Open in new window / Try shogun cloud
--- Log opened Mon Jul 16 00:00:17 2012
-!- heiko1 [~heiko@host86-181-80-163.range86-181.btcentralplus.com] has joined #shogun00:10
-!- heiko1 [~heiko@host86-181-80-163.range86-181.btcentralplus.com] has left #shogun []00:11
-!- blackburn [~blackburn@109.226.92.17] has quit [Quit: Leaving.]01:53
-!- audy- is now known as audy02:10
-!- zxtx [~zv@ool-43559feb.dyn.optonline.net] has quit [Ping timeout: 246 seconds]06:29
gsomixgood morning06:41
gsomixsonney2k, inplace arithmetic operators works, I will send PR at evening.06:42
-!- gsomix [~gsomix@188.168.5.232] has quit [Ping timeout: 265 seconds]07:33
-!- gsomix [~gsomix@r206-10.smr.ru] has joined #shogun08:39
-!- uricamic [~uricamic@2001:718:2:1634:f833:c09a:45c8:bbd3] has joined #shogun08:50
-!- uricamic [~uricamic@2001:718:2:1634:f833:c09a:45c8:bbd3] has quit [Ping timeout: 245 seconds]09:29
-!- uricamic [~uricamic@cmpgw-27.felk.cvut.cz] has joined #shogun09:30
-!- yoo [2eda6d52@gateway/web/freenode/ip.46.218.109.82] has joined #shogun09:54
yoohi09:55
-!- heiko [~heiko@host86-181-80-163.range86-181.btcentralplus.com] has joined #shogun10:29
@sonney2kgsomix, nice!11:17
-!- blackburn [~blackburn@109.226.92.17] has joined #shogun11:17
gsomixaha, workin on surface triangulation now11:19
@sonney2kgsomix, booring11:28
@sonney2kgsomix, looking forward to your PR this evening11:28
@sonney2kheiko, can you have a look at puffins PR?11:28
@sonney2kheiko, ahh and one question - do we have serialization support for SGSparseMatrix ?11:29
heikosonney2k, yes I will soon11:29
@sonney2kheiko, thanks11:29
heikosonney2k, I think no11:29
@sonney2kheiko, and for SGStringList?11:29
heikosonney2, I also just talked to this guy who found that x-val for multiclass machines doesnt work11:29
heikosonney2k, not that I know of, I never touched it11:29
@sonney2kok - so it is probably still the same code that we had with double* stuff right?11:30
heikosorry, didnt get that11:30
heikowhat do you mean now?11:30
@sonney2kheiko, before we had SGVector / SGMatrix11:31
@sonney2kwe had double* int11:31
heikoah I see11:31
heikoyes11:31
heikosame code11:31
@sonney2kk11:31
heikoactually, these vectors only register the array11:31
heikothe rest (the SGVector structure) stays with the class11:31
heikoso same thing as before SGVector for serialization11:32
@sonney2kso I guess the fix would be to do the same thing for SGSparseMatrix11:32
@sonney2kheiko, ohh one thing - what happens to the reference count of SGVector?11:32
heikoso far, nothing, its on my todo list :)11:32
heikoshould just be svaed11:33
heikosaved11:33
@sonney2kthat sounds like a crasher11:33
@sonney2kyeah11:33
heikoI think that should do it11:33
@sonney2kk11:33
@sonney2kheiko, I am asking because I am working on converting sparsefeatures -> SGSparseMatrix11:33
heikoI see11:34
heikomaybe this can be serialized11:34
heikoI just never used11:34
heikowas it possible before?11:34
@sonney2kI *think* so :D11:34
heikohehe :D11:35
@sonney2kheiko, ok thanks sorry for the interruption11:36
heikosonney2k, no worries11:36
heikosonney2k, btw as I already said11:36
heikothere is this guy who found out that x-val doesnt work for anything but libsvm11:37
heikois anybody aware of that?11:37
heikoI wasnt11:37
@sonney2kI never tried...11:40
@sonney2kheiko, ohh btw did you ask wrt. shogun toolbox foundation stuff?11:40
@sonney2kotherwise we might loose quite a bit of money.11:41
heikooh, yes, I forgot that11:41
heikoI talked to a guy who does that for a plane-friend-organization11:41
heikoHe told me the following:11:41
heikoits possible, however11:41
heikoits much work (we would have to do reports a few times a year, 4 or so)11:42
