| --- Log opened Sat Dec 28 00:00:25 2013 | ||
| -!- FSCV_ [~FSCV@fixed-203-250-138.iusacell.net] has quit [Quit: Leaving] | 00:39 | |
| -!- Jagged [~NULL@vtluug/member/jagged] has left #shogun [] | 02:57 | |
| shogun-buildbot | build #663 of nightly_default is complete: Failure [failed notebooks] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/663 | 04:20 |
|---|---|---|
| -!- zxtx [~zv@c-98-223-196-32.hsd1.in.comcast.net] has quit [Quit: Leaving] | 04:45 | |
| -!- zxtx [~zv@c-98-223-196-32.hsd1.in.comcast.net] has joined #shogun | 04:45 | |
| @wiking | besser82: ping? | 11:23 |
| besser82 | wiking: pong! | 11:24 |
| @wiking | besser82: ok so i've skid the article you've sent | 11:24 |
| besser82 | wiking: and what do you say? | 11:24 |
| besser82 | wiking: What's your oppinion | 11:24 |
| @wiking | besser82: one thing we should remember: there was a plan to move slowly from SGReferencedData to std::shared_ptr | 11:24 |
| besser82 | wiking: shared_ptr like non-copyable? | 11:25 |
| @wiking | besser82: it's like the shared pointer in boost | 11:25 |
| @wiking | besser82: that does automatic reference counting + freeing | 11:26 |
| besser82 | wiking: I see :D | 11:26 |
| besser82 | wiking: Nice idea | 11:26 |
| @wiking | so i dont know how that would work out with this idea | 11:26 |
| @wiking | as there it uses auto_ptr | 11:26 |
| besser82 | wiking: we possibly can change that, I need to look at the specs of shared_ptr && auto_ptr | 11:27 |
| @wiking | besser82: yeah i've had a quick look they not compatible :P | 11:28 |
| besser82 | wiking: damn | 11:28 |
| @wiking | but then again | 11:28 |
| besser82 | wiking: yes | 11:28 |
| @wiking | if we put every class into shared_ptr | 11:28 |
| @wiking | it's almost like auto_ptr<> | 11:28 |
| @wiking | shared_ptr is a template as well that can hold any type of object | 11:28 |
| @wiking | template< class T > class shared_ptr; | 11:28 |
| @wiking | with that we can throw out SGReferencedData | 11:29 |
| besser82 | wiking, so they are interchangeble ;) | 11:29 |
| @wiking | besser82: i would say so | 11:29 |
| besser82 | wiking: mostly | 11:29 |
| @wiking | but yeah if we do this | 11:29 |
| besser82 | wiking: Yes? | 11:29 |
| @wiking | we should do as well the shared_ptr thing as well | 11:29 |
| @wiking | as that would be another standardization of the code itself | 11:30 |
| @wiking | instead of using our own solution | 11:30 |
| besser82 | wiking: +1 :D | 11:30 |
| @wiking | i.e. possibly more interopability | 11:30 |
| @wiking | as it's part of c++11 | 11:30 |
| besser82 | wiking: so when shall we start? | 11:31 |
| @wiking | besser82: well i think we should finish your cmake developments | 11:31 |
| @wiking | since that already introduces some kind of a vertical split of the code | 11:32 |
| besser82 | wiking, yes, somehow. but just from build-pov | 11:32 |
| @wiking | since you've created CMakeLists.txt in almost each subdir | 11:32 |
| @wiking | besser82: yeah true but that would be required anyways for this transition | 11:32 |
| besser82 | wiking, which are in 98% just symlinks to some common CMakeLists :-P | 11:33 |
| besser82 | wiking, so my changes would be somewhat superfluous one this point of transition | 11:34 |
| besser82 | wiking, since on moving to plg.so we basically need to change virtually everything | 11:34 |
| besser82 | wiking, like modular_interfaces and all | 11:34 |
| @wiking | besser82: well yeah modular interfaces would need a total refactoring | 11:35 |
| @wiking | besser82: so yeah basically we would need a new feature branch | 11:36 |
