| @wiking | and if you merge the feature branch then you can update the doi | 19:00 |
| @wiking | ;P | 19:00 |
| @wiking | (tricks of publishing 101) | 19:00 |
| @wiking | HeikoS: is Rahul around you? | 19:06 |
| @HeikoS | yep | 19:06 |
| @HeikoS | hehe | 19:06 |
| @HeikoS | wiking: well the feature branch doesnt fullfull some shogun standards yet | 19:06 |
| @HeikoS | would be cool to have that | 19:06 |
| @wiking | well anyhow | 19:06 |
| @HeikoS | but I think we will just wait until we are done, and then put the DOI in for that | 19:07 |
| @HeikoS | NIPS takes like 6 months after submission before there is a final version online | 19:07 |
| @HeikoS | plenty of time | 19:07 |
| @wiking | yep | 19:07 |
| @wiking | you can even have the doi for the post gsoc release :) | 19:07 |
| @HeikoS | I still want to update the JMLR | 19:07 |
| @HeikoS | yeah ;) | 19:08 |
| lisitsyn | update? | 19:08 |
| @HeikoS | like re-submit | 19:09 |
| @HeikoS | new author list | 19:09 |
| @HeikoS | and maybe some more modern content ;) | 19:09 |
| lisitsyn | is it possible? | 19:09 |
| @HeikoS | the old one has 200+ citations | 19:09 |
| @HeikoS | lisitsyn: yes I think if new stuff is in | 19:09 |
| @HeikoS | lisitsyn: jo | 19:24 |
| @HeikoS | question: | 19:24 |
| @HeikoS | how to tell shogun the number of threads to use in openmp? | 19:24 |
| @HeikoS | because get_global_parallel->set_num_threads() doesnt do it | 19:24 |
| @HeikoS | wiking: or you know that? | 19:24 |
| @wiking | you can set an env var | 19:28 |
| @wiking | OMP_NUM_THREADS | 19:28 |
| @wiking | but i guess we should update set_num_threads :) | 19:29 |
| @wiking | to support openmp | 19:29 |
| @wiking | as well | 19:29 |
| @HeikoS | wiking: yes that would be the solution right? | 19:30 |
| @wiking | you mean ? | 19:30 |
| @HeikoS | #ifdef USE_OPENMP | 19:30 |
| @HeikoS | omp_set_num_threads(4); | 19:30 |
| @HeikoS | #endif | 19:30 |
| @wiking | yeah | 19:30 |
| @HeikoS | inside the set_num_threads | 19:30 |
| @wiking | i mean no | 19:30 |
| @HeikoS | of Parallel | 19:30 |
| @wiking | but something like this :) | 19:30 |
| @wiking | dont fix it to 4 | 19:30 |
| @wiking | :P | 19:30 |
| @HeikoS | why not? :D | 19:30 |
| @HeikoS | void Parallel::set_num_threads(int32_t n) | 19:32 |
| @HeikoS | { | 19:32 |
| @HeikoS | #ifndef HAVE_PTHREAD | 19:32 |
| @HeikoS | ASSERT(n==1) | 19:32 |
| @HeikoS | #endif | 19:32 |
| @HeikoS | num_threads=n; | 19:32 |
| @HeikoS | look what I found :) | 19:32 |
| @HeikoS | ah | 19:32 |
| @HeikoS | #ifndef | 19:32 |
| @HeikoS | I thought this was #ifdef | 19:32 |
| @HeikoS | haha | 19:32 |
| @HeikoS | Kernel.cpp explicitly asks Parallel for number of threads btw | 19:32 |
| @HeikoS | and then create pthreads stuff with that number | 19:32 |
| @HeikoS | but well | 19:33 |
| @HeikoS | thanks | 19:33 |
| -!- sanuj [~sanuj@117.203.5.121] has quit [Quit: Leaving] | 19:38 | |
| -!- HeikoS [~heiko@nat-243-249.internal.eduroam.ucl.ac.uk] has quit [Quit: Leaving.] | 19:39 | |
| -!- mizari [~mizari@95-174-213-100.nts.su] has quit [Ping timeout: 276 seconds] | 20:51 | |
| shogun-buildbot | build #2 of xenial - libshogun is complete: Failure [failed compile] Build details are at http://buildbot.shogun-toolbox.org/builders/xenial%20-%20libshogun/builds/2 blamelist: Viktor Gal <viktor.gal@maeth.com> | 21:25 |
| shogun-buildbot | build #2866 of bsd1 - libshogun is complete: Failure [failed configure] Build details are at http://buildbot.shogun-toolbox.org/builders/bsd1%20-%20libshogun/builds/2866 blamelist: Viktor Gal <viktor.gal@maeth.com> | 21:27 |
| -!- besser82_ [~besser82@fedora/besser82] has quit [Ping timeout: 260 seconds] | 22:22 | |
| -!- Saurabh7 [Saurabh7@gateway/shell/panicbnc/x-gbgxwuujaknpqmda] has quit [Ping timeout: 260 seconds] | 22:46 | |
| -!- Saurabh7 [Saurabh7@gateway/shell/panicbnc/x-oxzhsmcptzyjglej] has joined #shogun | 22:46 | |
| --- Log closed Wed May 18 00:00:01 2016 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!