Open in new window / Try shogun cloud
--- Log opened Fri Sep 20 00:00:06 2013
shogun-buildbotbuild #1501 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1501  blamelist: Soeren Sonnenburg <sonne@debian.org>00:06
-!- van51 [~van51@athedsl-409794.home.otenet.gr] has quit [Quit: Leaving.]00:07
shogun-notifier-shogun: Soeren Sonnenburg :develop * 14914b9 / examples/undocumented/python_modular/kernel_rationalquadratic_modular.py: https://github.com/shogun-toolbox/shogun/commit/14914b969758afa9dc40e16dc918a4971b1bc78000:21
shogun-notifier-shogun: fix rational quadratic00:21
-!- travis-ci [~travis-ci@ec2-23-22-84-25.compute-1.amazonaws.com] has joined #shogun00:40
travis-ci[travis-ci] it's Soeren Sonnenburg's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1156986200:40
-!- travis-ci [~travis-ci@ec2-23-22-84-25.compute-1.amazonaws.com] has left #shogun []00:40
shogun-buildbotbuild #1502 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1502  blamelist: Soeren Sonnenburg <sonne@debian.org>01:06
shogun-buildbotbuild #1850 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/185001:35
-!- travis-ci [~travis-ci@ec2-23-22-84-25.compute-1.amazonaws.com] has joined #shogun01:44
travis-ci[travis-ci] it's Soeren Sonnenburg's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1157177801:44
-!- travis-ci [~travis-ci@ec2-23-22-84-25.compute-1.amazonaws.com] has left #shogun []01:44
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 256 seconds]02:05
shogun-buildbotbuild #19 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/19  blamelist: Soeren Sonnenburg <sonne@debian.org>02:20
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun02:26
-!- mode/#shogun [+o iglesiasg] by ChanServ02:26
shogun-buildbotbuild #7 of clang34 - undefined behaviour analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/7  blamelist: Soeren Sonnenburg <sonne@debian.org>02:30
-!- pickle27 [~Kevin@208.138.34.236] has joined #shogun02:36
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Read error: Operation timed out]02:38
-!- pickle27 [~Kevin@208.138.34.236] has quit [Quit: Leaving]02:42
shogun-buildbotbuild #11 of clang34 - thread analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/11  blamelist: Soeren Sonnenburg <sonne@debian.org>02:45
shogun-buildbotbuild #473 of nightly_none is complete: Failure [failed compile]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_none/builds/47303:09
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]03:21
shogun-buildbotbuild #460 of nightly_all is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_all/builds/46003:32
-!- zxtx [~zv@149.165.244.2] has joined #shogun05:16
-!- sonne|osx [~sonne@82.113.121.195] has joined #shogun08:20
wikingsonne|osx: yo. have u managed to update the notebook update script?08:20
sonne|osxwiking: ?08:20
sonne|osxso no08:20
wiking:(08:20
sonne|osxwiking: is this from the automated tests?08:21
sonne|osxd08:21
sonne|osxd/home/buildslave/nightly_none/build/tests/unit/base/SGObject_unittest.cc: In function 'void* stress_test(void*)':08:21
sonne|osxSGObject_unittest.cc:56:17: warning: unused variable 'labs_2' [-Wunused-variable]08:21
wikingsonne|osx: no that's something handwritten08:22
sonne|osxok then08:24
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun08:24
shogun-notifier-shogun: Soeren Sonnenburg :develop * 687d5c8 / tests/unit/base/SGObject_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/687d5c83fc19e690988d261304250d76fcf03e7608:24
shogun-notifier-shogun: disable refcount test when refcounting is not available08:24
sonne|osxwiking: and the Serialization* tests from here http://buildbot.shogun-toolbox.org/builders/nightly_all/builds/460/steps/test/logs/stdio ?08:25
wikingSerialization{JSON, Ascii, XML, HDF5} are autogenerated from the jinja2 template08:26
sonne|osxwiking: what are those doing?08:26
wikingsonne|osx: serialize all the possible objects into a file08:26
wikingreads it08:26
sonne|osxwiking: we really need a gdb trace08:26
wikingand checks if it's equal08:26
wikingsonne|osx: couldn't find such option for ctest yet08:27
sonne|osxI have no clue how to debug this08:27
sonne|osxwiking: no we can do it when running our test08:27
sonne|osxwiking: like this08:27
wikingsonne|osx: how... do u know the ctest command for it?08:27
sonne|osxtest08:27
wikingbtw: do you know how to create a tmpfs?08:27
sonne|osxtest || gdb test08:27
sonne|osxwiking: isn't ctest just calling some commands08:28
sonne|osx?08:28
wikingafaik yes08:28
sonne|osxso then it is easy08:28
sonne|osxa test returns 0 if it worked fine08:29
sonne|osxif not we can re-run the test with gdb08:29
sonne|osxand voila!08:29
sonne|osxthis is how I did this when we had configure running08:30
sonne|osxmaybe you can do it a bit more beautiful by defining some cmake function that is encapsulating this?08:30
wikingdunno08:31
sonne|osxshogun-buildbot: force build branch=develop 'nightly_none'08:31
shogun-buildbotno such builder 'branch=develop'08:31
sonne|osxshogun-buildbot: force build --branch=develop 'nightly_none'08:31
shogun-buildbotThe build has been queued, I'll give a shout when it starts08:32
sonne|osxwiking: so can you do it?08:34
sonne|osxwiking: all that is needed is to do08:34
sonne|osxin: /bin/ls || gdb --command shogun/src/.gdb /bin/ls08:35
shogun-buildbotbuild #164 of osx1 - libshogun is complete: Failure [failed compile]  Build details are at http://buildbot.shogun-toolbox.org/builders/osx1%20-%20libshogun/builds/164  blamelist: Soeren Sonnenburg <sonne@debian.org>08:35
sonne|osxgtg brb08:35
-!- sonne|osx [~sonne@82.113.121.195] has quit [Quit: sonne|osx]08:35
shogun-buildbotbuild #474 forced08:38
shogun-buildbotI'll give a shout when the build finishes08:38
-!- besser82 [~besser82@fedora/besser82] has joined #shogun08:38
shogun-buildbotbuild #474 of nightly_none is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_none/builds/47408:41
shogun-notifier-shogun: Viktor Gal :develop * 83b821f / src/shogun/mathematics/Mosek.cpp: https://github.com/shogun-toolbox/shogun/commit/83b821f949c07762018309025686a852b709be9208:50
shogun-notifier-shogun: Update Mosek.cpp wrapper to support mosek 708:50
shogun-notifier-shogun: issue #128708:50
-!- lisitsyn [~lisitsyn@fb2-lo1.global63.net] has quit [Quit: Leaving.]09:06
shogun-buildbotbuild #1503 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1503  blamelist: Soeren Sonnenburg <sonne@debian.org>09:10
shogun-buildbotbuild #1222 of rpm1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/rpm1%20-%20libshogun/builds/1222  blamelist: Viktor Gal <viktor.gal@maeth.com>09:17
sonne|workwiking: could you please disable running of tests on the nightly_none bot?09:27
wikingsonne|work: why?09:27
-!- travis-ci [~travis-ci@ec2-54-242-103-128.compute-1.amazonaws.com] has joined #shogun09:27
travis-ci[travis-ci] it's Soeren Sonnenburg's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1158238209:27
-!- travis-ci [~travis-ci@ec2-54-242-103-128.compute-1.amazonaws.com] has left #shogun []09:27
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting]09:28
sonne|workI think it is enough to see that things compile. Many examples/tests have to be written differently if certain stuff is not avilable09:28
-!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun09:28
wiking^ disabled09:28
sonne|worksure all doable but what the heck is it worth the effort?09:28
sonne|workif we figure this is a problem we can enable it later09:29
sonne|workwiking: and can you do the test running stuff with gdb or not?09:31
