--- Log opened Sat Nov 05 00:00:13 2011 | ||
blackburn1 | damn I can't solve it :( | 00:42 |
---|---|---|
blackburn1 | oh.. it is a lapack bug | 01:10 |
-!- blackburn1 [~blackburn@188.122.237.76] has quit [Quit: Leaving.] | 01:21 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 06:44 | |
-!- in3xes_ [~in3xes@180.149.49.227] has joined #shogun | 06:58 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 245 seconds] | 07:01 | |
-!- in3xes_ is now known as in3xes | 07:12 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 252 seconds] | 08:20 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 08:31 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Remote host closed the connection] | 09:06 | |
-!- blackburn [~blackburn@109.226.81.30] has joined #shogun | 10:06 | |
-!- blackburn [~blackburn@109.226.81.30] has quit [Read error: Connection reset by peer] | 10:14 | |
-!- blackburn [~blackburn@188.122.224.207] has joined #shogun | 10:14 | |
-!- blackburn [~blackburn@188.122.224.207] has quit [Read error: Connection reset by peer] | 10:36 | |
-!- blackburn [~blackburn@85.114.187.132] has joined #shogun | 10:37 | |
-!- blackburn [~blackburn@85.114.187.132] has quit [Read error: Connection reset by peer] | 10:47 | |
-!- blackburn [~blackburn@188.122.228.165] has joined #shogun | 10:56 | |
-!- blackburn1 [~blackburn@31.28.36.115] has joined #shogun | 11:12 | |
-!- blackburn [~blackburn@188.122.228.165] has quit [Ping timeout: 258 seconds] | 11:14 | |
-!- blackburn [~blackburn@85.114.185.74] has joined #shogun | 11:21 | |
-!- blackburn1 [~blackburn@31.28.36.115] has quit [Ping timeout: 260 seconds] | 11:23 | |
blackburn | sonney2k: around? | 11:39 |
blackburn | sonney2k: my issue with thread unsafety was related to thread-unsafe DLAMCH routine, which determines machine eps | 11:54 |
blackburn | it is fucking awful - the whole LAPACK is thread-safe but DLAMCH | 11:55 |
-!- blackburn1 [~blackburn@109.226.124.12] has joined #shogun | 12:17 | |
-!- blackburn [~blackburn@85.114.185.74] has quit [Read error: Connection reset by peer] | 12:18 | |
CIA-37 | shogun: Sergey Lisitsyn master * r4b6f385 / (3 files in 2 dirs): Removed float lapack wrappers - http://git.io/N57b8A | 16:40 |
CIA-37 | shogun: Sergey Lisitsyn master * rdafece0 / src/shogun/converter/LocalTangentSpaceAlignment.cpp : Optimized LTSA - http://git.io/sWLt1w | 16:42 |
CIA-37 | shogun: Sergey Lisitsyn master * r5280282 / src/shogun/converter/KernelLocalTangentSpaceAlignment.cpp : Optimized KLTSA - http://git.io/6YBvFw | 16:42 |
CIA-37 | shogun: Sergey Lisitsyn master * r59c976a / examples/undocumented/python_modular/converter_localitypreservingprojections_modular.py : Added LPP python modular example - http://git.io/c4MKsw | 16:44 |
shogun-buildbot | build #312 of octave_static is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/octave_static/builds/312 | 16:48 |
@sonney2k | blackburn1, it is a bit surprising - thought that lapack is 100% thread safe... | 19:16 |
@sonney2k | a bit weird for a lib of this kind | 19:16 |
blackburn1 | sonney2k: 3.3 is so, but 3.2.1's dlamch is not | 19:16 |
@sonney2k | blackburn1, well then add a test for that version and force unicore if <3.3 | 19:17 |
blackburn1 | sonney2k: how can i test lapack version in runtime? | 19:17 |
@sonney2k | blackburn1, compile time - not runtime | 19:17 |
blackburn1 | ehh okay | 19:17 |
@sonney2k | blackburn1, but maybe the even have a api function for checking the version? | 19:18 |
blackburn1 | sonney2k: I don't know it yet | 19:18 |
@sonney2k | no idea - check the source :) | 19:18 |
@sonney2k | blackburn1, have you read Madeleines question on the mailinglist? | 19:26 |