heikothe leaders are repsonsible for finanzial stuff and also liable in case of a crash or so11:42
heikothey had this example where they bought some ground to do stuff and the leaders took the wrong decision and then a few hundred thousand euros were gone11:43
heikobut thats probably not the case for us as we dont deal with any "risk"11:43
@sonney2kno money no risk :D11:43
heikoexcept for ML risks ;)11:43
@sonney2kheh11:43
heikoyeah, I mean its quite some formal organizational stuff that we would have to do. Everybody would have to get together to vote about the Satzung etc.11:44
heikoand the guy told me that usually these Amt people are not well aware of internet based stuff11:44
heikobut that might have changed11:45
@sonney2kheiko, if you can easily find out what we have to do then please do otherwise it is what it is11:45
@sonney2kheiko, well I would just do it via internet11:45
@sonney2kI don't see any other possible way11:45
heikoyeah, that was more of a general discussion about it with that friend11:45
heikoyes Internet is the way.11:46
heikoI will try to call somebody this week and ask more questions, kind of forgot about that in the last weeks11:46
heikowhen is your tax thing due?11:46
@sonney2kheiko, btw blackburn is digging into the modsel stuff so he might know more what works ok what not :)11:46
heikosonney2k, good to hear11:47
@sonney2kheiko, I don't know - I am usually slow with this stuff11:47
@sonney2kheiko, and I actually would like gsomix to start with the new modsel typemaps rather soon (next week latest) so we should find out too :)11:48
shogun-buildbotbuild #151 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/15111:48
heikoyes, I will investigate this multiclass problem later today11:48
blackburnmulticlass is not modsel issue11:48
blackburnit is just multiple train call issue11:49
@sonney2kheiko, you heard the master :D11:49
heikoblackburn, yes, did you already check that out?11:49
heikosonney2k, I am feeling spied now :D11:49
blackburnwell what I know larank fails to work with modsel11:49
blackburnnot because of modsel11:49
heikoblackburn, GMNPSVM fails with cross-validation evaluation already11:50
blackburnI am pretty sure the same reason11:50
heikook11:50
heikowhat exactly is the reason? multiple training? whats that? shouldnt be bad11:51
blackburnheiko: train() train() yeah11:53
heikowhy is that bad?11:53
heikoshould be slow but not give false results?11:53
blackburnlarank doesn't clean up some stuff11:54
heikoa ok11:54
blackburnso kaboooms11:54
blackburnheiko: did you like my modseloutput stuff?11:54
heikoblackburn, where?11:54
heikothe debug messages?11:54
blackburnnot really - I added a class11:54
heikohavent seen yet11:55
heikowhat does it do?11:55
blackburnmodelselection/ModelSelectionOutput.h11:55
blackburnwell just outputs test/train indices/results, evaluation result11:55
heikoah cool11:55
heikoyes I will probably like it11:55
blackburnah and for linear machine it outputs W11:55
blackburnnext step would be to set up different output ways11:56
blackburnnow it just prints11:56
heikocool11:56
heikogood idea11:56
blackburnheiko: have you seen some eigen3 code is in?12:00
heikoI noticed it12:00
heikowhats it for?12:00
blackburnI bet you would like it12:01
blackburnwell blas without 75878 parameters :)12:01
blackburnand faster12:01
heikoah12:02
heikonice :)12:02
heikothats something I like :)12:02
blackburnheiko: trace(A*B') is (A*B.transpose()).trace()12:03
heikoehm? yes?12:03
heikoso thats how you use it or what?12:03
heikoour SGMatrices got these operators now?12:04
blackburnno12:04
blackburnheiko: well that wouldn't work with sgmatrices anyway12:04
heikowhat was that then?12:05
blackburnheiko: they have MatrixXd for that12:05