wikingsonne|work: not now09:31
wikingand i think not in the next 3 days09:31
shogun-notifier-shogun: Viktor Gal :develop * ac174a6 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/ac174a6b111d7f1d1457bec0e11f1c3ca8a09ea409:35
shogun-notifier-shogun: Add Mosek based libshogun examples if Mosek is enabled09:35
shogun-notifier-shogun: Fix Mosek env creation in CCSOSVM for mosek v709:35
shogun-notifier-shogun: add USE_MOSEK macro to config.h09:35
sonne|workshogun-buildbot: force build --branch=develop 'nightly_none'09:36
shogun-buildbotThe build has been queued, I'll give a shout when it starts09:36
sonne|workwiking: how can one run e.g. piped commands in cmake?09:38
sonne|workwiking:  such as echo "foo" | sed 's/o/a' ?09:39
wikingsometimes it works when u add two COMMAND  lines09:41
wikingsay09:41
-!- lambday [67157f37@gateway/web/freenode/ip.103.21.127.55] has joined #shogun09:42
wikingCOMMADN echo "foo"09:42
wikingCOMMAND sed ...09:42
shogun-buildbotbuild #475 forced09:43
shogun-buildbotI'll give a shout when the build finishes09:43
wikinglambday: hey09:44
lambdaywiking: hi09:44
lambdaywiking: thanks for the SGVector09:45
wikinglambday: could u use it?09:45
lambdaywiking: not yet tested09:45
lambdaychecking09:45
lambday:)09:45
wikingok09:46
wikinglet me know if something is not ok or u need something more09:47
wikingi've tested SGVector<complex>09:47
wikingbut not the rest09:47
lambdaywiking: I need SGSparseMatrix<complex>, but I wanna try with this09:48
lambdayso I need to add this in and out for sparse in swig_typemaps.i with NPY_CDOUBLE09:49
lambdayright?09:49
lambdayanything else?09:49
shogun-buildbotbuild #1504 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1504  blamelist: Viktor Gal <viktor.gal@maeth.com>09:50
wikinglambday: i've added SGSparseMatrix<complex09:50
lambdaywiking: awesome! I am testing then :)09:51
lambdayI thought we needed to add TYPEMAP_SPARSEFEATURES_IN/OUT(complex64_t,     NPY_CDOUBLE).. earlier I was trying with this one09:52
wikingthat's for SparseFeatures09:54
shogun-buildbotbuild #20 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/20  blamelist: Soeren Sonnenburg <sonne@debian.org>09:57
shogun-buildbotbuild #475 of nightly_none is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_none/builds/47509:57
-!- travis-ci [~travis-ci@ec2-54-234-21-192.compute-1.amazonaws.com] has joined #shogun10:06
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1158306210:06
-!- travis-ci [~travis-ci@ec2-54-234-21-192.compute-1.amazonaws.com] has left #shogun []10:06
shogun-buildbotbuild #8 of clang34 - undefined behaviour analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/8  blamelist: Soeren Sonnenburg <sonne@debian.org>, Viktor Gal <viktor.gal@maeth.com>10:08
shogun-buildbotbuild #12 of clang34 - thread analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/12  blamelist: Soeren Sonnenburg <sonne@debian.org>, Viktor Gal <viktor.gal@maeth.com>10:24
sonne|workwiking: this is how it works http://stackoverflow.com/questions/3065220/ctest-with-multiple-commands10:33
shogun-buildbotbuild #1223 of rpm1 - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/rpm1%20-%20libshogun/builds/122310:36
wikingthat's good10:37
wikingneeds implementing then10:37
shogun-buildbotbuild #109 of precise - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/precise%20-%20libshogun/builds/109  blamelist: Viktor Gal <viktor.gal@maeth.com>10:44
-!- hushell [~hushell@c-98-232-178-161.hsd1.or.comcast.net] has quit [Quit: WeeChat 0.3.7]10:48
wikingHeikoS: around?10:50
wikingHeikoS: i need to run something on a cluster :)10:50
-!- travis-ci [~travis-ci@ec2-23-22-84-25.compute-1.amazonaws.com] has joined #shogun10:57
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1158407410:57
-!- travis-ci [~travis-ci@ec2-23-22-84-25.compute-1.amazonaws.com] has left #shogun []10:57
shogun-buildbotbuild #1505 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1505  blamelist: Viktor Gal <viktor.gal@maeth.com>10:59
-!- lisitsyn [~lisitsin@mxs.kg.ru] has joined #shogun11:03
shogun-buildbotbuild #21 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/21  blamelist: Viktor Gal <viktor.gal@maeth.com>11:12
shogun-buildbotbuild #1853 of deb3 - modular_interfaces is complete: Failure [failed test ruby modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1853  blamelist: Viktor Gal <viktor.gal@maeth.com>11:20
wikinglambday: or u would need SparseFeatures<complex64_t> wrapper as well?11:30
wikinglambday: plz comment what you've find out about the mapping (works/does work) at: https://github.com/shogun-toolbox/shogun/issues/149311:32
lambdaywiking: seems like I needed to add the sparse features thing after all, although it says features it actually works on sparse matrix11:32
lambdayearlier I couldn't make it work11:32
lambdayI added it11:32
lambdayand now it works11:32
lambdaySGVector<complex64_t> works perfectly by the way :) :)11:33
lambdaythanks a lot man :)11:33
wikingok11:33
wikingso u need sparsefeatures as well?11:33
lambdaywiking: I will send a PR, with a few of my wrappers11:33
lambdaywiking: actually, no.. I only needed SGSparseMatrix<complex64_t> to work11:34
lambdaywiking: just a sec11:34
wikinglambday: that should work11:34
wikingalthough i have not tested it11:34
wikingComplexSparseMatrix11:34
lambdaywiking: https://github.com/shogun-toolbox/shogun/blob/develop/src/interfaces/python_modular/swig_typemaps.i#L121611:34
wiking?11:35
lambdaywhen I added TYPEMAP_SPARSEFEATURES_OUT(complex64_t,     NPY_CDOUBLE), then it worked11:35
wikingbefore that?11:35
lambdaydidn't :(11:35
lambdaythe macro is for SGSparseMatrix11:36
lambdaynot CSparseFeatures11:36
lambday:(11:36
wikingIn [1]: from modshogun import ComplexSparseMatrix11:36
wikingIn [2]: x = ComplexSparseMatrix()11:36
wikingthis works for me11:36
wikingw/o that line11:36
wikingmake up your mind11:36
wikingdo u need SparseMatrix11:36
wikingor SparseFeatures11:36
lambdaywiking: but I tried to pass a csc_matrix(complex) as an argument to a class which takes SGSparseMatrix<complex64_t> it didn't work11:37
lambdayI need sparse matrix11:37
wikingcsc_matrix(complex)?11:37
wikingwhat's this?11:37
lambdaythat's scipy.sparse matrix...11:38
lambdayfor real types, I can seamlessly pass a csc_matrix to a shogun method that takes SGSparseMatrix<float64_t>11:38
lambdayfor complex, I needed the same11:38
wikingok tell me the code that i can generate this scipy.sparse11:39
wikingcsc_matrix(complex11:39
lambdaywiking: wait..11:39
wikingas you are talking way out of context11:39
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun11:39
-!- mode/#shogun [+o iglesiasg] by ChanServ11:39
@iglesiasgHi people!11:39
wikingyo11:39
lambdaywiking: take any matrix, a=array(random.randn(10,10))11:40
lambdaywiking: then b=csc_matrix(a)11:40
@iglesiasgwiking: ipython notebooks ready then? :O11:40
lambdayoh and import csc_matrix from scipy.sparse11:40
wikinglambday: tell me the imports11:40
lambdayfrom scipy.sparse import csc_matrix11:40
wikingwhat's arra11:41
wikingarray11:41
wikingrandom11:41
lambdaynumpy array11:41
wikinga=np.array(random.randn(10,10))11:41
wiking---------------------------------------------------------------------------11:41
wikingNameError                                 Traceback (most recent call last)11:41
wiking<ipython-input-13-1536df4033d4> in <module>()11:41
wiking----> 1 a=np.array(random.randn(10,10))11:41
wikingNameError: name 'random' is not defined11:41
