IRC logs of #shogun for Tuesday, 2013-03-05

--- Log opened Tue Mar 05 00:00:59 2013
-!- hoijui [~hoijui@dslb-092-078-043-220.pools.arcor-ip.net] has quit [Ping timeout: 276 seconds]00:22
-!- heiko [~heiko@b019b9ef.bb.sky.com] has joined #shogun00:35
-!- heiko [~heiko@b019b9ef.bb.sky.com] has left #shogun []00:38
-!- FSCV [~FSCV@173.254.212.46] has quit [Read error: Connection reset by peer]01:03
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]01:06
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Ping timeout: 264 seconds]03:27
-!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun03:31
shogun-buildbotbuild #312 of nightly_default is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/31204:07
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun05:30
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Quit: Page closed]06:22
-!- hoijui [~hoijui@dslb-092-078-043-220.pools.arcor-ip.net] has joined #shogun07:42
-!- blackburn [~blackburn@185.6.239.124] has quit [Quit: Leaving.]08:08
-!- blackburn [~lisitsin@mxs.kg.ru] has joined #shogun08:35
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun09:27
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds]09:32
-!- nando [~nando@n145-p185.kthopen.kth.se] has joined #shogun10:52
-!- heiko [~heiko@pat-191-250.internal.eduroam.ucl.ac.uk] has joined #shogun12:19
-!- hoijui [~hoijui@dslb-092-078-043-220.pools.arcor-ip.net] has quit [Quit: Leaving]12:54
-!- nando [~nando@n145-p185.kthopen.kth.se] has quit [Ping timeout: 255 seconds]13:28
-!- nando [~nando@n145-p185.kthopen.kth.se] has joined #shogun13:44
-!- nando [~nando@n145-p185.kthopen.kth.se] has quit [Client Quit]13:44
-!- nando [~nando@n145-p185.kthopen.kth.se] has joined #shogun13:44
-!- FSCV [~FSCV@204.45.132.11] has joined #shogun15:28
-!- blackburn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.]16:33
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun17:34
shogun-notifier-shogun: Heiko Strathmann :master * 9a867fa / src/shogun/ (4 files): https://github.com/shogun-toolbox/shogun/commit/9a867faa9ec6f29fc207362a34b4eab607dfed7417:34
shogun-notifier-shogun: removed some warnings via adding documentation17:34
shogun-notifier-shogun: Heiko Strathmann :master * 605fa17 / / (4 files): https://github.com/shogun-toolbox/shogun/commit/605fa179c9afa6d1963f8d1c4d2c42292203eda517:34
shogun-notifier-shogun: -removed the data generator example and moved it to the unit-tests17:34
shogun-notifier-shogun: -renamed the unit test17:34
shogun-notifier-shogun: Heiko Strathmann :master * bc4c071 / src/shogun/base/SGObject.cpp: https://github.com/shogun-toolbox/shogun/commit/bc4c0711455abc12531eed79f468d2cb9d2a387317:34
shogun-notifier-shogun: code formating17:34
shogun-notifier-shogun: Heiko Strathmann :master * 1b0e0ed / src/NEWS: https://github.com/shogun-toolbox/shogun/commit/1b0e0ed84f8453b4791d053313f7c34d7ed8052017:34
shogun-notifier-shogun: added a note on Viktors cool new unit tests17:34
shogun-notifier-shogun: Heiko Strathmann :master * 3a81dd2 / src/shogun/io/streaming/InputParser.h: https://github.com/shogun-toolbox/shogun/commit/3a81dd2791f76322174dd6438f8e588db8028d0717:34
shogun-notifier-shogun: fixed a typo17:34
shogun-notifier-shogun: Heiko Strathmann :master * 4b774b1 / examples/undocumented/libshogun/classifier_mkl_svmlight_modelselection_bug.cpp: https://github.com/shogun-toolbox/shogun/commit/4b774b13bd36a440806dab7c06d50048ef0883f417:34
shogun-notifier-shogun: added example to reproduce the mkl model selection bug with SVMLight17:34
shogun-notifier-shogun: Heiko Strathmann :master * 77e75db / examples/undocumented/libshogun/Makefile: https://github.com/shogun-toolbox/shogun/commit/77e75db4e4e3165b60103ead5780770cde74545117:34
-!- blackburn [~blackburn@185.6.239.191] has joined #shogun17:41
heikoblackburn: I have done all my stuff for release17:55
heikoblackburn: there are some bugs open, but I think it would more sense if other people fixed then since they are more into the framework at that point17:55
heikoblackburn: I have details about the problems in the issues17:56
nandoheiko: I will have a look to the issues and see if there is something for me there17:57
-!- nando [~nando@n145-p185.kthopen.kth.se] has quit [Quit: leaving]17:58
blackburnheiko: okay I see18:02
