IRC logs of #shogun for Sunday, 2012-08-12

« Saturday, 2012-08-11 Index Monday, 2012-08-13 »
--- Log opened Sun Aug 12 00:00:17 2012
shogun-buildbotbuild #318 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/318  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>00:01
CIA-39shogun: Sergey Lisitsyn master * r346f445 / src/interfaces/modular/modshogun_ignores.i : Added ignore for risk_directed failing in csharp modular - http://git.io/L3gF1Q00:06
CIA-39shogun: Sergey Lisitsyn master * rbb9493f / (5 files in 4 dirs): A few doc warning fixes - http://git.io/aS8jYw00:19
shogun-buildbotbuild #319 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/319  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>00:23
CIA-39shogun: Sergey Lisitsyn master * r97c9a87 / src/interfaces/modular/modshogun_ignores.i : Fixed wrong type of directed risk function method ignore - http://git.io/L76wIQ00:25
CIA-39shogun: Sergey Lisitsyn master * r52369c9 / data : Updated data submodule - http://git.io/n8JysA00:38
shogun-buildbotbuild #320 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/320  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>00:50
-!- blackburn [~blackburn@109.226.80.43] has quit [Quit: Leaving.]01:07
-!- blackburn [~blackburn@109.226.80.43] has joined #shogun01:08
shogun-buildbotbuild #321 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/32101:31
CIA-39shogun: Sergey Lisitsyn master * r6cf79ef / (106 files in 6 dirs): Moved string kernels to separate dir - http://git.io/REOuiw01:59
-!- blackburn [~blackburn@109.226.80.43] has quit [Quit: Leaving.]02:38
shogun-buildbotbuild #260 of deb2 - static_interfaces is complete: Failure [failed test octave_static]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb2%20-%20static_interfaces/builds/260  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>02:55
-!- n4nd0 [~nando@112.Red-2-137-52.dynamicIP.rima-tde.net] has joined #shogun09:30
n4nd0blackburn: I found the thesis useful but it is not completely clear for me how to work out the risk function for hmm from what they say there09:35
n4nd0blackburn: I am spending some time now on the application for hm-svm though, I am reading quite a bit about speech recognition09:36
-!- n4nd0 [~nando@112.Red-2-137-52.dynamicIP.rima-tde.net] has quit [Read error: Connection reset by peer]11:34
-!- n4nd0 [~nando@224.Red-2-137-11.dynamicIP.rima-tde.net] has joined #shogun11:38
CIA-39shogun: Viktor Gal master * r529eea9 / tests/unit/classifier/svm/SVMOcas_unittest.cc : Add unit test for SVMOcas - http://git.io/OIomQA12:14
CIA-39shogun: Viktor Gal master * rbebc04e / (3 files in 2 dirs): Add Gaussian generator in DataGenerator - http://git.io/WxwEIA12:14
CIA-39shogun: Viktor Gal master * rcff8ff6 / src/shogun/features/DataGenerator.h : Fix doxygen descrition in the header of DataGenerator - http://git.io/2sEqoQ12:14
CIA-39shogun: Heiko Strathmann master * r5206ce1 / (3 files in 2 dirs): Merge pull request #700 from vigsterkr/utest - http://git.io/FVDETQ12:15
wikingoh cool thanx heiko12:17
wikingn4nd0: here?12:23
n4nd0wiking: yeah12:23
n4nd0tell me12:23
wikingn4nd0: i'm just about to send in the CData stuff we've agreed on12:24
wikingi'll send u the link just to say whether it's ok12:24
wikingso we can use it in structured* as well12:24
n4nd0ok12:26
-!- blackburn [~blackburn@109.226.80.43] has joined #shogun12:52
shogun-buildbotbuild #261 of deb2 - static_interfaces is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb2%20-%20static_interfaces/builds/26112:59
-!- gsomix [~gsomix@109.169.140.95] has joined #shogun13:23