heikoah ok12:05
blackburnhowever12:05
blackburnif you want to use some preallocated memory12:05
blackburnyou may use Map<MatrixXd> stuff12:05
blackburnthis way you can modify sgmatrix's data12:06
heikoI am a bit lost now12:06
heikowhat can one do nwo?12:07
heikonow12:07
blackburnheiko: well you may use eigen3 both for sgmatrices and some internal MatrixXd matrices12:07
heikoah now I see12:07
blackburnheiko: http://latex.codecogs.com/gif.latex?\min_\beta%20\frac{1}{2}%20\beta^T%20K%20\beta%20-%20\kappa%20\beta%20\\%20\beta_i%20\in%20[0,B],%20~~~%20\left|\sum_i%20\beta_i%20-%20n_s\right|%20\leqslant%20n_s%20\varepsilon12:23
blackburnany idea how to approach this problem?12:23
blackburnuricamic: may be you? :)12:25
heikoblackburn, I cannot read that12:30
blackburnheiko: everything except beta is given12:30
heikono the font12:30
blackburnehm12:30
blackburnagain? :D12:30
heikoyeah12:30
heikoblack on black12:30
blackburnok will try different latex generator12:30
blackburnheiko: how about that? http://www.numberempire.com/equation.render?\min_\beta%20\frac{1}{2}%20\beta^T%20K%20\beta%20-%20\kappa%20\beta\\\beta_i%20\in%20[0,B],%20~~~%20\left|\sum_i%20\beta_i%20-%20n_s\right|%20\leqslant%20n_s%20\varepsilon12:31
heikosame thing12:31
heikoust send the tex12:31
heikoIll insert it myself12:31
heikoah i have it12:32
blackburnheiko: \min_\beta \frac{1}{2} \beta^T K \beta - \kappa \beta12:32
blackburn\\12:32
blackburn\beta_i \in [0,B], ~~~ \left|\sum_i \beta_i - n_s\right| \leqslant n_s \varepsilon12:32
heikowithout this abs value thing it would be easy right?12:32
blackburnyeah12:32
heikothis looks a bit like a L1 norm regularizer, isnt there a lasso method that does a similar thing?12:33
heikoblackburn, otherwise. no idea, never saw one like this :)12:33
blackburnheiko: I don't know actually..12:34
blackburnit is a KMM12:34
blackburnkernel mean matching12:34
heikommh12:34
blackburnsome stuff by your supervisor btw :)12:34
heikoreally?12:35
heikowe compute the distance between the means12:35
heikowe dont match them :)12:35
blackburnhttp://listengine.tuxfamily.org/lists.tuxfamily.org/eigen/2012/07/msg00101.html12:36
blackburnhehe12:36
blackburnheiko: http://www.kyb.mpg.de/fileadmin/user_upload/files/publications/attachments/shift-book-for-LeEtAl-webversion_5376%5b0%5d.pdf12:37
heikooh well12:38
heikoat least not me :)12:38
blackburnv-trick12:42
blackburnhmm what is v-trick12:42
blackburnor rather \nu-trick12:43
-!- Netsplit *.net <-> *.split quits: wiking_13:43
-!- Netsplit over, joins: wiking_13:44
blackburn[DEBUG] 201 iteration passed, objective = 301058401603326972979179723582841701075978963558336672155439500039329112745427114261305300261979153027907701835932192738286119401824164274262496646950403644063744.00000014:18
blackburnROFL14:18
heikohehe :)14:19
CIA-18shogun: Sergey Lisitsyn master * r7b9ac25 / (8 files in 4 dirs): Added Multitask Trace Logistic Regression - http://git.io/r8n9Ww14:24
blackburnargh14:24
CIA-18shogun: Sergey Lisitsyn master * r4f42e08 / src/shogun/mathematics/Math.h : Guarded eigen3 include - http://git.io/XRlOLA14:25
shogun-buildbotbuild #141 of deb1 - libshogun is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb1%20-%20libshogun/builds/141  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>14:28
shogun-buildbotbuild #142 of deb1 - libshogun is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb1%20-%20libshogun/builds/14214:30
-!- os252 [~stegle@as-net.tuebingen.mpg.de] has joined #shogun14:31
-!- os252 [~stegle@as-net.tuebingen.mpg.de] has quit [Client Quit]14:32
-!- blackburn [~blackburn@109.226.92.17] has quit [Quit: Leaving.]14:35
-!- gsomix [~gsomix@r206-10.smr.ru] has quit [Quit: Ex-Chat]15:05
