Open in new window / Try shogun cloud
--- Log opened Sun Mar 01 00:00:40 2015
-!- HeikoS [~heiko@2.221.116.151] has joined #shogun00:09
-!- mode/#shogun [+o HeikoS] by ChanServ00:09
@lisitsynHeikoS: yeah you get it right I think :)00:20
@lisitsynHeikoS: the other thing - you know why the pointer is null00:20
@HeikoSbecause the thing was freed?00:20
-!- vortex_ape [~vortex_ap@120.59.205.250] has joined #shogun00:20
@lisitsynno I mean00:20
@lisitsynif you have some wrapper for the thing that is potentially absent00:21
@lisitsynyou can check whether it is absent00:21
@lisitsynand then if it is absent you can log something00:21
@HeikoSI sewe00:21
@HeikoSsee00:21
@HeikoScool thing00:21
@HeikoSlisitsyn: what about this: https://github.com/shogun-toolbox/shogun/wiki/GSoC_2015_plugin00:22
@lisitsynHeikoS: some very common use case is to find something in collection00:22
@HeikoSdescription is too short00:22
@HeikoStimeline, waypoints, entrance tasks, suggested reading, examples etc00:22
@lisitsynHeikoS: oh yeah let me extend it00:23
@lisitsynwill do it tomorrow00:23
@HeikoScool00:23
@HeikoSlisitsyn: and this one would also be good to have  adescription for00:23
@HeikoShttps://github.com/shogun-toolbox/shogun/wiki/GSoC_2015_project_parameters00:23
@HeikoSwith your new ideas for that00:23
@lisitsynok00:23
@lisitsynHeikoS: good00:24
@HeikoSbtw did you look at this user guide thing?00:24
@lisitsynsorry not yet00:24
@HeikoShttps://github.com/karlnapf/shogun_manual00:24
@lisitsynlooks interesting00:25
@HeikoSwe should get this before gsoc00:25
@HeikoSand the meta examples too00:25
@lisitsynHeikoS: get like get merged?00:25
@HeikoSso that every student can write like 1 entry each week, and translate 3 minimal examples00:25
@HeikoSlisitsyn: yes, and easy to extend00:25
@HeikoSmeta examples are in00:25
@HeikoSI am currently editing the travis file to include it in the build00:26
@HeikoSno idea how to change buildbot00:26
@lisitsynhmmm I thought someone knows how to change buildbot00:28
@lisitsynI guess only wiking00:28
@HeikoSlisitsyn: just emailed wiking :)00:43
kingsleyDo you happen to know if anyone has tried optimizing the hyperparameters passed to machine learning algorithms (like the number of trees to be used by a random forest) with pre-existing non-linear optimization code like gradient descent or integer programming?00:44
kingsleyI'd like to think that a guided search of hyperparaemters would be more efficient than guessing, trial and error.00:45
@lisitsynHeikoS: ^ I think you're expert here ;)00:45
@HeikoSkingsley: for sure00:45
@lisitsynkingsley: people like to model performance as gaussian processes00:46
kingsleyHeikoS: Please elaborate.00:46
@lisitsynthat's how MOE and hyperopt work00:46
@HeikoSkingsley: what people often do is to use stochastic black box optimisers for example, see cma00:46
@HeikoSkingsley: or look up bayesian optimisation, which is also about this idea00:46
@lisitsynhttps://github.com/Yelp/MOE00:46
@lisitsynhttps://github.com/hyperopt/hyperopt00:47
@lisitsynhttps://github.com/JasperSnoek/spearmint00:47
@HeikoSwhere the idea is to put a gaussian process on the surface of the function your are optimising, and then use information theoretic ideas to select the next point to look at using a exploration-exploitation tradeoff00:47
kingsleyOK, you have given me much to think about.00:48
@HeikoSkingsley: would be a cool thing to have for Shogun00:48
kingsleyI'll check them out.00:48
@lisitsynHeikoS: btw00:49
