--- Log opened Sun Mar 13 00:00:21 2016 | ||
--- Day changed Sun Mar 13 2016 | ||
@HeikoS | besser82: quite good | 00:00 |
---|---|---|
@HeikoS | some more free time these days | 00:00 |
@HeikoS | which is nice | 00:00 |
@HeikoS | but ends soon :) | 00:00 |
-!- sonne|osx [~sonne@x4db34b0d.dyn.telefonica.de] has quit [Quit: sonne|osx] | 00:02 | |
shogun-notifier- | shogun: Heiko Strathmann :develop * 7bd7f28 / examples/meta/cpp/CMakeLists.txt: https://github.com/shogun-toolbox/shogun/commit/7bd7f283d8246f5fa9b63373e55de52eb945d697 | 00:07 |
shogun-notifier- | shogun: turn off unused variables warning for cpp meta examples | 00:07 |
shogun-buildbot | build #514 of trusty - libshogun - viennacl is complete: Failure [failed configure] Build details are at http://buildbot.shogun-toolbox.org/builders/trusty%20-%20libshogun%20-%20viennacl/builds/514 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 00:08 |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Quit: Leaving.] | 00:08 | |
-!- will__ [cc2db69e@gateway/web/freenode/ip.204.45.182.158] has joined #shogun | 02:53 | |
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 03:08 | |
-!- BurningBear [79edb837@gateway/web/cgi-irc/kiwiirc.com/ip.121.237.184.55] has joined #shogun | 05:00 | |
-!- lambday [6a33e81f@gateway/web/freenode/ip.106.51.232.31] has quit [Ping timeout: 252 seconds] | 05:44 | |
-!- lambday [6a33f0ce@gateway/web/freenode/ip.106.51.240.206] has joined #shogun | 06:41 | |
-!- mode/#shogun [+o lambday] by ChanServ | 06:41 | |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has joined #shogun | 06:51 | |
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 246 seconds] | 07:02 | |
-!- besser82 [~besser82@fedora/besser82] has joined #shogun | 07:03 | |
-!- mode/#shogun [+o besser82] by ChanServ | 07:03 | |
mizari | heya guys. wanna join GSoC, sorry for being late :s | 07:45 |
mizari | I'm interested in following ideas: | 07:45 |
mizari | linux part of https://github.com/shogun-toolbox/shogun/wiki/GSoC_2015_project_installation | 07:45 |
mizari | https://github.com/shogun-toolbox/shogun/wiki/GSoC_2015_clean_up_infrastructure | 07:45 |
mizari | https://github.com/shogun-toolbox/shogun/wiki/GSoC_2015_cluster_shogun if or when time's left | 07:45 |
mizari | I have little to none knowledge of ML, but I really wanna get into it | 07:46 |
mizari | so from shogun's GSoC I expect myself to improve my C++ skills and ML, so I can help after :p | 07:46 |
mizari | can I get someone to talk about it and pick up some tasks? | 07:47 |
shogun-buildbot | build #25 of deb1 - libshogun - PR is complete: Failure [failed cookbook] Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun%20-%20PR/builds/25 blamelist: tbskm | 07:55 |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has quit [Ping timeout: 252 seconds] | 08:01 | |
-!- will__ [cc2db69e@gateway/web/freenode/ip.204.45.182.158] has quit [Ping timeout: 252 seconds] | 08:12 | |
BurningBear | hello, mizari. I am new here and I wanna join GSoC too.I am a student in China. maybe we can study and make progress together! | 08:44 |
mizari | thats nice, what is the idea you want to work on? | 08:47 |
BurningBear | I am interested in "Fundamental Machine Learning Algorithms" | 08:49 |
BurningBear | https://github.com/shogun-toolbox/shogun/wiki/GSoC_2016_project_fundamental_usual_suspects | 08:49 |
mizari | two students could be there, sounds like a good idea | 08:52 |
mizari | I also like the project itself | 08:52 |
BurningBear | :) haha, I have little knowledge of ML too, so I think we must try our best to learn it! | 09:00 |
BurningBear | may I get your E-mail? | 09:01 |
mizari | PM'ed | 09:13 |
-!- curiousguy13 [~curiousgu@120.59.109.116] has quit [Ping timeout: 264 seconds] | 09:20 | |
-!- curiousguy13 [~curiousgu@120.59.96.172] has joined #shogun | 09:32 | |
-!- mizari [~mizari@95-174-213-100.nts.su] has quit [Ping timeout: 248 seconds] | 09:58 | |
