--- Log opened Sun Oct 13 00:00:39 2013 | ||
-!- van51 [~van51@athedsl-399544.home.otenet.gr] has joined #shogun | 01:18 | |
-!- van51 [~van51@athedsl-399544.home.otenet.gr] has quit [Read error: No route to host] | 01:31 | |
shogun-buildbot | build #576 of nightly_default is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/576 | 04:21 |
---|---|---|
-!- new_lido_afk [~walid@41.218.176.105] has quit [Ping timeout: 245 seconds] | 07:04 | |
-!- new_lido_afk [~walid@41.218.172.131] has joined #shogun | 07:18 | |
-!- new_lido_afk is now known as new_lido | 07:25 | |
-!- hushell [~hushell@c-98-232-178-161.hsd1.or.comcast.net] has quit [Quit: WeeChat 0.3.7] | 08:26 | |
-!- new_lido [~walid@41.218.172.131] has quit [Quit: Leaving] | 09:23 | |
-!- iglesiasg [~iglesiasg@s83-179-44-135.cust.tele2.se] has joined #shogun | 10:36 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 10:36 | |
@iglesiasg | hey guys, good morning | 10:36 |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 10:44 | |
shogun-notifier- | shogun: hushell :develop * 9d942b1 / / (6 files): https://github.com/shogun-toolbox/shogun/commit/9d942b1b5425780f0a35207be602d0f1114879be | 10:44 |
shogun-notifier- | shogun: FGM notebook update: using subset of training data | 10:44 |
shogun-notifier- | shogun: Fernando Iglesias :develop * 4117c9d / / (6 files): https://github.com/shogun-toolbox/shogun/commit/4117c9d7e0b4de16b8114f09a26ecd081bb1ef3c | 10:44 |
shogun-notifier- | shogun: Merge pull request #1719 from hushell/develop | 10:44 |
shogun-notifier- | shogun: | 10:44 |
shogun-notifier- | shogun: FGM notebook update: using subset of training data | 10:44 |
-!- van51 [~van51@athedsl-399544.home.otenet.gr] has joined #shogun | 10:59 | |
shogun-buildbot | build #1960 of deb3 - modular_interfaces is complete: Failure [failed test python modular] Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1960 blamelist: hushell <dom343@gmail.com>, Fernando Iglesias <fernando.iglesiasg@gmail.com> | 11:22 |
@sonney2k | shogun-buildbot, force build --branch=develop nightly_default | 11:39 |
shogun-buildbot | build #577 forced | 11:39 |
shogun-buildbot | I'll give a shout when the build finishes | 11:39 |
van51 | sonney2k: I have a couple more ready | 11:41 |
van51 | sonney2k: shall I make a PR for those or all of them together? | 11:42 |
-!- van51 [~van51@athedsl-399544.home.otenet.gr] has quit [Read error: No route to host] | 11:54 | |
-!- travis-ci [~travis-ci@ec2-50-17-62-119.compute-1.amazonaws.com] has joined #shogun | 11:57 | |
travis-ci | [travis-ci] it's Fernando Iglesias's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/12478623 | 11:57 |
-!- travis-ci [~travis-ci@ec2-50-17-62-119.compute-1.amazonaws.com] has left #shogun [] | 11:57 | |
-!- van51 [~van51@athedsl-399544.home.otenet.gr] has joined #shogun | 11:58 | |
-!- votjakovr [~votjakovr@95-27-254-98.broadband.corbina.ru] has joined #shogun | 12:12 | |
shogun-buildbot | build #577 of nightly_default is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/577 | 12:13 |
@iglesiasg | sonney2k, ping? | 13:10 |
van51 | hey iglesiasg | 13:24 |
@iglesiasg | hi van51 | 13:24 |
van51 | iglesiasg: do you know anything about where the web demos will be hosted and/or when? | 13:25 |
@iglesiasg | van51, no sorry, no idea | 13:25 |
van51 | iglesiasg: there is a data mining course starting here in my uni and I know the TA and I was thinking of asking them to show some algorithms from shogun | 13:25 |
@iglesiasg | van51, that is an awesome idea | 13:26 |
van51 | I remember us viewing kmeans for instance from a java applet | 13:26 |
@iglesiasg | van51, so what do you think it is the state of the demos? | 13:26 |
@iglesiasg | van51, I poked a bit around the ones in the PR you just sent | 13:26 |
van51 | iglesiasg: most of them are pretty good | 13:27 |
van51 | iglesiasg: I don't like the toy generator when there are more than 1 classes | 13:27 |
@iglesiasg | van51, I found some glitches though :S | 13:27 |
@iglesiasg | van51, that was one of them | 13:27 |