wiking...11:41
lambdayfrom numpy import *11:41
wikingiglesiasg: well test it11:41
lambdayits in numpy11:42
wikingok11:42
wikingi have b now11:42
lambdayyeah11:42
wikingwhat now?11:43
lambdaywiking: now what I needed was I needed to construct one of my class CSparseMatrixOperator<complex64_t> that takes a SGSparseMatrix<complex64_t> as an arg11:43
lambdayshit11:43
lambdayI added it locally you won't be able to access it11:43
lambdaywhat I need is that, I need to pass that csc_matrix to my class11:43
lambdayin the constructor11:43
lambdaylocally it works now11:44
wikingyeah11:44
wikingCSparseMatrixOperator needs a mapping as well11:44
wiking:)11:44
lambdaywiking: yeah I added that locally11:44
lambdaysending PR :)11:44
wikingbut i dont know how you can set the values of ComplexSparseMatrix11:44
wiking:<11:44
wikingin fact i dont know how u do that with ComplexMatrix either11:44
lambdayno idea!11:44
wiking:P11:44
wikingor RealMatrix11:45
wikinghow do you convert a numpy.matrix -> RealMatrix11:45
lambdaybut if I pass a csc_matrix to a method that takes SGSparseMatrix as an arg, it works :D11:45
wikingor RealSparseMatrix11:45
wikingiglesiasg: btw i think it would be a good thing to enable director classes on that shogun install in the notebook server11:46
@iglesiasgwiking: using the image I guess, right?11:46
lambdaywiking: I am using wrappers for these, so in the constructor I pass csc_matrix, works fine11:46
wikingiglesiasg: image?11:46
wikingiglesiasg: what image do you mean?11:46
@iglesiasgwiking: to test the notebooks11:46
wikingno comprende11:46
@iglesiasghaha11:47
wikingelaborate plz11:47
wiking:)11:47
@iglesiasgwiking: 11:41 < wiking> iglesiasg: well test it11:47
wikingiglesiasg: ah no11:47
wikingiglesiasg: haven't u received my email to shogun-team@ ?11:47
@iglesiasgnope11:47
wikingweeeeird11:47
@iglesiasgthat was the part of11:47
@iglesiasgdevelopers have been sent blah blah I did not get11:48
@iglesiasgbut it might have missed it11:48
@iglesiasglet me double check...11:48
wikingok let me know11:48
wikingmaybe it's still in greylist11:48
wikingoooor i've sent from an address from which i'm not allowed to that list11:48
wikingoh ok11:49
wikingthat's what happend11:49
wikinglet me resend it11:49
wikingiglesiasg: sent again... let me know if u received now11:49
@iglesiasgwiking: received11:49
wikingcool11:49
wikingsorry about the fuckup11:49
@iglesiasgwiking: aah np11:51
@iglesiasgI love struggling with these little things :D11:51
wikingah shiiit11:53
wikingwe would need this11:53
wikingMagics for temporary workspace11:53
wikinghttps://github.com/tkf/ipython-tempmagic11:53
@iglesiasgtesting the notebooks atm...11:53
@iglesiasgcool stuff11:53
@iglesiasgwiking: to ensure, so every time I run a cell11:53
@iglesiasgwiking: it is being executed in 7nn, right?11:54
wikingyesp11:54
wiking*yeps11:54
@iglesiasgsweet!11:54
wikingthe problem is that currently11:54
wikingor not the problem11:54
wikingbut the fact is11:54
@iglesiasgit is saved11:54
wikingthat the directory where you enter (where the notebooks are)11:54
wikingis readonly11:54
@iglesiasgwhy problem?11:55
@iglesiasgdrawback?11:55
wikingwell11:55
wikingcan u add a new notebook now?11:55
wiking2013-09-20 09:50:58.338 [tornado.general] WARNING | 400 GET /new (91.64.72.127): Unexpected error while autosaving notebook: [Errno 30] Read-only file system: u'/home/notebook/notebooks/Untitled0.ipynb'11:55
wikingWARNING:tornado.general:400 GET /new (91.64.72.127): Unexpected error while autosaving notebook: [Errno 30] Read-only file system: u'/home/notebook/notebooks/Untitled0.ipynb'11:55
wikingfrom the logs11:55
wikingso i suspect that u cannot add a new notebook11:55
@iglesiasgno.. I cant11:55
wikingsee11:56
wikingthat's the thing11:56
@iglesiasg400: Bad Request11:56
wikingyou can run the ones that are there11:56
@iglesiasgbut hold on11:56
wikingbut you cannot add a new one11:56
@iglesiasghow do we want to handle that?11:56
@iglesiasgI mean11:56
wikingwell this magic shit11:56
wikingshould be able to do that11:56
wikingcreates a temp location11:56
wikingand there you can write whatever11:56
@iglesiasgdo we want people coming in creating notebooks and keeping them in our server?11:56
wikingbut then as soon as u log out11:56
@iglesiasgor just keeping them while they are logged in?11:56
wikingit's cleaned up11:57
wikingwell11:57
@iglesiasgall right11:57
wikingit'd be interesting to keep some11:57
wikingi would say11:57
@iglesiasgthat makes sense then11:57
@iglesiasgdo you think so?11:57
wikingmaybe somebody does some interesting stuff11:57
wikingbut then again11:57
wikingthat person can send it to us11:57
@iglesiasgmmm yeah it could be11:57
wikingsince one can download the notebooks11:57
@iglesiasgwe could train a classifier to keep the interesting ones :)11:57
wikinghehehe11:57
wikingbut anyhow i think 2013-09-20 09:50:58.338 [tornado.general] WARNING | 400 GET /new (91.64.72.127): Unexpected error while autosaving notebook: [Errno 30] Read-only file system: u'/home/notebook/notebooks/Untitled0.ipynb'11:57
wikingso11:58
wikingi think that we should try to use this magic shit11:58
wikingbut yeah test it and then send some feedbacks (this is not for you but *all*) and then i'll do the changes stilll11:58
@iglesiasgwell I just realized that I need to update my notebook...11:59
@iglesiasgwe deffs need to get custom data to the notebook somehow11:59
@iglesiasgwiking: btw, the first mail ipython notebook test mail you sent just arrived here xD12:00
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]12:35
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun12:38
shogun-notifier-shogun-data: hushell :master * 62b19b7 / ocr/ocr_taskar.mat: https://github.com/shogun-toolbox/shogun-data/commit/62b19b767d7e50d8e7d87466b61ba2015bd11db312:38
shogun-notifier-shogun-data: added an OCR dataset (http://www.seas.upenn.edu/~taskar/ocr/)12:38
shogun-notifier-shogun-data: Fernando Iglesias :master * a41ea33 / ocr/ocr_taskar.mat: https://github.com/shogun-toolbox/shogun-data/commit/a41ea33ee80ad01b8c690a09d98eebe37486526212:38
shogun-notifier-shogun-data: Merge pull request #23 from hushell/master12:38
shogun-notifier-shogun-data:12:38
shogun-notifier-shogun-data: added an OCR dataset (.mat)12:38
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: Lost terminal]12:55
lambdayHeikoS: there?12:56
lambdaywiking: sent the PR12:58
lambdayusing this in my notebook now12:58
lambdayHeikoS: please have a look - http://nbviewer.ipython.org/659199412:58
lambdayHeikoS: will be adding ozone12:59
-!- iglesias [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun13:21
* besser82 just runs some experimental SHOGUN-stuff with ViennaCL && CUDA and is heavily impressed!13:42
shogun-notifier-shogun: Fernando Iglesias :develop * 8d498b2 / examples/undocumented/libshogun/ (2 files): https://github.com/shogun-toolbox/shogun/commit/8d498b2154d0b490df6ca16ec406381e429bc3e313:47
shogun-notifier-shogun: Rename and update to new API HM-SVM Mosek libshogun example13:47
shogun-notifier-shogun: Fernando Iglesias :develop * 3d6e5f6 / examples/undocumented/libshogun/ (2 files): https://github.com/shogun-toolbox/shogun/commit/3d6e5f6bea4619dd2ec617ab7d0e3a248478315513:47
shogun-notifier-shogun: Merge pull request #1638 from iglesias/fix/hmsvm_mosek_example13:47
shogun-notifier-shogun:13:47
shogun-notifier-shogun: Rename and update to new API HM-SVM Mosek libshogun example13:47
-!- iglesias [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving]13:50