-!- heiko [~heiko@host86-181-80-163.range86-181.btcentralplus.com] has quit [Quit: Leaving.]15:18
-!- yoo [2eda6d52@gateway/web/freenode/ip.46.218.109.82] has left #shogun []15:22
-!- wiking_ is now known as wiking15:22
-!- wiking [~wiking@info2k1.hu] has quit [Changing host]15:23
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun15:23
-!- zxtx [~zv@ool-457e7550.dyn.optonline.net] has joined #shogun15:58
-!- zxtx [~zv@ool-457e7550.dyn.optonline.net] has quit [Ping timeout: 246 seconds]16:17
-!- yoo [2eda6d52@gateway/web/freenode/ip.46.218.109.82] has joined #shogun17:03
-!- zxtx [~zv@ool-457e7550.dyn.optonline.net] has joined #shogun17:03
-!- yoo [2eda6d52@gateway/web/freenode/ip.46.218.109.82] has quit [Quit: Page closed]17:35
-!- zxtx [~zv@ool-457e7550.dyn.optonline.net] has quit [Ping timeout: 246 seconds]17:59
-!- yoo [2eda6d52@gateway/web/freenode/ip.46.218.109.82] has joined #shogun18:00
-!- gsomix [~gsomix@188.168.5.70] has joined #shogun18:23
gsomixgood evening18:24
-!- yoo [2eda6d52@gateway/web/freenode/ip.46.218.109.82] has quit [Quit: Page closed]18:25
-!- uricamic [~uricamic@cmpgw-27.felk.cvut.cz] has quit [Quit: Leaving.]18:33
@sonney2kgsomix, hey there? doing shogun stuff now?18:53
-!- blackburn [~blackburn@109.226.92.17] has joined #shogun19:18
gsomixsonney2k, yes, of course19:25
-!- blackburn [~blackburn@109.226.92.17] has quit [Quit: Leaving.]19:30
-!- blackburn [~blackburn@109.226.92.17] has joined #shogun19:31
-!- blackburn [~blackburn@109.226.92.17] has quit [Quit: Leaving.]19:37
-!- blackburn [~blackburn@109.226.92.17] has joined #shogun19:38
gsomixsonney2k, ta-dam19:44
-!- heiko [~heiko@host86-185-15-105.range86-185.btcentralplus.com] has joined #shogun19:55
@sonney2kgsomix, ta-dam? ka-boom? crashing?19:56
gsomixthere is "da bum tss" for crashing.19:56
@sonney2kgsomix, adn ta-dam?19:56
gsomixfor success19:57
@sonney2kheh19:57
gsomixok, need some defines.19:57
gsomixand PR.19:57
heikoblackburn, around?20:04
blackburnyeah20:04
heikoI wanted to ask about that multiple train() bug again20:04
blackburnaham20:04
heikowe need to train multiple times while cv20:04
blackburnexactly20:04
heikoso how did you resolve the other one?20:04
blackburnI didn't20:04
blackburnit is not fixed yet20:05
heikoah ok20:05
heikoI will check whether its really the resion20:05
heikoreason20:05
blackburnheiko: just add double train call in20:05
blackburnclassifier_larank_modular20:05
heikoblackburn, just did :)20:05
blackburngmnp works though20:06
heikommh20:07
heikoyes20:07
heikobut I train on same features20:07
heikoIll train on different and then on original again20:07
heikono its fine20:08
heikoso thats not the problem for gmnp20:08
heikoblackburn, any other suspicions?20:09
blackburnheiko: no, not really20:09
blackburnwe need to check what fails20:09
blackburnand what works20:09
-!- KMcQuisten [d8338942@gateway/web/freenode/ip.216.51.137.66] has joined #shogun20:09
KMcQuistenHello all20:09
blackburnit's been a long, KMcQuisten20:10
blackburnhow are you?20:10
blackburnheiko: I will take care of larank20:10
KMcQuistenI'm doing ok, but I'm having some troubles.  Can't even get a simple 2kernel MKL to train.20:10
heikoblackburn, ok, wait a bit20:10
heikoperhaps the fault is in the CV20:10
blackburndo you think so?20:10
heikoblackburn, you mean me?20:10
blackburnheiko: what is wrong in CV?20:11
heikoif so: I think the recent multiclass-change might have caused problems20:11
heikoI remember fixing something when I did my multi-class bioinfo coursework on christmas20:11
blackburnhmm20:11
heikobut I only did this for libsvm20:11
heikoremember?20:11
blackburnnot really20:11
blackburnKMcQuisten: what is the problem?20:11