blackburn1 | sonney2k: yes, but I'm afraid I can't help she | 19:29 |
blackburn1 | sonney2k: don't you think alessandro is a bit offensive? :D | 19:30 |
@sonney2k | blackburn1, no why? | 19:32 |
@sonney2k | he is right | 19:32 |
@sonney2k | if it doesn't work w/o jblas then it doesn't make sense to just print a warning | 19:32 |
@sonney2k | and one should fail instead | 19:32 |
@sonney2k | blackburn1, do we have MeanAbsoluteError? | 19:40 |
@sonney2k | in evaluation I mean? | 19:40 |
blackburn1 | sonney2k: mean squared error | 19:42 |
blackburn1 | sonney2k: I'm about his writing style :) | 19:42 |
@sonney2k | but that cannot do absoute right? | 19:43 |
@sonney2k | I just started writing it... | 19:43 |
blackburn1 | sonney2k: what is the difference? | 19:43 |
@sonney2k | one is ||.||^2 | 19:43 |
@sonney2k | one ||.|| | 19:43 |
@sonney2k | | | | 19:44 |
blackburn1 | sonney2k: does it have some 'fundamental' difference? | 19:44 |
@sonney2k | fundamental? | 19:44 |
@sonney2k | it will punish differences close to zero much more | 19:45 |
blackburn1 | eh nevermind, bad word | 19:45 |
blackburn1 | ah I see | 19:45 |
blackburn1 | sonney2k: you could meansquarederror -> meanerror | 19:45 |
blackburn1 | and add options to square or not | 19:45 |
blackburn1 | sonney2k: about madeleine question - I really do not understand much about her problem, it is related to things you are better with than me | 19:47 |
@sonney2k | blackburn1, ok implemented that mean absolute error thing | 19:48 |
blackburn1 | http://www.netlib.org/lapack/lapack-3.3.0.html | 19:49 |
blackburn1 | sonney2k: Thread safe xLAMCH: SLAMCH and DLAMCH were the only two routines not thread-safe in LAPACK-3.2. This is fixed: all routines in LAPACK-3.3 are now thread-safe. | 19:49 |
blackburn1 | it is really awful | 19:49 |
blackburn1 | sonney2k: could you please describe how to link shogun with my own lapack instead of atlas' one& | 19:50 |
blackburn1 | ? | 19:50 |
blackburn1 | I spent too much time on it, decided to ask you instead | 19:50 |
@sonney2k | your own lapack? | 19:51 |
blackburn1 | hah | 19:51 |
@sonney2k | just set the library / include paths on configure? | 19:51 |
blackburn1 | again bad wording | 19:51 |
blackburn1 | sonney2k: yes a did a try | 19:51 |
blackburn1 | but it links with atlas' one :( | 19:52 |
blackburn1 | I did* | 19:52 |
blackburn1 | I'll make a try again | 19:52 |
blackburn1 | sonney2k: today I've tested kernel ltsa on 84x112 pictures | 19:53 |
blackburn1 | it was 80s faster than original LTSA with the same results | 19:54 |
blackburn1 | 80x* | 19:54 |
blackburn1 | in fact I started writing a paper, what do you think, does this kind of algo extension worths to be published in any journal? | 19:55 |
blackburn1 | I'm a total newbie in science affairs hah | 19:56 |
@sonney2k | blackburn1, if the extension is not trivial and ltsa is some established algo - sure | 20:00 |
blackburn1 | sonney2k: what is the trivial extension? | 20:00 |
CIA-37 | shogun: Soeren Sonnenburg master * r857311c / (4 files in 2 dirs): introduce MeanAbsoluteError - http://git.io/H8jXVw | 20:00 |
@sonney2k | blackburn1, something obvious for experts | 20:00 |
@sonney2k | blackburn1, btw look at the copyright line in the above ^ commit | 20:01 |
blackburn1 | sonney2k: can I consider you as expert? hehe | 20:01 |
blackburn1 | why me? | 20:01 |
@sonney2k | blackburn1, no - I don't know anything about LTSA | 20:01 |
@sonney2k | blackburn1, I used your MSE code | 20:02 |
blackburn1 | sonney2k: do you know anything about SVD and kernel matrix? | 20:02 |
@sonney2k | kernel matrix? no clue what that could be | 20:02 |