| @wiking | and start experimenting there | 11:36 |
| @wiking | first try to do the splitting only on the c++ level | 11:36 |
| @wiking | and then try to fix the modular interfaces | 11:36 |
| besser82 | wiking: would be a beginning ;) | 11:44 |
| besser82 | wiking: but if we are moving to plugin-style, I'd strongly recommend keeping every plugin inside a seperate git-repo | 11:44 |
| besser82 | wiking: and use some common-toolchain to checkout and build | 11:45 |
| besser82 | wiking: which would make freezes and stuff quite easier | 11:45 |
| besser82 | wiking: have a look at those suse YaST repo-style ;) | 11:46 |
| besser82 | wiking, https://github.com/yast | 11:46 |
| besser82 | wiking, they are using a git-repo for one thing each | 11:47 |
| besser82 | wiking, and have some common {build,dev}tools | 11:47 |
| besser82 | wiking, https://github.com/yast/yast-meta | 11:48 |
| besser82 | wiking, https://github.com/yast/yast-devtools | 11:48 |
| besser82 | wiking, with such a structure we could make things quite easier for user and developers a well | 11:49 |
| besser82 | wiking, we could create a tool like shogun-manager or sth. which checks out the needed repos and builds them | 11:51 |
| @wiking | besser82: mmm i wouldn't do that | 11:51 |
| besser82 | wiking, why? can you explain your pov? | 11:51 |
| @wiking | besser82: i feel that that's just too much of an overhead | 11:52 |
| besser82 | wiking, in the first place YES | 11:52 |
| besser82 | wiking, but when thinking about further progress NO | 11:52 |
| @wiking | besser82: there's just way tooo many plugins | 11:53 |
| besser82 | wiking, not everyone needs / uses the whole set of features | 11:53 |
| @wiking | besser82: if u think about Features | 11:53 |
| @wiking | or Kernel | 11:53 |
| @wiking | each of the kernel should be a separate plugin imho | 11:53 |
| @wiking | but for each of the kernel to make a separate repo | 11:54 |
| @wiking | that's just waaaay too much | 11:54 |
| besser82 | wiking, somehow | 11:54 |
| @wiking | there are for example 75 different kernel implementations | 11:54 |
| @wiking | and that's just really kernel | 11:54 |
| besser82 | wiking, i see | 11:55 |
| @wiking | think about all the others | 11:55 |
| @wiking | features, machine, classifiers | 11:55 |
| @wiking | etc etc | 11:55 |
| besser82 | wiking, so we would endup with ~400 repos or so | 11:55 |
| @wiking | or more like 1000 | 11:55 |
| besser82 | wiking, but with some `smart` managment that wouldn't be so much pain | 11:56 |
| besser82 | wiking, we would just need some metadata about what's needed for which purpose ;) | 11:57 |
| besser82 | wiking, and the manager-app could then resolve | 11:58 |
| besser82 | wiking, on the other hand we could manage the repos a bit task-based | 11:59 |
| besser82 | wiking, like having a repo for all kernel, classifier, etc .... | 12:00 |
| besser82 | wiking, a common repo | 12:00 |
| besser82 | wiking, so we would end with lemme guess ~20 repos | 12:01 |
| besser82 | wiking, have i been too aggressive on marketing my idea? :( | 12:11 |
| @sonney2k | besser82, hey good morning | 13:09 |
| @sonney2k | besser82, any news on cmake? | 13:09 |
| @sonney2k | or have you been thinking all day night about the split thing | 13:09 |
| besser82 | sonney2k: actually that plugin-split-action(tm) occupies all my mind ;) | 14:02 |
| besser82 | sonney2k: my currrent thoughts are: Shall I really spend more time in sth. we need to re-assemble, when running off with that split-stuff? | 14:03 |
