IRC logs of #shogun for Tuesday, 2013-08-06

« Monday, 2013-08-05 Index Wednesday, 2013-08-07 »
--- Log opened Tue Aug 06 00:00:00 2013
--- Day changed Tue Aug 06 2013
@iglesiasgthat is just the README.md00:00
shogun-notifier-shogun: Viktor Gal :feature/CMake * b180b08 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/b180b084be7d3b1b7d461a6556da477ad7421d8000:00
shogun-notifier-shogun: On the way to fix octave_modular interface00:00
hushelliglesiasg: yeah :(00:00
@iglesiasghushell, in the clone statement?00:00
hushelliglesiasg: so the clone_MAPInfeence unit-test failed I think00:00
hushellyep00:00
hushellIIRC, clone_ptype failed00:01
@iglesiasghushell, so it looks like you are not initializing the members in the constructors of CMAPInference, right?00:01
@iglesiasgfor instance after the default constructor00:01
@iglesiasgwell, not after, in the default constructor00:02
hushellokay, I see in default constructor the pointer member has not been initialized00:03
hushelllet me try to fix this00:03
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has joined #shogun00:05
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/987479100:05
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has left #shogun []00:05
@sonney2kHeikoS, yeah so it is nicely possible. We just use README.md and then have links to README_DEVELOPER.md via https://help.github.com/articles/relative-links-in-readmes00:05
@iglesiasghushell, well none of the members is initialized, right?00:05
@HeikoSsonney2k: yeah this would be cool00:05
@HeikoSthen its synched *and*  looks nice/easy to access00:05
@sonney2kHeikoS, etc ... not really a wiki though00:05
@sonney2kyou can only edit the text document from git00:06
@HeikoSsonney2k: well a list of things/links would already be good00:06
@HeikoSsonney2k: ah .... true00:06
@sonney2kHeikoS, btw doxygen will also render the README.md etc pages - it also understands markdown syntax00:07
@sonney2kso we could have it on the github front page and in our doxygen doc00:07
@sonney2kjust editing a single file00:08
shogun-buildbotbuild #1472 of deb3 - modular_interfaces is complete: Failure [failed compile java_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1472  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>00:08
hushelliglesiasg: hmm, a big mistake, now I set them to NULL initially00:09
@iglesiasghushell, did it fix it?00:09
hushelliglesiasg: compiling00:10
hushelliglesiasg: you could close the PR, I'll send new ones00:10
@sonney2kiglesiasg, I would just rename log_pdf to log_pdf_single or so to fix this00:10
@iglesiasghushell, all right00:10
@iglesiasgsonney2k, but why is it that SGVector sometimes goes to RealVector and sometimes to DoubleMatrix in java?00:11
@HeikoSsonney2k: I renamed that thing already00:13
@iglesiasgHeikoS, cool!00:14
@iglesiasgalso the other error that gave problems in java_modular?00:14
@sonney2kiglesiasg, no clue00:15
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has joined #shogun00:17
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/987538300:17
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has left #shogun []00:17
@sonney2kHeikoS, and the error is gone or?00:21
@HeikoSsonney2k: dont know but should00:21
@HeikoSname is different now00:21
@sonney2kHeikoS, seems not http://shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1472/steps/compile%20java_modular/logs/stdio00:22
@iglesiasgI cam compile and check00:22
@iglesiasgno need then00:22
@HeikoSthis error mkaes no sense00:23
@iglesiasgsonney2k, HeikoS, it compiled fine there this time00:25
@HeikoSbuildbot weirdness?00:27
@sonney2kshogun-buildbot, force build "deb3 - modular_interfaces" --branch=develop00:27
shogun-buildbotbuild forced [ETA 38m22s]00:27
shogun-buildbotI'll give a shout when the build finishes00:27
@sonney2klets see00:27
@HeikoSwhoooooooooo00:27
@HeikoSwe now have a working script that runs and stores and ipython notebook00:28
@HeikoSwiking:  ^00:28
@wiking\o/00:29
@sonney2kHeikoS, where?00:30
@sonney2khow ?00:30
@sonney2kmore details!00:30
@HeikoShttps://gist.github.com/minrk/616006700:30
@HeikoSI got in contact with one of the devs00:30
@HeikoShe updated the old ones for us :)00:30
@sonney2kvery good00:31
@HeikoSjust tried it, and it works00:31
hushelliglesiasg: Thanks! clone is good after initialization :)00:31
@iglesiasghushell, good to know00:31
@HeikoSso we should integrate this into old system but also cmake00:31
@sonney2kHeikoS, so how would we want that to work?00:31
@HeikoSsonney2k: my idea is: we have a dir with couple of notebooks00:31
@sonney2krun it on buildbot in nightly_default?00:31
@HeikoSsonney2k: yes00:31
@HeikoSit runs them all and stores resulting ones, and uploads those to our website00:32
@sonney2kand then have the python code run to check stuff works and then put this on website somewhere?00:32
@HeikoSyep00:32
@HeikoSand then on our website, we have a manually created site that lists all of the examples in some order00:32
@HeikoSand provides links to the rendered version00:32
@HeikoSvia http://nbviewer.ipython.org/00:33
@HeikoShttp://nbviewer.ipython.org/url/herrstrathmann.de00:33
@HeikoSpretty easy to use00:33
@sonney2kHeikoS, hmhh doesn't work here00:34
@sonney2kI always get the error00:35
@sonney2k    km = KernelManager()00:35
@sonney2kNameError: global name 'KernelManager' is not defined00:35
@HeikoSsonney2k: it only works on current git ipython00:35
@HeikoSyou have 0.1300:35
@sonney2khmmhh00:35
@HeikoSbut its very easy to install00:35
@HeikoSjust clone and run setup.py build00:35
@HeikoStakes 2 s00:35
@iglesiasgHeikoS, I am having a look at the GaussianDistribution unit test that hangs on my machine (and maybe in travis too). It happens in an Eigen statement. What version of eigen do you use?00:35
@HeikoSthen sudo setup.py install00:35
@iglesiasgit looks pretty weird00:36
@HeikoSiglesiasg: yeah I had issues with old eigen version00:36
@HeikoSiglesiasg: I have 3.100:36
@HeikoSwhich was released last year00:36
@iglesiasgaaham I see00:36
@iglesiasgHeikoS, did you just download it from the webpage?00:36
@HeikoSiglesiasg: ubuntu00:36
@iglesiasgmmm I have that one too00:37
@iglesiasgHeikoS,  PPA?00:37
@HeikoSgnhcyj8ju!00:37
@HeikoSnice00:37
@HeikoSthats my root password00:37
@sonney2khttps://github.com/ipython/ipython.git00:37
@iglesiasghahahaha00:37
@HeikoScan I delete that somehow from the logs?00:37
@sonney2klets all ssh to HeikoS machine!00:38
@HeikoSsonney2k:  yes00:38
@sonney2kHeikoS, better change it NOW00:38
@HeikoSjust did00:38
@HeikoSah annoying00:39
@HeikoSiglesiasg: just because I opened synaptic ;)00:40
@HeikoS3.1.0~beta1-1ubuntu100:40
@iglesiasgHeikoS, sorry that I caused that :(00:40
@HeikoSdont worry00:40
@HeikoStoo late00:40
@HeikoSstill at work00:40
@HeikoSso I will go home now :)00:40
@iglesiasgdpkg -l | grep next time00:41
@sonney2kwiking, when will you have a nice uplink?00:41
@HeikoSiglesiasg: I like synaptic :)00:41
@iglesiasghehe ok :)00:41
@iglesiasgmaybe we should update travis' Eigen as well00:43
@HeikoSiglesiasg: yes that would be good00:43
@iglesiasgI have the feeling it is continuously dying in there00:43
@HeikoSI got problems with the old old version00:43
@sonney2kHeikoS, ok the thing runs now00:43
@HeikoSbut -- cmake will fix that00:43
@HeikoSsonney2k: nice00:43
@sonney2kHeikoS, how do I get it on a website?00:43
@iglesiasgcool then00:43
@HeikoSyou need to have a fixed http link to the file00:43
@HeikoSso upload to shogun - server directory and pass that link to the nbviwer00:44
@sonney2kHeikoS, ahh so we can just use the nbviewer then?00:44
@HeikoSyes00:44
@HeikoSthen we need to generate a little script that  creates a website in some fashion00:44
@sonney2kwhich should I upload the run_saved.ipynb?00:44
@HeikoSbut I think I would do this by hand, i.e. we have a growing list of notebooks somewhere for each category00:45
@sonney2kor the one in our dir?00:45
@HeikoSsonney2k:  yes00:45
@HeikoSno the saved00:45
@HeikoScan also be deleted afterwards00:45
@sonney2kwhat is the difference?00:45
@HeikoSsonney2k: the saved one contains output00:45
@HeikoSsonney2k:  we should not store output in the notebooks00:45
@HeikoSoh there is also a script for removing output00:45
@HeikoShttps://gist.github.com/minrk/371984900:46
@HeikoSwhich we should call on the existing ones to clean up00:46
@HeikoSsonney2k: if we have this list of notebooks, we could then generate a website based on a template00:47
@HeikoSthat inserts notebooks into categories or whatever00:47
@HeikoSlike GSoC 201300:48
@HeikoSok I really gotta go now00:48
@HeikoSsee you!00:48
-!- HeikoS [~heiko@nat-160-11.internal.eduroam.ucl.ac.uk] has quit [Quit: Leaving.]00:48
@sonney2khttp://nbviewer.ipython.org/url/shogun-toolbox.org/static/notebook/svm_binary_classification_run_saved.ipynb00:48
@iglesiasgnice00:48
@iglesiasglot of import errors00:49
@iglesiasgI guess that is my thing though00:49
@iglesiasgnice! I got unit tests working good again00:57
shogun-buildbotbuild #1473 of deb3 - modular_interfaces is complete: Failure [failed install test ruby_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/147301:07