-!- BurningBear [79edb837@gateway/web/cgi-irc/kiwiirc.com/ip.121.237.184.55] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] | 10:05 | |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has joined #shogun | 10:09 | |
-!- mizari [~mizari@95-174-213-100.nts.su] has joined #shogun | 10:11 | |
-!- BurningBear [79edb837@gateway/web/cgi-irc/kiwiirc.com/ip.121.237.184.55] has joined #shogun | 10:19 | |
sanuj | lisitsyn: there? | 11:07 |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun | 11:18 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 11:18 | |
shogun-buildbot | build #515 of trusty - libshogun - viennacl is complete: Failure [failed configure] Build details are at http://buildbot.shogun-toolbox.org/builders/trusty%20-%20libshogun%20-%20viennacl/builds/515 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 11:23 |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 11:23 | |
shogun-notifier- | shogun: Heiko Strathmann :develop * 25eacf4 / CMakeLists.txt: https://github.com/shogun-toolbox/shogun/commit/25eacf4bc5362f8e2a25e39352f816ab2dfe1d78 | 11:23 |
shogun-notifier- | shogun: fix HAVE_LINALG_LIB not being set if ViennaCL is not available (but eigen3 is) | 11:23 |
shogun-buildbot | build #26 of deb1 - libshogun - PR is complete: Success [build successful] Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun%20-%20PR/builds/26 | 11:27 |
shogun-notifier- | shogun: youssef :develop * 261d224 / src/shogun/machine/LinearMachine.cpp: https://github.com/shogun-toolbox/shogun/commit/261d2246c9a66d355b56d42d535a27a717fabf8b | 11:28 |
shogun-notifier- | shogun: rename params to avoid warnings | 11:28 |
shogun-notifier- | shogun: Heiko Strathmann :develop * 197fe44 / src/shogun/machine/LinearMachine.cpp: https://github.com/shogun-toolbox/shogun/commit/197fe44b19f47f512f61c4808672899b481b243e | 11:28 |
shogun-notifier- | shogun: Merge pull request #3065 from youssef-emad/develop | 11:28 |
shogun-notifier- | shogun: | 11:28 |
shogun-notifier- | shogun: Avoid compiler warnings in LinearMachine | 11:28 |
shogun-buildbot | build #516 of trusty - libshogun - viennacl is complete: Failure [failed configure] Build details are at http://buildbot.shogun-toolbox.org/builders/trusty%20-%20libshogun%20-%20viennacl/builds/516 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, youssef <youssef.emad.attia@gmail.com> | 11:28 |
-!- BurningBear [79edb837@gateway/web/cgi-irc/kiwiirc.com/ip.121.237.184.55] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] | 11:41 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Quit: Leaving.] | 11:43 | |
-!- sonne|osx [~sonne@x4db453b4.dyn.telefonica.de] has joined #shogun | 11:49 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun | 11:52 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 11:52 | |
sanuj | HeikoS: how to generate least_squares_regression.dat? | 11:53 |
@HeikoS | sanuj: hi | 11:53 |
@HeikoS | run `make test` | 11:53 |
sanuj | oh | 11:53 |
@HeikoS | and then check build/examples/meta/cpp/classifier/knn.dat | 11:53 |
@HeikoS | if this is not clear from the Readme, please add it :) | 11:53 |
sanuj | HeikoS: i tried generate.py | 11:54 |
@HeikoS | this file has to go to data/testsuite/meta/classifier/knn.dat | 11:54 |
sanuj | HeikoS: cool | 11:54 |
@HeikoS | sanuj: no thats old | 11:54 |
@HeikoS | the meta examples use something else | 11:54 |
@HeikoS | please update any readme if you find something outdated | 11:54 |
sanuj | HeikoS: | 11:54 |
sanuj | yeah | 11:54 |
@HeikoS | sanuj: great that you are using it, I am looking forward to see these cookpage example tests in action | 11:54 |
sanuj | i'll send a PR to shogun data for this | 11:55 |
sanuj | start a new PR for linear ridge regression cookbook | 11:55 |
@HeikoS | sanuj: yes you have to update data | 11:59 |
@HeikoS | and then use this updated data revision in your PR where you send the example, it will be two PRs | 11:59 |