| besser82 | sonney2k: And it is not that plugin-things, only, which moves me brains | 14:16 |
| besser82 | sonney2k: the is sooooo much stuff about shogun, it's future, our goals, the possibilities, possible movements, getting people involved, making it THE one | 14:18 |
| besser82 | sonney2k: I'm thinking about stuff centred making SHOGUN the B.I.G. thing | 14:20 |
| besser82 | sonney2k: making it easier for people to contrib | 14:21 |
| besser82 | sonney2k: creating community | 14:21 |
| besser82 | sonney2k: all such | 14:21 |
| besser82 | sonney2k: my mind currently shoots ideas like bulletts out of an uzi | 14:24 |
| besser82 | sonney2k: creating sth. like pypi or rubygems, but around shogun | 14:25 |
| -!- zxtx [~zv@c-98-223-196-32.hsd1.in.comcast.net] has quit [Remote host closed the connection] | 16:30 | |
| -!- zxtx [~zv@c-98-223-196-32.hsd1.in.comcast.net] has joined #shogun | 16:32 | |
| besser82 | lisitsyn: ^^^ What do you think? | 17:07 |
| lisitsyn | besser82: yes that's how I see it | 17:07 |
| besser82 | lisitsyn: all the same I do? | 17:07 |
| lisitsyn | besser82: actually I see it as something like cross-language platform | 17:07 |
| lisitsyn | let me explain | 17:08 |
| besser82 | lisitsyn: sure | 17:08 |
| lisitsyn | from linguistics position :D | 17:08 |
| lisitsyn | shogun is like a commander ;) | 17:08 |
| lisitsyn | I see it would be wonderful to have cross-lang inteface for cross-lang stuff | 17:08 |
| besser82 | lisitsyn: that can be implemented by a plugin :D | 17:09 |
| lisitsyn | yes | 17:09 |
| besser82 | lisitsyn: python is easy as dell ;) | 17:09 |
| lisitsyn | I mean | 17:09 |
| lisitsyn | we have scikits, weka | 17:09 |
| lisitsyn | we can rule them all | 17:09 |
| lisitsyn | in a cross-platform manner | 17:09 |
| lisitsyn | err cross-language | 17:09 |
| besser82 | lisitsyn: but ruby or other lang might be some more tricky | 17:09 |
| lisitsyn | why? | 17:10 |
| lisitsyn | lets have some ruby standart interface for classifier et | 17:10 |
| lisitsyn | when once you get some code you do some glueing | 17:10 |
| lisitsyn | to support that interface | 17:10 |
| lisitsyn | and our runner runs and gets results back | 17:10 |
| besser82 | lisitsyn: ok, ok, but python can be intergrated like full interpreter with really few code ;) | 17:11 |
| lisitsyn | yeah | 17:11 |
| lisitsyn | same for matlab etc | 17:11 |
| besser82 | lisitsyn: ruby and others are more tough | 17:11 |
| besser82 | lisitsyn: from that pov | 17:11 |
| besser82 | lisitsyn: are you sure? | 17:12 |
| lisitsyn | besser82: sure about what? :) | 17:12 |
| besser82 | lisitsyn: embedding interpreters being easy as baking a pie? | 17:13 |
| lisitsyn | why to embed them? | 17:13 |
| lisitsyn | it is enough to run it as a process and get results back | 17:13 |
| besser82 | lisitsyn: like with pipes ? | 17:13 |
| lisitsyn | yes | 17:13 |
| besser82 | lisitsyn: would be nice to get that a bit more elaborated :-P | 17:14 |
| lisitsyn | besser82: okay I download some code by famous-researcher-of-machine-learning | 17:14 |
| besser82 | lisitsyn: which code from sonney2k ?!? | 17:14 |
| lisitsyn | it has function do_stuff(...) | 17:15 |
| besser82 | lisitsyn: go on | 17:15 |
| lisitsyn | we set up COMMON interface for matlab classifiers we support | 17:15 |
| lisitsyn | lets say it is train(X,Y) | 17:15 |
| lisitsyn | you write | 17:15 |
| lisitsyn | train(X,Y) | 17:15 |
| lisitsyn | do_stuff(....) | 17:15 |
| lisitsyn | with proper dispatching of parameters | 17:15 |