besser82eigenvectors do get _REALLY_ speedy with that :)13:53
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun13:57
iglesiasgsonne|work: hi!14:02
sonne|workhi14:02
iglesiasghave you checked the notebooks?14:02
sonne|worka bit14:02
iglesiasgwiking did some stuff and we can already execute them14:02
iglesiasgcool14:02
iglesiasgsonne|work: any thought regarding getting data into the notebook?14:03
iglesiasgI think we really need to provide that14:03
iglesiasgthe API could look like14:04
iglesiasgfeatures = RealFeatures(LocalFile('path/to/file/in/client/machine'))14:04
iglesiasgand the same for the labels14:05
sonne|workiglesiasg: we have that already14:05
iglesiasgoh really?14:05
sonne|workiglesiasg: see examples RealFeatures(CSVFile('/path/to/file'))14:05
sonne|worksame with .h514:05
sonne|workand libsvm files14:05
iglesiasgsonne|work: yeah yeah14:05
sonne|worketc14:05
iglesiasgbut the ipython runs in 7nn!14:05
iglesiasgso CSVFile('path/blahblah')14:06
iglesiasgwould look in 7nn's FS14:06
sonne|worksure14:06
iglesiasgwhat I am thinking about is that someone using the notebook14:06
iglesiasgcan choose what data to use from her machine14:06
sonne|workand then?14:07
iglesiasg?14:08
iglesiasgwell then she can do stuff in the notebook with that data14:08
sonne|workwhat does he do if he has chosen some data14:08
sonne|workthat doesn't explain how to get the data14:08
sonne|workto fatbot14:08
iglesiasgyeah that's why I said API :)14:09
sonne|workno idea what you say :/14:09
sonne|workwhat does  features = RealFeatures(LocalFile('path/to/file/in/client/machine')) do14:10
sonne|workwhere is this run?14:10
iglesiasgLocalFile does stuff under the hood14:11
iglesiasgsends the file14:11
iglesiasgmakes security checks14:11
iglesiasgcheks for format14:11
iglesiasgand finally load it in numpy array if everthing went fine14:12
sonne|workiglesiasg: well this is maybe 2 lines python then14:13
iglesiasgsonne|work: even better then14:14
sonne|work        import urllib214:14
sonne|work        req = urllib2.urlopen(url)14:14
sonne|work        content = req.read()14:14
shogun-buildbotbuild #110 of precise - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/precise%20-%20libshogun/builds/11014:14
iglesiasgsonne|work: what do you put in the url there?14:15
sonne|workhttp://somewhere.where/the/file/is14:15
iglesiasgI guess that file:/// ... won't work14:16
iglesiasgbecause that referes to 7nn again14:16
iglesiasgthat would require the client to upload the data to a server14:16
shogun-buildbotbuild #1854 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/185414:24
shogun-buildbotbuild #1855 of deb3 - modular_interfaces is complete: Failure [failed test libshogun]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1855  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>14:25
-!- lambday [67157f37@gateway/web/freenode/ip.103.21.127.55] has quit []14:25
shogun-buildbotbuild #1506 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1506  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>14:30
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving]14:46
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun14:55
-!- mode/#shogun [+o iglesiasg] by ChanServ14:55
shogun-buildbotbuild #22 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/22  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:08
shogun-buildbotbuild #1507 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1507  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:11
-!- lisitsyn [~lisitsin@mxs.kg.ru] has left #shogun []15:11
shogun-buildbotbuild #9 of clang34 - undefined behaviour analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/9  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:19
@HeikoSwiking: nice! just tried the could shogun :) aweseom!15:24
@iglesiasgHeikoS: so in the example evaluation_cross_validation_classification.py15:30
@iglesiasgHeikoS: it says that StratifiedCrossValidation is better15:31
@HeikoSyes15:31
@iglesiasgbut standard StratifiedCrossValidatoinSplitting is also available15:31
@iglesiasgI don't quite a get it15:31
@HeikoSyou mean CrossValidationSplitting15:31
@iglesiasgok15:31
@HeikoSsometimes one cannot do stratified15:31
@iglesiasgthen you meant that :P15:31
@HeikoSfor example in regression15:31
@iglesiasgI will update the example15:31
@HeikoSstratified should always be used with classification15:31
@HeikoSactually, we could think about doing something similar to stratified for regression, (grouping y values) but that would be later15:32
@HeikoSiglesiasg: cool thanks!15:32
@iglesiasgoh I see15:32
@iglesiasgI guess this regression thing is related (though not limited) to GPs, right?15:32
shogun-buildbotbuild #13 of clang34 - thread analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/13  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:33
@iglesiasgHeikoS: so stratified basically maintains the same ratio of classes in the folds, right?15:34
-!- travis-ci [~travis-ci@ec2-23-22-50-150.compute-1.amazonaws.com] has joined #shogun15:39
travis-ci[travis-ci] it's Fernando Iglesias's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1159033115:39
-!- travis-ci [~travis-ci@ec2-23-22-50-150.compute-1.amazonaws.com] has left #shogun []15:39
@HeikoSiglesiasg: yes15:45
@HeikoSiglesiasg: this reduces the variance of the test error estimate15:45
@HeikoSquite a bit15:45
@iglesiasgHeikoS: aham, stupid question: the ratio wrt to the whole data set, I guess15:46
@HeikoSiglesiasg: yeah as far as this is possible15:46
@HeikoSiglesiasg: it just distributes the labels to the folds in turns15:46
@HeikoSgiving one class to a fold a t a time15:47
-!- lambday [67157f37@gateway/web/freenode/ip.103.21.127.55] has joined #shogun15:47
@iglesiasgoh I understand15:49
@iglesiasgit is a simple idea, like the great ones! hehe15:50
@HeikoSiglesiasg: there is a nice paper about these things15:50
@HeikoSI think its mentioend in the source15:50
-!- zxtx [~zv@149.165.244.2] has quit [Ping timeout: 245 seconds]16:00
lambdayHeikoS: checking your mail :)16:08
@HeikoSlambday: I kind of wrote a lot of things, but most of them are small16:08
@HeikoSlambday: main thing you should work on is to make it flow a bit more and look a bit nicer, but its already very cool16:09
@HeikoSlambday: and its a lot of stuff, thats why I would put the most interesting parts with the prettiest pictures on top :)16:09
-!- lambday_ [67157e37@gateway/web/freenode/ip.103.21.126.55] has joined #shogun16:10
lambday_HeikoS: sorry I got dc16:11
@HeikoSlambdayno worries16:11
-!- lambday [67157f37@gateway/web/freenode/ip.103.21.127.55] has quit [Ping timeout: 250 seconds]16:12
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]16:47
-!- votjakovr [~votjakovr@host-46-241-3-209.bbcustomer.zsttk.net] has joined #shogun17:26
-!- FSCV [~FSCV@65.19.131.247] has joined #shogun17:28
-!- lambday_ [67157e37@gateway/web/freenode/ip.103.21.126.55] has quit []17:32
-!- van51 [~van51@athedsl-409794.home.otenet.gr] has joined #shogun17:35
-!- FSCV_ [~FSCV@187.191.50.146] has joined #shogun17:46
-!- FSCV [~FSCV@65.19.131.247] has quit [Ping timeout: 264 seconds]17:48
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun18:12
shogun-notifier-shogun: Heiko Strathmann :develop * c82e050 / src/shogun/mathematics/linalg/ (4 files): https://github.com/shogun-toolbox/shogun/commit/c82e05033eca8e96a625c60dd020ab2291aa842c18:12
shogun-notifier-shogun: turned some debug messages into info messages18:12
