Open in new window / Try shogun cloud
--- Log opened Mon Mar 11 00:00:10 2013
heikoblackburn: I think I will add a data generator for this one00:02
heikothe toy multiclass data problem seems to be weird, but I dont know what it is00:02
heikoblackburn: I will update the example so that you can see whats going on in there00:02
blackburnheiko: okay00:02
shogun-notifier-shogun: Heiko Strathmann :master * 5131bb5 / examples/undocumented/libshogun/classifier_larank.cpp: https://github.com/shogun-toolbox/shogun/commit/5131bb50ee2e46d9965f21d94a68c4e6557c3f2c00:03
shogun-notifier-shogun: made example way better (random input data means non-sense output) and prepared for unit-test00:03
shogun-notifier-shogun: Heiko Strathmann :master * 4b7ded4 / tests/unit/multiclass/LaRank_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/4b7ded480f37fdefe091557d6f264f25757f5c6700:03
shogun-notifier-shogun: added unit-test for LaRank which asserts basic functionality of classifier00:03
shogun-notifier-shogun: Heiko Strathmann :master * d3f1eee / examples/undocumented/python_modular/classifier_larank_modular.py: https://github.com/shogun-toolbox/shogun/commit/d3f1eeefc1d4ac709b8f4879654a77eb3a08f78600:03
shogun-notifier-shogun: used a more sensible data set to actually see whether the method works00:03
shogun-notifier-shogun: Heiko Strathmann :master * 4c9352a / examples/undocumented/python_modular/classifier_larank_modular.py: https://github.com/shogun-toolbox/shogun/commit/4c9352a018c5c5adfc91c11b02110f5d5ae6b58300:03
shogun-notifier-shogun: empty line to make things look nicer00:03
shogun-notifier-shogun: Heiko Strathmann :master * 11d764e / examples/undocumented/python_modular/classifier_multiclassocas_modular.py: https://github.com/shogun-toolbox/shogun/commit/11d764e59461c41a833bb18ea156614519c267e400:03
shogun-notifier-shogun: new easy toy data data for example00:03
shogun-notifier-shogun: Heiko Strathmann :master * 1e89f0c / / (4 files): https://github.com/shogun-toolbox/shogun/commit/1e89f0c64f6ccc62ed242c2d7d25b190efcc2d2700:03
shogun-notifier-shogun: Merge pull request #915 from karlnapf/master00:03
shogun-notifier-shogun:00:03
shogun-notifier-shogun: example updates in order to make test-suite work00:03
heikoblackburn: ok thats it from me today, good night!00:06
-!- heiko [~heiko@b019b9ef.bb.sky.com] has left #shogun []00:06
blackburnheiko: good night00:06
shogun-buildbotbuild #633 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/633  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>00:09
shogun-buildbotbuild #634 of cyg1 - libshogun is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/63400:24
shogun-buildbotbuild #274 of nightly_none is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_none/builds/27403:01
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]03:03
shogun-buildbotbuild #318 of nightly_default is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/31803:38
-!- blackburn [~blackburn@92.240.132.81] has quit [Quit: Leaving.]03:42
-!- ashar799 [4e61ecae@gateway/web/freenode/ip.78.97.236.174] has joined #shogun08:20
-!- ashar799 [4e61ecae@gateway/web/freenode/ip.78.97.236.174] has quit [Client Quit]08:25
-!- blackburn [~lisitsin@mxs.kg.ru] has joined #shogun08:33
-!- sonne|work [~sonnenbu@sams-office-nat.tomtomgroup.com] has joined #shogun09:01
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun10:33
-!- blackburn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.]10:40
-!- blackburn [~lisitsin@mxs.kg.ru] has joined #shogun10:52
-!- aamir is now known as syst3mw0rm13:36
-!- Netsplit *.net <-> *.split quits: wiking15:31
-!- wiking [~wiking@info2k1.hu] has joined #shogun15:31
-!- FSCV [~FSCV@206.225.135.24] has joined #shogun16:04
-!- blackburn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.]16:29
-!- FSCV [~FSCV@206.225.135.24] has quit [Ping timeout: 256 seconds]16:53
-!- FSCV [~FSCV@187.210.54.166] has joined #shogun17:04
-!- heiko1 [~heiko@nat-162-236.internal.eduroam.ucl.ac.uk] has joined #shogun17:17
-!- FSCV [~FSCV@187.210.54.166] has quit [Ping timeout: 245 seconds]17:29
-!- FSCV [~FSCV@206.225.135.24] has joined #shogun17:37
-!- blackburn [~blackburn@83.234.54.98] has joined #shogun18:07
blackburnheiko1: what's up with mds?18:09
blackburnheiko1: there is a unit-test alrady18:09
heiko1blackburn:  mds?18:09
heiko1I am currently trying to fix the GPs18:09
heiko1simplest cases dont work18:09
blackburnheiko1: multidimensional_scaling18:09
heiko1crash/return nonsense18:09
heiko1blackburn: just check whether it does the correct thing18:10
heiko1I cannot do that18:10
heiko1and integration test fails18:10
blackburnheiko1: all converter examples would fail with integration tests though18:10
heiko1blackburn: yeah, can you do something against this? :)18:10
blackburnheiko1: no idea..18:11
heiko1I will try to fix the gps since this is something that should work18:12
heiko1at least the part without model selection18:12
heiko1blackburn: sonney2k, sonne|work next gsoc, we should really force people to use valgrind. Some of the gsoc examples from last year massively leak, depend on uninitialised memory etc. This is a hell to fix18:29
heiko1maybe we could make make tests mandatory with valgrind18:30
heiko1and also always require a proper examples AND a proper test18:30
heiko1so example illustrates and ensures memory stuff is clean18:30
heiko1unit test makes sure that all modules work18:30
heiko1integration test ensures that the method works on toy data18:31
blackburnheiko1: gtest has some stuff to replace new and count memory leaks18:31
heiko1blackburn: we have that already with trace-mallocs18:31
heiko1and valgrind18:31
heiko1just have to use it18:31
heiko1or more specific:make other people use it18:31
blackburnheiko1: I just mean we could automate that with gtest too18:31
heiko1the gp stuff is a mess, I get a few hundred memory errors when doing it18:31
blackburnbe back in 20 mins18:32
heiko1blackburn:  we could, but why use a new system when the old one works18:32
heiko1I wont be here in 20, see you later then!18:32
blackburnheiko1: yes makes sense18:32
blackburnheiko1: I'll check mds and ocas18:32
heiko1blackburn:  lol now I see why gp return nonsense: it is uninitialised memory what they return :)18:33
heiko1or depends on it18:34
-!- heiko1 [~heiko@nat-162-236.internal.eduroam.ucl.ac.uk] has quit [Quit: Leaving.]18:37
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving]18:49
@sonney2kblackburn, https://www.youtube.com/watch?feature=player_embedded&v=NicfLmpYuy820:27
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun21:51
@sonney2kn4nd0, could you put your menu stuff somewhere (even emailing is enough...)22:14
@sonney2kthanks!22:14
n4nd0sonney2k: hey!22:22
n4nd0I can make a PR if you prefer??22:22
n4nd0nah mail is probably better22:25
@sonney2kn4nd0, i don't care how you do it :)22:30
n4nd0sonney2k: ok, I will send it to you by mail soon22:31
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 245 seconds]22:56
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun22:57
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 245 seconds]23:04
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun23:06
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving]23:56
--- Log closed Tue Mar 12 00:00:11 2013