heikonadn0, cool, I think its just a point of how much time one is going to spend on these bugs, and someone who has written the code will probably be faster in doing this18:05
heikoblackburn: what do you think of adding this automatic example description? we could have some meta tags in the files for this18:05
blackburnheiko: reading your mail18:05
blackburnheiko: well we have something (recall we have documentation in examples)18:06
shogun-notifier-shogun: Heiko Strathmann :master * a499fcd / src/shogun/statistics/LinearTimeMMD.h: https://github.com/shogun-toolbox/shogun/commit/a499fcd75ccbb3b0939760a5800bc1524b28d1b818:06
shogun-notifier-shogun: removed warnings by adding documentation18:06
blackburnheiko: like that http://shogun-toolbox.org/doc/en/current/python_modular_examples.html#python_modular_kernel_examples18:06
shogun-notifier-shogun: Heiko Strathmann :master * 6bc71bb / src/shogun/statistics/LinearTimeMMD.h: https://github.com/shogun-toolbox/shogun/commit/6bc71bb61422fa361079a55986d047a7789762c718:06
shogun-notifier-shogun: Merge pull request #909 from karlnapf/master18:06
shogun-notifier-shogun:18:06
shogun-notifier-shogun: more warnings removed18:06
heikoblackburn:  I know, but this just pastes the code into the website18:07
blackburnheiko: doc + code18:07
heikowhere is that?18:07
heikoblackburn: so I had some kind of list in mind which one can easily browse through18:07
blackburnheiko: http://shogun-toolbox.org/doc/en/current/python_modular_examples.html#python_modular_classifier_examples18:07
blackburn# This example shows how to use a custom defined kernel function for training a # two class Support Vector Machine (SVM) classifier on a randomly generated # examples. The SVM regularization constant is set to C=1.18:08
blackburn# In this example we demonstrate how to use SVMs in a domain adaptation # scenario. Here, we assume that we have two problem domains, one with # an abundance of training data (source domain) and one with only a few # training examples (target domain).18:08
blackburnblablba18:08
heikoI see18:08
heikoso just comments18:08
heikocool18:08
heikoI never saw that :)18:09
heikomaybe I will just add comments to the existing examples in that case18:09
heikosorry for always not knowing about things :D18:09
heikoblackburn:  syntax highlighting would be cool :)18:09
blackburnheiko: yes18:10
blackburnheiko: ipython notebooks actually18:10
blackburnis the thing I would like to see18:10
heikoblackburn, this would be ultra cool18:10
blackburnheiko: takes time and work..18:10
heikotrue18:10
heikoI mean just having a few scentences is already worth more than nothing18:11
heikosince once can search that18:11
heikolike: custom kernel18:11
heikoor similar18:11
shogun-buildbotbuild #866 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/866  blamelist: Sergey Lisitsyn <lisitsyn.s.o@gmail.com>18:15
blackburnheiko: about active subset stack18:16
blackburnheiko: it failed in GPs18:16
blackburnheiko: I added a NULL check in some method a few weeks ago18:17
shogun-buildbotbuild #867 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/867  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>18:23
heikoblackburn, I think this should be fine18:26
heikoif the parameter is null, it just wont build a parameter dictionary18:27
heikobut parameters need to be allowed to be null18:27
heikoI actually dont really know that jacob did with these parameter dictionaries18:27
heikoblackburn: Perceptron algorithm did not converge after 1000 iterations.18:28
heikowhat about this thing18:28
heikothat happens from time to time18:28
blackburnheiko: it is random iirc18:42
heikoyes, probably18:43
blackburnheiko: our test data is not linearly separable18:43
blackburnor not easy sepeerable18:44
blackburn:)18:44
shogun-buildbotbuild #868 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/86819:09
shogun-buildbotbuild #869 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/869  blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>19:25
-!- heiko [~heiko@pat-191-250.internal.eduroam.ucl.ac.uk] has quit [Quit: Leaving.]20:41
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]21:06
--- Log closed Wed Mar 06 00:00:04 2013

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