@HeikoSdiscrete case is another thing00:49
@HeikoSmuch much harder to do00:49
@lisitsynI was curious about MOE00:49
@lisitsynhave you seen how it works?00:49
@lisitsynwanted to ask you whether we can do the same in shogun00:49
@HeikoSMOE uses bayesian optimisation00:49
@HeikoSI invited Matthew Hoffman to give a talk here at UCL on Friday00:50
@lisitsynyes with GPs00:50
@HeikoShttp://mlg.eng.cam.ac.uk/hoffmanm/00:50
@lisitsynso can we do this with shogun and if not what is missed?00:50
@HeikoSlisitsyn, kingsley https://github.com/mwhoffman/pybo00:50
@HeikoSthats the way to do this00:50
kingsleyHeikoS and lisitsyn: Thank you very much for saving me from re-inventing the wheel.00:51
@HeikoSlisitsyn: would be cool to use MOE in shogun, what do you think?00:51
@lisitsynHeikoS: yes or do the same thing with our own GPs00:51
@HeikoSlisitsyn: though its not parallelisable00:51
@lisitsynthat's why I am asking ;)00:51
@HeikoSlisitsyn: cma is better in that respect00:51
@HeikoSwe could use our own gps sure00:52
@HeikoSbut why do that?00:52
@HeikoSpeople have done it, its just a matter of plugging things together00:52
@HeikoSthe bayesian thing is hard to parallelise00:52
@HeikoScma is easier to00:52
@HeikoSbased on particles00:52
@lisitsynwhat's cma?00:53
@lisitsyncovariance matrix adaptation?00:53
@HeikoSyep00:53
@HeikoSthat works very nicely00:53
@HeikoSI use it for some things sometimes00:53
@HeikoSlisitsyn: though the thing is00:54
@HeikoScma is a one-guy library and also method00:54
@HeikoSwhereas bayesina optimisation is a large field with lots of excellent people doing research on00:54
@HeikoSand yelp is a kind of plus too00:54
@lisitsynwe should have both00:54
@lisitsynHeikoS: interfacing with MOE is easy00:55
@HeikoSlisitsyn: thats cool00:55
@HeikoSwe should do that00:55
@HeikoSits important00:55
@HeikoSbut next year00:55
@HeikoSthats part of making model-selection easier and offer full pipelines00:55
@HeikoSbut we are not there yet00:55
@HeikoSneed other things first00:55
@lisitsynI'd do that in a few hours I guess00:55
@HeikoSmaybe00:56
@HeikoSyeah cool thing in fact00:56
@HeikoSthough I wonder... maybe you should spend these hours on something else currently??00:56
@HeikoSwe have so many things that are not working properly00:56
@lisitsynyes :)00:56
@lisitsynI just mean it is not a big deal to interface with that00:57
@lisitsynthey have a few examples00:57
@HeikoShyperparameters are not really an issue currently00:57
@HeikoSyeah I have seen00:57
@HeikoS but then we need00:57
@HeikoScmake options00:57
@HeikoSa build00:57
@HeikoStests00:57
@HeikoSetc00:57
@HeikoSso not sure00:57
@HeikoSwhether its a good time now00:57
@HeikoShackathon thing00:57
@HeikoSactually00:57
@lisitsynno it is not :)00:57
@HeikoSthat'd be cool00:57
@lisitsyndo you think hackathon thing is possible still?00:58
@HeikoSyes00:59
@lisitsynHeikoS: UK won't let me in :D00:59
@HeikoSbut I cannot do this alone00:59
@HeikoSwe need:00:59
@HeikoS1.) BSD00:59
@HeikoS2.) foundation00:59
@HeikoS3.) user guide00:59
@HeikoSwithout that it will not work00:59
@HeikoSoh and00:59
@HeikoS0.) installation00:59
@HeikoSlisitsyn: why wont they let you in?00:59
@lisitsynHeikoS: cold war :)01:00
@lisitsynwell actually this list is doable01:00
@HeikoSlisitsyn: it IS doable!01:01
@lisitsynHeikoS: there is a wisdom I've learnt at some point01:01
@HeikoSwhich is?01:01
@lisitsynto develop something it is usually enough to set up a cycle01:01