wikinganybody had this error before: UnicodeEncodeError: 'ascii' codec can't encode character u'\u201c' in position 426236: ordinal not in range(128)13:36
n4nd0:D13:36
n4nd0I think it looks pretty similar to an error that drove Soeren mad some time ago13:37
n4nd0it was something related to a very weird character that Heiko introduced not in purpose13:37
wikinglol13:37
wikingbtw: i cannot use latex symbols in doxygen?13:38
wikinglike \hat ?13:38
wikingshould that be rather mathml?13:38
n4nd0are you using the latex syntax for doxygen?13:39
n4nd0something like f[ or so13:39
n4nd0don't really remember by heart13:39
wikingaaah ok i see13:40
n4nd0wiking: check for example how it is done here13:43
n4nd0https://github.com/shogun-toolbox/shogun/blob/master/src/shogun/classifier/LDA.h13:43
n4nd0you have both examples there13:43
n4nd0to do equations on separate lines with \f[ .... \f]13:43
n4nd0adn on the same line with \f$...\f$13:44
wikingyep i'm just testing it :)13:44
wikingbut still i have to find out what's with that codec problem :D13:45
n4nd0mmm I am not sure if it is the same that happened that time I told you about13:45
n4nd0let me search in the logs a bit13:45
n4nd0maybe I can find something13:45
n4nd0don't find it :S13:49
wikingmmm no worries13:49
wikingi'm trying now again with some changes... let's see if that solves it13:49
wikingheheh no help :)13:50
n4nd0have you found out wich character is it?13:51
n4nd0try to make your editor show you non printable characters too, it may help13:51
wikingyeah i'm just checking the xml13:52
wiking <compound refid="dir_41918c6c0cd8700fe74209d6eeba587a" kind="dir"><name>/Users/wiking/shogun/src/shogun/lib/malsar/</name>13:52
wiking  </compound>13:52
wikingthat's the line :)))13:52
n4nd0oh13:52
n4nd0what are you using that for?13:53
wikingmeeee?13:53
wikingnothing13:53
n4nd0haha13:53
n4nd0ok13:53
wikingit's generated by something in that index.ml13:53
wikingindex.xml13:53
wikingok i'm not sure if that's true... actually13:54
wikingsince i thought it's a line13:54
wikingbut rather a position...13:54
wikingdamn how can i search a char position ? :)13:54
n4nd0eeeh13:55
n4nd0looks like a philosophical question :D13:55
CIA-39shogun: Sergey Lisitsyn master * r5598224 / (14 files in 9 dirs): A few doc fixes - http://git.io/WWb2aA14:11
-!- n4nd0 [~nando@224.Red-2-137-11.dynamicIP.rima-tde.net] has quit [Ping timeout: 268 seconds]14:17
shogun-buildbotbuild #262 of deb2 - static_interfaces is complete: Failure [failed test cmdline_static]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb2%20-%20static_interfaces/builds/262  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>14:19
CIA-39shogun: Sergey Lisitsyn master * ra2eff59 / (4 files in 4 dirs): Fixed static KRR examples (KRR -> KernelRidgeRegression) - http://git.io/2Tneig14:23
-!- gsomix_ [~gsomix@109.169.140.245] has joined #shogun15:06
-!- gsomix [~gsomix@109.169.140.95] has quit [Ping timeout: 248 seconds]15:09
wikingblackburn: here?15:20
blackburnwiking: yeah15:21
blackburnwhat's up?15:21
wikingblackburn: i'm having some character encoding problem in the generated index.xml15:21
wikingwhen creating the doxygen file... any ideas?15:21
wikingUnicodeEncodeError: 'ascii' codec can't encode character u'\u201c' in position 426236: ordinal not in range(128)15:22
wikingsomething like this15:22
blackburnwell heiko tends to add such great characters :D15:22
blackburnactually no idea how sonney2k managed to solve that twice15:22
wikingmmm was it coming from the doxygen description part?15:23
blackburnyes I think so15:23