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has joined #shogun01:15
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/987674601:15
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has left #shogun []01:15
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has joined #shogun01:29
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/987910101:29
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has left #shogun []01:29
hushelliglesiasg: Do you know why after git commit --amend, git push always say error: failed to push some refs, and I have to use --force01:43
-!- foulwall [~user@110.16.165.220] has quit [Ping timeout: 264 seconds]01:43
@iglesiasghushell, it tries to prevent you from rewriting past history that is already written in there01:43
hushelliglesiasg: so --force is correct?01:45
@iglesiasghushell, in this case, it is fine01:45
@iglesiasghushell, another possible thing is to delete the remote branch and push again01:47
@iglesiasgalthough I suspect that git push --force might do something like that01:47
hushelliglesiasg: hmm, I prefer to keep every branch currently01:49
@iglesiasghushell, you delete it, and then you push again01:50
@iglesiasgwithout doing --force01:50
hushelliglesiasg: do you think will be okay to put IGNORE_IN_CLASSLIST in front of CMosek?01:54
hushellso it will not be in the clone test01:54
@iglesiasghushell, yeah sure01:54
@iglesiasgit is crashing there?01:54
hushellstill01:54
@iglesiasgoh yeah, I never looked at it at the end01:54
@iglesiasgsorry01:54
hushellno matter01:54
hushellI am curious how you prevent LMNNImpl from class_list01:55
@iglesiasghushell, why did you have to remove const in get_factor_type?01:55
@iglesiasghushell, you mean why is LMNNImpl not included in SWIG interfaces?01:56
hushelliglesiasg: I know this is because you add a line in modshogun_ignores.i01:57
hushellbut they are not it base/class_list.cpp01:57
hushellbut you do not put IGNORE_IN_CLASSLIST in front01:57
hushelliglesiasg: let me check the get_factor_type01:58
@iglesiasghushell, maybe because LMNNImpl is not a SGObject? I have no idea really01:58
@iglesiasghushell, I am just curious about it :)01:58
@iglesiasghushell, and btw, LMNNImpl does not need to be in modshogun_ignores.i, I just don't include it anyway in interfaces/modular/*.i01:58
@iglesiasghushell, it seems  that our SWIG is configured not to compile everything by default, so I just didn't include LMNNImpl anywhere there01:59
@iglesiasgjust LMNN01:59
@iglesiasgall right, time to get to the bed02:00
@iglesiasggood night!02:00
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: Ex-Chat]02:01
hushellgood night!02:01
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]03:00
-!- foulwall [~user@110.16.165.220] has joined #shogun03:09
-!- foulwall [~user@110.16.165.220] has quit [Read error: Connection reset by peer]03:14
shogun-buildbotbuild #415 of nightly_all is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_all/builds/41503:40
shogun-buildbotbuild #480 of nightly_default is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/48004:08
shogun-buildbotbuild #425 of nightly_none is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_none/builds/42504:20
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has joined #shogun05:16
-!- foulwall [~user@110.16.165.220] has joined #shogun06:58
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun07:41
shogun-notifier-shogun: Viktor Gal :feature/CMake * c64099b / .travis.yml,CMakeLists.txt,cmake/FindOctave.cmake: https://github.com/shogun-toolbox/shogun/commit/c64099bc9a05f9603cc2257eef4475a1359370aa07:41
shogun-notifier-shogun: Travis: fix FindOctave and remove python_static test07:41
shogun-notifier-shogun: Add python interpreter as required07:41
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has joined #shogun08:00
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/988799708:00
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has left #shogun []08:00
@sonney2kfoulwall, could you please add an option in the dictionary where one can explain the parameters so that one can display some information below (maybe some '?' sign to click that pulls out some help)08:18
shogun-notifier-shogun: Soeren Sonnenburg :develop * ad16e5d / src/shogun/ (4 files): https://github.com/shogun-toolbox/shogun/commit/ad16e5d32607a7b2caf4dfdb014c9ae11e5de47808:34
shogun-notifier-shogun: rename get_log_probability_f to get_log_probability_fmatrix to fix java modular compile error08:34
shogun-notifier-shogun: Viktor Gal :feature/CMake * a7d019a / CMakeLists.txt,cmake/FindR.cmake,src/interfaces/r_modular/CMakeLists.txt: https://github.com/shogun-toolbox/shogun/commit/a7d019a2b630f4feff05a054d916e3a6df42bdf908:35
shogun-notifier-shogun: Refactor R detection08:35
shogun-notifier-shogun: Viktor Gal :feature/CMake * bdb942b / .travis.yml,cmake/FindR.cmake,src/interfaces/r_modular/CMakeLists.txt: https://github.com/shogun-toolbox/shogun/commit/bdb942bdfef102bf32d8f793cb3abf5b279d728608:35
shogun-notifier-shogun: Fix R includes and try to force python3 on travis08:35
shogun-buildbotbuild #1667 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb1%20-%20libshogun/builds/1667  blamelist: Soeren Sonnenburg <sonne@debian.org>08:39
foulwallok sonney2k08:54
@wikingfuck08:56
@wikingdont fucking DO THIS!!!08:56
@wikinghttps://github.com/shogun-toolbox/shogun/pull/1370/files#diff-008:56
sonne|workwiking: haha whats wrong?08:57
@wikingrealised that travis has preinstalled ruby and python08:58
@wikingso somehow now i have to deal with this08:58
shogun-notifier-shogun-data: Zhengyang Liu :master * a4c963e / tapkee/mml.txt,tapkee/words.dat: https://github.com/shogun-toolbox/shogun-data/commit/a4c963e96e51453c33fef9a90ac0f0b247614d1308:58
shogun-notifier-shogun-data: add origin dataset for dimred08:58
shogun-notifier-shogun-data: Soeren Sonnenburg :master * 2a44741 / tapkee/mml.txt,tapkee/words.dat: https://github.com/shogun-toolbox/shogun-data/commit/2a44741e27c4dfa7a35d74f333340ed267a07dcc08:58
shogun-notifier-shogun-data: Merge pull request #11 from ZhengyangL/master08:58
shogun-notifier-shogun-data:08:58
shogun-notifier-shogun-data: add origin dataset for dimred08:58
@wikingand apt-get -ing stuff (python and ruby) is a big nono08:58
@wikingon travis08:58
shogun-notifier-shogun-demo: Zhengyang Liu :master * 9509e80 / / (8 files): https://github.com/shogun-toolbox/shogun-demo/commit/9509e8039b95775562d5d3444e0fdcddc49c9b0008:59
shogun-notifier-shogun-demo: 1. make k argument available to change for tapkee/words demo. 2.little fixes08:59
shogun-notifier-shogun-demo: Soeren Sonnenburg :master * d1e39eb / / (8 files): https://github.com/shogun-toolbox/shogun-demo/commit/d1e39eb5403ddae80c06172cc4d031a847ebd3d408:59
shogun-notifier-shogun-demo: Merge pull request #19 from ZhengyangL/master08:59
shogun-notifier-shogun-demo:08:59
shogun-notifier-shogun-demo: interactive tapkee demos08:59
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has joined #shogun09:00
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/988915109:00
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has left #shogun []09:00
sonne|workwiking: but he refers to eigen3?!09:00
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun09:02
-!- mode/#shogun [+o iglesiasg] by ChanServ09:02
shogun-notifier-shogun: Soeren Sonnenburg :develop * c2fa818 / tests/unit/machine/gp/LogitLikelihood_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/c2fa818be63cf402060266c61f354c5131e1b69509:03
shogun-notifier-shogun: fix renaming also in unit test09:03
@wikingsonne|work: yes09:04
@wikingsonne|work: he should do the eigen3 fetcher for cmake09:04
@wikinglet's fucking freeze *now* hacking with travis and ./configure09:04
@wikingplz09:04
@wikingotherwise i'll never catch up09:04
sonne|workwiking: well merge your stuff and we can help09:05
@wikingsonne|work: the branch is freely writeable by any developer09:06
sonne|workwiking: yes but converting the build bots to use cmake will make the difference09:06
@wikingsonne|work: why?09:07
@wikingsonne|work: i mean now not even travis runs fine09:07
@wikingso why would be good now to even add the buildbots09:07
@wikingi prefer to fix at least the compilation on travis09:07
@wikingso at least that works09:07
sonne|workwhat is wrong with travis?09:07
@wikingRModular, RubyModular, PythonModular (python3)09:08
@wikingand OctaveModular09:08
@wikingin these builds09:08
@wikingthere's still something wrong09:08
@wikingnot test wise09:08
@wikingcmake wise...09:08
@wikingeither some env variable is missing or something09:08
sonne|workwiking: I am not talking about travis09:08
@wikingcmake09:08
@wikingmostly these are problems with cmake09:08
sonne|workwiking: I want to convert the buildbots to slowly use this09:08
@wikingbut can be seen that it's broken on travis09:08
sonne|workfirst the libshogun based ones09:09
@iglesiasgwiking, Travis build freezes all the time because there is a test that does not finish with the version of Eigen in Travis09:09