shogun-notifier-shogun: Heiko Strathmann :develop * da19157 / src/shogun/mathematics/linalg/ (4 files): https://github.com/shogun-toolbox/shogun/commit/da191575dad9865be95ea51fbbd8c2c3524dbce918:12
shogun-notifier-shogun: Merge pull request #1641 from karlnapf/develop18:12
shogun-notifier-shogun:18:12
shogun-notifier-shogun: turned some debug messages into info messages18:12
@HeikoSiglesiasg: https://travis-ci.org/shogun-toolbox/shogun/jobs/1158795918:13
@HeikoSlmnn still fails18:13
shogun-notifier-shogun: lambday :develop * e39fa9e / src/interfaces/ (3 files): https://github.com/shogun-toolbox/shogun/commit/e39fa9ed85d4b310a407472b345fc80049dfb7d418:14
shogun-notifier-shogun: added ComplexSparseMatrixOperator to modular interface (log-det)18:14
shogun-notifier-shogun: Heiko Strathmann :develop * c6e84fd / src/interfaces/ (3 files): https://github.com/shogun-toolbox/shogun/commit/c6e84fd1c432cc30f30d9b65465f0cc3a070f42e18:14
shogun-notifier-shogun: Merge pull request #1637 from lambday/feature/log_determinant18:14
shogun-notifier-shogun:18:14
shogun-notifier-shogun: added ComplexSparseMatrixOperator to modular interface (log-det)18:14
@HeikoSvotjakovr: looking at the PR now18:14
votjakovrHeikoS: thanks18:15
@HeikoSvotjakovr: so whats the status? any questions on the PR? when is it ready to merge? what is missing?18:16
votjakovrHeikoS: i've used some trick there to make it possible to call SG_ADD with multiply arguments18:16
@HeikoSvotjakovr: very nice! we could even do this for MS_NOT_AVAILABLE then18:16
@HeikoSso that one only has to add it if modelselection is available18:17
@HeikoSbut no need to change that now18:17
@HeikoSvotjakovr: pls dont do all these whitespace changes, makes it hard to look through PRs18:17
votjakovrHeikoS: yep, i know, sorry18:17
@iglesiasgHeikoS: will take a look at LMNN again18:18
@iglesiasgcan it be that it was for a commit where develop was not rebased?18:19
@iglesiasggtg now18:19
@iglesiasgbut will check later18:19
@HeikoSiglesiasg: I dont know, but then this is on travis since its not possible to merge otherwise18:20
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: Lost terminal]18:22
-!- van51 [~van51@athedsl-409794.home.otenet.gr] has quit [Quit: Leaving.]18:32
@HeikoSvotjakovr: what is missing now?18:44
@HeikoSvotjakovr: I will be away on the weekend, can maybe help a bit with the notebook on late Sat or Sun, but I need your initial draft for that18:45
@HeikoSvotjakovr: its impossible for me to read through all of these changes, especially since there so many unreleated changes, so what are you working on currently?18:45
votjakovrHeikoS: i'm still working on CParameterCombination18:48
@HeikoSvotjakovr: whats after that?18:48
votjakovrHeikoS: gradient model selection will be working18:48
votjakovrHeikoS: then notebook18:49
@HeikoSvotjakovr: okay nice, I just skipped though the PR, seems al ok18:49
@HeikoSvotjakovr: but the problem is I cannot read this properly since my browser can't handle such a large page ;)18:49
shogun-buildbotbuild #1856 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/185618:49
@HeikoSvotjakovr: so let me know when the thing is final and all tests pass ( I will have  alook into the tests then) and then OK for merge18:50
@HeikoSI will be around for another 2hrs18:50
@HeikoSvotjakovr: what the time schedule you have?18:50
@HeikoSvotjakovr: actually travis passed18:50
votjakovrHeikoS: cool, i'd like to work until model selection is done18:51
@HeikoSokay18:52
@HeikoSvotjakovr: so I will update your notebook a bit on Sun then18:52
@HeikoSvotjakovr: let me know how the state is in 1-2 hrs ok?18:52
votjakovrHeikoS: ok, about examples: i drop them, because they are using old gradient model selection18:53
votjakovrHeikoS: it's not a problem to back them18:53
@HeikoSvotjakovr: I see, then pls rather deactivate them for now18:53
@HeikoSsince all these method need to be illustrated atsome point18:54
@HeikoSjust do something like #ifdef 018:54
@HeikoSand add a comment that they are currently disabled due to api changes but soon coming again18:54
votjakovrHeikoS: ok, i'll do18:54
shogun-buildbotbuild #1508 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1508  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>18:55
-!- lambday [67157c37@gateway/web/freenode/ip.103.21.124.55] has joined #shogun18:58
lambdayHeikoS: sorry I had to go because had a meeting with prof'18:59
lambdayHeikoS: I have some doubts regarding your suggestions.. mailing it19:00
shogun-buildbotbuild #1857 of deb3 - modular_interfaces is complete: Failure [failed test ruby modular]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/1857  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>19:09
@HeikoSlambday: ok let me know19:09
@HeikoSnothing there yet19:09
lambdayHeikoS: just a few things.. incorporating the rest of the things19:13
@HeikoSlambday: okay, keep in mind that these are just suggestions :)19:13
lambdayHeikoS: yeah but this will make it look cool :)19:13
lambdayHeikoS: how do I display the cholesky factors? using cholmod?19:14
lambdayusing shogun I cannot do that19:14
@HeikoSlambday: turn the matrix into dense and compute dense cholesky19:15
@HeikoSfor the small one, that should work19:15
lambdayHeikoS: so, for west0479 I can do that19:16
lambdayand you want the putting it all together part pulled above19:16
@HeikoSlambday: yeah exactly19:16
@HeikoSlambday: yeah I think so19:16
@HeikoSall these solvers are not crucial for using the method19:16
lambdaynah19:16
@HeikoSmore like an illustration how to use the tools you created19:17
lambdayyes19:17
lambdayalright..19:17
@HeikoSyou can introduce that as "All the methods needed to estimate the log-determinants in the above way can also be used on their own19:17
lambdayah that sounds great'19:17
lambdayHeikoS: oh one thing.. I didn't get that part when you said that the matrix I used for CG may not be psd19:18
lambdayI already did m^Tm+I19:18
lambdayand applied cg on that one19:18
@HeikoSah19:18
@HeikoSmaybe I missed that19:18
@HeikoSsorry then19:18
lambdaywhere m is randomly generated19:18
lambdayHeikoS: np.. oh and where would these notebooks go? (so that I can put a relative path to the data, gotta add apache2 and ozone though)19:19
@HeikoSthere is a directory for them ...19:19
@HeikoSin doc19:20
lambdayHeikoS: doc/ipython-notebooks19:20
lambdayfound it19:20
lambdayLD19:20
lambday:D19:20
@HeikoSlambday: just respect the directory structure for yours19:20
lambdayalright19:21
lambdayHeikoS: I wanted to add something for the ozone mcmc19:21
lambdayHeikoS: oh and one more thing... combining the normal and probing samplers is exactly what I did earlier19:21
lambdayHeikoS: but probing samplers have so less variance,that when I put it with Normal, it almost looks aligned with the actual logdet19:22
@HeikoSlambday: I see19:22
lambdaythat's why kept these two separate19:22
lambday:-/19:22
@HeikoSthen do subplot(121) #plot normal subplot(122) #plot probing19:22
@HeikoSjust do have them next to each other19:22
@HeikoSand plot histograms on top of each other19:23
@HeikoSwould be cool to have a histogram like on the poster19:23
@HeikoSwith both samples and the true value19:23
lambdayHeikoS: alright19:23
lambdaystill new to this matplotlib, fighting a bit :-/19:23
lambdayHeikoS: can I add external images (that are not generated plots from the code) in the notebook?19:24
@HeikoSlambday: yes, but why do you want to do that?19:24
@HeikoSlambday: ask me if you have problems with plotting19:24
