IRC logs of #shogun for Monday, 2012-01-09

« Sunday, 2012-01-08 Index Tuesday, 2012-01-10 »
--- Log opened Mon Jan 09 00:00:19 2012
-!- blackburn [~qdrgsm@109.226.69.156] has quit [Read error: Connection reset by peer]01:48
-!- CIA-1 [~CIA@cia.atheme.org] has quit [Read error: Operation timed out]02:54
-!- CIA-29 [~CIA@cia.atheme.org] has joined #shogun03:20
-!- Netsplit *.net <-> *.split quits: @sonney2k, sonne|work07:46
-!- Netsplit over, joins: sonne|work, @sonney2k07:49
-!- sonne|work [~sonnenbu@194.78.35.195] has left #shogun []09:26
-!- sonne|work [~sonnenbu@194.78.35.195] has joined #shogun10:47
-!- sonne|work [~sonnenbu@194.78.35.195] has quit [Read error: Connection reset by peer]11:07
-!- sonne|work1 [~sonnenbu@194.78.35.195] has joined #shogun11:07
-!- policecar [~policecar@141.23.70.195] has joined #shogun11:21
-!- blackburn [~qdrgsm@109.226.69.156] has joined #shogun12:08
CIA-29shogun: Sergey Lisitsyn master * r1f52747 / examples/undocumented/libshogun/mathematics_arpack.cpp : Fixed arpack example - http://git.io/x7uXPA12:10
shogun-buildbotbuild #446 of cmdline_static is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cmdline_static/builds/44612:20
shogun-buildbotbuild #426 of r_static is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/r_static/builds/42612:25
shogun-buildbotbuild #430 of octave_static is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/octave_static/builds/43012:30
-!- policecar [~policecar@141.23.70.195] has quit [Ping timeout: 240 seconds]13:03
-!- policecar [~policecar@vpn-0199.dialin.tu-berlin.de] has joined #shogun13:07
policecarhey hey, it seems like the Gaussian Naive Bayes classifier segfaults when used with SparseRealFeatures #fyi13:28
sonne|work1policecar: mind creating a github issue for that?13:44
sonne|work1thanks13:44
-!- sonne|work1 [~sonnenbu@194.78.35.195] has left #shogun []13:44
-!- sonne|work [~sonnenbu@194.78.35.195] has joined #shogun13:51
policecark, done.13:51
-!- policecar [~policecar@vpn-0199.dialin.tu-berlin.de] has left #shogun []15:12
-!- heiko [~heiko@host86-148-55-14.range86-148.btcentralplus.com] has joined #shogun15:53
blackburnsonne|work: around?16:59
-!- heiko [~heiko@host86-148-55-14.range86-148.btcentralplus.com] has quit [Ping timeout: 240 seconds]17:21
-!- axitkhurana [b49531e7@gateway/web/freenode/ip.180.149.49.231] has joined #shogun18:28
-!- in3xes [~in3xes@49.249.161.125] has joined #shogun18:35
CIA-29shogun: Sergey Lisitsyn master * r681147a / (2 files): Fixes gaussian naive bayes classifier memory handling, related to #351 - http://git.io/DK0Xqw19:37
CIA-29shogun: Sergey Lisitsyn master * r1309149 / (2 files): Added libshogun test for the GNB classifier - http://git.io/RbYY5A19:38
@sonney2kblackburn, around now20:24
@sonney2kwhat's up?20:24
blackburnnot actual I hope20:24
blackburnI really forgot what is get_computed_dot_feature_vector20:24
blackburnor so20:24
blackburnsonney2k: btw different suggestion20:25
blackburnabout committing20:25
@sonney2kget_computed_dot_feature_vector gives you a vector that is computed / you have to free it afterwards20:26
blackburnI think we could use issues fully: then implementing new feature or changing something one could create new issue20:26
blackburnthen commit with mentioning it20:26
blackburnwith just writing #37820:27
blackburnor any other number :)20:27
blackburnthen we wouldn't have to work with different branches, etc20:27
@sonney2kwe used to do that already with trac... but when you do lots of things it is easy to forget that20:27
blackburnand no need to wait20:27
@sonney2kno need to wait?20:28
blackburnyes, commit is already here20:28
blackburnbut there would be some thread for discussion20:28
@sonney2kdoes it change anything except that we put fixed in #231 the issue / commit20:28
blackburnnothing, but gives us ability to group commits by issues20:29