@HeikoS | sanuj: and please update the readme if you think it is not clear, so we dont have to explain this to 100 people again and again ;) | 11:59 |
sanuj | HeikoS: sure i'll update the readme | 12:00 |
@HeikoS | sanuj: great thanks a lot | 12:00 |
sanuj | :) | 12:00 |
sanuj | HeikoS: In this wiki https://github.com/shogun-toolbox/shogun/wiki/Example_Generation | 12:08 |
-!- curiousguy13 [~curiousgu@120.59.96.172] has quit [Remote host closed the connection] | 12:08 | |
sanuj | 'real' is not mentioned. it was used at some point of time in knn.sg | 12:08 |
@HeikoS | sanuj: this is not relevant fot the meta examples | 12:08 |
@HeikoS | I see | 12:09 |
sanuj | HeikoS: using 'float' gives an error | 12:09 |
@HeikoS | I see | 12:09 |
@HeikoS | good catch | 12:10 |
@HeikoS | sanuj: that is in fact defined in the target jsons | 12:10 |
@HeikoS | if you want to re-add this, check cpp.json, csharp.json and java.json | 12:10 |
@HeikoS | they tell how the convertion works | 12:10 |
@HeikoS | I just put in real for now | 12:10 |
@HeikoS | could put in float as well | 12:11 |
sanuj | HeikoS: okay | 12:11 |
@HeikoS | sanuj: but not important for now | 12:11 |
@HeikoS | ill take care of it later | 12:11 |
sanuj | HeikoS: cool | 12:11 |
-!- lambday [6a33f0ce@gateway/web/freenode/ip.106.51.240.206] has quit [Ping timeout: 252 seconds] | 12:13 | |
-!- sonne|osx [~sonne@x4db453b4.dyn.telefonica.de] has quit [Quit: sonne|osx] | 12:20 | |
sanuj | HeikoS: https://github.com/shogun-toolbox/shogun-data/pull/80 | 13:01 |
sanuj | is it correct? | 13:01 |
sanuj | i don't see any serialized data in that file | 13:01 |
@HeikoS | sanuj: yes see my comment | 13:10 |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Quit: Leaving.] | 13:11 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun | 13:25 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 13:25 | |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has quit [Ping timeout: 252 seconds] | 13:27 | |
-!- sonne|osx [~sonne@x4db453b4.dyn.telefonica.de] has joined #shogun | 13:28 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Client Quit] | 13:28 | |
-!- curiousguy13 [~curiousgu@120.59.96.172] has joined #shogun | 13:37 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun | 13:38 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 13:38 | |
-!- sonne|osx [~sonne@x4db453b4.dyn.telefonica.de] has quit [Quit: sonne|osx] | 13:52 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Quit: Leaving.] | 13:52 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun | 13:52 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 13:52 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Client Quit] | 13:53 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun | 14:07 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 14:07 | |
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 14:28 | |
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 268 seconds] | 14:50 | |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has joined #shogun | 15:09 | |
sanuj | HeikoS: can you review the content in updated lsr cookbook page? I'll put almost the same content in LinearRidgeRegression page also. | 15:31 |
@HeikoS | sanuj: will do, but I have to fix the integration test file first. Not sure what the problem there is | 15:32 |
sanuj | https://github.com/shogun-toolbox/shogun/pull/3064 | 15:32 |
@HeikoS | sanuj: I am currently preparing a talk for tomorrow, but will have a look a bit later | 15:32 |
sanuj | HeikoS: sure :) | 15:32 |
@HeikoS | http://papers.nips.cc/paper/6006-learning-structured-densities-via-infinite-dimensional-exponential-families.pdf | 15:32 |
@HeikoS | in case you like kernels ;) | 15:32 |
sanuj | HeikoS: nice! | 15:34 |
sanuj | you are going to talk about this? | 15:35 |
-!- lambday [6a33852b@gateway/web/freenode/ip.106.51.133.43] has joined #shogun | 15:39 | |