@lisitsynwhen you change, compile and test01:01
@lisitsynand to make people use software you just have to make it installable :)01:02
@HeikoShehe01:02
@HeikoSyes01:02
@lisitsyneven with scarce functionality01:02
@lisitsynit will be used01:02
@lisitsynif it just works01:02
@lisitsynyou know people use VW so much01:03
@HeikoSyes I agree01:03
@HeikoSit is cool01:03
@HeikoSshogun is not installable01:03
@HeikoSpeople fail all the time01:03
@HeikoShalf my institude tried it01:03
@HeikoSinstitute01:03
@lisitsynhaha01:03
@HeikoSsmart people you know01:03
@lisitsynyeah that's the priority one thing01:03
@HeikoSnot linux experts but know things01:03
@HeikoSbut they just are overwhealmed01:03
@HeikoSso the install project is a must01:04
@lisitsynwe can even have two students on that01:04
@HeikoSif we dont get a student for that, I refuse to do GSoC :D01:04
@lisitsynyou're still phd01:04
@lisitsynstudent I mean01:04
@HeikoSI cannot do it01:04
@lisitsyn:)01:04
@lisitsynwhy? :D01:04
@lisitsynadmin?01:04
@lisitsyn:)01:04
@HeikoSI dont know these things01:04
@lisitsynah01:04
@HeikoSalso no time01:04
@lisitsynyeah sure :)01:04
@HeikoSviktor could do it01:05
@lisitsynI am not student and corporate slave01:05
@HeikoSbut he also doesnt have time01:05
@HeikoSyou could do it01:05
@HeikoSbut also no time01:05
@HeikoSso we need a student who somehow knows01:05
@lisitsynremember I am not a student  :)01:05
@HeikoSI do01:05
@lisitsyngood time to sleep01:06
@HeikoSsame here01:06
@HeikoSgood night!01:06
@lisitsynHeikoS: I'll extend these gsoc ideas tomorrow01:06
@lisitsynsee you! gntie01:06
-!- vortex_ape [~vortex_ap@120.59.205.250] has quit [Ping timeout: 246 seconds]01:20
-!- vortex_ape [~vortex_ap@120.59.77.110] has joined #shogun01:36
-!- vortex_ape [~vortex_ap@120.59.77.110] has quit [Quit: Leaving]02:02
-!- HeikoS [~heiko@2.221.116.151] has quit [Ping timeout: 245 seconds]02:10
-!- HeikoS [~heiko@2.221.116.151] has joined #shogun02:15
-!- mode/#shogun [+o HeikoS] by ChanServ02:15
-!- HeikoS [~heiko@2.221.116.151] has quit [Quit: Leaving.]03:54
-!- HeikoS [~heiko@2.221.116.151] has joined #shogun03:57
-!- mode/#shogun [+o HeikoS] by ChanServ03:57
-!- rajul [~rajul@node-wns.pool-125-24.dynamic.totbb.net] has joined #shogun04:04
-!- HeikoS [~heiko@2.221.116.151] has quit [Quit: Leaving.]04:06
-!- curiousguy13_ [~curiousgu@120.59.228.160] has quit [Ping timeout: 256 seconds]04:38
-!- curiousguy13_ [~curiousgu@59.177.205.235] has joined #shogun04:51
-!- prakhar2511 [~prakhar25@202.78.172.162] has joined #shogun04:58
-!- rajul [~rajul@node-wns.pool-125-24.dynamic.totbb.net] has quit [Ping timeout: 272 seconds]05:06
-!- Floatingman [~Floatingm@c-68-52-34-232.hsd1.tn.comcast.net] has quit [Remote host closed the connection]05:19
-!- rajul [~rajul@node-wns.pool-125-24.dynamic.totbb.net] has joined #shogun05:20
-!- Floatingman [~Floatingm@c-68-52-34-232.hsd1.tn.comcast.net] has joined #shogun05:23
-!- kshitijk [~kshitijk@115.249.18.25] has joined #shogun06:05
-!- rajul [~rajul@node-wns.pool-125-24.dynamic.totbb.net] has quit [Read error: Connection reset by peer]06:13
-!- rajul [~rajul@node-6gf.pool-180-180.dynamic.totbb.net] has joined #shogun06:29
-!- prakhar2511 [~prakhar25@202.78.172.162] has quit [Ping timeout: 255 seconds]06:43
-!- kshitijk [~kshitijk@115.249.18.25] has quit [Ping timeout: 256 seconds]07:49
-!- prakhar2511 [~prakhar25@202.78.172.162] has joined #shogun07:50
-!- kshitijk [~kshitijk@115.249.18.25] has joined #shogun08:11