-!- axitkhurana [b49531e7@gateway/web/freenode/ip.180.149.49.231] has quit [Quit: Page closed]20:29
blackburnand discuss it20:29
blackburnbtw scikits-learn do it ~ this way20:29
blackburnwell branches are good, but I am afraid are slow20:30
blackburnto work with20:30
@sonney2kI don't favour branches20:32
@sonney2kI favor pull requests for everyone :)20:33
blackburnsonney2k: I meant that20:34
blackburnsonney2k: heh Gunnar moved too?20:35
@sonney2kanyways, using issues is a good idea - it is only difficult to keep track of them when there are too many20:35
blackburnI thought only Chris did20:35
blackburnsonney2k: but I do not know how to combine issues and pull requests20:37
-!- in3xes [~in3xes@49.249.161.125] has quit [Ping timeout: 276 seconds]21:15
blackburnsonney2k: 'the Locally Linear Embedding' vs 'Locally Linear Embedding'? what is right?21:26
@sonney2kw/o the21:27
blackburnnot right w/ the?21:27
blackburnthanks21:27
blackburnsonney2k: and 'the ARPACK' vs 'APPACK'?21:27
blackburnnever thought I'll have doubts about it21:28
@sonney2ksame21:28
@sonney2kw/o the21:28
blackburnthanks21:28
blackburnsonney2k: oh I had another question. how did you get citing for Plearn in your 2007 paper?21:30
blackburnI tend to cite url but..21:30
@sonney2kI don't understand the question21:30
blackburnsonney2k: they do not provide any cite information21:30
blackburnI need to cite it somehow I guess21:30
@sonney2kyou need to create some citation for that21:31
@sonney2kjmlr doesn't like urls21:31
blackburnsonney2k: url citation?21:31
@sonney2kargh! http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/99/steps/configure/logs/stdio21:33
blackburnargh what?21:33
@sonney2kit didn't fail upon configure21:34
@sonney2kI mean it needs swig 2.0.4 nowadays21:34
blackburnah21:34
@sonney2kbut I hopefully fixed the new buildbot21:38
@sonney2kit was always failing (didn't detect numpy though it is installed)21:38
blackburnsonney2k: yeah seen it21:38
blackburnsonney2k: is web-site citing @article?21:39
@sonney2khow can you?21:39
blackburnsonney2k: well I have seen it fails with numpy not detected :)21:39
@sonney2kahh ok21:39
@sonney2know I hope you observe that it works too :)21:40
blackburnI observe wrong things21:40
blackburne.g. missed arpack failure21:40
blackburnsonney2k: don't you know if PLearn died already?21:48
@sonney2khmmhh, when was the last update?21:49
blackburnlast docs was updated 200621:49
blackburnhmm no trunk was updated 2011-0921:49
blackburnbut I guess it slowed down totally21:49
blackburnsonney2k: the problem is I can't select authors21:50
blackburnworst doc ever21:50
blackburnhmm there is a hotel sonnenburg in osterreich21:57
blackburn:D21:57
CIA-29shogun: Soeren Sonnenburg master * rc0d0193 / src/configure :22:05
CIA-29shogun: When swig >= 2.0.4 is not detected - disable modular interfaces22:05
CIA-29shogun: To this end22:05
CIA-29shogun: - return 1 in assert_version22:05
CIA-29shogun: - don't set $_interface_modular to yes but keep auto state - http://git.io/Tkmwug22:05
@sonney2kblackburn, there is even a city named sonnenburg in germany22:12
blackburnheh22:12
@sonney2kto bad the Nazis abused it22:12
blackburnwell nazis did strange things :)22:13
@sonney2kwelcome to concentration camp sonnenburg ...22:13
blackburnreally?22:14
blackburnwhat a 'gift' to you22:14
@sonney2khttp://en.wikipedia.org/wiki/Sonnenburg22:14
blackburnah poland22:15
blackburnsonney2k: were in germany any camps?22:15
blackburninternally I mean22:15
@sonney2kof course22:16
@sonney2kok swig 2.0.4 on the buildbot22:16
blackburnI wonder cause it looks like most of them were in poland22:19
blackburnsonney2k: which buildbot is running win 7?22:19
@sonney2knone - I didn't find the time to install all dependencies22:20
blackburnI see22:22
blackburnsonney2k: http://dl.dropbox.com/u/10139213/shogun/lisitsyn12a.pdf could you please take a look on url citations on last page?22:23