| lisitsyn | I mean you extract all the data you need whatever original algorithm requires | 17:16 |
| besser82 | lisitsyn: I see | 17:24 |
| besser82 | lisitsyn: My thought on this particular topic was more like having the possibility to be able to develop plugins in other lang, which directly interface with shogun | 17:26 |
| lisitsyn | besser82: yeah if possible | 17:26 |
| lisitsyn | but what about matlab shit? | 17:26 |
| besser82 | lisitsyn, sure for python easy as spending 500$ | 17:26 |
| besser82 | lisitsyn, :-P | 17:26 |
| besser82 | lisitsyn, just a bit of code and some `Cython`-magic and here we go | 17:27 |
| besser82 | lisitsyn, so for python we just need a bit of native python-code and some Cython cppdefs ---> convert to *.cpp ---> compile a plugin ---> go | 17:29 |
| besser82 | lisitsyn, that wrapper-plugin is then able to "embed" the whole python-interpreter and load-in shogun-plugins written in pure-python | 17:30 |
| besser82 | lisitsyn, ruby, java, ... would be some more work I guess | 17:31 |
| besser82 | lisitsyn, matlab? I dunno, I don't own, I don't use | 17:31 |
| besser82 | lisitsyn, but should be perfectly the same as octave | 17:31 |
| lisitsyn | yes would be better | 17:32 |
| besser82 | lisitsyn, better what? | 17:32 |
| lisitsyn | to compile it as plugins | 17:32 |
| besser82 | lisitsyn, lang-embed-plugins you mean? | 17:33 |
| lisitsyn | yes | 17:33 |
| lisitsyn | than piping them | 17:33 |
| * besser82 throws an instance of confused-exception (What?) | 17:33 | |
| lisitsyn | :D | 17:33 |
| lisitsyn | besser82: making them plugins (your suggestion) would be better than piping (my suggestion) | 17:34 |
| besser82 | lisitsyn, now i got it :D | 17:34 |
| besser82 | lisitsyn, yes piping is somewhat fragile, I'm afraid | 17:34 |
| besser82 | lisitsyn, and i dunno whether there is protobuf support for matlab :-P | 17:35 |
| besser82 | lisitsyn, which would be another way | 17:35 |
| besser82 | lisitsyn, but having plugins "embedding" the interp would be the best i guess | 17:35 |
| besser82 | lisitsyn, how do like that idea of having sth. like pypi or gems for shogun? | 17:54 |
| lisitsyn | besser82: yeah that sounds good | 17:54 |
| lisitsyn | besser82: though I'd make it as minimal as possible | 17:55 |
| besser82 | lisitsyn, which minimal? | 17:55 |
| lisitsyn | besser82: just some basic thing - not like whole new packaging system | 17:55 |
| besser82 | lisitsyn, no nothing just stealing metadata stuff && needs from gems or sth ;) | 17:56 |
| besser82 | lisitsyn, like having a wrapper which is able to resolve deps between plugins, fetches the needed one (e.g. git checkout), builds && registers them | 18:49 |
| besser82 | lisitsyn, with the shogun-interface-lib | 18:49 |
| -!- besser82 is now known as besser82|mobile | 19:01 | |
| -!- besser82|mobile is now known as group | 19:01 | |
| -!- group is now known as besser82|mobile | 19:02 | |
| -!- besser82|mobile is now known as besser82 | 19:02 | |
| lisitsyn | besser82: actually such repo is what I want to develop :D | 19:48 |
| besser82 | lisitsyn: Nice! then go ahead with that :D | 19:49 |
| lisitsyn | I think I'd like to learn a bit how to design this type of software | 19:49 |
| besser82 | lisitsyn, pkg-manager? | 19:49 |
| lisitsyn | yeah somewhat like that | 19:49 |
| besser82 | lisitsyn, if you need some help or design ideas you can ask Ale? Kozumpl?k <ales@redhat.com> | 19:52 |
| lisitsyn | ha you have contacts right | 19:52 |
| lisitsyn | :D | 19:52 |
| besser82 | lisitsyn, yes :D | 19:52 |