-!- kshitijk [~kshitijk@115.249.18.25] has quit [Ping timeout: 244 seconds]08:19
-!- prakhar2511 [~prakhar25@202.78.172.162] has quit [Ping timeout: 240 seconds]08:34
-!- kshitijk [~kshitijk@115.249.18.25] has joined #shogun08:50
-!- joergkurtwegner [~joergkurt@d54c5db3e.access.telenet.be] has joined #shogun09:41
joergkurtwegnerhi, does anyone have a clean ubuntu 14.04 install routine?09:42
-!- joergkurtwegner [~joergkurt@d54c5db3e.access.telenet.be] has quit [Client Quit]09:42
-!- joergkurt [~joergkurt@178.162.199.93] has joined #shogun09:45
joergkurtping is online...09:46
-!- sonne|osx [~sonne@f049254171.adsl.alicedsl.de] has joined #shogun09:50
-!- smallgraph [~textual@vpn-cl-194-80.scc.kit.edu] has joined #shogun10:07
joergkurthi, I am looking for a clean ubuntu install instruction for python-modular, will be back in 2-3h10:12
-!- curiousguy13_ [~curiousgu@59.177.205.235] has quit [Ping timeout: 245 seconds]10:39
-!- kshitijk [~kshitijk@115.249.18.25] has quit [Ping timeout: 264 seconds]10:40
-!- vortex_ape [~vortex_ap@59.177.140.38] has joined #shogun10:44
joergkurtwith clean I mean fresh Ubuntu14.04 in VirtualBox and then a native from GIT source, if all dependencies are properly defined in the install docu, all should just compile through, I guess we neeed some refinement on this process, since my naive attempt does not work right away10:48
-!- curiousguy13_ [~curiousgu@120.59.226.121] has joined #shogun10:52
-!- kingsley_ [~kingsley@174-21-235-71.tukw.qwest.net] has joined #shogun11:07
-!- kingsley [~kingsley@174-21-171-80.tukw.qwest.net] has quit [Ping timeout: 255 seconds]11:10
@lisitsynjoergkurt: hey11:21
-!- smallgraph [~textual@vpn-cl-194-80.scc.kit.edu] has quit [Ping timeout: 256 seconds]11:22
joergkurthoi11:22
@lisitsynjoergkurt: looks like you've already installed shogun :)11:22
@lisitsynI mean from you mail I see you've run an example and it worked11:23
joergkurtnope, its compiled, but not running :(11:23
@lisitsynwhat do you mean?11:23
@lisitsynwhat if you run python and write 'from modshogun import KMeans'?11:23
joergkurtimport works, kinda busy right now, can I get back to you in 2h...?11:25
joergkurtsorry, just have to do something else ...will get back to it later...11:25
@lisitsynjoergkurt: sure. but if import works probably everything is probably ok anyway :)11:26
joergkurtok, lets figure it later :)11:26
-!- kshitijk [~kshitijk@115.249.18.25] has joined #shogun12:54
-!- prakhar2511 [~prakhar25@202.78.172.162] has joined #shogun12:58
joergkurtok, my bet, it works, I was kinda expecting distance_tanimoto_modular.py returns some console printout, its not13:06
joergkurtso, seems to works just fine, one other question13:06
joergkurtif loading a dat set like feats_train=RealFeatures(CSVFile(train_fname))13:06
joergkurthow can I get this into a python object I can also process on the Python side, and vice versa?13:07
joergkurtSecond question, how can I load a sparse file, and which formats are accepted, also for usage in Python?13:07
-!- vortex_ape [~vortex_ap@59.177.140.38] has quit [Ping timeout: 264 seconds]13:23
@lisitsynok first13:43
@lisitsynjoergkurt: what do you mean python object?13:44
@lisitsynI am not sure I get what you want to do with that13:44
-!- kshitijk [~kshitijk@115.249.18.25] has quit [Ping timeout: 250 seconds]14:15
-!- exploreshaifali [~shaifali@182.70.196.201] has joined #shogun14:22
joergkurtmaybe I just dont know all details, the CSVFile is loaded, how to I access the elements in this object?14:25