van51 | iglesiasg: what else? | 13:27 |
@iglesiasg | van51, in KMeans the circunferences are pretty weird | 13:28 |
@iglesiasg | van51, they become way too large | 13:28 |
@iglesiasg | I just tried a couple of examples | 13:28 |
van51 | iglesiasg: yeah indeed | 13:28 |
@iglesiasg | and in KMeans also | 13:28 |
@iglesiasg | if you click with the right button | 13:28 |
@iglesiasg | it creates points of another class | 13:28 |
@iglesiasg | that shouldn't be allowed here I guess | 13:28 |
-!- iglesiasg [~iglesiasg@s83-179-44-135.cust.tele2.se] has left #shogun ["Leaving"] | 13:29 | |
-!- iglesiasg [~iglesiasg@s83-179-44-135.cust.tele2.se] has joined #shogun | 13:29 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 13:29 | |
@iglesiasg | ups.. wrong hotkey and close the chat window heh | 13:29 |
van51 | iglesiasg: so about the radius it comes down to what is calculated in the KMeans class as radius | 13:29 |
van51 | iglesiasg: and the code mentions something about variance and not radius | 13:29 |
@iglesiasg | van51, mm I see | 13:29 |
van51 | iglesiasg: so we may have to calculate the max distance from the points and draw that | 13:29 |
@iglesiasg | van51, but hold on a second | 13:30 |
@iglesiasg | van51, I used that KMeans not so long ago for my thesis | 13:30 |
@iglesiasg | and the circunferences were fine IIRC | 13:30 |
@iglesiasg | let me double check | 13:30 |
@iglesiasg | in case I was using another method | 13:30 |
@iglesiasg | nope | 13:31 |
@iglesiasg | kMeans indeed | 13:31 |
van51 | iglesiasg: get_radius()? | 13:31 |
@iglesiasg | let me check the code | 13:31 |
@iglesiasg | yes | 13:31 |
@iglesiasg | get_radiuses | 13:31 |
@iglesiasg | van51, https://github.com/iglesias/linal/blob/master/graphical/cluster_kmeans.py | 13:31 |
@iglesiasg | van51, the figure generated in on page 4 here https://dl.dropboxusercontent.com/u/11020840/pfc/fjig_pfc.pdf | 13:32 |
van51 | iglesiasg: hmm I see | 13:33 |
van51 | iglesiasg: I will have another look then | 13:33 |
van51 | iglesiasg: could be I overlooked something | 13:33 |
van51 | iglesiasg: thx for showing me that | 13:33 |
@iglesiasg | van51, hehe no problem | 13:34 |
@iglesiasg | I make publicity of the things I do this way :P | 13:34 |
van51 | iglesiasg: hah true ;) | 13:34 |
van51 | iglesiasg: as for the secondary class, it plays no role idd, butI think the label coloring may good for demonstration purposes | 13:35 |
van51 | may be* | 13:35 |
@iglesiasg | van51, but, how are the blue points used in the clustering demo? | 13:35 |
van51 | iglesiasg: they are all treated the same way | 13:36 |
@iglesiasg | van51, are they ignored? are they considered to be of the same class as the red ones? | 13:36 |
van51 | iglesiasg: the latter | 13:37 |
@iglesiasg | van51, I see | 13:37 |
van51 | iglesiasg: from what I see in the demo code, it parses the labels but never passes them to the kmeans object | 13:38 |
@iglesiasg | van51, I see | 13:38 |
van51 | as expected thatis | 13:38 |
@iglesiasg | also | 13:38 |
@iglesiasg | van51, what about the toy data in the classification demos? | 13:39 |
@iglesiasg | van51, because the generator only creates examples of one class, right? | 13:39 |
van51 | iglesiasg: yeah | 13:39 |
van51 | iglesiasg: I was thinking of extending it to allow assigning more than 1 labels | 13:40 |
van51 | iglesiasg: but maybe it needs a new toy generator so that the classes are pretty much seperated | 13:40 |
@iglesiasg | van51, aham | 13:41 |
@iglesiasg | van51, what do you think is best to do then? | 13:41 |
@iglesiasg | van51, I don't think it is a good idea to keep the generator that just generates one-class data for classification demos | 13:41 |
van51 | iglesiasg: yeah I agree. I also prefer the second approach | 13:42 |
lisitsyn | hello guyz | 13:42 |
van51 | lisitsyn: hey | 13:43 |
van51 | iglesiasg: I'll finish with the descriptions and start after that on a new generator | 13:43 |
lisitsyn | who's got time? ;) | 13:43 |
van51 | iglesiasg: if you spot anything else let me know :) | 13:43 |