sonne|workthen static and test09:09
@wikingiglesiasg: ok good09:09
@iglesiasgwiking, not because of the change I am trying to do... but whatever09:09
@iglesiasgI am actually trying to fix that09:09
@wikingiglesiasg: idontcare ./configure and .travis.yml IS DO NOT FUCKING TOUCH IT09:09
@iglesiasgyeah yeah.. I just said whatever09:09
@wikingsonne|work: either everything or nothing09:09
@iglesiasglet's get more and more commits inside09:09
@iglesiasgwithout seeing what travis said >(09:10
@wikingsonne|work: you cant have cmake libshogun but python_modular by ./configure09:10
sonne|workwiking: no that won't work09:10
sonne|worksmall steps09:10
sonne|workwhy not?09:10
@wikingsonne|work: sorry but there has been just too much changes09:10
@wikingthere will be missing files ets09:10
@wikingetc09:10
@wikinglike i dont have anymore shogun/lib/config.h in feature/CMake barnch09:11
@wiking*branch09:11
@wikingthat's like one major problem you'll start having...09:11
sonne|workwiking: what do you do instead of config.h then?09:12
@wikingsonne|work: i just have a config.h.in09:12
@wikingand then cmake will generate it for you09:12
sonne|workyeah but where is the issue?09:12
@wikingbut not in src/shogun/lib/config.h09:12
sonne|workwhere then?09:13
@wikingit'll be in build/src/shogun/lib/config.h09:13
@wikingor something similar09:13
@wikingi cannot tell by heart09:13
@wikinganyhow... cmake is still not ready09:13
sonne|workbut that is not a problem then09:13
@wikingit needs fucking tweaking on cmake + travis side09:13
@wikingto get it at least to some level09:13
@wikingonce that's done i can do the rebase09:13
sonne|workwell cmake with libshogun  is or not?09:13
sonne|workmy migration plan would be09:14
sonne|work1) get libshogun & cmake to run on buildbot09:14
sonne|work2) get static interfaces work with cmake09:14
sonne|work3) do modular09:14
sonne|workand then one by one move the buildbot stuff over09:14
@wikingsonne|work: i dont have any static in cmake09:15
sonne|workwe cannot stop development just because we are transitioning09:15
sonne|workwell it is trivial to add when you already have (parts of) modular09:15
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has joined #shogun09:16
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/988916109:16
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has left #shogun []09:16
sonne|workwiking: only when everything works we can start using travis09:16
sonne|workwith cmake09:17
@wikingsonne|work: well then good luck cherrypicking09:17
sonne|workso the fixes iglesiasg is doing make sense09:17
@wikingfrom the feature/CMake branch09:17
@wikingsonne|work: no it doesn't because it's not right anyways...09:17
@wikingbut ok09:17
sonne|workwiking: well just explain what the issue is then09:17
sonne|workI mean config.h.in is not an issue09:17
sonne|workit could just be there09:17
@iglesiasgwiking, it is not right agree. But it is not like I am trying to fuck up your work doing it09:18
sonne|workwiking: I might be missing sth but you need to explain more then09:18
@wikingiglesiasg: a) it's not correct the way you try to do it b) instead of tryign to fix that you are doing wrong, use that energy and gimme a cmake patch that automagically downloads eigen3 (the version you want)09:19
@wikingiglesiasg: it's not that hard, i've even showed an example yesterday09:19
@iglesiasgwiking, b) sure. I am glad to help you with that09:20
@iglesiasgwiking, a) it is in the same way you did the gtest and other things in there, it is just not working yet because I got some details wrong09:20
@wikingiglesiasg: it's not the same...09:21
@iglesiasgwiking, about b, when is it going to be CMake ready? Even if I do this patch? One day, two days, one week?09:21
@iglesiasgwiking, what do we do until then? continue merging all the stuff when travis is hanging up on it?09:21
@wikingone of the reason is because it's not working. if it'd be the same it would be working or?09:21
@iglesiasgwiking, did you read the part that I got some details wrong?09:21
@iglesiasgwiking, the idea is the same (I think)09:21
@iglesiasgwiking, wget, unpack, give the flags to configure09:22
@iglesiasgwiking, I am getting some directory wrong yeah, and I actually think it is a stupid mistake09:22
@iglesiasgbut whatever09:22
@iglesiasgIMHO I don't think the mistake I have done is stopping you from seeing your changes in cmake09:22
@iglesiasgjust check the build history in travis and see it by yourself09:23
shogun-notifier-shogun: Viktor Gal :feature/CMake * 51d498f / .travis.yml: https://github.com/shogun-toolbox/shogun/commit/51d498f223842c4252544d23dd86df0e23c0335209:23
shogun-notifier-shogun: Travis: switch to virtualenv and pip for python builds09:23
@iglesiasgbut I mean, I can live without that change... it doesn't pay off for me to have this argument due to it, so I just skip it09:23
@wikingiglesiasg: i see it as a double waste time if you try to fix this09:23
@wikingiglesiasg: as the energy u use for fixing it could be used much better and it would be more beneficial in a long run09:24
@wikingbut even in a short one09:24
@wikinglike within a day...09:24
@wikingwe are not using travis the right way09:24
@wikingand that's mostly my fault09:24
@wikingas i did it09:24
@wikingbut now following this path is purely wrong09:25
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has joined #shogun09:41
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/988959009:41
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has left #shogun []09:41
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: Ex-Chat]09:43
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has joined #shogun09:51
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/989008409:51
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has left #shogun []09:51
-!- lisitsyn [~lisitsin@mxs.kg.ru] has joined #shogun09:56
shogun-buildbotbuild #1668 of deb1 - libshogun is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb1%20-%20libshogun/builds/166809:56
shogun-buildbotbuild #1169 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/1169  blamelist: Soeren Sonnenburg <sonne@debian.org>09:57
shogun-notifier-shogun: Viktor Gal :feature/CMake * 6340a2e / .travis.yml,CMakeLists.txt,cmake/FindEigen3.cmake: https://github.com/shogun-toolbox/shogun/commit/6340a2e2da2f4577996539e190b26921626749b510:01
shogun-notifier-shogun: Fix Eigen3 version detection10:01
shogun-notifier-shogun: remove gcc+python_modular builds from travis matrix10:01
-!- foulwall [~user@110.16.165.220] has quit [Ping timeout: 245 seconds]10:11
lisitsynhaha! mr. agressive wiking10:24
-!- iglesiasg [~iglesias@2001:6b0:1:1da0:18b5:d71d:f59f:b8b6] has joined #shogun10:40
-!- mode/#shogun [+o iglesiasg] by ChanServ10:40
shogun-buildbotbuild #1474 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/1474  blamelist: Soeren Sonnenburg <sonne@debian.org>10:40
-!- hushell [~hushell@c-24-21-169-136.hsd1.or.comcast.net] has quit [Quit: WeeChat 0.3.7]10:52
shogun-notifier-shogun: Viktor Gal :feature/CMake * f7abfd0 / CMakeLists.txt: https://github.com/shogun-toolbox/shogun/commit/f7abfd068c0e9bc32df364321dbcd1fc306691d711:20
shogun-notifier-shogun: require 3.1.2 eigen11:20
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has quit [Ping timeout: 256 seconds]11:45
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has joined #shogun11:50
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/989272611:50
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has left #shogun []11:50
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * 512d94f / src/shogun/lib/DataType.cpp,src/shogun/lib/DataType.h: https://github.com/shogun-toolbox/shogun/commit/512d94f6a6abb1a1fe43eb22d4dbe65e2347c85511:54
shogun-notifier-shogun: refactored sizeof_ptype, sizeof_stype to make them available from static context11:54
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * a5435a5 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/a5435a543f2265fc4409ac487f33fc72d7e7b66311:54
shogun-notifier-shogun: first parts of copy_ptype with unit tests11:54
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * 883ac63 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/883ac6316c1401812621067b8890205715c6b68411:54
shogun-notifier-shogun: -refactored compare_stype a bit11:54
shogun-notifier-shogun: -added copy_stype method with first bits11:54
shogun-notifier-shogun: -added first unit test for copy_stype11:54
shogun-notifier-shogun: lambday :feature/CMake * 68690e2 / tests/unit/io/Serialization_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/68690e2d868e0bec77db017256f36c05bfacc00211:54
shogun-notifier-shogun: addded unit-tests for serialization11:54
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * b0873e1 / tests/unit/io/Serialization_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/b0873e1a83200071d7ac4cfc145b94e9a372856311:54
shogun-notifier-shogun: Merge pull request #1162 from lambday/develop11:54
shogun-notifier-shogun:11:54