heikolabels were not set correctly to all submachine or something20:11
KMcQuistenblackburn:  Same kind of thing as before.  Using multiple string kenrels in MKLregression, setup up the proper features and kernels, then when I train, sometimes it will train, and sometimes I get:20:13
KMcQuistenSystemError: [ERROR] Assertion R >= 0 failed!20:13
KMcQuistenAnd it seems random20:13
KMcQuisteni think we dealt with this once before, but it's become unfixed, as it were20:14
KMcQuistenI have a test script I can email you if you want to reproduce20:14
blackburnKMcQuisten: sure20:15
KMcQuistenSent20:19
KMcQuistenDid that make it through to you?20:25
blackburnKMcQuisten: yeah fails here too20:25
blackburnchecking code20:25
KMcQuistenRighto20:25
blackburnKMcQuisten: for now you may disable linadd again20:26
KMcQuistenAh, yes20:26
blackburnslower but seems to work20:26
KMcQuistenSounds like a plan for now20:27
-!- KMcQuisten [d8338942@gateway/web/freenode/ip.216.51.137.66] has quit [Ping timeout: 245 seconds]20:31
-!- audy [~audy@heyaudy.com] has quit [Changing host]20:41
-!- audy [~audy@unaffiliated/audy] has joined #shogun20:41
-!- os253 [~stegle@as-net.tuebingen.mpg.de] has joined #shogun20:43
blackburnhey os25320:44
blackburnheiko: any thoughts?20:44
os253hey Blackburn20:44
blackburnos253: do you know kernel mean matching method?20:45
blackburnI ask everyone today :D20:45
wikingboooyaaa20:46
blackburnawakening of wiking20:46
wikingheheh yeps20:46
os253I have compared to this at some point: it's very simple and works ok.20:46
os253http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.165.892920:46
wikingblackburn: new pr20:46
wiking!20:46
wikingso take it20:46
wikingoh no just a sec to pull it20:47
wikingblackburn: btw have u had time to check on larank and cv?20:47
blackburnos253: I'd like to implement it but stucked with formulation20:47
blackburnos253: can you help me there?20:47
@sonney2kblackburn, heiko is the expert for that isn't he?20:47
wikingblackburn: and do you have a hog implementation at hand?20:47
blackburnwiking: I have python hog implementation only actually20:47
wikingblackburn: what are the dependencies?20:47
wikingapart from python ;)20:47
os253blackburn: with the exact identical method?20:48
os253I.e. you need an implementation of this?20:48
blackburnos253: well I am confused with |\sum \beta_i - n_tr | < e n_tr20:48
wikingsonney2k: i think u could run libshogun test on freebsd bot20:48
blackburnos253: how one can handle this?20:48
blackburnwiking: I didn't get to larank yet20:48
wikingsonney2k: i've installed some extra dependencies on that machine for shogun20:48
wikingblackburn: link for hog?20:49
blackburnwiking: I'll send you a link in a sec20:49
wikingblackburn: cool thnx20:49
blackburnwiking: I used this one http://code.google.com/p/python-cvgreyc/source/browse/trunk/cvgreyc/features/HOG.py?r=8320:49
os253To be honest I looked at this years ago and compared to it only. The implementation was from elephant (an ML toolbox), I think. Maybe I also coded it up, I can take a look.20:49
blackburnos253: formulation is20:50
blackburnhttp://latex.codecogs.com/gif.latex?\min_\beta%20\frac{1}{2}%20\beta^T%20K%20\beta%20-%20\kappa%20\beta%20\\%20\beta_i%20\in%20[0,B],%20~~~%20\left|\sum_i%20\beta_i%20-%20n_s\right|%20\leqslant%20n_s%20\varepsilon20:50
wikingblackburn: thnx i'll give it a go20:50
blackburnsonney2k: may be you can help too?20:50
blackburn:)20:50
os253get's scrambled in this IRC client. Give me a bit, need to finish 1,2 things and then will take a look.20:51
blackburnok sure20:51
gsomixsonney2k, PR21:04
blackburnwiking: I  can't see any PR :)21:07
* wiking mumblesbadshit21:09
-!- yoo [575b08cb@gateway/web/freenode/ip.87.91.8.203] has joined #shogun21:15
