--- Log opened Wed Jul 12 00:00:51 2017
-!- travis-ci [~travis-ci@ec2-54-224-193-41.compute-1.amazonaws.com] has joined #shogun00:10
travis-ciit's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/25256763900:10
-!- travis-ci [~travis-ci@ec2-54-224-193-41.compute-1.amazonaws.com] has left #shogun []00:10
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/3924 opened by micmn00:44
-!- travis-ci [~travis-ci@ec2-54-224-193-41.compute-1.amazonaws.com] has joined #shogun00:55
travis-ciit's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/25256763900:55
-!- travis-ci [~travis-ci@ec2-54-224-193-41.compute-1.amazonaws.com] has left #shogun []00:55
-shogun-buildbot:#shogun- Build nightly_all #15 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/22/builds/1504:06
-!- mikeling [uid89706@gateway/web/irccloud.com/x-siynttqijkgauifg] has joined #shogun04:52
-!- geektoni [c1cdd220@gateway/web/freenode/ip.] has joined #shogun09:23
-!- geektoni [c1cdd220@gateway/web/freenode/ip.] has quit [Ping timeout: 260 seconds]10:50
-!- geektoni [c1cdd220@gateway/web/freenode/ip.] has joined #shogun11:04
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/3906 synchronized by MikeLing11:57
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/3906 synchronized by MikeLing12:01
-!- geektoni [c1cdd220@gateway/web/freenode/ip.] has quit [Ping timeout: 260 seconds]12:14
-!- HeikoS [~heiko@eduroam-int-pat-8-96.ucl.ac.uk] has joined #shogun13:18
-!- mode/#shogun [+o HeikoS] by ChanServ13:18
@wikingHeikoS, type in "machine learning library" into goog13:19
Trixiswiking: cant see shogun on the first 2 pages :(13:24
@wikingTrixis, but on the top :)13:24
Trixisoh its because i skipped the google suggestions thing13:24
@wikinggood stuff :)13:28
@wikinglisitsyn, if u check our analytics webpage it is obvious that we need win packages :)13:31
@wikinglisitsyn, do u know anybody familiar wiht msvc?13:32
-!- HeikoS [~heiko@eduroam-int-pat-8-96.ucl.ac.uk] has quit [Quit: Leaving.]13:32
-!- besser82_ [~besser82@fedora/besser82] has joined #shogun14:42
-!- mode/#shogun [+o besser82_] by ChanServ14:42
-!- rcurtin [~ryan@knife.lugatgt.org] has quit [Ping timeout: 240 seconds]14:43
-!- besser82 [~besser82@fedora/besser82] has quit [Excess Flood]14:43
-!- besser82_ is now known as besser8214:43
mikelingwiking: Hi, for the prng things, do we want to ensure all the meta tests still the same? I mean get same result wit h the reference data14:44
-!- geektoni [c1cdd220@gateway/web/freenode/ip.] has joined #shogun14:46
-!- rcurtin [~ryan@knife.lugatgt.org] has joined #shogun14:50
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun15:02
-!- mode/#shogun [+o HeikoS] by ChanServ15:02
-!- rcurtin [~ryan@knife.lugatgt.org] has quit [Ping timeout: 248 seconds]15:08
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/3925 opened by geektoni15:15
geektoniping HeikoS15:24
@HeikoSgeektoni: jojo15:25
geektoniHeikoS: regarding the xvalidation thing15:25
@HeikoSgeektoni: yes!15:25
geektoniwhat do you want to monitor exactly? :)15:25
@HeikoSso what would be cool would be to get the results of15:26
@HeikoS-each fold15:26
@HeikoS-weights/parameters of the model for each fold15:26
@HeikoS(e.g. MKL weights)15:26
@HeikoS-average fold performance over multiple runs15:26
@HeikoSgeektoni: and then there is a kernel section framework in the statistical_testing impl code15:27
@HeikoSthat has similar things, there we want to extract certain parameters when doing the selection15:27
geektonimmh I see15:27
geektoniso I'll start digging into xvalidation code15:27
@HeikoSgeektoni: it is quite simple15:28
@HeikoSyou have a CMachine15:28
@HeikoSand CFeatures15:28
@HeikoSand CLabels15:28
@HeikoSthen you split the data into say 5 folds15:28