-!- mode/#shogun [+o lambday] by ChanServ | 15:39 | |
-!- lambday [6a33852b@gateway/web/freenode/ip.106.51.133.43] has quit [Client Quit] | 15:40 | |
-!- lambday [6a33852b@gateway/web/freenode/ip.106.51.133.43] has joined #shogun | 15:49 | |
-!- mode/#shogun [+o lambday] by ChanServ | 15:49 | |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has quit [Ping timeout: 252 seconds] | 16:01 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Quit: Leaving.] | 16:03 | |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has joined #shogun | 16:44 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 16:44 | |
-!- besser82 [~besser82@fedora/besser82] has joined #shogun | 17:10 | |
-!- mode/#shogun [+o besser82] by ChanServ | 17:10 | |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 17:42 | |
shogun-notifier- | shogun-data: Heiko Strathmann :master * 591f9d5 / testsuite/meta/classifier/knn.dat: https://github.com/shogun-toolbox/shogun-data/commit/591f9d567a9fb3de2354627b3c81053d6321e383 | 17:42 |
shogun-notifier- | shogun-data: correct knn meta example output | 17:42 |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has joined #shogun | 17:47 | |
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 250 seconds] | 17:50 | |
@HeikoS | sanuj: I sent a fix for the integration text | 17:51 |
@HeikoS | test | 17:51 |
@HeikoS | once it is merged, you can use the example as inspiration again | 17:51 |
@HeikoS | sanuj: note I changed my mind on evaluation, a sinmple case should be in there | 17:51 |
sanuj | HeikoS: yeah i saw | 17:51 |
@HeikoS | all real valued results are stored to file | 17:52 |
@HeikoS | the problem was a wrong method call on the labels | 17:52 |
@HeikoS | get_values rather than get_labels | 17:52 |
@HeikoS | http://buildbot.shogun-toolbox.org/cookbook_pr/fed6c4075a52028fb11206ce2aafcdbf09411bce/examples/classifier/knn.html | 17:52 |
sanuj | HeikoS: how do you get the links to the buildbot cookbook builds? :P | 17:53 |
sanuj | HeikoS: and why do we have a buildbot when we have travis? | 17:53 |
@HeikoS | sanuj: wiking did soem magic | 17:53 |
@HeikoS | sanuj: we now generate the html output of the cookbook on the buildbot for every PR sent | 17:54 |
@HeikoS | http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun%20-%20PR | 17:54 |
sanuj | HeikoS: yes | 17:54 |
@HeikoS | sanuj: and we have the buildbot because travis does not give us unlimited computation time or architectures | 17:54 |
@HeikoS | we run shogun build and tests in all sorts of flavours, update the notebooks and doc online, etc | 17:55 |
@HeikoS | http://buildbot.shogun-toolbox.org/waterfall | 17:55 |
@HeikoS | you can see all the differnt OS there | 17:55 |
@HeikoS | the cookbook feature is the best, since I can read the patches in rendered html rather than the source file itself | 17:55 |
sanuj | HeikoS: yeah i have seen those before..... | 17:55 |
sanuj | HeikoS: agreed about the cookbook build features | 17:56 |
sanuj | HeikoS: it's sweet | 17:56 |
sanuj | buildbot looks vast | 17:56 |
sanuj | HeikoS: i will have to update the PR before you review it....add back mean squared error etc | 17:59 |
-!- besser82 [~besser82@fedora/besser82] has joined #shogun | 18:09 | |
-!- mode/#shogun [+o besser82] by ChanServ | 18:09 | |
-!- sonne|osx [~sonne@x4db453b4.dyn.telefonica.de] has joined #shogun | 18:09 | |
shogun-notifier- | shogun: Heiko Strathmann :develop * e1c923c / / (7 files): https://github.com/shogun-toolbox/shogun/commit/e1c923c7114a675a6192a23294d3d4504b0ef370 | 18:14 |
shogun-notifier- | shogun: general update of knn example, in particular integration test | 18:14 |
@HeikoS | sanuj: merged it | 18:14 |
shogun-notifier- | shogun: Heiko Strathmann :develop * fed6c40 / doc/cookbook/README.md: https://github.com/shogun-toolbox/shogun/commit/fed6c4075a52028fb11206ce2aafcdbf09411bce | 18:14 |