yoogood evening there21:22
blackburnevening21:22
@sonney2kgsomix, I don't get it21:24
heikoyoo, still looking at the problem21:25
@sonney2kgsomix, are these operators for += ?21:25
heikoquick fix didnt work ;)21:25
@sonney2ketc21:25
yooheiko: brb21:25
gsomixsonney2k, +=, *=, -=. more operators are needed?21:26
@sonney2kgsomix, but isn't this doing x+=y where x are CDenseFeatures and y is a numpy matrix?21:26
@sonney2kgsomix, why don't you write instead of21:28
@sonney2kresult_mat[num_feat*i + j] = lhs[num_feat*i + j] ## operator ## buf[num_feat*i + j]21:28
@sonney2klhs[num_feat*i + j] ## operator ## buf[num_feat*i + j]21:28
@sonney2kand operator then is += etc?21:28
@sonney2kinstead of creating a new one?21:29
@sonney2kgsomix, ^21:29
gsomixsonney2k, what do you mean? it works for DenseFeature+(*, -)=DenseFeatures and for DenseFeatures+(*,-)=numpy.array21:30
gsomixsonney2k, oops, my fault21:30
gsomixone moment21:30
blackburngsomix: but you treat it inplace21:30
blackburn:)21:30
heikoblackbrun, is there a test for the store_model_features of multi class kernel machine?21:30
blackburnheiko: nein21:30
blackburnit is not even implemented21:31
blackburnI am afraid21:31
heikoit is21:31
heikoI bet this is causing the errors21:31
heikowrong sv indices21:31
blackburnheiko: ohh21:32
blackburnthat can be21:32
wikingheiko: that has been implemented in a rush afaik21:32
heikoChiyuan Zhang wrote it21:32
blackburnheiko: ok this describes gmnp failure21:33
blackburnbut larank still to fix21:33
heikoThe only thing is that why works it for libsvm multiclass21:33
blackburninteresting21:33
wikingheiko: when the new machines were introduced somebody realised that cv doesnt work for  kernelmachines21:33
wikingi mean kernel multiclass machines21:33
yooheiko: one thg I havnt checked is the sv indices. I m double checking.21:33
wikingand then it got implemented fast...21:33
wikingthat was like 2-3 months ago21:33
wikingrather 321:33
heikommmh21:34
wikingheiko: you can check the correspoinding chat logs for the git commit date...21:34
heikowe need more tests ...21:34
heikothis should be captured automatically21:34
heikobut it works for libsvm so I will search for the differences21:35
wikingheiko: CMulticlassLibLinear?21:37
heikowiking, have a look at MultiClassLibSVM21:37
heiko.cpp21:37
heikoline 19721:37
heikothere I wrote a fix for this problem21:37
heikowhich is rather unefficient, but works21:37
heikoI think this would be needed for the other multi class svms too21:37
gsomixcompling...21:38
heikoits almost the only difference between working and not-working implementations21:38
heikoand whats this?21:39
heikosvm_svs()=SGVector<index_t>(m_kernel->get_num_vec_lhs());21:39
-!- blackburn [~blackburn@109.226.92.17] has quit [Quit: Leaving.]21:39
wikingdunno21:40
-!- blackburn [~blackburn@109.226.92.17] has joined #shogun21:43
heikoyoo, its a bit more complicated, Ill have to dive a bit into it so it may take some time ...21:46
yooheiko: ok. guide me to help if needed. what type of fix did u do for MultiClassLibSVM ?21:48
heikoyoo, there was a simila problem and I just put in a hack where I just stored *all* SV21:49
heikoso the problem is probably related to wrongly set SVs21:49
@sonney2kwiking, can you test whether make tests works on your buildslave?21:49
heikobut I currently cannot oversee that all21:50
heikohas to be debugged step by step unfortunately21:50
wikingsonney2k: sec21:50
@sonney2kwiking, to do so just go into the build/src directory21:50
heikowhich is a lot of work21:50
@sonney2kthen run make install21:50
@sonney2kmake tests21:50
yooheiko: arf21:50
heikoyoo, yes annoying21:51
wikingsonney2k: ah usual stuff: #!/bin/bash21:52
wikingsonney2k: in check.sh21:52