joergkurtfeats_train=RealFeatures(CSVFile(train_fname)), aka how does a data object look like and what are access functions?14:26
joergkurtsince its SWIG I assume the data is staying at the C++ side and Python is just pointing to it14:34
@lisitsynjoergkurt: yes14:36
@lisitsynjoergkurt: it depends what you want to do14:37
@lisitsynsimple access like per vector is easy14:38
@lisitsynuse get_feature_vector14:38
@lisitsynif you want to change data you'd probably have to get vector, modify it and set_feature_vector14:38
joergkurtthe fecture vector requires an index, so do I need to do this element-wise?14:42
@lisitsynjoergkurt: that's the index of vector14:43
@lisitsynso vector-wise14:43
joergkurtiterator or loop 0length ?14:43
joergkurtvector is here a feature, right? (not an instance)14:44
@lisitsynmight be confusing but it is an instance :)14:44
@lisitsynor example14:44
@lisitsynjoergkurt: you can access whole matrix as well14:44
@lisitsynget_feature_matrix14:44
@lisitsynhttp://shogun-toolbox.org/doc/en/latest/classshogun_1_1CDenseFeatures.html14:45
joergkurt(checking...)14:46
joergkurtvectors=features14:48
joergkurtfeatures=instances14:48
joergkurtright?14:48
@lisitsynjoergkurt: example is feature vector14:50
@lisitsynits components are features14:50
@wikingmorning14:58
@lisitsynwiking: hmm morning.. are you in us? :D15:00
@wikingheheh no15:05
@wikingactually it's middle of night here15:05
-!- prakhar2511 [~prakhar25@202.78.172.162] has quit [Ping timeout: 246 seconds]15:06
@lisitsynwiking: looks like you've been having busy times lately?15:08
@wikingyes15:09
@wikingsoon over15:09
joergkurtis there an overview about data formats, especially sparse ones, have seen libSVM is supported...15:17
@lisitsynjoergkurt: http://shogun-toolbox.org/doc/en/4.0.0/classshogun_1_1CLibSVMFile.html15:23
@lisitsynthat's for libsvm15:23
@lisitsynjoergkurt: hdf5 and protobuf are also supported15:23
-!- iglesiasg [~iglesias@524B8E0B.cm-4-4c.dynamic.ziggo.nl] has joined #shogun15:29
-!- mode/#shogun [+o iglesiasg] by ChanServ15:29
@lisitsyniglesiasg: who's there15:30
joergkurthdf5 support spare and dense right?15:30
joergkurtsparse and dense?15:30
@lisitsynjoergkurt: looks like only dense15:31
joergkurtimports SparseFEatures.h15:31
joergkurtehm, includes15:32
-!- prakhar2511 [~prakhar25@202.78.172.162] has joined #shogun15:32
joergkurtprobably last question for today, if I wanna bind from C++ directly, any algorithm and property parser available, or do we need to hard bind all algorithms and algo properties?15:33
@lisitsynjoergkurt: something like using classifier based on some configuration file?15:34
joergkurtyes, something just defining data+config file15:34
@lisitsynif yes then the answer is no but sounds like a feature request15:34
joergkurt:)15:35
@iglesiasglisitsyn: surprise! xD15:35
@lisitsyniglesiasg: how is it going?15:35
@iglesiasglisitsyn: very well, with you?15:36
joergkurtkinda factory pattern on C++ side or compiler-compiler for some to be defined algo+property language15:36
joergkurtBTW, where are you folks from?15:36
@lisitsynjoergkurt: yeah I'll think about it and describe as an issue15:36
@lisitsynjoergkurt: different places15:36
@lisitsynI am from the most wild country in the world15:37
joergkurtme:German:nowBelgium15:37
@lisitsynjoergkurt: you're close to iglesiasg which is now in netherlands15:37
@iglesiasgyep, me:Spanish:inNL :)15:38
joergkurtah, hi, iglesiasg, hoe gaat het?15:38