| besser82 | lisitsyn, another one might be Klaus K?mpf <kkaempf@suse.de> | 19:53 |
| lisitsyn | besser82: I am trying to learn piano, any contacts? :D | 19:53 |
| besser82 | lisitsyn, mmmhhhhhh.... | 19:54 |
| lisitsyn | besser82: it is Heiko! ;) | 19:54 |
| besser82 | lisitsyn, some local mi?usic-student, perhaps :-P | 19:54 |
| lisitsyn | he is kind of musician | 19:54 |
| besser82 | lisitsyn, yes, but not piano, i'm afraid | 19:54 |
| lisitsyn | besser82: every musician can play piano at least a bit ;) | 19:55 |
| lisitsyn | besser82: Heiko plays guitar mainly afaik | 19:55 |
| besser82 | lisitsyn, yes, and yes a bit for sure. But enough to teach? | 19:55 |
| lisitsyn | haha no idea | 19:55 |
| besser82 | lisitsyn, and mind the distance :-P | 19:55 |
| lisitsyn | besser82: I am experimenting of self-taught learning | 19:55 |
| besser82 | lisitsyn, me too ;) | 19:56 |
| besser82 | lisitsyn, all me skillz are mostly from self-studies | 19:56 |
| lisitsyn | besser82: yeah so why not to try to learn music this way too | 19:56 |
| lisitsyn | :D | 19:56 |
| besser82 | lisitsyn, my greatest problem with that is: I'm left-handed | 19:57 |
| lisitsyn | besser82: me too | 19:57 |
| lisitsyn | besser82: what is the problem? | 19:57 |
| besser82 | lisitsyn, most instruments are made for right-handed | 19:58 |
| lisitsyn | besser82: well.. kind of | 19:58 |
| lisitsyn | besser82: but what is the instrument which is that inconvenient for left-handed guys? | 19:59 |
| besser82 | lisitsyn, guitar e.g. | 19:59 |
| besser82 | lisitsyn, basically all string instr | 19:59 |
| lisitsyn | I feel it makes no difference when you don't know how to play ;) | 19:59 |
| besser82 | lisitsyn, kk but if you try to make it sound nice?!? | 20:00 |
| lisitsyn | I mean it is harder to write with right hand because you have been doing it left hand all the way | 20:00 |
| lisitsyn | but when you learn instrument it should be ok ;) | 20:00 |
| lisitsyn | besser82: there is a long way to make it sound nice anyway haha | 20:00 |
| besser82 | lisitsyn, actually it hasn't to do much with the way you learned.... | 20:01 |
| lisitsyn | besser82: I love upright bass, want to get it once and start learning | 20:01 |
| lisitsyn | it is symmetric so no problem with left-handness | 20:02 |
| lisitsyn | besser82: what do you mean? | 20:02 |
| besser82 | lisitsyn, it's the way the "wiring" in ya brain has grown ;) | 20:02 |
| lisitsyn | besser82: I am not sure, I believe you can change that wiring enough | 20:03 |
| besser82 | lisitsyn, raise a lefthanded child writing with the right hand and it will surely get depressions and other psychologic problem | 20:03 |
| besser82 | lisitsyn, known fact | 20:04 |
| lisitsyn | oh yeah people used to do that before | 20:04 |
| besser82 | lisitsyn, some 70 y back | 20:04 |
| lisitsyn | besser82: but that's true for any force applied to child | 20:04 |
| besser82 | lisitsyn, not with force ;) | 20:05 |
| besser82 | lisitsyn, just gentle teaching is enough | 20:05 |
| besser82 | lisitsyn, one simply cannot reorder the two halves of the brain | 20:06 |
| lisitsyn | besser82: I am not sure ambidexterity is not possible to train | 20:06 |
| besser82 | lisitsyn, nope it's born that way | 20:07 |
| besser82 | lisitsyn, there are actually a lot of things which is decided @~3rd month of pregnancy | 20:07 |
| besser82 | lisitsyn, sexual orientation, left / right handed, ..... | 20:08 |
| lisitsyn | besser82: oh no sexual orientation could be decided much later | 20:08 |