blackburn1 | matrix of pairwise kernels? | 20:02 |
blackburn1 | k11 k12 k13 ... | 20:02 |
blackburn1 | sonney2k: in the ltsa algorithm there are right singular vectors of centered feature matrix [X-[mean feature vector]] are being computed | 20:04 |
blackburn1 | I proved that it is the same to find eigenvectors of centered gram matrix | 20:05 |
blackburn1 | => faster, makes possible to use kernels | 20:06 |
blackburn1 | sonney2k: it is obvious for me, but I don't know if it is for anyone else | 20:07 |
blackburn1 | sonney2k: do not leave me alone when I'm ask stupid Qs :D | 20:20 |
-!- heiko [~heiko@host86-160-249-224.range86-160.btcentralplus.com] has joined #shogun | 22:23 | |
heiko | sonney2k, just read the mails regarding the modsel problem, will try to figure our what is going on now. | 22:24 |
blackburn1 | howdy! | 22:24 |
blackburn1 | saturday night fixing model selection show | 22:25 |
heiko | hi blackburn | 22:27 |
heiko | hehe, well just read the mail | 22:27 |
heiko | and i just got a minute | 22:28 |
heiko | very busy the rest of the time | 22:28 |
heiko | everything ok in russia? :) | 22:28 |
blackburn1 | heiko: sure, we are vodka-enough | 22:29 |
heiko | hehe :) | 22:29 |
heiko | what about your studies, everything going well? | 22:29 |
blackburn1 | heiko: oh studies are going ok, though I have a little spare time because I started working at netcracker | 22:30 |
heiko | netcracker? | 22:30 |
blackburn1 | heiko: yeah, http://netcracker.com/en/ | 22:31 |
heiko | ah ok, new job | 22:32 |
heiko | cool | 22:32 |
heiko | is it nice? | 22:32 |
blackburn1 | heiko: it would be better to say first job :) | 22:32 |
blackburn1 | not very nice but 2 pros: a) I'm paid b) I work part-time :) | 22:33 |
heiko | ok :) | 22:33 |
blackburn1 | heiko: what about you? what do you do? | 22:34 |
heiko | I am started with a master in machine learning in london | 22:34 |
heiko | http://www.csml.ucl.ac.uk/courses/msc_ml/ | 22:34 |
heiko | really cool | 22:34 |
heiko | but consumes all of my time | 22:34 |
blackburn1 | heiko: so sad you can't do shoguning :) | 22:36 |
heiko | yes, indeed | 22:36 |
heiko | but thigns will change | 22:36 |
heiko | but for the moment it is like it is | 22:36 |
heiko | think i will have some free time around january | 22:36 |
heiko | and then in march | 22:36 |
heiko | and then in the summer also | 22:36 |
blackburn1 | I guess we will never have much free time as we had in summer | 22:36 |
heiko | yes i think so | 22:38 |
heiko | btw current head does not compile here | 22:38 |
blackburn1 | heiko: error? | 22:38 |
heiko | In file included from base/class_list.cpp:66:0: | 22:39 |
heiko | ./kernel/MultitaskKernelMaskPairNormalizer.h: In member function 'virtual bool shogun::CMultitaskKernelMaskPairNormalizer::init(shogun::CKernel*)': | 22:39 |
heiko | ./kernel/MultitaskKernelMaskPairNormalizer.h:76:7: error: 'string' is not a member of 'std' | 22:39 |
heiko | guess include is missing | 22:39 |
blackburn1 | I would personally kick chris' ass for usage of std but he is moving to NY | 22:39 |
heiko | ;) | 22:39 |
blackburn1 | that would be hard to catch him | 22:40 |
blackburn1 | okay I'll take a look | 22:40 |
heiko | blackburn can you help me with a little git problem? | 22:41 |
blackburn1 | heiko: | 22:41 |
blackburn1 | sure | 22:41 |
heiko | I still have some work in my master branch and i committed it already to my github repo because i wanted to backup it | 22:41 |
heiko | however, the code is not yet really working | 22:41 |
heiko | but if i want to do a bugfix i cannot send pullrequests | 22:41 |
heiko | because all my changes are in there | 22:42 |
