| --- Log opened Mon May 29 00:00:50 2017 | ||
| @sukey | Issue #3823 "The problem about interface of matlab" closed by karlnapf - https://github.com/shogun-toolbox/shogun/issues/3823 | 00:05 |
|---|---|---|
| -!- HeikoS [~heiko@x4e37c3c0.dyn.telefonica.de] has quit [Ping timeout: 255 seconds] | 00:08 | |
| -!- besser82 [~besser82@fedora/besser82] has quit [Quit: “Freedom, Friends, Features, First” ---> https://fedoraproject.org] | 01:52 | |
| -!- besser82 [~besser82@fedora/besser82] has joined #shogun | 01:55 | |
| -!- mode/#shogun [+o besser82] by ChanServ | 01:55 | |
| mikeling | wiking: ping | 05:07 |
| mikeling | Why we need to make CBinaryLabels as {-1, 1} but CMultilabelLabels define as {0,1...num_classs-1}? | 07:24 |
| mikeling | I mean why we just define just define it as sort of 'DigitalLabels' which can accept {-1, 0, 1, ...num_classes} | 07:26 |
| mikeling | I'm wondering why we need CBinaryLabels as {-1, 1} rather than {0, 1} | 07:28 |
| mikeling | so we could fix this issuehttps://github.com/shogun-toolbox/shogun/issues/3686 directly rather than cast one label to another | 07:28 |
| -!- HeikoS [~heiko@x55b6020d.dyn.telefonica.de] has joined #shogun | 07:40 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 07:40 | |
| -!- HeikoS [~heiko@x55b6020d.dyn.telefonica.de] has quit [Ping timeout: 268 seconds] | 07:45 | |
| shogun-buildbot | build #74 of nightly_all is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_all/builds/74 blamelist: MikeLing <sabergeass@gmail.com>, Viktor Gal <vigsterkr@gmail.com> | 07:48 |
| -!- HeikoS [~heiko@x55b6020d.dyn.telefonica.de] has joined #shogun | 07:57 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 07:57 | |
| mikeling | HeikoS: hey! | 07:57 |
| @HeikoS | hi | 07:57 |
| @HeikoS | all good? | 07:57 |
| mikeling | Yep, nice to see you online :) | 07:58 |
| @HeikoS | yeah, I am travelling right now, so that was luck :) | 07:58 |
| mikeling | you could tell me something about CBinaryLabels and CMultilabelLabels? | 07:58 |
| @HeikoS | I will be online more from the next few days | 07:58 |
| @HeikoS | sure | 07:58 |
| mikeling | I had notice this issue https://github.com/shogun-toolbox/shogun/issues/3686 | 07:59 |
| mikeling | Why we need to make CBinaryLabels as {-1, 1} but CMultilabelLabels define as {0,1...num_classs-1}? | 07:59 |
| mikeling | could we just define it as sort of 'DigitalLabels' which can accept {-1, 0, 1, ...num_classes} | 07:59 |
| mikeling | And could I cast CBinaryLabels to CMultilabelLabels? | 08:00 |
| @HeikoS | I think the distinction doesnt make sense | 08:01 |
| @HeikoS | I mean sure internally these representations are good | 08:01 |
| @HeikoS | but from an API perspective they are nonsense | 08:01 |
| @HeikoS | the algorithms should just check at runtine | 08:01 |
| @HeikoS | mikeling: I can talk more about this, but have to run now | 08:01 |
| mikeling | ok, later :) | 08:01 |
| @HeikoS | sorry, but feel free to start an email thread on the list | 08:01 |
| @HeikoS | because I can reply to emails | 08:01 |
| mikeling | sure, thank you | 08:02 |
| -!- HeikoS [~heiko@x55b6020d.dyn.telefonica.de] has quit [Ping timeout: 260 seconds] | 08:06 | |
| -!- HeikoS_mobile [~Mutter@82.132.187.64] has joined #shogun | 08:23 | |
| HeikoS_mobile | mikeling: hi | 08:23 |
| mikeling | HeikoS_mobile: hi, I'm writing to email | 08:23 |
| HeikoS_mobile | So I think there should be no distinction between label classes, just one class | 08:23 |
| HeikoS_mobile | API wise | 08:24 |
| HeikoS_mobile | Internally these classes offer ways to check whether the numbers stored do. make sense, and how to convert them to each other if possible | 08:24 |