shogun-notifier- | shogun: added comment on integration test reference file locations | 18:14 |
shogun-notifier- | shogun: Heiko Strathmann :develop * edd688b / / (8 files): https://github.com/shogun-toolbox/shogun/commit/edd688be75ffa2d5a9905c72114264318c7ee0b2 | 18:14 |
shogun-notifier- | shogun: Merge pull request #3068 from karlnapf/develop | 18:14 |
shogun-notifier- | shogun: | 18:14 |
shogun-notifier- | shogun: general update of knn example, in particular integration test | 18:14 |
shogun-buildbot | build #517 of trusty - libshogun - viennacl is complete: Failure [failed configure] Build details are at http://buildbot.shogun-toolbox.org/builders/trusty%20-%20libshogun%20-%20viennacl/builds/517 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 18:14 |
@HeikoS | sanuj: ^ | 18:14 |
shogun-buildbot | build #518 of trusty - libshogun - viennacl is complete: Failure [failed configure] Build details are at http://buildbot.shogun-toolbox.org/builders/trusty%20-%20libshogun%20-%20viennacl/builds/518 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 18:15 |
sanuj | HeikoS: cool, i'll rebase and update my pR | 18:15 |
lisitsyn | hey | 18:25 |
lisitsyn | HeikoS: no news yet | 18:25 |
@HeikoS | lisitsyn: hi! | 18:25 |
@HeikoS | news on website you mean? | 18:25 |
lisitsyn | yeap | 18:26 |
@HeikoS | lisitsyn: ok | 18:26 |
@HeikoS | about to merge first PR for cookbook | 18:26 |
@HeikoS | lisitsyn: hey just also wanted to follow up on this "show" listing thing | 18:26 |
lisitsyn | HeikoS: oh I could focus on that now | 18:27 |
lisitsyn | HeikoS: but! | 18:27 |
lisitsyn | HeikoS: need your advice ;) | 18:28 |
lisitsyn | HeikoS: what's best short GP introduction you know? | 18:28 |
@HeikoS | lisitsyn: the rasmussen book | 18:28 |
@HeikoS | it is wonderful | 18:28 |
@HeikoS | chapter 3 | 18:29 |
@HeikoS | ah no | 18:29 |
@HeikoS | 2 | 18:29 |
@HeikoS | the one on regression | 18:29 |
lisitsyn | HeikoS: yeah that's true | 18:29 |
lisitsyn | so nothing better you know? | 18:29 |
@HeikoS | 3 is classification | 18:29 |
@HeikoS | lisitsyn: i dont think one can do this better | 18:29 |
lisitsyn | hah I see cool | 18:29 |
lisitsyn | ok will use it then | 18:29 |
@HeikoS | lisitsyn: let me know if you have questions | 18:29 |
lisitsyn | HeikoS: I have to put GPs in a max 3 hour long lecture | 18:30 |
@HeikoS | thats fine | 18:30 |
@HeikoS | you need 3 slides | 18:30 |
@HeikoS | http://www.gaussianprocess.org/gpml/chapters/RW2.pdf | 18:30 |
@lambday | lisitsyn: you're giving lectures? where? | 18:31 |
@HeikoS | figure 2.2 | 18:31 |
@lambday | I might use some of the slides to get introduced to GPs | 18:31 |
@HeikoS | figure 2.5 | 18:31 |
lisitsyn | lambday: yes, hse | 18:31 |
@HeikoS | and one slide with the math | 18:31 |
@HeikoS | eq 2.21 | 18:32 |
@HeikoS | thats all people need to know ;) | 18:32 |
lisitsyn | HeikoS: hah | 18:32 |
@HeikoS | we also have this wonderful video lecture from the shogun workshop ;) | 18:32 |
lisitsyn | HeikoS: btw my student is currently researching this spectral mixture kernel | 18:32 |
@HeikoS | lisitsyn: I see | 18:32 |
@HeikoS | and? | 18:32 |
@HeikoS | did he already realise it is crap? :) | 18:33 |
lisitsyn | its working but | 18:33 |
lisitsyn | well | 18:33 |
lisitsyn | I don't want it to be crap | 18:33 |
@HeikoS | btw esben hopefully soon fixes his PR on that | 18:33 |
@HeikoS | he impoemented it in shogun | 18:33 |
lisitsyn | HeikoS: we had some daily activity dataset | 18:33 |
lisitsyn | and this thing made good enough regression | 18:33 |
@HeikoS | lisitsyn: it is impossible to optimise the parameters | 18:34 |
@HeikoS | other approaches are way more stable | 18:34 |
lisitsyn | what other approaches you could mention? | 18:34 |
@HeikoS | 1.) hand chosen kernel composition | 18:34 |
@HeikoS | check the rassmussen bok | 18:35 |
@HeikoS | http://www.gaussianprocess.org/gpml/chapters/RW5.pdf | 18:35 |