@sonney2klooks good to me22:24
blackburnok thanks22:24
shogun-buildbotbuild #421 of python_static is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/python_static/builds/42122:24
@sonney2khaha :-) http://shogun-toolbox.org/buildbot/waterfall22:25
@sonney2kgreen all green22:25
blackburnsonney2k: sorry if I bother you with my paper efforts :)22:25
blackburnnightly_default?22:25
blackburnsonney2k: should address under name be fully available for me?22:26
@sonney2kthat was still when my configure did fail22:26
blackburnI see22:26
@sonney2k(fixed above)22:26
@sonney2kblackburn, I wish I was still at univiersity + your supervisor :)22:27
blackburngsoc would be announced in two weeks22:27
blackburnsonney2k: why?22:27
@sonney2kgsoc already?22:27
@sonney2kwas it last year?22:28
blackburnsonney2k: last year it was announced on jan 2422:28
@sonney2kI could push you more :)22:28
@sonney2kuh!22:28
@sonney2kthen we have to hurry up producing ideas and announce things on the list!22:28
blackburnsonney2k: well start for organizations was on 28 feb22:29
blackburnJanuary 24:22:29
blackburnProgram announced.22:29
blackburnFebruary 28:19:00 UTC22:29
blackburnMentoring organizations can begin submitting applications to Google.22:29
@sonney2kdoesn't matter - it is important to get people interested22:29
@sonney2kand maybe we get some feedback22:29
@sonney2kwe should try hard to update the ideas page until end of this week22:30
@sonney2kand then ask for feedback22:30
blackburnupdate with?22:30
@sonney2knew ideas22:30
@sonney2kblackburn, regarding your abstract:22:30
@sonney2kcriterion an implemented -> criterion and implemented22:31
@sonney2kit22:31
blackburnthanks22:31
blackburnit?22:31
@sonney2kcriterion an implemented -> criterion and implemented it as part of the SHOGUN ...22:31
blackburnah22:31
blackburnnot the, right?22:32
blackburnno 'the' I mean22:32
@sonney2kthe in this case22:32
blackburnsonney2k: damn I do not understand then :(22:32
blackburnprovided by the SHOGUN?22:33
@sonney2kAll methods were implemented using the state-of-the-art algorithms -> All methods were implemented using the respective state-of-the-art algorithms.22:33
@sonney2kyes22:33
@sonney2kAll methods were implemented using the state-of-the-art algorithms -> All methods were implemented using the respective state-of-the-art algorithms and were parallelized where possible.22:34
@sonney2kperformance and distributed under GPLv3 license. -> performance and distributed under the GPLv3 license.22:34
blackburnsonney2k: where is 'efficient method formulations' part?22:34
@sonney2kblackburn, isn't chris working with you on this?22:35
blackburnsonney2k: ehm yes he does22:35
@sonney2kHe should do the final pass checking for language problems22:35
blackburnwell I rewrote much things22:35
blackburnin last days22:35
blackburnsonney2k: I'll try to do my best with language :(22:35
blackburnI try22:35
@sonney2kdon't worry - the paper won't be rejected because of that22:36
@sonney2k(if you let chris do one final pass)22:36
blackburnsonney2k: there are a lot of chris' sentences22:36
blackburnabstract was written by him, but reworked by me22:36
blackburnsonney2k: what about address?22:37
@sonney2kit should be an address you can be reached under22:37
blackburnsonney2k: I'm afraid I can't be reached with my institution address, what should I do?22:38
blackburnno idea how students can be reached with their instituion..22:39
@sonney2kkeep it22:39
@sonney2kin the end no one will ever write you to your postal address22:39
@sonney2kjust keep the email addr current22:40
blackburnok22:40
blackburnI registered better email :D22:40
blackburnblackburn91@gmail.com is not too serious haha22:40
blackburnsonney2k: ok thanks22:40
@sonney2kI hope you can help with the gsoc ideas list22:42
@sonney2khttp://www.shogun-toolbox.org/gsoc-ideas.html22:42