@iglesiasghet gaat heel goed, en met jou?15:38
joergkurtah, hola :)15:38
@iglesiasghahaha hola, hola15:38
joergkurthihi :P15:39
@iglesiasgwhere is Belgium are you?15:40
@iglesiasgin*15:40
joergkurtliving in Mechelen (between Brussels and Antwerp), working in Beerse/Turnhout (Janssen Pharma)15:40
@lisitsynyou have bad weather there, do you?15:41
joergkurtnever :P15:41
-!- prakhar2511 [~prakhar25@202.78.172.162] has quit [Ping timeout: 250 seconds]15:42
@lisitsynthough I am not a person to speak about bad weather :D15:42
joergkurtlisisyn-whats the wildest country?15:42
@lisitsynjoergkurt: this big bear to the east of you15:42
joergkurtrussia?15:43
@lisitsynyeap15:43
joergkurtah, cool, some of our developers are from Russi ...15:44
joergkurtRussia15:44
@lisitsynjoergkurt: somehow bad environment produces good math and software guys :)15:44
joergkurt(best dev of the world, honestly!)15:44
joergkurtok, sold now, we will integrate shogun as soon as we get to it ;)15:45
@lisitsynhaha15:45
@lisitsynjoergkurt: what do you want to apply it to?15:45
joergkurtwhere are you located (please tell me StPetersburg ;)15:45
@lisitsynjoergkurt: unfortunately moscow :)15:46
joergkurthihi, ok :P15:46
joergkurtkinda big compute ML stuff, with tendency to multi-class and MKL15:46
@lisitsynjoergkurt: svms?15:47
joergkurt...possible...its still cubic, and we will see if it can handle our data sizes15:47
@lisitsynjoergkurt: shogun is good but we lack ease of installation and may be better API15:48
@lisitsynbut if you push us enough and tell us what's better be fixed it would be fixed ;)15:48
joergkurtnever mind, so far things look good15:48
@lisitsynjoergkurt: one thing is that we have pointer based api15:49
@lisitsynbut that's going to be fixed soon15:49
joergkurtright now we are aiming at writing single train-test instances and just wanna call them via algoXZY with parameter setting XYZ15:50
joergkurtthen we fetch the results and estimate profiling (time,space) and scientific performance15:50
@lisitsynI see15:52
-!- exploreshaifali [~shaifali@182.70.196.201] has quit [Quit: Leaving]15:52
joergkurtok, will poke around a bit more and sketch some plans ... we will need a while ... ttyl15:54
joergkurtcheers, moving out now ... have a nice weekend15:55
-!- joergkurt [~joergkurt@178.162.199.93] has quit [Quit: WeeChat 1.1.1]15:58
-!- kshitijk [~kshitijk@115.249.18.25] has joined #shogun15:58
-!- prakhar2511 [~prakhar25@202.78.175.199] has joined #shogun16:08
-!- curiousguy13_ [~curiousgu@120.59.226.121] has quit [Ping timeout: 246 seconds]16:40
-!- exploreshaifali [~shaifali@182.70.196.201] has joined #shogun16:42
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has joined #shogun16:44
-!- iglesiasg [~iglesias@524B8E0B.cm-4-4c.dynamic.ziggo.nl] has quit [Ping timeout: 246 seconds]16:48
-!- curiousguy13_ [~curiousgu@59.177.203.40] has joined #shogun16:53
-!- smallgraph [~textual@scc-wkit-clx-208-236.scc.kit.edu] has joined #shogun16:59
smallgraphHi, I'm trying to build shogun from the develop branch and I get errors on "symbol(s) not found for architecture x86_64" for several methods (first is "shogun::CGaussianARDFITCKernel::CGaussianARDFITCKernel()") when linking libshogun.dylib. Building from master worked fine and I'm working on Yosemite with clang 3.5.0. Did anyone had those problems before?17:00
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has quit [Ping timeout: 246 seconds]17:15