| besser82 | lisitsyn, possibly yes, but some scientist found out the basics are made then | 20:09 |
| besser82 | lisitsyn, being gay or a dyke isn't just a decision | 20:10 |
| lisitsyn | besser82: haha yes most of the times yes it is not a decision | 20:10 |
| lisitsyn | but I would believe it is quite related to environment and stresses you have in childhood | 20:10 |
| besser82 | lisitsyn, that can be possibly some "weak" circumstances, but I know a lot of gays which didn't have any suspicious things in their childhood | 20:12 |
| lisitsyn | I mean just like that anecdotical thing - if girl had difficult relationships with her father you can expect some kind of sexual behaviour | 20:12 |
| besser82 | lisitsyn, they are just that way | 20:12 |
| lisitsyn | besser82: I don't speak about great suspicious things - some small things could change the game | 20:13 |
| besser82 | lisitsyn, which ones?!? | 20:13 |
| lisitsyn | besser82: if I knew I said but that's how I see it ;) | 20:13 |
| lisitsyn | brain and all these mental things are kinda complex you know ;) | 20:14 |
| besser82 | lisitsyn, yes | 20:14 |
| besser82 | lisitsyn, god, why did you not make brain && mental stuff like boolean???? | 20:14 |
| besser82 | :-P | 20:15 |
| lisitsyn | I am quite biased to the thought that any human is like tabula rasa once born | 20:15 |
| besser82 | lisitsyn, nope | 20:15 |
| lisitsyn | well not physically | 20:15 |
| lisitsyn | and not for the things that are related to genes | 20:16 |
| besser82 | lisitsyn, even not for psychologic stuff | 20:16 |
| lisitsyn | no I can't accept that ;) | 20:16 |
| besser82 | lisitsyn, baby catches more in mama's belly than we believe | 20:16 |
| lisitsyn | well we all know baby can't speak and understand | 20:18 |
| lisitsyn | speak and understand words not emotions | 20:18 |
| besser82 | lisitsyn, and emotions can be enough | 20:18 |
| lisitsyn | so I am not sure it matter what is spoken around | 20:18 |
| besser82 | lisitsyn, not the word, but tone ;) | 20:19 |
| lisitsyn | yes I can agree human like any other animal can catch the emotion beneath | 20:19 |
| besser82 | lisitsyn, jupp | 20:19 |
| lisitsyn | besser82: what's the jupp about? ;) | 20:20 |
| besser82 | lisitsyn, s=eval(jupp) ; print $s ---> Yes | 20:20 |
| lisitsyn | ahh | 20:20 |
| besser82 | just the german way to ack / yes ;) | 20:21 |
| lisitsyn | besser82: one more thing you could not agree with me is that thing about humans being animals ;) | 20:21 |
| lisitsyn | I suspect you won't accept that | 20:21 |
| besser82 | lisitsyn, as long as we did not find the missing link... | 20:22 |
| besser82 | lisitsyn, I can whether denie nor agree | 20:22 |
| besser82 | lisitsyn, but everything pretty leads to the asumption the common evolution theory is correct ;) | 20:23 |
| besser82 | lisitsyn, so yes, in fact we are animals | 20:23 |
| besser82 | lisitsyn, that's one thing i totally agree with | 20:23 |
| besser82 | lisitsyn, we have enough basic-instincts in us to prove that | 20:24 |
| besser82 | lisitsyn, you know why one cannot stop eating crisps once started? | 20:24 |
| lisitsyn | besser82: haha | 20:29 |
| besser82 | lisitsyn, it has to do with our basic-instrincts | 20:34 |
| lisitsyn | be back soon | 20:35 |
| -!- lisitsyn [~lisitsyn@80.252.20.67] has quit [Quit: Leaving.] | 20:35 | |
| -!- lisitsyn [~lisitsyn@80.252.20.67] has joined #shogun | 20:38 | |
| lisitsyn | besser82: having troubles with my notebook ac adapter haha | 20:38 |
| besser82 | lisitsyn, the sound of biting down on some crisps triggers sth in the Medulla oblongata | 20:38 |