heiko | do you know how to move this stuff to another branch or something so that i can work on head without any of my other changes involved? | 22:42 |
blackburn1 | you can stash your changes for now | 22:42 |
blackburn1 | git stash | 22:43 |
blackburn1 | it will keep your changes until you will git stash pop it | 22:43 |
heiko | does that even work if they are already in my github repo? | 22:43 |
blackburn1 | heiko: ah did you commit it? | 22:44 |
heiko | yes | 22:44 |
blackburn1 | hmm | 22:44 |
heiko | its also already pushed | 22:44 |
heiko | you know i wanted to backup itonline, kind of stupid | 22:45 |
blackburn1 | not a stupid | 22:45 |
heiko | but i was in a hurry and i did not want to loose all of my changes | 22:45 |
heiko | moving to london, plane, laptop broken or anything...you know | 22:45 |
heiko | but now I want to put them away to work on them later | 22:45 |
blackburn1 | huh that's difficult | 22:46 |
heiko | yes. | 22:46 |
heiko | i thought of perhaps to the brute force way and save my repo copy here and just reset my github account | 22:46 |
heiko | and then work the stuff in again later | 22:46 |
heiko | i have to reconsider it anyway | 22:46 |
heiko | think i will jsut do this | 22:47 |
blackburn1 | heiko: I'm sorry but I have no idea how to change your base in this way | 22:48 |
blackburn1 | I'm still lame with some git issues | 22:48 |
heiko | me too :) | 22:48 |
heiko | reforking ... | 22:49 |
blackburn1 | heiko: I would know more if I was using forks and etc | 22:49 |
blackburn1 | but I commit right into the shogun and it is pretty simplified | 22:50 |
heiko | yes, with all the forks and stuff -- thats pretty messy sometimes | 22:50 |
heiko | well i just cloned it | 22:50 |
blackburn1 | heiko: from my experience it comes easy to have some dev branch | 22:51 |
heiko | yes, true true, i normally do that | 22:52 |
blackburn1 | but even this becomes messy sometimes | 22:52 |
heiko | but not this time :) | 22:52 |
blackburn1 | oh I really need to stop procrastinate and compiling lapack | 22:53 |
blackburn1 | and start research on road sign recognition | 22:53 |
blackburn1 | :D | 22:53 |
heiko | for studies? | 22:53 |
blackburn1 | yeah, it is my bachelor's work or so | 22:54 |
blackburn1 | but if I go through it with high accuracy I would even get paid for that | 22:54 |
heiko | that sounds cool | 22:55 |
heiko | blackburn1, the file with the error is full of std:: | 22:57 |
heiko | i will just add an include or so | 22:57 |
blackburn1 | heiko: are you going to fix it? I can | 22:57 |
heiko | i can do it | 22:57 |
blackburn1 | heiko: if you have time please do, cause it is not being reproduced on my | 22:57 |
blackburn1 | machine | 22:58 |
heiko | done | 23:02 |
blackburn1 | heiko: does it work? | 23:06 |
blackburn1 | heiko: there are vector too, is it ok? | 23:06 |
heiko | yes, but they are included in Kernel.h | 23:06 |
blackburn1 | ah ok | 23:06 |
CIA-37 | shogun: Heiko Strathmann master * r54afa01 / src/shogun/kernel/MultitaskKernelMaskPairNormalizer.h : | 23:07 |
CIA-37 | shogun: temporary added include to compile again | 23:07 |
CIA-37 | shogun: (string include was recently removed out of Kernel.h) - http://git.io/qOdR6w | 23:07 |
-!- heiko [~heiko@host86-160-249-224.range86-160.btcentralplus.com] has quit [Quit: Leaving.] | 23:40 | |
-!- heiko [~heiko@host86-160-249-224.range86-160.btcentralplus.com] has joined #shogun | 23:51 | |
heiko | blackburn1 goodn8 | 23:51 |
-!- heiko [~heiko@host86-160-249-224.range86-160.btcentralplus.com] has left #shogun [] | 23:51 | |
blackburn1 | heiko see you | 23:52 |
--- Log closed Sun Nov 06 00:00:19 2011 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!