wikingsonney2k: should i replace it or will you to #!/usr/bin/env bash21:53
wiking?21:53
yooheiko: I will then check the mailing list to be aware of any fix for this bug.21:53
yoothx21:53
heikoyoo, I fear there currently is not21:54
heikois has to be "really" fixed21:54
@sonney2kwiking, I am doing... but just replace it locally and try to get it to work...21:54
@sonney2kmaybe there is more21:55
wikingok trying21:55
wikingError expanding embedded variable.22:01
wikingwht is this?22:01
wikingah ok i know22:02
wikingso the problem is that ./check.sh calls make22:02
wikingbut it should call gmake on freebsd22:02
wikingwill replace it locally just to test it22:02
wikingok that solved the problem22:04
wikingwoah but there's a problem with compilation22:04
wiking//usr/home/wiking/shogun-buildbot/bsd1_-_libshogun/build/src/build/lib/libshogun.so: undefined reference to `std::__detail::_List_node_base::_M_hook(std::__detail::_List_node_base*)@GLIBCXX_3.4.15'22:04
CIA-18shogun: Soeren Sonnenburg master * r2f6ce63 / (25 files in 24 dirs): replace #!/bin/bash with #!/usr/bin/env bash - http://git.io/lxgOMQ22:04
wiking//usr/home/wiking/shogun-buildbot/bsd1_-_libshogun/build/src/build/lib/libshogun.so: undefined reference to `std::ctype<char>::_M_widen_init() const@GLIBCXX_3.4.11'22:05
yooheiko: need to leave. thks for the info. its too bad that the easy fix didnt work but the bug is identified now. hope some1 is going to work on this (?) cya.22:05
heikoyoo, I will work on it this week22:06
heikohope to fix it in a few days22:06
yooheiko: great I ll come back to support ^^ ++22:06
-!- yoo [575b08cb@gateway/web/freenode/ip.87.91.8.203] has quit [Quit: Page closed]22:06
@sonney2kwiking, wait your gcc is 3.X sth?22:07
gsomixsonney2k, fixed22:08
wikingsonney2k: noup.. i think it's more that it tries to use the wrong lib22:08
@sonney2kgsomix, looking22:08
@sonney2kwiking, do you run make check-examples from the build/src dir?22:10
wikingsonney2k: gmake tests22:12
wikingi did this22:12
CIA-18shogun: Evgeniy Andreev master * rb5127eb / (2 files in 2 dirs): inplace arithmetic operators for dense features - http://git.io/Av3-fQ22:13
CIA-18shogun: Soeren Sonnenburg master * rdb85eb2 / (2 files in 2 dirs): Merge pull request #650 from gsomix/buffer_protocol - http://git.io/aTWcrA22:13
@sonney2kwiking, yeah that is the same22:13
@sonney2kgsomix, alright - could you then tomorrow do slicing support?22:14
wikingsonney2k: apparently not ;)22:14
wikingi mean make != gmake on bsd22:14
gsomixsonney2k, ok22:14
gsomixsonney2k, tnx for review22:15
@sonney2kgsomix, thanks for the code :)22:17
@sonney2kbtw wiking, gsomix did you send your weekly report yet?22:17
gsomixsonney2k, I'm writing now.22:17
wikingsonney2k: so there should be a variable in the check.sh for make22:17
wikingor seomthing22:17
@sonney2kwiking, sure but you can patch this locally for now22:19
wikingsonney2k: i did it22:19
CIA-18shogun: Sergey Lisitsyn master * r6b69c17 / src/shogun/multiclass/LaRank.cpp : Fixed larank multiple training issue - http://git.io/je1UZA22:19
@sonney2kwiking, I am rather interested in understanding why you get such weird erros22:20
wikingi just need to find outnow about those build errors22:20
@sonney2kwiking, grep M_widen_init /usr/lib/x86_64-linux-gnu/libs* -r22:21
@sonney2kBinary file /usr/lib/x86_64-linux-gnu/libstdc++.so.6 matches22:21
-!- ckwidmer [8ca3fe9d@gateway/web/freenode/ip.140.163.254.157] has joined #shogun22:33
CIA-18shogun: Soeren Sonnenburg master * r98f4788 / (2 files in 2 dirs): call $MAKE from configure... - http://git.io/cwgqMA22:46
@sonney2kwiking, that ^ could do it... please test22:47
@sonney2kwiking, and again please don't forget about your weekly report22:47