| besser82 | lisitsyn, hehe | 20:38 |
| lisitsyn | basically it is mechanically broken | 20:38 |
| besser82 | lisitsyn, get a new one ;) | 20:39 |
| lisitsyn | besser82: yeah that's what I am thinking of | 20:39 |
| lisitsyn | new notebook or new ac adapter :D | 20:39 |
| besser82 | lisitsyn, wanna buy a real mobile workstation? | 20:39 |
| lisitsyn | besser82: oh no I am fan of small devices :D | 20:39 |
| besser82 | lisitsyn, kk | 20:40 |
| lisitsyn | you can suggest me i19 48-core CPU | 20:40 |
| lisitsyn | ;) | 20:40 |
| besser82 | lisitsyn, :-P | 20:40 |
| lisitsyn | besser82: http://images.amazon.com/images/G/01/electronics/samsung/samsung-12q2bts-series5-13-main-lg.jpg | 20:40 |
| lisitsyn | my current one | 20:40 |
| lisitsyn | piece of crap | 20:40 |
| besser82 | lisitsyn, no but i could offer i7-2680QM 4-core / HT | 20:41 |
| besser82 | lisitsyn, those are the shitty ones from samsung | 20:41 |
| besser82 | lisitsyn, my old "Q45" is still alive ;) | 20:41 |
| lisitsyn | besser82: it is made mostly of plastic and my way of using it makes it broken day by day | 20:41 |
| lisitsyn | and it has already dead battery | 20:42 |
| lisitsyn | 1.5yr old | 20:42 |
| besser82 | lisitsyn, the q45 was made of magnesium ;) | 20:42 |
| besser82 | lisitsyn, 7 yo | 20:42 |
| lisitsyn | besser82: that's why I like macbook air | 20:42 |
| besser82 | lisitsyn, don't buy things with bitten fruit on it ;) | 20:43 |
| besser82 | http://business.chip.de/bii/149460082_270cd35a4e.jpg <-- Q45 | 20:43 |
| lisitsyn | besser82: you've got to convince sonney2k macs are bad too ;) | 20:43 |
| besser82 | lisitsyn, on it | 20:44 |
| besser82 | sonney2k, ^^^ Stuff with bitten fruit on it is bad, you know ;) | 20:44 |
| besser82 | lisitsyn, ^^^ :-P | 20:45 |
| besser82 | lisitsyn, those are good ;) ---> http://shop.lenovo.com/de/de/laptops/thinkpad/x-series/x240/ | 20:47 |
| lisitsyn | besser82: yeah | 20:48 |
| besser82 | lisitsyn, but quite expensive for so less power | 20:50 |
| lisitsyn | besser82: I do not need any power ;) | 20:50 |
| besser82 | lisitsyn, I can sell you my q45 for ~200EUR | 20:50 |
| lisitsyn | it would be enough to have ssd and not that bad cpu to handle | 20:50 |
| lisitsyn | hahaha | 20:50 |
| lisitsyn | why would I buy it | 20:51 |
| lisitsyn | ;) | 20:51 |
| besser82 | lisitsyn, has ssd and Core2Duo 2,5 GHz | 20:51 |
| besser82 | lisitsyn, 4GB RAM | 20:51 |
| besser82 | lisitsyn, 12" | 20:51 |
| besser82 | lisitsyn, DVD-RW | 20:51 |
| lisitsyn | besser82: 8 gb, ssd and 1.7 i5 here | 20:51 |
| lisitsyn | ;) | 20:51 |
| lisitsyn | sounds like downgrade | 20:51 |
| besser82 | lisitsyn, but it doesn't break ;) | 20:51 |
| besser82 | lisitsyn, like the one you currently have | 20:52 |
| lisitsyn | besser82: well it had a few flights | 20:52 |
| besser82 | lisitsyn, mine too | 20:52 |
| besser82 | lisitsyn, nothing broken | 20:52 |
| lisitsyn | like hard landing to floor :D | 20:52 |
| besser82 | lisitsyn, mine too | 20:52 |
| besser82 | lisitsyn, nothing broken | 20:52 |
| lisitsyn | heh yeah metal shouldn't suffer that much | 20:53 |
| besser82 | lisitsyn, only screen needed replacement once | 20:53 |
| lisitsyn | yeah I replaced samsung screen once too | 20:53 |
| lisitsyn | not on this machine though | 20:53 |
| lisitsyn | russian winter (tm) | 20:54 |
| lisitsyn | I got back to home and turn it on | 20:54 |
| lisitsyn | I guess it was quite cold yet and a few seconds after it cracked :D | 20:54 |
| besser82 | lisitsyn, hehe | 20:54 |