shogun-notifier-shogun: addded unit-tests for serialization11:54
shogun-notifier-shogun: lambday :feature/CMake * 8acb5bc / tests/unit/io/Serialization_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/8acb5bc56796bb41c63c884ce6b582a7aeeb3ebb11:54
shogun-notifier-shogun: added more unit-tests serialization for ptype and ctype11:54
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * 24b2ef0 / tests/unit/io/Serialization_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/24b2ef02075a7e1bef2fed30ac6de35a4db13a7711:54
shogun-notifier-shogun: Merge pull request #1163 from lambday/develop11:54
shogun-notifier-shogun: van51 :feature/CMake * 35a2ea0 / / (2 files): https://github.com/shogun-toolbox/shogun/commit/35a2ea0e34d59176a865d3508a687643f02fe15511:54
shogun-notifier-shogun: NNZ: added unit test and fixed get_next_feature()11:54
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * f63dadf / / (2 files): https://github.com/shogun-toolbox/shogun/commit/f63dadf8ce331f02b6695f6a6ed2b4e437ad135d11:54
shogun-notifier-shogun: Merge pull request #1168 from van51/develop11:54
shogun-notifier-shogun:11:54
shogun-notifier-shogun: NNZ: added unit test and fixed get_next_feature()11:54
shogun-notifier-shogun: Roman Votyakov :feature/CMake * b8b2e6f / src/shogun/regression/gp/ (6 files): https://github.com/shogun-toolbox/shogun/commit/b8b2e6f9cf99a1055429945f5d0ef309dfcba9c111:54
shogun-notifier-shogun: Fixed recomputing of posterior parameters11:54
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * a640e47 / src/shogun/regression/gp/ (6 files): https://github.com/shogun-toolbox/shogun/commit/a640e473fae1e5e608457e2d0a5cc242c8ca53a711:54
shogun-notifier-shogun: Merge pull request #1170 from votjakovr/develop11:54
shogun-notifier-shogun:11:54
shogun-notifier-shogun: Fixed recomputing of posterior parameters11:54
shogun-notifier-shogun: lambday :feature/CMake * d8dc5b5 / tests/unit/base/SGObject_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/d8dc5b55611729a83213b460e4984ca959a7ffd511:54
shogun-notifier-shogun: minor fix for eigen3 check in SGObject_unittest11:54
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * 8181acb / src/shogun/io/SGIO.cpp,src/shogun/io/SGIO.h: https://github.com/shogun-toolbox/shogun/commit/8181acb60da0d7182497d30594223a929794454111:54
shogun-notifier-shogun: enable location info (as in object / function name printing before SGIO11:54
shogun-notifier-shogun: prints) with config option to disable / show line and file11:54
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * e2142df / src/shogun/io/SGIO.cpp,src/shogun/io/SGIO.h: https://github.com/shogun-toolbox/shogun/commit/e2142df4a17e39d4ec94023be6198c59d370a96311:54
shogun-notifier-shogun:11:54
shogun-notifier-shogun: added get_width and obtain_from_generic for Chi2 kernel11:54
shogun-notifier-shogun: lambday :feature/CMake * b019018 / src/shogun/ (4 files): https://github.com/shogun-toolbox/shogun/commit/b019018843af6218904f04bfb28a7af25105cfc811:54
shogun-notifier-shogun: CJobResult added, CDenseMatrixOperator documentation extended11:54
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * dae693c / src/shogun/ (4 files): https://github.com/shogun-toolbox/shogun/commit/dae693c9d4accaf9bdbbc9475b05a2ac35b8766c11:54
shogun-notifier-shogun: Merge pull request #1177 from lambday/feature/log_determinant11:54
shogun-notifier-shogun:11:54
shogun-notifier-shogun: CJobResult added, CDenseMatrixOperator documentation extended11:54
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * 1f705d4 / src/configure: https://github.com/shogun-toolbox/shogun/commit/1f705d4baba8f87fa007b220191f229edf9b539a11:54
shogun-notifier-shogun: require json-c version >=1111:54
shogun-notifier-shogun:11:54
shogun-notifier-shogun: in version 0.10 json_object_get_int is b0rken causing shogun11:54
shogun-notifier-shogun: serialization / tests to fail11:54
shogun-notifier-shogun: iglesias :feature/CMake * bf430bc / src/shogun/distance/ (6 files): https://github.com/shogun-toolbox/shogun/commit/bf430bc660897b5419fb7a2527e3fab2a964a57e11:54
shogun-notifier-shogun: Minor changes in Custom, Dense and base distances.11:54
shogun-notifier-shogun: Mainly move a few methods from headers to implementation files.11:54
shogun-notifier-shogun: Fernando Iglesias :feature/CMake * d0e18e5 / src/shogun/distance/ (6 files): https://github.com/shogun-toolbox/shogun/commit/d0e18e515154d0eeace86fa2f90719f58945dc6711:54
shogun-notifier-shogun: Merge pull request #1178 from iglesias/develop11:54
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * 430d49e / / (8 files): https://github.com/shogun-toolbox/shogun/commit/430d49ee93ab6e334a867240ebf2c34dd1aa1b1811:55
shogun-notifier-shogun: Merge pull request #1180 from lambday/feature/log_determinant11:55
shogun-notifier-shogun:11:55
shogun-notifier-shogun: Removed setter for linear operator/vector for CDenseExactLogJob, computation engine added11:55
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * 3b2dd4d / / (3 files): https://github.com/shogun-toolbox/shogun/commit/3b2dd4ddc0f01105dfca6e4048624faa02de7af211:55
shogun-notifier-shogun: Merge pull request #1181 from tklein23/develop11:55
shogun-notifier-shogun:11:55
shogun-notifier-shogun: Enabled/fixed streaming_onlineliblinear example11:55
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * 58fe6ca / / (4 files): https://github.com/shogun-toolbox/shogun/commit/58fe6ca2a03c4d0e238d4be62b0953d471a16b0b11:55
shogun-notifier-shogun: Merge pull request #1182 from lambday/feature/log_determinant11:55
shogun-notifier-shogun:11:55
shogun-notifier-shogun: added COperatorFunction base class, one implementation CDenseMatrixExactLog, and a unit-test11:55
shogun-notifier-shogun: lambday :feature/CMake * 409995e / src/shogun/ (12 files): https://github.com/shogun-toolbox/shogun/commit/409995e338ae25d0faf2e62107970f5976bf2c0011:55
shogun-notifier-shogun: minor documentation fixes in log-det framework11:55
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * 4137923 / src/shogun/ (12 files): https://github.com/shogun-toolbox/shogun/commit/413792389e5c94a6fc09b249bc30248c1a9461fc11:55
shogun-notifier-shogun: Merge pull request #1183 from lambday/feature/log_determinant11:55
shogun-notifier-shogun:11:55
shogun-notifier-shogun: minor documentation fixes in log-det framework11:55
shogun-notifier-shogun: iglesias :feature/CMake * ad6fed6 / src/shogun/distance/ (3 files): https://github.com/shogun-toolbox/shogun/commit/ad6fed6bb703099f6a8b845c7cbde2bf34eb87d611:55
shogun-notifier-shogun: Add CustomMahalanobisDistance11:55
shogun-notifier-shogun: iglesias :feature/CMake * ccf16f7 / tests/unit/distance/Distance_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/ccf16f7629f1f7107658bdc27d773466ab2939d411:55
shogun-notifier-shogun: Very simple unit test for CustomMahalanobisDistance11:55
shogun-notifier-shogun: iglesias :feature/CMake * 70d9fca / src/shogun/metric/LMNN.cpp,src/shogun/metric/LMNN.h: https://github.com/shogun-toolbox/shogun/commit/70d9fca923a8f5822f00de50500af98d324f3d7411:55
shogun-notifier-shogun: [WIP] Add LMNN without train method11:55
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * 9ccdc30 / src/shogun/multiclass/KNN.h: https://github.com/shogun-toolbox/shogun/commit/9ccdc3094289093631cf2a5ea710935e3c52c6ec11:55
shogun-notifier-shogun: remove commented out function11:55
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * 29e2882 / src/.version.sh: https://github.com/shogun-toolbox/shogun/commit/29e2882cac5c3c6de8e569c697daebc60d75a23211:55
shogun-notifier-shogun: use the current branch as merge base when creating version strings11:55
shogun-notifier-shogun: Fernando Iglesias :feature/CMake * 384185f / / (6 files): https://github.com/shogun-toolbox/shogun/commit/384185f9c1aeb1f8213e63173f9407e27ea1099d11:55
shogun-notifier-shogun: Merge pull request #1188 from iglesias/feature/metric11:55
shogun-notifier-shogun:11:55
shogun-notifier-shogun: Feature/metric11:55
shogun-notifier-shogun: lambday :feature/CMake * 5bcdd54 / / (25 files): https://github.com/shogun-toolbox/shogun/commit/5bcdd547fab1c0ba3cad5d915d9b67a27e62f29c11:55
shogun-notifier-shogun: params registered, sample modified (log-det), gitignore modified11:55
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * 2aa7804 / / (25 files): https://github.com/shogun-toolbox/shogun/commit/2aa78041e8717f39419c83f6ad8a52b50c0bd92811:55
shogun-notifier-shogun: Merge pull request #1189 from lambday/feature/log_determinant11:55
shogun-notifier-shogun: iglesias :feature/CMake * f366d03 / tests/unit/mathematics/Math_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/f366d03906fe7ad24961157fa442130855ab081e11:55
shogun-notifier-shogun: linspace unit test11:55
shogun-notifier-shogun: Fernando Iglesias :feature/CMake * 27c9a1c / tests/unit/mathematics/Math_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/27c9a1cd0cb53183b7d86c2b6529c16ce133688111:55
shogun-notifier-shogun: Merge pull request #1193 from iglesias/develop11:55
shogun-notifier-shogun:11:55
shogun-notifier-shogun: linspace unit test11:55