@HeikoStrain on 4, test on 115:28
@HeikoSthats a fold result (say accuracy)15:29
@HeikoSthen you do that for all combinations, so you get 5 fold results15:29
@HeikoS(you migh want to store the fold incdices as well)15:29
@HeikoSthen you might repeat with a different random splitting15:29
@HeikoSin addition to the say test performance, you might also want to look at say the w vector of a linear model for each fold15:29
@HeikoSgeektoni: but start simple: result of each fold :)15:29
geektoniHeikoS: aha yeah, one thing at a time15:31
@HeikoSgeektoni: so the thing is15:31
@HeikoSI believe you register model parameters15:31
@HeikoSgeektoni: so the fold results might not be a parameter of x-validation15:31
@HeikoSso it might need some refactoring15:31
@HeikoSbut that should be easy15:31
geektoniactually no, there is no such thing. There is a general method which emits a value, without worrying about if the value emitted comes from a registered parameter or not.15:32
geektoniHeikoS: ^15:33
@HeikoSgeektoni: ah cok15:33
@HeikoSthen it might be easier15:33
@HeikoSwanna start with the results of each fold?15:33
@HeikoSthat is most useful15:33
@HeikoS(and we have API for that I think)15:33
@HeikoSthere is a class like CrossValidationResults or so15:34
@HeikoSthis should be deleted15:34
@HeikoSand the mean should be returned15:34
@HeikoSand if one wants more (standard deviation for example)15:35
@HeikoSone registers an observer15:35
@HeikoSand boonm15:35
-!- iglesiasg [~iglesiasg@] has joined #shogun15:35
-!- mode/#shogun [+o iglesiasg] by ChanServ15:35
@wikingmikeling, nono we can change it15:36
@wikingmikeling, but not yet15:36
@wikingmikeling, because as soon as we change to c++11 random15:36
@wikinganyways many things will change15:36
@HeikoSah man I should rewrite the xvalidation class15:37
geektoniHeikoS: ok, I'll start doing things then ;)15:37
@HeikoSgeektoni: https://github.com/shogun-toolbox/shogun/blob/develop/src/shogun/evaluation/CrossValidationMKLStorage.h15:38
@wikingmikeling, here?15:38
@HeikoSplease please delete this! :)15:38
@HeikoSthis is exactly the kind of sh** we were doing before the nice observables15:38
@wikingHeikoS, :D15:38
@HeikoSand there is more of those15:39
@HeikoSetc etc15:39
@HeikoSgeektoni: please put them on some todo list15:39
geektoniHeikoS: sure15:40
@HeikoSwiking: whats up with some?15:41
@HeikoSis there any perspective on this?15:41
@wikingsome is some :(15:41
@HeikoSwiking: how to SWIG it?15:44
@HeikoShow to get rid of raw pointers?15:44
mikelingwiking: hi, yes15:44
@HeikoSwiking: you have some ideas there? :D15:44
@HeikoSwiking: some thing?15:44
@wikinggeektoni, https://travis-ci.org/shogun-toolbox/shogun/jobs/252804570#L97415:44
Trixishaha cross validation15:46
Trixiscoding grid search w/ cross validation is more work than i expected. (because custom kernels :|)15:46
geektoniwiking: ahhh that's clang fault :>15:47
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/3925 synchronized by geektoni15:50
-!- rcurtin [~ryan@knife.lugatgt.org] has joined #shogun16:23
-!- geektoni [c1cdd220@gateway/web/freenode/ip.] has quit [Quit: Page closed]16:48
-!- dennyjooo [~dennyjooo@] has joined #shogun16:49
-!- OXPHOS [92bd305b@gateway/web/freenode/ip.] has joined #shogun17:41
-!- dennyjooo [~dennyjooo@] has quit []17:42
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 248 seconds]18:32
-!- sanuj [~sanuj@] has joined #shogun19:34
-!- HeikoS [~heiko@host-92-0-169-11.as43234.net] has joined #shogun21:53
-!- mode/#shogun [+o HeikoS] by ChanServ21:53
-!- sanuj [~sanuj@] has quit [Remote host closed the connection]23:15
-!- HeikoS [~heiko@host-92-0-169-11.as43234.net] has quit [Quit: Leaving.]23:33
--- Log closed Thu Jul 13 00:00:53 2017