@sonney2kthis one is what we have to update22:43
blackburnsonney2k: sure, but can we do it not this week?22:43
@sonney2kwhen?22:43
blackburnsonney2k: I have last exam on 23..22:44
blackburnweek would be enough22:44
blackburnso to the end of january22:44
blackburnis it ok for you?22:44
@sonney2kI will talk to my ex-colleagues at TU if they want to help / work on this or not22:44
blackburnsonney2k: I could virtually take 2-3 students with not very complicated tasks22:45
@sonney2kI can only do 122:46
blackburnsonney2k: then take 2-3 and I will be helping you22:46
@sonney2kI will probably be on paternal leave again and have a few weeks of holidays this time22:46
@sonney2kwell first we have to see if someone is interested in co-mentoring22:47
blackburnsonney2k: I will be available almost fully in july and august22:47
@sonney2kthen we have to apply again22:47
blackburnfully free22:47
@sonney2kand then we will see if we will be accepted22:47
blackburnyeah sure22:48
@sonney2kand then we can discuss how many students we can really effort to take22:48
blackburnsonney2k: I think I should show carol my paper once it is published hah22:48
blackburnnice adv for shogun as gsoc org22:48
@sonney2kI really hope you manage to get this published.22:48
@sonney2kand no, it doesn't help that I initiated that jmlr-mloss track22:49
@sonney2kthe barrier is defined by the reviewers and action editor you get22:50
blackburnwell yes, I meant you know much what is good for this track22:50
blackburnI did not mean you can help22:50
@sonney2kit is a moving target... and I haven't been active for a while now22:50
blackburnsonney2k: will you be active with researches in next years?22:51
@sonney2kunclear22:51
blackburndo you do any research at job?22:51
@sonney2kno public research at least22:51
blackburnI see22:52
blackburnsonney2k: I'm not the worst russian english writer : http://www.waset.org/journals/waset/v13/v13-13.pdf23:00
blackburnI bet you won't understand anything hah23:01
-!- kate__ [59d20f7c@gateway/web/freenode/ip.89.210.15.124] has joined #shogun23:01
-!- kate__ [59d20f7c@gateway/web/freenode/ip.89.210.15.124] has quit [Client Quit]23:03
-!- kate__ [59d20f7c@gateway/web/freenode/ip.89.210.15.124] has joined #shogun23:03
kate__anybody here?23:03
kate__blackburn?23:04
blackburnkate__: yes23:04
kate__hello23:04
kate__remember me?23:04
blackburnkate__: haha sure23:04
kate__had a swig problem23:04
kate__ok23:04
kate__lets say i have linux23:04
kate__whould it be easier to do an example?23:05
blackburnkate__: would be easier to install shogun23:05
@sonney2kkate__, btw I managed to get shogun to work under cygwin - but I had to manually install swig 2.0.4 for that... (as in compile from source and install dependencies)23:05
shogun-buildbotbuild #266 of python_modular is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/python_modular/builds/26623:06
@sonney2kbut linux is the easier option :)23:06
@sonney2kall hail the autobuild :)23:06
kate__@sonney2k i tried to install manually swig 2.0.1  but i couldn't23:06
blackburn2.0.1 will not fit23:07
blackburnI thought so but not23:07
kate__ok but 2.0.4 didn't have an .exe file and no instrunctions23:07
@sonney2kkate__, download the .tar.gz. from source23:07
kate__sorry sonney2k but i m gonna need some more info on that23:08
kate__the what now?23:08
blackburnkate__: http://www.swig.org/download.html23:09
blackburndownload swigwin-2.0.423:09
@sonney2kno - don't download swigwin23:09
blackburn??23:09
blackburn:D23:09
@sonney2kjust the normal swig-2.0.4.tar.gz23:09
blackburnok download normal one23:09
@sonney2kIIRC one has to install pcre or so in cygwin23:10
@sonney2kthen compile/install swig23:10
@sonney2kand then shogun again23:10
@sonney2kand then finally make install DESTDIR=23:10
@sonney2kanyways - I have to sleep now23:10
@sonney2kcu23:10
blackburnleft me alone :(23:11