shogun-notifier-shogun: hushell :feature/CMake * 3142600 / / (33 files): https://github.com/shogun-toolbox/shogun/commit/3142600b5951013bee76e42f4ec7e63b2a5349a411:55
shogun-notifier-shogun: move risk() from StructuredModel to DualLibQPSOSVM11:55
shogun-notifier-shogun: Fernando Iglesias :feature/CMake * 2135513 / / (33 files): https://github.com/shogun-toolbox/shogun/commit/2135513c8e74c6a40a830036ef469debf27a2ab511:55
shogun-notifier-shogun: Merge pull request #1185 from hushell/develop11:55
shogun-notifier-shogun:11:55
shogun-notifier-shogun: move risk() from StructuredModel to StructuredOutputMachine11:55
shogun-notifier-shogun: iglesias :feature/CMake * b83fa92 / .gitignore: https://github.com/shogun-toolbox/shogun/commit/b83fa92d34d5fd6e0689e57e1c37e635ce7f83c411:55
shogun-notifier-shogun: Update gitignore with files created by unit tests11:55
shogun-notifier-shogun: json files and combined_kernel.weights11:55
shogun-notifier-shogun: iglesias :feature/CMake * 9464e9e / tests/unit/io/Serialization_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/9464e9e5d7d52a9e4f2b912074459fb5dbaac68811:55
shogun-notifier-shogun: Fix typo in file extension: from hfd5 to hdf511:55
shogun-notifier-shogun: Fernando Iglesias :feature/CMake * e5d5b16 / .gitignore,tests/unit/io/Serialization_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/e5d5b1674cb2cd9e50fd4443ff3ab714019783df11:55
shogun-notifier-shogun: update documentation of gp framework11:56
shogun-notifier-shogun: Fernando Iglesias :feature/CMake * c6e3416 / src/shogun/machine/OnlineLinearMachine.h: https://github.com/shogun-toolbox/shogun/commit/c6e3416ac3cca42514762c974747a47ec2fad15811:56
shogun-notifier-shogun: Merge pull request #1196 from tklein23/double_free_hotfix11:56
shogun-notifier-shogun:11:56
shogun-notifier-shogun: Fixed double-free when using "SGVector<float32_t> OnlineLinearMachine::get_w()"11:56
shogun-notifier-shogun: Viktor Gal :feature/CMake * b90b3e3 / .travis.yml: https://github.com/shogun-toolbox/shogun/commit/b90b3e34132352ddddc184851861a4545a5a393111:56
shogun-notifier-shogun: Fix json version on travis11:56
shogun-notifier-shogun: Viktor Gal :feature/CMake * 77e3c59 / .travis.yml: https://github.com/shogun-toolbox/shogun/commit/77e3c595b7fb4d8be48b31f62eb1a1fbcafbbee611:56
shogun-notifier-shogun: Travis: fix cwd before building11:56
shogun-notifier-shogun: Roman Votyakov :feature/CMake * 7103b5c / src/shogun/machine/gp/ (8 files): https://github.com/shogun-toolbox/shogun/commit/7103b5c9f8650186c6cbe0c8ddbca9f31f0950e111:56
shogun-notifier-shogun: refactor likelihood models11:56
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * 667f2f3 / src/shogun/machine/gp/ (8 files): https://github.com/shogun-toolbox/shogun/commit/667f2f32640e8348b3f404d9027725b5e653ddf011:56
shogun-notifier-shogun: Merge pull request #1200 from votjakovr/feature/gp_refactoring11:56
shogun-notifier-shogun:11:56
shogun-notifier-shogun: refactor likelihood models11:56
shogun-notifier-shogun: Roman Votyakov :feature/CMake * 550de70 / src/shogun/machine/gp/ (3 files): https://github.com/shogun-toolbox/shogun/commit/550de70e7cb7bd017e29267309e58cd6b0f910dc11:56
shogun-notifier-shogun: change default value of the derivative wrt hypers11:56
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * f19478d / src/shogun/machine/gp/ (3 files): https://github.com/shogun-toolbox/shogun/commit/f19478d8e38e5191a72b717cc309fdadcd1f02fa11:56
shogun-notifier-shogun: minor documentation fix in log-det framework11:56
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * c764ec8 / src/shogun/mathematics/logdet/ (2 files): https://github.com/shogun-toolbox/shogun/commit/c764ec8c9285f24a272b3aa610e5aeeb9ce175fe11:56
shogun-notifier-shogun: Merge pull request #1204 from lambday/feature/log_determinant11:56
shogun-notifier-shogun:11:56
shogun-notifier-shogun: minor documentation fix in log-det framework11:56
shogun-notifier-shogun: Kevin :feature/CMake * ffb662e / src/shogun/mathematics/ajd/ (2 files): https://github.com/shogun-toolbox/shogun/commit/ffb662eb4d8572b73e6ccd64aa3d2066db27f6dd11:56
shogun-notifier-shogun: added base class for AJD and an example of a derived class header11:56
shogun-notifier-shogun: Kevin :feature/CMake * 3c3592a / src/shogun/mathematics/ajd/ (3 files): https://github.com/shogun-toolbox/shogun/commit/3c3592ae10b782195e5f8983259a05285415e4b811:56
shogun-notifier-shogun: added JADiag.cpp and made some of the other suggested edits11:56
shogun-notifier-shogun: Kevin :feature/CMake * 37cb7ba / src/shogun/mathematics/ajd/JADiag.cpp,src/shogun/mathematics/ajd/JADiag.h: https://github.com/shogun-toolbox/shogun/commit/37cb7ba3a8e423605ea3549163626e6458b6fd5511:56
shogun-notifier-shogun: fixed some minor mistakes so that it builds11:56
shogun-notifier-shogun: Kevin :feature/CMake * f0abf7e / src/shogun/mathematics/ajd/ (3 files): https://github.com/shogun-toolbox/shogun/commit/f0abf7e4a4e803d8d2d05e45f48eae9fe6ed28c411:56
shogun-notifier-shogun: swithed to cmath for machine eps and changed includes format11:56
shogun-notifier-shogun: Roman Votyakov :feature/CMake * 971d4fb / src/shogun/ (8 files): https://github.com/shogun-toolbox/shogun/commit/971d4fbb6186ca10b168f76f999e474e31918a5011:56
shogun-notifier-shogun: refactor evaluation of posterior and predictive distributions11:56
shogun-notifier-shogun: Heiko Strathmann :feature/CMake * 2c0f087 / src/shogun/ (8 files): https://github.com/shogun-toolbox/shogun/commit/2c0f08749fb47800f8a5ef038b0ada76e585b13c11:56
shogun-notifier-shogun: Merge pull request #1206 from votjakovr/feature/gp_refactoring11:56
shogun-notifier-shogun:11:56
shogun-notifier-shogun: Merge pull request #1208 from lambday/feature/log_determinant11:56
shogun-notifier-shogun:11:56
shogun-notifier-shogun: bugfixes in log-det framework11:56
shogun-notifier-shogun: Sergey Lisitsyn :feature/CMake * 2664590 / / (4 files): https://github.com/shogun-toolbox/shogun/commit/26645905424e53272b20196fd623d66bcc076db311:56
shogun-notifier-shogun: Merge pull request #1205 from pickle27/develop11:56
shogun-notifier-shogun:11:56
shogun-notifier-shogun: added base class for AJD and an example of a derived class header11:56
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * ded6e97 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/ded6e973a11d4cb7b7e023cc80c90f6efc65810b11:56
shogun-notifier-shogun: Merge pull request #1209 from van51/feature/hashing11:56
shogun-notifier-shogun:11:56
shogun-notifier-shogun: Changed CDelimiterTokenizer behavior on consecutive delimiters11:56
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * 0ac043b / src/shogun/lib/external/libocas.cpp: https://github.com/shogun-toolbox/shogun/commit/0ac043b75386152ffca7e7ad985a86c1a6c9d90511:56
shogun-notifier-shogun: don't use reserverd keyword _C11:56
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * 7062837 / src/interfaces/modular/modshogun_ignores.i: https://github.com/shogun-toolbox/shogun/commit/7062837ed457c6765530e430620b2615549d0dfe11:56
shogun-notifier-shogun: ignore evaluate_means to fix csharp compile error11:56
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * e726a47 / src/interfaces/modular/modshogun_ignores.i: https://github.com/shogun-toolbox/shogun/commit/e726a470fe9a9592fc6a8d6f1a74934027efcb3211:56
shogun-notifier-shogun: blacklist also evaluate_variance11:56
shogun-notifier-shogun: Soeren Sonnenburg :feature/CMake * bceddcf / applications/ocr/data/ocr.svm.gz: https://github.com/shogun-toolbox/shogun/commit/bceddcf2fa3173b3cb5a2b3a1724d165e39413cb11:56
-!- shogun-notifier- was kicked from #shogun by wiking [shogun-notifier-]11:57
@wikingi think i fucked up this branch :)11:57
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has joined #shogun12:01
sonne|workwiking: hurray!12:03
shogun-notifier-shogun: Viktor Gal :feature/CMake * b672846 / examples/undocumented/libshogun/CMakeLists.txt: https://github.com/shogun-toolbox/shogun/commit/b6728464de3a00c2f3d7e20fe7e7081ea52d95d012:04
shogun-notifier-shogun: Rebase the libshogun example list12:04
@wikingsonne|work: /home/travis/build.sh: line 141: [: too many arguments12:05
@wikingon a line like this12:05
@wikingif [ ! -z ${PYTHON_PKG} ] ; then ${PYTHON_PKG} ; fi12:05
@wikingany ideas?12:05
sonne|workwhat?12:05
@wikingi have a bash script like this12:06
@wikingand i'm getting that error12:06
@wikingbasically i want it to run the given env variable if set12:07
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has joined #shogun12:07
travis-ci[travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/989306112:07
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has left #shogun []12:07
@iglesiasgsonne|work, do you happen to know Kilian Weinberger? I just mailed him asking a couple of doubts about his LMNN implementation12:09
@iglesiasgI saw you are in his google+ network so that's why I wondered12:09
sonne|workwiking: ahh ok then maybe it is not in the path? you could try12:10
@wikingwiking@snf-77841:~$ export LO="ls -la"12:10
@wikingwiking@snf-77841:~$ if [ ! -z ${LO} ] ; then ${LO} ; fi12:10