wikingblackburn: do u know any tool to find out which char is a given position in the file?15:30
blackburnhmm not really15:31
blackburnmay be some snippet could do that15:31
shogun-buildbotbuild #263 of deb2 - static_interfaces is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb2%20-%20static_interfaces/builds/26315:38
-!- blackburn [~blackburn@109.226.80.43] has quit [Quit: Leaving.]15:39
wikinglol i've found the char15:51
wikingok i guess one should never ever copy past a paper reference from a pdf into doxygen comment... otherwise you'll end up with bad char encoding16:35
-!- n4nd0 [~nando@224.Red-2-137-11.dynamicIP.rima-tde.net] has joined #shogun17:20
-!- n4nd0 [~nando@224.Red-2-137-11.dynamicIP.rima-tde.net] has quit [Ping timeout: 268 seconds]17:37
-!- n4nd0 [~nando@224.Red-2-137-11.dynamicIP.rima-tde.net] has joined #shogun18:09
-!- n4nd0 [~nando@224.Red-2-137-11.dynamicIP.rima-tde.net] has left #shogun []18:35
-!- n4nd0 [~nando@224.Red-2-137-11.dynamicIP.rima-tde.net] has joined #shogun18:36
-!- n4nd0 [~nando@224.Red-2-137-11.dynamicIP.rima-tde.net] has left #shogun []18:36
-!- blackburn [~blackburn@109.226.80.43] has joined #shogun19:05
blackburnwiking: so where is that bad char?19:19
-!- zxtx [~zv@ool-457e7550.dyn.optonline.net] has quit [Ping timeout: 246 seconds]20:30
-!- puffin444 [62e3926e@gateway/web/freenode/ip.98.227.146.110] has joined #shogun20:39
-!- gsomix_ [~gsomix@109.169.140.245] has quit [Ping timeout: 244 seconds]21:06
blackburnhey puffin44421:12
blackburnI noticed there are a few memory leaks in gps21:12
puffin444yes21:12
puffin444The problem lies with CMap21:12
blackburnaahh21:13
puffin444I posted a reproduction of the bug on the mailing list a week or so ago21:13
blackburnpuffin444: hmm I just valgrinded21:15
blackburngaussian_processes_ard21:15
puffin444yes?21:15
blackburnthere are a few sgvector leaks21:15
puffin444Yes. CMap does something weird when its CMap<SGVector, TParameter*>21:15
blackburnokay got it21:16
puffin444I've been corresponding with gsomix about it and he hasn't figured it out yet.21:17
blackburnpuffin444: okay I will take a look21:20
puffin444okay. Sorry if I have been AWOL a bit this week.21:22
blackburnpuffin444: I am not worried about it ;)21:24
blackburnyou did quite a lot things this summer21:24
blackburnpuffin444: is it worth to use gaussian process regressor as classifier?21:30
puffin444I am not sure. There may be a few things I need to do to get it ready for classification.21:31
puffin444I know it is possible - the GPML has code to do it.21:31
blackburnI see21:31
-!- puffin444 [62e3926e@gateway/web/freenode/ip.98.227.146.110] has quit [Quit: Page closed]23:14
wikingblackburn: it was in the doxygen part that i've written...23:19
blackburnheh23:19
wikingit was some weird quotation mark coming from a pdf file23:22
wikingok i'm back to finish the doxygen for the code23:22
wikingpr is about to land \o/23:23
blackburnthat's cool23:23
CIA-39shogun: Sergey Lisitsyn master * ra805d7f / (3 files in 2 dirs): Fixed map memory handling for sgvector value case - http://git.io/AhMS5A23:38
wikingblackburn: have u seen that awesome unittest? :D23:42
blackburnwiking: no, I missed it actually23:42
blackburnwhat is so awesome in it? :)23:42
blackburnwiking: checked ocas test - so you require to linearly separate data23:44
wikingheheh that it actually works smoothly :)23:46
blackburnwiking: do you remember what was wrong with google mock under ubuntu?23:53
--- Log closed Mon Aug 13 00:00:17 2012
« Saturday, 2012-08-11 Index Monday, 2012-08-13 »

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