@iglesiasg | van51, ok, that sounds great | 13:43 |
van51 | lisitsyn: me if I can help | 13:43 |
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 13:44 | |
lisitsyn | van51: okay let me check what can you help us with | 13:44 |
van51 | lisitsyn: ah, if it's not urgent then I'll start something else after the demos | 13:45 |
lisitsyn | van51: argh - I don't know | 13:48 |
lisitsyn | please check https://github.com/shogun-toolbox/shogun/issues?direction=desc&milestone=3&sort=updated&state=open and see if you can resolve any of these | 13:48 |
lisitsyn | there are a lot of infrastructure issues unfortunately | 13:48 |
lisitsyn | like cmake etc | 13:48 |
van51 | lisitsyn: not sure how much I can help with cmake, but I will have a look later or tonight | 13:50 |
lisitsyn | van51: https://github.com/shogun-toolbox/shogun/issues/1548 may be this one? | 13:50 |
van51 | lisitsyn: yeah that sound better | 13:52 |
-!- iglesiasg [~iglesiasg@s83-179-44-135.cust.tele2.se] has quit [Quit: Leaving] | 14:03 | |
-!- votjakovr [~votjakovr@95-27-254-98.broadband.corbina.ru] has quit [Quit: WeeChat 0.4.0] | 14:52 | |
-!- lisitsyn [~lisitsyn@109-226-118-24.clients.tlt.100megabit.ru] has left #shogun [] | 15:04 | |
van51 | sonney2k: can you give me some info on the application/arts and application/asp demos? | 15:15 |
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun | 15:31 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 15:31 | |
-!- van51 [~van51@athedsl-399544.home.otenet.gr] has quit [Read error: Connection reset by peer] | 16:03 | |
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Read error: Operation timed out] | 16:55 | |
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun | 19:02 | |
-!- lisitsyn [~lisitsyn@fb2-lo1.global63.net] has joined #shogun | 19:05 | |
-!- lisitsyn [~lisitsyn@fb2-lo1.global63.net] has quit [Ping timeout: 240 seconds] | 19:22 | |
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: Ex-Chat] | 19:27 | |
-!- lisitsyn [~lisitsyn@fb2-lo1.global63.net] has joined #shogun | 19:32 | |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 20:31 | |
shogun-notifier- | shogun-demo: van51 :master * c679642 / / (6 files): https://github.com/shogun-toolbox/shogun-demo/commit/c679642cd522e3a179faa9f752a0113075adb725 | 20:31 |
shogun-notifier- | shogun-demo: Updated description in more examples and fixed a couple errors | 20:31 |
shogun-notifier- | shogun-demo: Soeren Sonnenburg :master * e09d850 / / (6 files): https://github.com/shogun-toolbox/shogun-demo/commit/e09d850c6100bf228b0beee31ca4cbf15018738b | 20:31 |
shogun-notifier- | shogun-demo: Merge pull request #27 from van51/feature/docs | 20:31 |
shogun-notifier- | shogun-demo: | 20:31 |
shogun-notifier- | shogun-demo: Updated description in more examples and fixed a couple errors | 20:31 |
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun | 20:59 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 20:59 | |
@iglesiasg | sonney2k, hey | 20:59 |
@sonney2k | iglesiasg, hi | 21:03 |
@iglesiasg | sonney2k, writing a mail to you about it right now... | 21:05 |
@iglesiasg | sonney2k, sent! | 21:05 |
@sonney2k | iglesiasg, well links on the info page are broken for many refs - you could correct them though with proper ones pointing to the refernces page | 21:11 |
@iglesiasg | sonney2k, ok, but it is not just the links | 21:11 |
@iglesiasg | I mean | 21:11 |
@iglesiasg | sonney2k, [1] in info page | 21:11 |
@iglesiasg | does not reference to [1] in the reference pages I think | 21:12 |
@iglesiasg | sonney2k, do you see what I mean? | 21:12 |
@iglesiasg | sonney2k, and still, even if they were correct, I'd rather see the references after the description in the info page too | 21:14 |
@iglesiasg | I don't see the point of having this References page honestly | 21:14 |
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: Ex-Chat] | 22:08 | |
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 23:31 | |
--- Log closed Mon Oct 14 00:00:40 2013 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!