@wiking-bash: [: ls: binary operator expected12:10
sonne|workbash -c "$PYTHONPKG"; fi12:10
@wikingwhy/12:10
@wiking?12:10
sonne|workahh sure12:10
sonne|workls -la is taken as a command12:10
sonne|workso it tries to find "ls -la"12:11
@wikingsonne|work: how can i do it12:11
@wiking?12:11
@wikingi need to execute a long command like source ~/virtualenv/python3.3/bin/activate && pip install numpy scipy --use-mirrors12:11
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has joined #shogun12:11
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/989380312:11
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has left #shogun []12:11
sonne|workwiking: use the bash thing12:11
sonne|workwiking: bash -c "$LO"12:12
sonne|workiglesiasg: yes sure I met him at nips12:12
@wikingwiking@snf-77841:~$ export LO="ls -la"12:12
@wikingwiking@snf-77841:~$ if [ ! -z ${LO} ] ; then bash -c "${LO}" ; fi12:12
@wiking-bash: [: ls: binary operator expected12:12
sonne|workexport LO="ls -lad /tmp"$12:14
sonne|workbash -c "$LO"12:14
sonne|workdrwxrwxrwt. 17 root root 323584 Aug  6 12:11 /tmp12:14
sonne|workwiking: and $LO also works here12:15
sonne|worknot "$LO" of course12:15
sonne|workwiking: ohh I see12:15
sonne|workuse if [ ! -z "${LO}" ]12:16
shogun-notifier-shogun: Viktor Gal :develop * d666811 / tests/unit/mathematics/Integration_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/d666811ea0a96f484994e971a9204d7da933b5db12:17
shogun-notifier-shogun: Unit-test should only test modules that are available!12:17
shogun-notifier-shogun: ode for Roman Votyakov12:17
@wikingsonne|work: ok that works, lets see...12:17
@wikingthnx!12:17
shogun-notifier-shogun: Viktor Gal :feature/CMake * 3cdf28a / .travis.yml: https://github.com/shogun-toolbox/shogun/commit/3cdf28afc0540e9b30f09817ef401a00aff605ff12:19
shogun-notifier-shogun: PYTHON_PKG test fix in travis script12:19
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has joined #shogun12:24
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/989421912:24
-!- travis-ci [~travis-ci@ec2-50-16-141-3.compute-1.amazonaws.com] has left #shogun []12:24
@wikingsonne|work: ccache would make faster things on travis? i mean since it would compile shogun always only once... but still the precompiler ccache could make things faster (e.g. eigen headers cached) or?12:30
shogun-buildbotbuild #1359 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/1359  blamelist: Viktor Gal <viktor.gal@maeth.com>12:31
shogun-buildbotbuild #1170 of cyg1 - libshogun is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/1170  blamelist: Viktor Gal <viktor.gal@maeth.com>12:36
@wikingsonne|work lisitsyn iglesiasg what should we do with external projects, should be bundled them within shogun, i.e. if one chooses to have a library compiled by/with shogun's cmake should that be part of shogun and ship it in <prefix>/include/shogun/third_party/... ? see the case of eigen12:41
@iglesiasgwhat is easiest for the user, which I guess is to bundle them12:41
@wikingiglesiasg: imho bundling12:42
-!- HeikoS [~heiko@nat-179-196.internal.eduroam.ucl.ac.uk] has joined #shogun12:42
-!- mode/#shogun [+o HeikoS] by ChanServ12:42
@wikingiglesiasg: so ship all the headers under shogun/third_party/ and static link the library with libshogun.so12:43
@wikingany objections ^ sonne|work ?12:43
@wikingiglesiasg: u use ccache right?12:44
@iglesiasgyes12:44
@wikingiglesiasg: is the clean compilation faster for you?12:44
@wikingiglesiasg: i.e. when your ccache is empty?12:45
@wikingthen w/o ccache12:45
@wiking*than12:45
@iglesiasgwiking, I think that clean compilation is the same12:45
@iglesiasgI have not timed it though12:45
@wikingiglesiasg: i hoped that in case of EIGEN3 present it would be a bit faster12:46
@wikingwiking@snf-77841:~$ ccache -s12:48
@wikingcache directory                     /home/wiking/.ccache12:48
@wikingcache hit (direct)                  104312:48
@wikingcache hit (preprocessed)            151712:48
@wikingcache miss                          112212:48
@wikingcalled for link                      73412:48
@wikingpreprocessor error                    1812:48
@wikingfiles in cache                       33812:48
@wikingcache size                          21.9 Mbytes12:48
@wikingmax cache size                       1.0 Gbytes12:49
@wikingwith a clean ccache and a clean compile12:49
@wikingso i suppose it should be faster even in this case (having these stats)12:49
-!- foulwall [~user@110.16.165.220] has joined #shogun12:52
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has joined #shogun12:58
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/989463012:58
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has left #shogun []12:58
shogun-notifier-shogun: Viktor Gal :develop * 11bd317 / tests/unit/classifier/GaussianProcessBinaryClassification_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/11bd3178b1fe5549c994a6567ebbbf4ab83295f313:00
shogun-notifier-shogun: Only check modules that are compiled/available part213:00
-!- van51 [~van51@ppp-94-66-42-68.home.otenet.gr] has joined #shogun13:03
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has joined #shogun13:05
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/989472313:05
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has left #shogun []13:05
shogun-buildbotbuild #1360 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/136013:12
shogun-buildbotbuild #1171 of cyg1 - libshogun is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/1171  blamelist: Viktor Gal <viktor.gal@maeth.com>13:15
-!- van51 [~van51@ppp-94-66-42-68.home.otenet.gr] has quit [Ping timeout: 256 seconds]13:19
sonne|workwiking: what exactly is the question?13:24
-!- foulwall` [~user@116.114.154.68] has joined #shogun13:39
-!- foulwall [~user@110.16.165.220] has quit [Ping timeout: 256 seconds]13:41
shogun-buildbotbuild #1476 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/1476  blamelist: Viktor Gal <viktor.gal@maeth.com>13:44
@wikingsonne|work: bundled third party libs13:47
@wikingsonne|work: i.e. we want to have at least eigen 3.1.2 for shogun... if it's not available in the distro we could give the user the option to tell shogun's make to download and use 3.1.2 eigen... and we could ship this in third_party/...13:48
sonne|workwiking: but why ship it if cmake downloads it?14:02
-!- lambday [67157c4f@gateway/web/freenode/ip.103.21.124.79] has joined #shogun14:10
@wikingsonne|work: well even if it downloads it we have to install it somewhere14:21
@wikingsonne|work: and i dont really think it's a good idea to install it to the system14:21
sonne|workwiking: ahh you mean we download and then extract compile somewhere14:24
sonne|workyeah sure create the folder third_party then and do all that in there14:25
sonne|work(on-the-fly)14:25
@wikingsonne|work: yeah but when installing?14:28
@wikingsonne|work: this is exactly the problem of eigen14:28
@wikingit's just a bunch of headers14:29
@wikinglets say somebody not only wants to run some c++ examples but as well compile them... i.e. he needs all the shogun headers14:29
sonne|workwiking: errm we don't need eigen3 stuff installed14:29
@wikinghence we need to ship all the headers somehow14:29
sonne|workwiking: he will only need shogun headers14:29
sonne|workno eigen14:29
@wikingsonne|work: ?14:29
sonne|workwiking: we have no eigen requirements in headers!14:30
sonne|workso only needed at compile time14:30
@wikingsonne|work: examples/undocumented/libshogun/converter_diffusionmaps.cpp14:31
@wikingwhy is there HAVE_EIGEN3 then?14:31
sonne|workwell libshogun won't hae CDiffusionMaps if EIGEN3 was not availabe at compile time14:32
@wikingand if available?14:32
@wikingthat's what i mean...14:32
@wikingwe supply eigen3 with cmake during compilation... then ship that libshogun.so/a14:33
@wikingand what if somebody will want to develop for that libshogun a binary that needs CDiffusionMaps for example14:34
@wikinghe should be able to do it as it's in libshogun....14:34
sonne|workwiking: then he can just use it with libshogun14:34
sonne|workwiking: OK?14:35
@wikingsonney2k: mmm i dont see that it wont start erroring with the #includes of shogun but lets see14:36
@wikingunless we dont ship the eigen headers with shogun14:36
sonne|workwiking: but you would agree that if no header file in shogun uses eigen3 that we are fine?14:36
@wikingsonne|work: indeed14:37
@wikingif we can do the opaque shit then it should b fine14:37
sonne|workwiking: exaclty14:37
sonne|workthat is the plan14:37
@wikingok14:37
@wikingthen i'll just download and compile shogun with that... and dont add the headers into the install target...14:38
sonne|workwe should somehow define clear interfaces and hide all the implementation details to the extent possible14:38
sonne|workwiking: what is problematic would be e.g libglpk.so etc14:38
sonne|workthese we would somehow need14:38
sonne|workat runtime14:38
@wikingsonne|work: statically compile within libshogun.so?14:38
@wikingi mean s/within/into/g14:38