wikingsonney2k: for both: yep just doing them!22:51
@sonney2kwiking, thanks...22:51
wikingsonney2k: just had to organize soemthing about my wedding...22:51
wikingcrazy times22:51
blackburnwedding?22:52
blackburn:)22:52
wikingblackburn: yes :)22:52
blackburnmy brother is going to have wedding on 27th Jul22:52
wikingcongrtz22:52
gsomixgood night guys22:54
heikoblackburn, my sister maried a few weeks ago :)22:56
blackburnchain22:56
wikingsonney2k: here?23:09
shogun-buildbotbuild #154 of deb3 - modular_interfaces is complete: Failure [failed test python_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/154  blamelist: Evgeniy Andreev <gsomix@gmail.com>23:10
wikingsonney2k: the problem about the compiling: the c++ has not been set in the right way...23:10
wikingsonney2k: the configure script doesn't set it23:10
wikingsonney2k: in examples/undocumented/libshogun/Makefile the CC variable should be set by ./configure script23:11
wikingas now it's always CC=c++ which in this box will fail as the whole shogun has been built with c++4623:11
wikingwhich is set by the configure script23:12
wikingsonney2k: if i replace CC=c++46 in that Makefile then it compiles23:12
-!- heiko1 [~heiko@host86-181-81-54.range86-181.btcentralplus.com] has joined #shogun23:16
wikingsonney2k: ok almost... there's still some weird errors... just trying with a clean build...23:17
-!- heiko [~heiko@host86-185-15-105.range86-185.btcentralplus.com] has quit [Ping timeout: 265 seconds]23:17
wikingok this part i dont get yet:23:22
wikingrunning basic_minimal .. ERROR23:22
wiking/usr/lib/libstdc++.so.6: version GLIBCXX_3.4.11 required by //usr/home/wiking/shogun-buildbot/bsd1_-_libshogun/build/src/build/lib/libshogun.so.12 not found23:22
wikingok i think there's a problem with linking23:25
wikingsince it's actually according to ldd libshogun.so.12 is requiring /usr/lib/libstdc++.so.6 but it should use /usr/local/lib/gcc46/libstdc++.so.6 instead since that's the one for c++46 on my system...23:26
shogun-buildbotbuild #155 of deb3 - modular_interfaces is complete: Failure [failed test python_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/155  blamelist: Soeren Sonnenburg <sonne@debian.org>, Sergey Lisitsyn <lisitsyn.s.o@gmail.com>23:33
shogun-buildbotbuild #136 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/136  blamelist: Evgeniy Andreev <gsomix@gmail.com>, Soeren Sonnenburg <sonne@debian.org>, Sergey Lisitsyn <lisitsyn.s.o@gmail.com>23:36
-!- rieck [~rieck@paul.ml.tu-berlin.de] has joined #shogun23:37
rieckmoin23:37
rieckanyone home23:37
blackburnhey23:37
riecksonney2k: still alive?23:38
rieckthat's a pitty, got an IM 40 mins ago23:39
shogun-buildbotbuild #137 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/13723:43
rieckblackburn: still working on shogun extensions?23:44
blackburnrieck: yeah breaking shogun still ;)23:44
rieckblackburn: did gsoc work out last year? did you have fun and get some money?23:45
blackburnrieck: yes, I am participating this year too23:45
blackburnrieck: this year we've got 8 students23:46
wikingbzzzz23:48
* wiking does not understand freebsd's moves in this linking...23:48
rieckcool.23:49
riecksupport from google is clearly a boost for shogun23:49
rieckit's also good to see people working again in gsoc23:49
blackburnrieck: yeah things have changed here23:50
riecki am tired. i have to get some sleep, tomorrow is a written exam23:51
blackburnnot for you I guess :D23:51
rieckno. its my students and my exam. still, i need to be somewhat prepared and awake, otherwise it would not be fair23:52
riecknite23:52
blackburnheh good luck to you students then23:52
blackburnnight23:52
--- Log closed Tue Jul 17 00:00:17 2012