lambdayHeikoS: I wanted to put a CG-M convergence plot based on the residuals19:24
@HeikoSlambday: can you actually access the residuals?19:25
@HeikoSlambday: thats a nice idea in fact19:25
@HeikoSCant you do that for a small matrix, extract the residuals and then just plot?19:25
lambdayHeikoS: externally, run the c++ in debug more, collect the residuals (thanks to you)19:25
lambdayHeikoS: how can I externally insert an image to a notebook.. it works locally, but while uploading it in the gist to the nbviewer, I don't find an option to upload any image19:26
@HeikoSlambday: I dont like this so much with the external picture19:27
@HeikoSlambday: though I like the idea about plotting convergence19:27
lambdayHeikoS: then I have to add something in the solvers itself to store the residuals :-/19:27
@HeikoSlambday: yeah, in fact that is useful19:28
@HeikoSlambday: but it should be off by default19:28
lambdayHeikoS: is it :-/19:28
lambdayalright19:28
@HeikoSlambday: yes, in cases one is interested in acessing residuals19:28
besser82HeikoS, wiking, sonne|work: Anyone of you interested in adding ViennaCL-support to SHOGUN?19:28
@HeikoSlambday: can this be done once? and then used form all solvers?19:28
lambdayHeikoS: alright then19:28
@HeikoSlambday: but maybe do other things first19:28
@HeikoSlambday: its not too important19:28
@HeikoSlambday: but the external picture is not really useful if there is not code how to generate it19:29
lambdayHeikoS: could you please explain a bit regarding the ozone model19:29
@HeikoSand since convergence diagnostics are always useful, should be in there19:29
lambdayHeikoS: yeah I understand19:29
@HeikoSlambday: ok, what to explain?19:29
lambdayHeikoS: I didn't get what these matrices represent?19:29
lambdayGiCG, C0, G A19:29
@HeikoSbesser82: maybe at some point :)19:30
@HeikoSbesser82: would be cool to have these kind of backend19:30
@HeikoSbesser82: though we are currently heavily using eigen319:30
@HeikoSlambday: ah19:30
@HeikoSlambday: thats not too important, it is the model19:30
@HeikoSso A is a piecewise linear triangulation of a sphere (the globe)19:30
lambday(with Q, we are not even using A, right?)19:31
@HeikoSit takes the latent gaussian field of the model (which lives on a sphere) and maps it to a plane19:31
@HeikoSlambday: there are two log-dets in the log likelihood19:31
@HeikoSonce involves A^T A19:31
besser82HeikoS, wiking, sonne|work: GTX285-SLI would slightly outperform recent Xeons, but on affordable $. :)19:31
-!- FSCV_ [~FSCV@187.191.50.146] has quit [Quit: Leaving]19:31
besser82HeikoS: eigenvector runs really speedy on ViennaCL. Just tried :)19:32
@HeikoSbesser82: thats really nice, but the problem is all the interfacing19:32
@HeikoSbesser82: we dont want to change everything again, just moved to eigen319:32
@HeikoSif eigen3 had a backend like that, that would be coo19:32
@HeikoScool19:32
@HeikoSand is rather the way to go19:33
@HeikoSah I just see there is eigen stuff in there19:33
besser82HeikoS: wouldn't be much for eigen3 --> ViennaCL.  ViennaCL has interface for eigen3  :)19:33
@HeikoSbesser82: so that might be an idea19:33
@HeikoSyep, that sounds cool19:33
-!- van51 [~van51@athedsl-409794.home.otenet.gr] has joined #shogun19:33
@HeikoSbesser82: maybe a gsoc project or so :)19:33
lambdaywhat's ViennaCL?19:34
@HeikoSlambday: dont worry for now :)  but here http://viennacl.sourceforge.net/viennacl-about.html19:34
besser82lambday: ViennaCL is BLAS, LAPACK/ATLAS, eigen3 on OpenCL/CUDA :)19:34
lambdayHeikoS: just curious :D19:34
@HeikoSlambday: so you could add a little section before or after the tools section19:34
@HeikoSwhich states the ozone model19:35
lambdayHeikoS: tools?19:35
@HeikoSsolvers etc19:35
@HeikoSand kind of motivates why we need the log-determinants19:35
@HeikoSand then runs the thing with very low max iterations19:35
@HeikoSto illustrate that it works on desktop computers19:35
lambdayHeikoS: why we need log-determinants --- is coming from the requirement of computing the log-likelihood itself, right?19:36
lambdayHeikoS: sorry I am not getting what you mean here :(19:36
-!- sonne|osx [~sonne@f053037117.adsl.alicedsl.de] has joined #shogun19:36
@HeikoSlambday: exactly, every GMRF needs log.determinants19:36
@HeikoSin fact every Gaussian distribution19:36
shogun-buildbotbuild #1509 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1509  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, lambday <heavensdevil6909@gmail.com>19:36
shogun-buildbotbuild #23 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/23  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>19:37
lambdayHeikoS: yes19:37
lambdaydarn you cyg1 :'(19:37
@HeikoSbut the precision matrix is sparse in every GMRF19:37
@HeikoSso thats just an example19:37
@HeikoSyou can point to the Aune et al paper19:37
@HeikoSor even to this one19:37
lambdaybesser82: sounds really interesting19:38
lambday:D19:38
lambdaybesser82: I have no clue regarding the last two OpenCL/.CUDA :D19:38
lambdayHeikoS: :-/19:38
@HeikoShttp://arxiv.org/pdf/1104.3436.pdf19:39
@HeikoSpage 22 bottom19:39
@HeikoSyou could reference this19:39
besser82lambday: it surely is. openCL / CUDA is distributed computing using your GPU for performing.19:39
lambdayHeikoS: checking19:39
besser82lambday: CUDA is sth. specific to nvidia-cards19:39
@HeikoSbesser82: this stuff is only interesting if existing libraries do this in the back19:39
@HeikoSbesser82: otherwise its way too technical to use19:40
@HeikoSbesser82: i played with GPUs a bit a few years ago19:40
@HeikoSand without std libs this is pure horror ;=19:40
besser82HeikoS: ViennaCL should give simple interfaces to use for sci-tech tasks :)19:40
@HeikoSbesser82: so if we can keep our eigen3 calls as they are and just do some magic in the background, I am really up for this!19:41
besser82HeikoS: http://viennacl.sourceforge.net/viennacl-examples-eigen.html19:41
besser82HeikoS: nothing much to change :)19:41
lambdayHeikoS: I need to study :-/ what are these parameters :-/19:41
lambdayHeikoS: I thought giving only your paper as reference for ozone19:42
@HeikoSlambday: so, what I suggest, is just to state the model (as in the poster), mention the log-likelihood, and then show that one can compute the likelihood using shogun19:42
@HeikoSlambday: https://github.com/karlnapf/ozone-roulette/blob/master/ozone/distribution/OzonePosterior.py19:43
@HeikoSjust take the log_likelihood method and use log_det_estimate_shogun and solve_sparse_linear_system_shogun19:43
lambdayHeikoS: and the outputs/images?19:44
lambdayoh okay19:44
@HeikoSand say that for illustration, you only estimate the log-det but solve the linear system with cholesky19:44
besser82HeikoS: so just copying in/out eigen3-datatypes and performing tasks on them.19:44
@HeikoSlambday: images would be nice19:44
@HeikoSlambday: the plotting code is in matlab though, let me check how we can get the map19:44
lambdayHeikoS: no no.. I only want to put that much part that I myself understand :(19:46
@HeikoSlambday: ok then, just state the model, it has an easy form since it is a std model, it just looks technical19:47
@HeikoSso if you tell the distributions of x, y|y, tau, kappa, and then state the log-likelihood that is ok19:47
@HeikoSand then you can plot y if you want19:47
@HeikoSlambday: just if you want of course19:47
@HeikoSlambday: ill send you code to plot the observations of the ozone model19:47
lambdayHeikoS: ah thanks.. :)19:48