@HeikoS | section 5.4.3 | 18:35 |
@HeikoS | this is easier to tune | 18:35 |
@HeikoS | or do something like they do in the automated statistician | 18:35 |
@HeikoS | http://www.automaticstatistician.com/index/ | 18:35 |
@HeikoS | they have a grammar of kernel functions and automatically compose them | 18:36 |
@HeikoS | http://www.automaticstatistician.com/static/abcdoutput/01-airline.pdf | 18:36 |
@HeikoS | interpretable | 18:36 |
@HeikoS | and automatic | 18:36 |
lisitsyn | HeikoS: oh but that's still kernel | 18:36 |
sanuj | HeikoS: I'm still getting empty least_squares_regression.dat | 18:36 |
lisitsyn | this makes me happy enough | 18:36 |
sanuj | HeikoS: using get_labels() | 18:36 |
@HeikoS | sanuj: run make meta_examples | 18:36 |
sanuj | i did make examples | 18:37 |
sanuj | and make test | 18:37 |
@HeikoS | sanuj: is the cpp code updated? | 18:37 |
lisitsyn | HeikoS: the thing is that these standard models like ARMA don't do this kool enough | 18:37 |
sanuj | yes | 18:37 |
@HeikoS | lisitsyn: if you wanna be kool choose the SM kernel and witness it failing ;) | 18:38 |
@HeikoS | lisitsyn: no honestly, I dont reccomend using it | 18:38 |
@HeikoS | sanuj: its works here | 18:38 |
sanuj | HeikoS: this is what i have in cpp => auto output = labels_test->get_labels(); | 18:38 |
@HeikoS | and you are checking the correct .dat file? | 18:38 |
sanuj | HeikoS: okay, i'll see | 18:39 |
@HeikoS | it is generated in the build dir | 18:39 |
@HeikoS | not in data | 18:39 |
@HeikoS | check build/tests/meta/generated_results/cpp/classifier | 18:39 |
sanuj | HeikoS: /home/sanuj/Projects/shogun/build_new/examples/meta/cpp/regression/least_squares_regression.dat | 18:39 |
sanuj | i deleted the old one just to make sure | 18:39 |
@HeikoS | and still empty? | 18:40 |
sanuj | HeikoS: knn.dat is also empty | 18:40 |
sanuj | yeah | 18:40 |
@HeikoS | not for me | 18:40 |
@HeikoS | git pull? | 18:40 |
sanuj | yeah i did git pull | 18:41 |
sanuj | i'll build a fresh build | 18:41 |
@HeikoS | and make test fails? | 18:41 |
@HeikoS | because the data file is NOT empty | 18:41 |
sanuj | HeikoS: all the generated tests passed | 18:42 |
@HeikoS | and the knn.dat in data/testsuite/meta/classifier/? | 18:43 |
sanuj | HeikoS: is empty | 18:45 |
@HeikoS | you have to do git pull in data then | 18:45 |
@HeikoS | https://github.com/shogun-toolbox/shogun-data/blob/master/testsuite/meta/classifier/knn.dat | 18:45 |
sanuj | oh wait | 18:46 |
sanuj | HeikoS: 160 - integration_meta_cpp-classifier_knn (Failed) | 18:46 |
@HeikoS | you missed some git pull, pretty sure | 18:46 |
@HeikoS | I gotta run now | 18:46 |
@HeikoS | see you later :) | 18:46 |
-!- HeikoS [~heiko@host-92-0-162-192.as43234.net] has quit [Quit: Leaving.] | 18:46 | |
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 250 seconds] | 18:55 | |
-!- lambday [6a33852b@gateway/web/freenode/ip.106.51.133.43] has quit [Ping timeout: 252 seconds] | 20:17 | |
-!- sonne|osx [~sonne@x4db453b4.dyn.telefonica.de] has quit [Quit: sonne|osx] | 20:45 | |
-!- besser82 [~besser82@fedora/besser82] has joined #shogun | 20:53 | |
-!- mode/#shogun [+o besser82] by ChanServ | 20:53 | |
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 248 seconds] | 21:11 | |
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 21:14 | |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has quit [Quit: Page closed] | 21:31 | |
-!- bsxfun [9d8b131c@gateway/web/freenode/ip.157.139.19.28] has joined #shogun | 22:01 | |
-!- bsxfun [9d8b131c@gateway/web/freenode/ip.157.139.19.28] has quit [Client Quit] | 22:06 | |
-!- arunreddy [46a25370@gateway/web/freenode/ip.70.162.83.112] has joined #shogun | 22:51 | |
-!- arunreddy [46a25370@gateway/web/freenode/ip.70.162.83.112] has quit [Ping timeout: 252 seconds] | 23:38 | |
--- Log closed Mon Mar 14 00:00:29 2016 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!