sonne|workdidn't manage to do that14:39
@wikingmmm ok i'll try14:39
sonne|worksuspect that this will be 200MB in size14:39
@wikingbut now i realised that i completly fucked up the whole travis config14:39
@wikingwe were using it veeeeryt wrong14:39
@wikingthere are already precompiled envs for like all python and ruby versions14:40
@wikingand as well java14:40
@wikingone doesn't need to apt-get it14:40
@wikingjust it needs some extra care in the config file of traviss14:44
thoralfwiking: But it's working - it can't be that bad. :)14:59
-!- foulwall` [~user@116.114.154.68] has quit [Ping timeout: 260 seconds]15:03
shogun-buildbotbuild #1475 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/1475  blamelist: Viktor Gal <viktor.gal@maeth.com>15:49
sonne|workthoralf: Amen :D15:54
thoralfsonne|work: I had kind of a bad dream about "non-functional requirements"... and now a deja vu ;)15:57
thoralfsonne|work: "I could refactor it, so it uses 1% less memory!"15:58
sonne|workthoralf: well if it lowers travis' load it is actually good. it crazily often builds shogun...16:05
thoralfsonne|work: Sure, but depends of how much work it is (and how much it saves).16:07
thoralfAnyway, it's not my job to worry about this. :)16:07
-!- gsomix_ [~gsomix@178.45.86.56] has quit [Ping timeout: 246 seconds]16:10
thoralfOne feature hashing question: Is there a way to (a priori/posteriori) estimate a "good" number of dimensions for hashing?16:12
thoralfFor example: Having trained classifiers x and y with hashed features of dimensions d_x and d_y...16:13
thoralfOr a measure which tells when the hashed solution is somehow optimal/a good trade-off between time/space/accuracy?16:14
-!- gsomix_ [~gsomix@95.67.140.176] has joined #shogun16:24
sonne|workthoralf: not really. I guess your best bet is to increase hash size slowly and then you should observe a plateau at some point16:25
thoralfsonne|work: That's what I'm doing, but this is quite expensive.  Especially the plateau appears when I'm too close to the unhashed number of dimensions.16:28
-!- vgorbati [~vgorbati@91.216.173.29] has joined #shogun16:29
sonne|workthoralf: well hashing only makes things faster not better :)16:31
thoralfsonne|work: I'm aware of that.  But hashing size is yet another parameter to estimate - and estimating makes it too expensive.16:33
thoralfsonne|work: At least I know that I didn't forget something obvious.16:33
sonne|workthoralf: I usually used the biggest possible hash size (for my architecture)16:34
-!- foulwall [~user@greymonkey.im] has joined #shogun16:36
@iglesiasgall right, see you later guys!16:37
thoralfsonne|work: Hehe.  This is at least less expensive then "searching" the right size. :)16:38
sonne|workiglesiasg: cu!16:38
thoralfiglesiasg: Have fun.16:38
-!- iglesiasg [~iglesias@2001:6b0:1:1da0:18b5:d71d:f59f:b8b6] has quit [Quit: Ex-Chat]16:38
sonne|workthoralf: well I only use this when I have hard resource (memory / computation time) limits. So I can set it to the memory limit16:38
-!- foulwall [~user@greymonkey.im] has quit [Remote host closed the connection]16:42
-!- foulwall [~user@greymonkey.im] has joined #shogun16:57
-!- foulwall` [~user@116.114.154.68] has joined #shogun17:02
-!- foulwall [~user@greymonkey.im] has quit [Ping timeout: 276 seconds]17:04
-!- foulwall` [~user@116.114.154.68] has quit [Ping timeout: 245 seconds]17:07
-!- shogun-notifier- [~irker@94.23.237.10] has joined #shogun17:12
shogun-notifier-shogun: Viktor Gal :feature/CMake * fad52d4 / .travis.yml,CMakeLists.txt: https://github.com/shogun-toolbox/shogun/commit/fad52d4bffe62bb97260932bded4631934dd85c217:12
shogun-notifier-shogun: Refactor travis build config17:12
shogun-notifier-shogun: Require that the pythonlib found has the same version as the interperter17:12
shogun-notifier-shogun: found17:12
@wikingbtw: the reason our travis builds are slow because there's a hard limit on 5 concurrent builds... if it's higher (like our case) it'll take double time to build on travis17:12
@wikingso maybe removing some interfaces would be a good idea17:12
@sonney2kwiking, time is not yet the issue IMHO17:14
@wikingthis is the new way travis looks like17:15
@wikinghttps://travis-ci.org/shogun-toolbox/shogun/builds/990481117:15
@wikingby this refactor we can set allowed failures etc.17:15
@sonney2knice17:16
@wiking13c90bac (Fernando Iglesias 2013-07-24 14:12:13 +0200 1) ../modular/Metric.i17:18
@wikingwhy is a symlink commmited to the repo?17:18
* sonney2k off17:20
shogun-notifier-shogun: Viktor Gal :feature/CMake * d109cfa / cmake/FindRubyNArray.cmake: https://github.com/shogun-toolbox/shogun/commit/d109cfa9d7096c5444391070cca5f2cf656fab8217:21
shogun-notifier-shogun: Refactor RubyNarray package detector17:21
shogun-notifier-shogun: Viktor Gal :feature/CMake * 0ee7ccd / src/interfaces/python_modular/Metric.i: https://github.com/shogun-toolbox/shogun/commit/0ee7ccd1a756b79a0d2fe51795adc65bd3e9ce7d17:21
shogun-notifier-shogun: this should not be here...17:21
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has joined #shogun17:35
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/990481117:35
-!- travis-ci [~travis-ci@ec2-23-22-133-41.compute-1.amazonaws.com] has left #shogun []17:35
-!- pickle27 [~Kevin@d67-193-243-174.home3.cgocable.net] has joined #shogun17:40
-!- lisitsyn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.]17:41
-!- travis-ci [~travis-ci@ec2-54-227-8-185.compute-1.amazonaws.com] has joined #shogun17:57
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/990532917:57
-!- travis-ci [~travis-ci@ec2-54-227-8-185.compute-1.amazonaws.com] has left #shogun []17:57
@wikinganybody knows a non https mirror of eigen?17:58
-!- vgorbati [~vgorbati@91.216.173.29] has quit [Quit: vgorbati]18:38
-!- lisitsyn [~lisitsyn@213.87.128.141] has joined #shogun18:43
-!- lisitsyn [~lisitsyn@213.87.128.141] has quit [Read error: Connection reset by peer]18:47
-!- lisitsyn [~lisitsyn@fb2-lo1.global63.net] has joined #shogun18:49
-!- vgorbati [~vgorbati@91.216.173.29] has joined #shogun18:58
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun19:01
-!- mode/#shogun [+o iglesiasg] by ChanServ19:01
vgorbatilisitsyn: hello, are you here?19:01
lisitsynvgorbati: yes19:01
vgorbatilisitsyn: Is there an implementation of functions like logistic sigmoid, linear function etc. in shogun (I think it is, but I can't find it)?19:03
vgorbatilisitsyn: at least I remember some discussions in chat or email regarding logistic sigmoid function..19:06
lisitsynvgorbati: hmm I don't really know if there is something modular for that19:06
lisitsynbut that's easy to implement right?19:06
vgorbatilisitsyn: I've found only this: https://github.com/shogun-toolbox/shogun/blob/develop/src/shogun/mathematics/Function.h, and this: https://github.com/shogun-toolbox/shogun/blob/develop/src/shogun/evaluation/DifferentiableFunction.h (strange that these files seem to be 'disconnected' from each other)19:07
lisitsynhehe yeah19:08
lisitsynthey are both from GP19:08
lisitsynbut developed by different people19:08
vgorbatilisitsyn: I will need basic interface for differentiable function - like operator() (float64_t x) and derivative(float64_t x) functions, and the existing DifferentiableFunction interface seems to be a bit overcomplicated, while Function interface lacks derivative method:)19:22
lisitsynvgorbati: what is it for?19:23
vgorbativgorbati: neural networks. A neuron has its activation function, the most used are logistic sigmoid, tanh and linear, but there can be others, like polynomial etc. I thought, that the same interface for them all will be a good approach19:25
vgorbatioops19:25
lisitsynvgorbati: ah just add a new class you need then19:25
vgorbatilisitsyn: then there will be 3 different classes for functions:) Ok, I will name my class 'ActivationFunction' :)19:26
lisitsynthat's ok I think19:32
vgorbatilisitsyn: even with functions, that seem to have pretty accurate definition - it's really hard to make a class suitable for all:)19:32
lisitsyndifferentiable functio is some model selection related stuff19:32
lisitsynquite complicated19:32
pickle27iglesiasg: lisitsyn Im having trouble rebasing here, its not fixing the problem19:59
lisitsynpickle27: what happens?20:00
pickle27Previous HEAD position was 11bd317... Only check modules that are compiled/available part220:00
lisitsynpickle27: can't say it tells me much ;)20:03
pickle27lisitsyn: found the problem I think20:06
pickle27iglesiasg: lisitsyn okay fixed it up and made a new pr20:14
@iglesiasgpickle27, nice, thank you20:14
pickle27iglesiasg: the diff for Converter.i is a bit weird it should only say I added one line but I now get credit for 3 more20:15
pickle27iglesiasg: wait20:15
pickle27iglesiasg: I did add extra20:15
pickle27let me fix this20:15
@iglesiasgpickle27, ok20:15
pickle27iglesiasg: okay there we go20:18
@iglesiasgpickle27,  ready?20:18
pickle27iglesiasg: yup!20:18
@iglesiasgI would say let's wait for travis20:18
@iglesiasgbut hey, it doesn't work!20:18
pickle27iglesiasg: kk20:18