@HeikoSlambday: the idea here is that you just say: "Here is a sparse Gaussian model for which this stuff is useful"19:48
lambdayyes19:48
@HeikoSand the likelihood function in the file I sent you is just directly turning the log-likelihood math into numbers, no tricks there19:48
shogun-buildbotbuild #10 of clang34 - undefined behaviour analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/10  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>19:48
shogun-buildbotbuild #1858 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/185819:49
lambdayHeikoS: the log_likelihood method in the OzonePosterior?19:50
@HeikoSyes19:50
@HeikoSlambday: check it, it just computes the log-likelihood on the poster19:50
@HeikoSlambday: I have not plotting code in python and its non-trivial to plot19:54
@HeikoSlambday: thinking on what to do ....19:54
lambdayHeikoS: let's chuck it19:54
lambdayHeikoS: I'll add these things later on this notebook when I'll work on more on this..19:54
@HeikoSlambday: okay19:55
lambdayHeikoS: I really liked the issues that you posted, parallel computation engine and one other19:55
lambdayHeikoS: I would like to work on that19:55
@HeikoSlambday: yeah we can work on that19:55
lambdayHeikoS: plus there are many many tricks that the paper suggested, I'd like to explore these too19:55
shogun-buildbotbuild #165 of osx1 - libshogun is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/osx1%20-%20libshogun/builds/16519:55
@HeikoSlambday: but I think it still would be nice to say: Here is a std GMRF model and its log likelihood. Here we load some ozone data, see reference paper, and here we can compute the log-likelihood19:56
@HeikoSlambday: its ultra cool even without images19:56
@HeikoSso I would totally suggest to add that19:56
lambdayHeikoS: so, computing the likelihood thing, I can use your code? :D19:56
@HeikoSlambday: yeah sure19:56
@HeikoSits GPL :)19:57
lambdayhehe19:57
lambdayalright19:57
lambdayyeah it would be really cool19:57
lambdayso, the structure would be ---intro --- log-det on apache2, walking through the components --- ozone --reference19:58
lambdaynot just apache2, but also west047919:58
lambdaywith fill-in19:58
lambdayI really want to put apache2 fill in cause as per source that shows worst fill-in19:59
shogun-buildbotbuild #17 of debian wheezy - memcheck is complete: Failure [failed compile]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/17  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, lambday <heavensdevil6909@gmail.com>20:02
shogun-buildbotbuild #14 of clang34 - thread analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/14  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>20:03
@HeikoSlambday: okay!20:03
@HeikoSmaybe even first ozone and then components20:03
@HeikoSlambday: but as you like20:03
-!- lambday [67157c37@gateway/web/freenode/ip.103.21.124.55] has quit [Ping timeout: 250 seconds]20:03
@HeikoSlambday: I have to go now, will be in touch on Sunday20:03
@HeikoSvotjakovr: I am going home now, checking later today, pls send me an email with the status today when you stop working. Good luck!20:06
-!- HeikoS [~heiko@nat-163-62.internal.eduroam.ucl.ac.uk] has left #shogun []20:06
-!- zxtx [~zv@149.165.244.2] has joined #shogun20:21
-!- votjakovr [~votjakovr@host-46-241-3-209.bbcustomer.zsttk.net] has quit [Quit: Good bye!]20:24
shogun-buildbotbuild #24 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/24  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, lambday <heavensdevil6909@gmail.com>20:42
shogun-buildbotbuild #11 of clang34 - undefined behaviour analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/11  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, lambday <heavensdevil6909@gmail.com>20:52
-!- zxtx [~zv@149.165.244.2] has quit [Ping timeout: 245 seconds]20:53
shogun-buildbotbuild #15 of clang34 - thread analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/15  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, lambday <heavensdevil6909@gmail.com>21:05
-!- lisitsyn [~lisitsyn@109-226-79-195.clients.tlt.100megabit.ru] has joined #shogun21:09
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]21:14
-!- travis-ci [~travis-ci@ec2-23-23-31-73.compute-1.amazonaws.com] has joined #shogun21:33
travis-ci[travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1160025521:33
-!- travis-ci [~travis-ci@ec2-23-23-31-73.compute-1.amazonaws.com] has left #shogun []21:33
-!- iglesiasg [~iglesias@2001:6b0:1:1da0:cc71:7583:fe5b:b7d1] has joined #shogun21:37
-!- mode/#shogun [+o iglesiasg] by ChanServ21:37
@iglesiasgsonney2k, lisitsyn ping ping21:39
lisitsynja21:39
lisitsyniglesiasg: hello21:39
@iglesiasglisitsyn, how are you?21:39
lisitsyniglesiasg: fine21:39
lisitsyniglesiasg: what about you?21:39
@iglesiasgLMNN is giving headache :D21:39
lisitsyniglesiasg: what's with it?21:39
@iglesiasghttps://travis-ci.org/shogun-toolbox/shogun/jobs/11599254#L477221:39
@iglesiasgintegration test21:40
@iglesiasgthe 16th decimal or so is different :D21:40
lisitsynhah21:40
@iglesiasglisitsyn, can that be Eigen related?21:41
lisitsyniglesiasg: no clue :)21:41
lisitsyniglesiasg: is that ok accuracy?21:42
lisitsynI mean 16th sign is a bit too much21:42
lisitsynbut who knows21:43
@iglesiasglisitsyn, a bit too much?21:43
@iglesiasgI think the difference is insignificant21:43
lisitsyniglesiasg: looks too precise21:43
@iglesiasglisitsyn, I have just run it more than 100 times here in my machine and it is always OK.21:49
@iglesiasglisitsyn, I guess I could just deactivate de integration test21:49
lisitsyniglesiasg: I am not sure what to do about it21:50
@iglesiasgsome of the parts are already tested in unit tests21:50
@iglesiasgit is fine in the  buildbot as well21:51
sonne|osxlisitsyn: please disable integration tests on travis21:53
sonne|osxit is sufficient to have this on the buildbots21:54
lisitsynsonne|osx:21:54
lisitsynok21:54
@iglesiasgI like that initiative!21:54
@iglesiasgthank you guys21:54
lisitsynI like to disable/delete21:54
lisitsyn:D21:54
sonne|osxlisitsyn: can you do the gdb thing?21:58
lisitsynsonne|osx: yes21:58
sonne|osxlisitsyn: https://github.com/shogun-toolbox/shogun/issues/164321:58
lisitsynI am on it just after integration tests21:58
sonne|osxthis would help *a lot* to fix e.g. cyg121:58
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun22:00
shogun-notifier-shogun: Sergey Lisitsyn :develop * 1564b47 / .travis.yml,CMakeLists.txt: https://github.com/shogun-toolbox/shogun/commit/1564b47cc11cc5c1a3c6a9de6a8027a0522dcc2322:00
shogun-notifier-shogun: Disabled integration tests on travis22:00
@iglesiasglisitsyn, thanks!22:06
lisitsyniglesiasg: willkommen22:07
@iglesiasglisitsyn, how is your German going? :)22:08
lisitsyniglesiasg: schlecht22:08
* iglesiasg translates22:08
@iglesiasghehe why so?22:08
lisitsyniglesiasg: well I am having some time management troubles22:13
lisitsyn;)22:13
lisitsyniglesiasg: I have absolutely no idea how sonne|osx cope with that22:13
-!- travis-ci [~travis-ci@ec2-23-23-31-73.compute-1.amazonaws.com] has joined #shogun22:16
travis-ci[travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1160036522:16
-!- travis-ci [~travis-ci@ec2-23-23-31-73.compute-1.amazonaws.com] has left #shogun []22:16
-!- zxtx [~zv@149.165.244.2] has joined #shogun22:35
lisitsynoh gosh wiking you have done crazy amount of work with cmake22:40
besser82lisitsyn: araound?22:40
lisitsynbesser82: yes what's up?22:40
besser82lisitsyn: I've seen you're adding ViennaCL to tapkee...22:41