| HeikoS_mobile | lisitsyn: thoughts? | 08:24 |
| HeikoS_mobile | But to the outside user api, none of this should be visible | 08:25 |
| HeikoS_mobile | Should just take a bunch of numbers and them interpret them as labels in the desired way | 08:25 |
| HeikoS_mobile | wiking: | 08:25 |
| HeikoS_mobile | This is also how the features api should look like inside the algorithms | 08:26 |
| HeikoS_mobile | But start with labels | 08:26 |
| HeikoS_mobile | mikeling: you get what I mean? | 08:26 |
| HeikoS_mobile | mikeling: how's gsoc preparations going btw, all set for tomorrow? :) | 08:26 |
| HeikoS_mobile | rcurtin: how do you manage all the projects you accepted? | 08:27 |
| mikeling | sorry, I'm writing the email. Yeah, everything going well I think | 08:27 |
| HeikoS_mobile | Cool | 08:28 |
| HeikoS_mobile | mikeling: I think it's a good idea to refactor the labels | 08:28 |
| HeikoS_mobile | So yeah you know my opinion now | 08:28 |
| mikeling | Ok, Done | 08:29 |
| HeikoS_mobile | Nice I'll check | 08:30 |
| HeikoS_mobile | See you later | 08:30 |
| mikeling | Thank you | 08:30 |
| mikeling | later ;) | 08:30 |
| -!- HeikoS_mobile_ [~Mutter@82-132-221-149.dab.02.net] has joined #shogun | 08:32 | |
| -!- geektoni [~geektoni@93-34-234-213.ip52.fastwebnet.it] has joined #shogun | 08:32 | |
| -!- HeikoS_mobile [~Mutter@82.132.187.64] has quit [Ping timeout: 268 seconds] | 08:35 | |
| -!- HeikoS_mobile_ [~Mutter@82-132-221-149.dab.02.net] has quit [Ping timeout: 246 seconds] | 08:37 | |
| -!- micmn_ is now known as micmn | 09:02 | |
| -!- iglesiasg [~iglesiasg@217.119.234.214] has joined #shogun | 09:12 | |
| -!- mode/#shogun [+o iglesiasg] by ChanServ | 09:12 | |
| -!- HeikoS [~heiko@213.182.143.74] has joined #shogun | 11:03 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 11:03 | |
| -!- HeikoS [~heiko@213.182.143.74] has quit [Ping timeout: 240 seconds] | 11:07 | |
| -!- iglesiasg [~iglesiasg@217.119.234.214] has quit [Ping timeout: 240 seconds] | 12:48 | |
| -!- iglesiasg [~iglesiasg@wlan072115.mobiel.utwente.nl] has joined #shogun | 12:50 | |
| -!- mode/#shogun [+o iglesiasg] by ChanServ | 12:50 | |
| -!- geektoni [~geektoni@93-34-234-213.ip52.fastwebnet.it] has left #shogun [] | 13:01 | |
| -!- HeikoS [~heiko@213.182.143.74] has joined #shogun | 14:41 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 14:41 | |
| -!- HeikoS [~heiko@213.182.143.74] has quit [Ping timeout: 240 seconds] | 14:45 | |
| -!- iglesiasg [~iglesiasg@wlan072115.mobiel.utwente.nl] has quit [Ping timeout: 255 seconds] | 14:47 | |
| -!- iglesiasg [~iglesiasg@217.119.234.214] has joined #shogun | 14:49 | |
| -!- mode/#shogun [+o iglesiasg] by ChanServ | 14:49 | |
| -!- iglesiasg [~iglesiasg@217.119.234.214] has quit [Quit: leaving] | 15:57 | |
| @sukey | Pull Request #3824 "Replace DynArray with std::vector for io module" opened by MikeLing - https://github.com/shogun-toolbox/shogun/pull/3824 | 16:43 |
| -!- olinguyen [81615ad9@gateway/web/freenode/ip.129.97.90.217] has joined #shogun | 17:32 | |
| -!- geektoni [~Mutter@93-34-234-212.ip52.fastwebnet.it] has joined #shogun | 21:31 | |
| -!- geektoni [~Mutter@93-34-234-212.ip52.fastwebnet.it] has quit [Client Quit] | 21:35 | |
| -!- geektoni [~Mutter@93-34-234-212.ip52.fastwebnet.it] has joined #shogun | 21:36 | |
| -!- geektoni [~Mutter@93-34-234-212.ip52.fastwebnet.it] has quit [Client Quit] | 21:36 | |
| @sukey | Issue #3822 "The problem about install shogun 2.0.0" closed by karlnapf - https://github.com/shogun-toolbox/shogun/issues/3822 | 22:54 |
| --- Log closed Tue May 30 00:00:51 2017 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!