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has joined #shogun17:25
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has quit [Ping timeout: 246 seconds]17:35
-!- joergkurt [~joergkurt@2a02:1810:1c19:ea00:b9e8:4dc2:fd9:1e69] has joined #shogun17:44
joergkurtfeature questions: I see numeric, boolean, and String as features, if we want other nominal features they are represented as?17:45
-!- prakhar2511 [~prakhar25@202.78.175.199] has quit [Ping timeout: 246 seconds]17:51
joergkurtrephrasing question, in CDense the types are listed bool, 8bit, etc17:59
joergkurthttp://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CDenseFeatures.html18:00
joergkurtwhat are the equivalent sparse features being allowed in CSparse18:00
joergkurthttp://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CSparseFeatures.html18:00
joergkurtif I interpret this correctly the template allows largely numeric, and boolean as input18:01
joergkurtthis means all integers are in a numeric relationship, right? Nominal classes are not allowed as input space, just as multiclass labels for prediction?18:02
-!- sonne|osx [~sonne@f049254171.adsl.alicedsl.de] has quit [Quit: sonne|osx]18:12
-!- smallgraph [~textual@scc-wkit-clx-208-236.scc.kit.edu] has quit [Quit: Textual IRC Client: www.textualapp.com]18:15
-!- pickle27 [~pickle27@192-0-134-235.cpe.teksavvy.com] has joined #shogun18:20
-!- joergkurt [~joergkurt@2a02:1810:1c19:ea00:b9e8:4dc2:fd9:1e69] has quit [Quit: WeeChat 1.1.1]18:31
-!- kshitijk [~kshitijk@115.249.18.25] has quit [Ping timeout: 255 seconds]18:44
-!- abinashm [uid66179@gateway/web/irccloud.com/x-hkeeqnqajahrlnip] has joined #shogun19:01
-!- prakhar2511 [~prakhar25@202.78.172.162] has joined #shogun19:27
-!- exploreshaifali [~shaifali@182.70.196.201] has quit [Quit: Leaving]19:33
-!- smallgraph [~textual@p4FDCCD96.dip0.t-ipconnect.de] has joined #shogun19:43
-!- PirosB3 [~pirosb3@host72-217-dynamic.42-79-r.retail.telecomitalia.it] has quit [Quit: PirosB3]20:04
smallgraphHi, can someone help me, please? I'm having problems to compile shogun.20:05
-!- sonne|osx [~sonne@f049254171.adsl.alicedsl.de] has joined #shogun20:14
-!- rajul [~rajul@node-6gf.pool-180-180.dynamic.totbb.net] has quit [Quit: Leaving]20:16
-!- sonne|osx [~sonne@f049254171.adsl.alicedsl.de] has quit [Client Quit]20:19
-!- sonne|osx [~sonne@f049254171.adsl.alicedsl.de] has joined #shogun20:20
-!- kshitijk [~kshitijk@115.249.18.25] has joined #shogun20:34
-!- kshitijk [~kshitijk@115.249.18.25] has quit [Ping timeout: 246 seconds]20:46
-!- exploreshaifali [~shaifali@182.70.164.213] has joined #shogun20:52
-!- HeikoS [~heiko@2.221.116.151] has joined #shogun21:03
-!- mode/#shogun [+o HeikoS] by ChanServ21:04
-!- pickle27 [~pickle27@192-0-134-235.cpe.teksavvy.com] has quit [Remote host closed the connection]21:14
-!- exploreshaifali [~shaifali@182.70.164.213] has quit [Quit: Leaving]21:19
-!- f0z [67157d4e@gateway/web/freenode/ip.103.21.125.78] has joined #shogun21:25
f0zhey21:26
f0zCan anybody guide me through how debugging is done in shogun?21:26
f0zAny...body?21:28
f0z@lisitsyn Do you mind?21:31
@lisitsynf0z: hey21:32
f0zhey21:32
@lisitsynf0z: well, we mostly use gdb and valgrind21:32
@lisitsynwhat do you want to debug?21:32
f0zNo no21:32
f0zI am not talking about that21:32
f0zI am talking about SG_DEBUG flags and the likes21:32
@lisitsynsg_debug is basically a printf21:32
@lisitsynthat shows something is the level is set to debug21:33