lisitsynbesser82: I've made some experiments22:41
besser82lisitsyn: Any motivation to bring this up to SHOGN?22:41
lisitsynbesser82: well I should update tapkee in shogun yeah22:41
lisitsynbesser82: the problem is that speedup is not really worth the game22:42
besser82lisitsyn: for tapkee or shogun?22:42
lisitsynbesser82: the only thing I have seen is some 1.5x speedup with diffusion maps22:42
lisitsynbesser82: for shogun it makes sense22:42
lisitsynbut needs proper design22:43
besser82lisitsyn: eigenvectors do get really fast with it :)22:43
lisitsynbesser82: dense?22:43
besser82lisitsyn: both sparse and dense22:43
lisitsynbesser82: full eigendecomposition?22:43
besser82lisitsyn:  not tried yet22:43
lisitsynbesser82: I mean is it full or partial?22:43
besser82lisitsyn:  should be full22:44
lisitsynbesser82: that's the critical point because in tapkee full eigendecomposition is never used22:44
lisitsynsay you are computing LLE embedding22:44
lisitsynyou have N x N sparse matrix22:44
lisitsynbut to compute 2d embedding you need *only* 3 smallest eigenvectors22:44
shogun-buildbotbuild #1510 of cyg1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/cyg1%20-%20libshogun/builds/1510  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>22:45
besser82lisitsyn: so partial...22:45
lisitsynyeah so I need partial22:45
lisitsynthat's why I am tied to arpack :)22:45
besser82lisitsyn: but ViennaCL should support both22:45
lisitsynlet me check22:45
lisitsynI don't mind to put it if it is here22:45
lisitsynI tried to make it modular22:45
lisitsynbesser82: true, lanczos is here22:46
besser82lisitsyn: :D22:46
lisitsynbesser82: ah one more problem - I don't have any gpu ;)22:47
besser82lisitsyn: pocl is your freind :)22:47
besser82lisitsyn: pocl is openCL on cpu only :)22:47
lisitsynbesser82: well no, my cpu supports opencl natively22:47
lisitsynthat's sandy bridge22:47
lisitsynbut hey that's essentially the same22:47
lisitsynI mean I can't see the speedup22:47
besser82lisitsyn: because of no GPU :)22:48
besser82lisitsyn: GTX285 SLI nearly oupterforms recent XEON !!!22:48
besser82lisitsyn: and is really affordable $22:48
lisitsynbesser82: okay thanks for the tip, I'll try to put lanczos to tapkee22:48
besser82lisitsyn: :D22:48
lisitsynbesser82: well I am kind of mobile so I used to use notebook22:49
lisitsynI am switching cities sometimes :D22:49
lisitsynlike today22:49
besser82lisitsyn: me, too. But mine has GTX580 SLI :D22:49
besser82lisitsyn: in notebook :D22:49
lisitsynbesser82: .. and your notebook weights like 8 kg I guess ;)22:50
besser82lisitsyn: nope ~5 kg  :)22:50
lisitsynoh damn22:50
besser82but with 17" screen :D22:50
besser82and 3 HDD22:50
besser82if i need ultra-mobility i use my old q45 from samsung  ---> ~1.5 kg22:51
lisitsynwell you can save some money on fitness carrying tihs bag of potato22:51
lisitsyn:D22:51
besser82sure :D22:51
lisitsynbesser82: anyway I just don't need gpu computing22:52
lisitsynthat's the main reason22:52
lisitsyn:)22:52
besser82if you don't need then all is fine :)22:52
lisitsynI played with it out of curiousity but it doesn't look I need it22:52
besser82but in shogun22:52
besser82it would be a real option :D22:52
lisitsynI don't mind to spend some time on that if there is real demand22:53
besser82combined with sth. like openShift I it would be great...22:53
-!- zxtx [~zv@149.165.244.2] has quit [Ping timeout: 240 seconds]22:53
lisitsynbesser82: I see a few problems there22:53
besser82lisitsyn: which?22:54
lisitsyn1) I don't know how to design heterogeneous software22:54
lisitsyn2) I don't know how to deal with limited memory here22:54
lisitsynI mean say you have kernel svm22:54
lisitsynyou need to have all the features in gpu memory22:54
lisitsynotherwise you'll spend most of the time copying stuff back and forth22:55
besser82won't 1.5 Gbytes be enough?22:55
lisitsynbesser82: 1.5gb seem to be enough yeah22:55
shogun-buildbotbuild #18 of debian wheezy - memcheck is complete: Failure [failed compile]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/18  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>22:55
besser82lisitsyn: any recent gtx285+ has at least 1gig22:56
lisitsynoh I have no clue about these numbers ;)22:56
lisitsynwell yeah 1gb should be ok22:56
besser82lisitsyn: gtx285 is from ~ 4years ago ;)22:56
besser82lisitsyn: today's GTX{5,6,7,8}** are all <= 2Bytes of mem22:57
lisitsynbesser82: I haven't had any non-notebook computer nearly the same time22:57
besser82lisitsyn: today's GTX{5,6,7,8}** are all >= 2Bytes of mem22:57
lisitsynlike once I started my studies22:57
lisitsynheh22:57
besser82desktops have these specs as well22:58
lisitsynbesser82: I have to get back to the fact I don't need it :D22:58
besser82lisitsyn: ok, ok :D22:58
lisitsynbesser82: but if I get some hardware like that I'll try it for sure22:58
lisitsynbesser82: we should employ gpu to compile shogun to make it fast :D23:06
besser82lisitsyn: using gpu on compiling or compiling with support for gpu-use?23:07
lisitsynbesser82: jk :) it is so slow23:08
besser82lisitsyn: jk?23:08
lisitsynbesser82: just kidding23:08
besser82lisitsyn: kk23:08
besser82lisitsyn: but compiling isn't that slow takes ~7 minutes23:09
besser82lisitsyn: try compiling Mahout :D23:09
besser82lisitsyn: will run ~40 minutes on i723:09
lisitsynbesser82: really?23:12
besser82lisitsyn: yes23:12
lisitsynbesser82: but its java, how can it be that slow?23:13
lisitsynI am totally surprised23:14
besser82lisitsyn: because of maven23:14
besser82lisitsyn: hadoop takes ~1h23:14
lisitsynbesser82: what makes it that slow with maven?23:14
besser82lisitsyn: not sure...23:15
besser82lisitsyn: no idea23:15
lisitsynbesser82: sounds so weird!23:15
lisitsynI am really surprised because I was used to think java = rather fast compilation23:15
besser82lisitsyn: not even close :D23:16
lisitsynC++ is terribly slow when it comes to expression templates like in eigen323:16
besser82lisitsyn: that's true, but java is always slow :)23:17
lisitsynwell at my job we have quite good performance with java23:18
lisitsynwe are heavily using message passing with akka and I can't say it is slow23:18
lisitsynwell and I used to compile terribly big modules at my previous job and it was ok too23:19
lisitsynthat's why I am surprised23:19
besser82lisitsyn: I really don't know what the apache-people did to it, making it that slow :(23:20
lisitsynbesser82: when I hear apache I think of cemetary23:20
lisitsynlike they have zillions of dead projects23:21
besser82lisitsyn: but Mahout is one of the recent ones :) still in beta23:21
besser82lisitsyn: or RC now23:21
lisitsynyeah that's not dead23:21
lisitsynI have used http://commons.apache.org/proper/commons-scxml/23:22
lisitsynbut I didn't know what to think with last commit in ~200723:22
lisitsynsonne|osx: did you mean both integration and unit tests?23:27
-!- besser82 [~besser82@fedora/besser82] has quit [Quit: freedom, friends, features, first ---> fedoraproject.org]23:30
shogun-buildbotbuild #25 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/25  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>23:33
shogun-buildbotbuild #12 of clang34 - undefined behaviour analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/12  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>23:43
shogun-buildbotbuild #16 of clang34 - thread analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/16  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>23:57
--- Log closed Sat Sep 21 00:00:08 2013