blackburnkate__: okay lets start with downloading23:11
blackburnkate__: I am afraid I will have to go sleep in 30-40 mins too23:11
kate__ok23:11
kate__lets do this quick23:11
kate__write me the steps23:11
kate__will you be on tomorrow morning if something goes wrong?23:12
blackburnkate__: just download it now, I'm not going to sleep just now :)23:13
kate__ok23:13
kate__question23:13
kate__i download swig 2.0.4 right?23:13
blackburnkate__: yes, not swigwin23:14
kate__ok23:14
kate__done23:14
blackburnkate__: then install pcre package and remove swig in your cygwin setup23:16
kate__ok23:17
kate__done23:17
blackburnkate__: unpack swig tar to some dir23:18
blackburnthen ./configure23:18
kate__no directory found23:19
blackburn./configure in unpacked swig dir23:19
kate__aahhhh23:19
blackburnjust like it is done with shogun23:20
blackburnbrb23:21
-!- blackburn [~qdrgsm@109.226.69.156] has quit [Remote host closed the connection]23:21
-!- blackburn [~qdrgsm@109.226.69.156] has joined #shogun23:23
kate__ok,i m doing something wrong with the command, sorry but i m tired from the trip last night and my brain doesn't connect properly, lets say that the swig file is at the desktop under the name swig what do i do?23:24
blackburnkate__: no, there will be no file23:25
blackburnok I'll describe what you should do23:25
kate__ok23:25
blackburnyou have to run ./configure in unpacked swig dir23:25
blackburnthen once it is done right23:25
blackburn(if something goes wrong it will say it lacks some package - just install it)23:25
blackburndo make and make install23:26
blackburnbut I am afraid you have to specify install dir23:26
kate__ok23:26
kate__yes i did23:26
kate__its running23:26
kate__for some reason it doesnt read pcre23:31
kate__checking for pcre-config... no configure: error:         Can't find pcre-config script from PCRE (Perl Compatible Regular Express              ions)         library package. This dependency is needed for configure to complete,         Either:23:32
kate__i installed it...i really did23:32
blackburnI guess you have to install everything related to pcre23:32
blackburnpcre dev or so23:32
kate__it has some libs23:33
blackburnlibpcre may be I don't remember exactly23:33
blackburnjust install everything found with 'pcre'23:33
-!- kate__ [59d20f7c@gateway/web/freenode/ip.89.210.15.124] has quit [Quit: Page closed]23:34
-!- kate__ [59d20f7c@gateway/web/freenode/ip.89.210.15.124] has joined #shogun23:34
kate__ok i believe its going well23:37
kate__completed succefull23:39
blackburnkate__: configure, right?23:40
kate__yes23:40
kate__./configure --interfaces=python_modular23:41
kate__i do that now?23:41
blackburnkate__: no you have to make swig23:41
kate__ok23:44
kate__done23:44
blackburnmake install23:45
kate__ok23:46
blackburndone?23:46
kate__almost i believe23:46
kate__but say the rest23:46
blackburnonce it is done cd to shogun and make it23:48
kate__so you can go to sleep23:48
blackburnwell I'm interested if you manage to install it :)23:48
-!- kate___ [59d20f7c@gateway/web/freenode/ip.89.210.15.124] has joined #shogun23:51
kate___sorry but it keeps logging of23:52
-!- kate__ [59d20f7c@gateway/web/freenode/ip.89.210.15.124] has quit [Ping timeout: 258 seconds]23:52
-!- kate___ [59d20f7c@gateway/web/freenode/ip.89.210.15.124] has quit [Client Quit]23:53
-!- kate__ [59d20f7c@gateway/web/freenode/ip.89.210.15.124] has joined #shogun23:53
kate__:D23:53
kate__complete instalation23:57
kate__with no problem so ever!!!!23:57
blackburninstallation of?23:58
kate__swig23:58
kate__:D23:58
blackburnnow make shogun23:58
kate__i did that last time didn't i?23:59
blackburnI don't think you did it w/o swig23:59
kate__ok23:59
kate__how do i make it?23:59
--- Log closed Tue Jan 10 00:00:15 2012
« Sunday, 2012-01-08 Index Tuesday, 2012-01-10 »

Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!