-!- abinashm [uid66179@gateway/web/irccloud.com/x-hkeeqnqajahrlnip] has quit [Quit: Connection closed for inactivity]21:33
@lisitsynif you are asking how to see sg_debug messages then the answer is21:33
@lisitsynjust get global io21:33
@lisitsynanyshogunobject->io21:33
@lisitsynand call set_loglevel21:33
@lisitsynanyshogunobject->io->set_loglevel(MSG_DEBUG)21:33
f0zwhat's the default state?21:34
@lisitsyniirc it is msg_info21:34
@lisitsynso only SG_INFO are shown21:35
f0zso this isssue21:35
f0zhttps://github.com/shogun-toolbox/shogun/issues/272321:35
f0zI get how to compute the avg. gradient and all21:36
f0zi mean absolute value21:36
f0zbut do I have to put explicitly that info somewhere using these flags?21:36
@lisitsynhmm not sure I get the question21:37
@lisitsynbut currently SG_INFO is used here21:37
@lisitsynso the gradient stuff should be output as info as well21:37
f0zby SG_DEBUG or SG_INFO? the latter one right?21:38
@lisitsynSG_INFO yeah21:38
@lisitsynit is not debug info I think as watching at gradients is normal for neural networks training21:39
-!- prakhar2511 [~prakhar25@202.78.172.162] has quit [Ping timeout: 246 seconds]21:39
f0zthanks@@21:40
f0zi meant thanks!!!21:40
@lisitsynyou're welcome :)21:40
smallgraphyeah, if I can't see those numbers change I nearly die because of impatience^^21:40
smallgraphlisitsyn, do you also have time for my question?21:41
@lisitsynsmallgraph: sure21:42
smallgraphI'm trying to build from develop branch but get this error:21:42
smallgraphLinking CXX shared library libshogun.dylib21:42
smallgraphUndefined symbols for architecture x86_64:21:42
smallgraph  "shogun::CGaussianARDFITCKernel::get_parameter_gradient(shogun::TParameter const*, int)", referenced from:21:42
smallgraph      vtable for shogun::CGaussianARDFITCKernel in GaussianARDFITCKernel.cpp.o21:43
@lisitsynsmallgraph: oh21:43
@lisitsynshould be easy to fix21:43
@lisitsynsmallgraph: let me update and reproduce21:43
smallgraphwell the method is there, so I don't see why the compiler does not generate it21:44
@lisitsynsmallgraph: could be due to no eigen available21:44
@lisitsyndo you have eigen3?21:45
smallgraphok, no, not yet^^21:46
smallgraphOh, I did install it, but cmake did not find it^^21:47
smallgraphlet me check why21:47
@lisitsynsmallgraph: could be the reason21:47
-!- HeikoS [~heiko@2.221.116.151] has quit [Quit: Leaving.]21:57
-!- f0z [67157d4e@gateway/web/freenode/ip.103.21.125.78] has quit [Ping timeout: 246 seconds]21:59
smallgraphok, now it finds it and compiles but it still dosen't find OpenCL22:05
smallgraphthanks :)22:07
-!- sonne|osx [~sonne@f049254171.adsl.alicedsl.de] has quit [Quit: sonne|osx]22:18
-!- curiousguy13_ [~curiousgu@59.177.203.40] has quit [Ping timeout: 264 seconds]22:40
-!- smallgraph [~textual@p4FDCCD96.dip0.t-ipconnect.de] has quit [Quit: My Mac has gone to sleep. ZZZzzz...]22:49
-!- curiousguy13_ [~curiousgu@120.59.227.130] has joined #shogun22:54
-!- sonne|osx [~sonne@f049254171.adsl.alicedsl.de] has joined #shogun23:12
-!- vortex_ape [~vortex_ap@120.59.74.178] has joined #shogun23:13
-!- sonne|osx [~sonne@f049254171.adsl.alicedsl.de] has quit [Quit: sonne|osx]23:44
-!- sonne|osx [~sonne@f049254171.adsl.alicedsl.de] has joined #shogun23:50
-!- PirosB3 [~pirosb3@host72-217-dynamic.42-79-r.retail.telecomitalia.it] has joined #shogun23:52
-!- xAndreLopesx [ThrashIRC@179.34.60.135] has joined #shogun23:59
-!- xAndreLopesx [ThrashIRC@179.34.60.135] has left #shogun []23:59
--- Log closed Mon Mar 02 00:00:41 2015