shogun-notifier-shogun: Kevin :develop * 035a21d / src/shogun/converter/ica/ (3 files): https://github.com/shogun-toolbox/shogun/commit/035a21d0402da1375871b9ca8eaa21d0fb6de69c20:19
shogun-notifier-shogun: added FFSep ICA algorithm20:19
shogun-notifier-shogun: Kevin :develop * 7c44bb6 / / (10 files): https://github.com/shogun-toolbox/shogun/commit/7c44bb663f5646485cdce854de134be83e19894320:19
shogun-notifier-shogun: ICA methods, added doxygen comment lines, added assert for input to apply, added ffsep unit test, made sobi unit test consitent20:19
shogun-notifier-shogun: Kevin :develop * e050440 / src/interfaces/modular/Converter.i,src/interfaces/modular/Converter_includes.i: https://github.com/shogun-toolbox/shogun/commit/e050440a5c7b337879d7292e1870681c05e69bc720:19
@iglesiasgthere we go20:19
shogun-notifier-shogun: some re org and cleanup in Converter.i and Converter_includes.i20:19
shogun-notifier-shogun: Fernando Iglesias :develop * 4aa4f62 / / (10 files): https://github.com/shogun-toolbox/shogun/commit/4aa4f62ee4f923aadd9a18e49ce1e32c9c1a387a20:19
shogun-notifier-shogun: Merge pull request #1372 from pickle27/develop20:19
shogun-notifier-shogun:20:19
shogun-notifier-shogun: added FFSep ICA algorithm20:19
@iglesiasgall right, I am off now for a while. See you!20:21
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: Ex-Chat]20:21
-!- hushell [~hushell@c-24-21-169-136.hsd1.or.comcast.net] has joined #shogun20:32
shogun-buildbotbuild #1172 of cyg1 - libshogun is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/117220:42
shogun-buildbotbuild #1173 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/1173  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>20:43
-!- lisitsyn1 [~lisitsyn@fb2-lo1.global63.net] has joined #shogun21:02
-!- lisitsyn [~lisitsyn@fb2-lo1.global63.net] has quit [Ping timeout: 246 seconds]21:03
shogun-buildbotbuild #1174 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/1174  blamelist: Kevin <kevinhughes27@gmail.com>21:14
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has quit [Ping timeout: 240 seconds]21:19
shogun-buildbotbuild #1477 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/1477  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>, Kevin <kevinhughes27@gmail.com>21:20
shogun-buildbotbuild #1373 of bsd1 - libshogun is complete: Failure [failed test]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/bsd1%20-%20libshogun/builds/1373  blamelist: Kevin <kevinhughes27@gmail.com>21:25
@sonney2khey gsomix! how is it going?21:39
-!- HeikoS [~heiko@nat-179-196.internal.eduroam.ucl.ac.uk] has quit [Quit: Leaving.]21:40
shogun-notifier-shogun: Viktor Gal :feature/CMake * 76d640b / .travis.yml,CMakeLists.txt,cmake/external/Eigen3.cmake,cmake/external/JSON.cmake: https://github.com/shogun-toolbox/shogun/commit/76d640baf523b9e163ea6b295630d5a005877bcb21:40
shogun-notifier-shogun: Add BUNDLE_EIGEN and BUNDLE_JSON options21:41
shogun-notifier-shogun: download compile as a bundle both of the libraries during compilation of shogun21:41
shogun-notifier-shogun: Dedicated to Fernando21:41
@wikingok so i think cmake is done21:41
@sonney2kwiking, then merge man21:42
@wikinglets wait for the travis output21:47
@wikingat least21:47
@wikingbut basically from now on we could bundle the whole f*ing dependency into libshogun21:48
@wikingoh noooo... jinja2 is missing..21:53
shogun-buildbotbuild #1478 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/1478  blamelist: Kevin <kevinhughes27@gmail.com>22:00
-!- lambday [67157c4f@gateway/web/freenode/ip.103.21.124.79] has quit [Ping timeout: 250 seconds]22:03
-!- travis-ci [~travis-ci@ec2-54-227-8-185.compute-1.amazonaws.com] has joined #shogun22:06
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/991241222:06
-!- travis-ci [~travis-ci@ec2-54-227-8-185.compute-1.amazonaws.com] has left #shogun []22:06
-!- vgorbati [~vgorbati@91.216.173.29] has quit [Ping timeout: 264 seconds]22:07
-!- pickle27 [~Kevin@d67-193-243-174.home3.cgocable.net] has quit [Quit: Leaving]22:09
-!- travis-ci [~travis-ci@ec2-54-227-8-185.compute-1.amazonaws.com] has joined #shogun22:39
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/991518022:39
-!- travis-ci [~travis-ci@ec2-54-227-8-185.compute-1.amazonaws.com] has left #shogun []22:39
@wikingbu22:44
-!- lisitsyn1 [~lisitsyn@fb2-lo1.global63.net] has quit [Quit: Leaving.]22:50
-!- lisitsyn [~lisitsyn@fb2-lo1.global63.net] has joined #shogun22:50
@sonney2kwiking, bu!22:55
@sonney2kgsomix_, I just figured out that the data is already in C_ORDER on the disc the csv reader is just wrong22:56
-!- lisitsyn [~lisitsyn@fb2-lo1.global63.net] has quit [Quit: Leaving.]22:57
shogun-notifier-shogun: Viktor Gal :feature/CMake * cf0fe33 / / (6 files): https://github.com/shogun-toolbox/shogun/commit/cf0fe33bdd91680a4d39227efeba9f6b1192d88023:00
@wikingthis should do it23:00
shogun-notifier-shogun: Add Jinja2 detection23:00
shogun-notifier-shogun: fix dependency for shogun when using bundles23:00
shogun-notifier-shogun: install python-jinja2 package and remove libjson package23:00
-!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun23:01
-!- mode/#shogun [+o iglesiasg] by ChanServ23:02
@wikingwe need an eigen mirror in the us... travis build servers are there and my server where currently the eigen source resides is in eu...hence slow download23:13
@sonney2kgsomix_, ping ping ping23:16
@sonney2kgsomix_, something is wrong and I cannot figure it out23:16
@sonney2kgsomix_, the C_ORDER works almost but of the 92x92 matrix about 700 entries are wrong23:17
@sonney2kno idea why23:17
shogun-notifier-shogun: Viktor Gal :feature/CMake * 1293b5a / CMakeLists.txt,cmake/FindJinja2.cmake: https://github.com/shogun-toolbox/shogun/commit/1293b5a019d29ea0cf4097e969bd22a09a2e0aa823:17
shogun-notifier-shogun: revoke c++11 flags as it utterly breaks bundle compilation23:17
shogun-notifier-shogun: remove debugging output from FindJinja223:17
shogun-notifier-shogun: Soeren Sonnenburg :develop * 083a21f / / (4 files): https://github.com/shogun-toolbox/shogun/commit/083a21ff7f8ae745a7dbda005276b18c235beec123:26
shogun-notifier-shogun: change c/fortran conversion of csv file reader and convert gaussian kernel23:26
@sonney2kwiking, I have a server in the US23:26
@wikingsonney2k: can u put there this tar.bz2: http://bitbucket.org/eigen/eigen/get/3.1.4.tar.bz223:27
@wikingand give me the url23:27
@sonney2kwiking, I have to check first if I have anything running that can serve files23:28
@wikingdoh /usr/bin/ld: ../../../third_party/libs/json/libjson-c.a(json_object.o): relocation R_X86_64_32S against `.text' can not be used when making a shared object; recompile with -fPIC23:28
@wiking../../../third_party/libs/json/libjson-c.a: could not read symbols: Bad value23:29
@sonney2kwiking, only some https stuff23:29
@sonney2kwiking, omfg why is it not compiled with -fPIC?23:30
@wikingsonney2k: dunno... i mean it's json which is not compiled.. but yeah i'm adding now the --with-pic for it's ./configure script23:31
@wikingsonney2k: i need that file on http://23:31
@wikingw/o ssl23:31
@sonney2kwiking, is ftp also fine?23:32
-!- travis-ci [~travis-ci@ec2-54-234-6-103.compute-1.amazonaws.com] has joined #shogun23:34
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/991783723:34
-!- travis-ci [~travis-ci@ec2-54-234-6-103.compute-1.amazonaws.com] has left #shogun []23:34
shogun-notifier-shogun: Viktor Gal :feature/CMake * 76a3786 / cmake/external/JSON.cmake: https://github.com/shogun-toolbox/shogun/commit/76a3786a5bd204925f94dc460d7029c00a2bd72623:37
shogun-notifier-shogun: Build JSON bundle with -fPIC23:37
@wikingsonney2k: afaik yes23:41
@sonney2kwiking, ok then let me install vsftp23:41
-!- travis-ci [~travis-ci@ec2-54-227-8-185.compute-1.amazonaws.com] has joined #shogun23:49
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/991835623:49
-!- travis-ci [~travis-ci@ec2-54-227-8-185.compute-1.amazonaws.com] has left #shogun []23:49
@wikingmmm we need to change the gmock/gtest version in order to support unit testing :P23:50
@sonney2kwiking, why that? What feature is it missing?23:51
@wiking/Users/wiking/shogun/build/GoogleMock/src/GoogleMock/gtest/include/gtest/internal/gtest-port.h:499:13: fatal error: 'tr1/tuple' file not found23:51
@wiking#   include <tr1/tuple>  // NOLINT ^23:51
@wikingthat's why23:51
@wikingit's fixed in gtest's repo23:51
@wikingbut not in the 1.6.0 release23:52
@wikingwhich is not so surpriseing23:53
@wikingas it was released in 201123:53
@wiking:23:53
@wiking:P23:53
@sonney2kwiking, try wget ftp://nn7.de/eigen3/3.1.4.tar.bz223:55
@sonney2kwiking, urgs23:55
@sonney2kwhat is tr1/tuple?23:55
@wikingtr1 is pre c++11 ;)23:57
@sonney2kwiking, but where does this appear?23:57
@sonney2kI cannot find it in the shogun tree23:57
@wikingsonney2k: in gtest framework23:58
@sonney2kwiking, ahh it is inside gtest?23:59
@sonney2kwiking, so std=c++11 will fail?23:59
@wikingyes23:59
@wikingindeed23:59
--- Log closed Wed Aug 07 00:00:04 2013
« Monday, 2013-08-05 Index Wednesday, 2013-08-07 »

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