| besser82 | lisitsyn, was an lcd, right? | 20:55 |
| lisitsyn | yes | 20:55 |
| besser82 | lisitsyn, those are known to crack when getting too cold and powered | 20:55 |
| besser82 | lisitsyn, the other offer i can make is i7-2680QM 32GB-RAM, SSD, 2 HDD, BlueRay-burner, nvidia 580GTX | 20:56 |
| besser82 | lisitsyn, 1,4 Gbit/s WLAN | 20:56 |
| lisitsyn | haha whaat a monstah | 20:56 |
| besser82 | lisitsyn, 17" ~6kgs | 20:56 |
| lisitsyn | hahahah | 20:56 |
| besser82 | lisitsyn, real workstation ;) | 20:56 |
| lisitsyn | well you don't have to go to gym | 20:56 |
| lisitsyn | if you have such a thing | 20:57 |
| besser82 | lisitsyn, no, but how do know i don't care about my overweight?!? | 20:57 |
| besser82 | :-P | 20:57 |
| lisitsyn | besser82: me too though I prefer to not carry such things ;) | 20:57 |
| besser82 | lisitsyn, let's say so: That Workstation suits me "size" | 20:58 |
| lisitsyn | haha I get what you mean | 20:58 |
| besser82 | lisitsyn, 140 kgs pure erotic :-P | 20:58 |
| lisitsyn | haha yeah then I totally get what I mean | 20:59 |
| lisitsyn | a bit less than 100 here :D | 20:59 |
| besser82 | lisitsyn, you must "train" harder then to get closer to the 140 :-P | 20:59 |
| lisitsyn | besser82: with my kind of genetic properties it should be easy enough :D | 21:00 |
| lisitsyn | besser82: I guess you haven't trained hard to achieve that too ;) | 21:01 |
| besser82 | lisitsyn, not really | 21:01 |
| lisitsyn | it is a gift haha | 21:01 |
| lisitsyn | of nature | 21:01 |
| besser82 | lisitsyn, yes :D | 21:01 |
| besser82 | lisitsyn, but some girls don't understand that :( | 21:02 |
| lisitsyn | hehe yeah | 21:02 |
| -!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 21:02 | |
| shogun-notifier- | shogun: Saurabh7 :develop * db77e9f / / (3 files): https://github.com/shogun-toolbox/shogun/commit/db77e9f7de9d7323c882d16406ca53dc7d40a088 | 21:02 |
| shogun-notifier- | shogun: add LOOxval | 21:02 |
| shogun-notifier- | shogun: Saurabh7 :develop * 1010cba / src/shogun/evaluation/ (2 files): https://github.com/shogun-toolbox/shogun/commit/1010cbabf8325379b035d3ebfc100008f8faf240 | 21:02 |
| shogun-notifier- | shogun: add stdxval inheritance to loo | 21:02 |
| shogun-notifier- | shogun: Saurabh7 :develop * abd777f / examples/undocumented/libshogun/splitting_LOO_crossvalidation.cpp: https://github.com/shogun-toolbox/shogun/commit/abd777fe72889a606f6d624921a160e1681179af | 21:02 |
| shogun-notifier- | shogun: update looxval example | 21:02 |
| shogun-notifier- | shogun: Heiko Strathmann :develop * 79a2b45 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/79a2b4574082d096a3e5628ac1378daf0c452629 | 21:02 |
| shogun-notifier- | shogun: Merge pull request #1799 from Saurabh7/loocrossval | 21:02 |
| shogun-notifier- | shogun: | 21:02 |
| shogun-notifier- | shogun: add LOOxval | 21:02 |
| besser82 | lisitsyn, ^^^ merge alert!!!! | 21:03 |
| lisitsyn | besser82: alarm | 21:03 |
| besser82 | lisitsyn, AUSWEIS BITTE! | 21:03 |
| lisitsyn | besser82: I remember there was a game with germans yell | 21:03 |
| lisitsyn | ALARM | 21:03 |
| lisitsyn | I think it is commandos | 21:04 |
| besser82 | lisitsyn, yes some 15 years back or so ;) | 21:04 |
| lisitsyn | cool game | 21:04 |
| lisitsyn | I should not take a look at it to not ruin my memories | 21:04 |
| besser82 | lisitsyn, yes | 21:04 |
| besser82 | lisitsyn, you know Wolfenstein 3D? | 21:05 |
| lisitsyn | yeah sure | 21:05 |
| besser82 | lisitsyn, the Germans there are yelling stuff too | 21:05 |
| besser82 | lisitsyn, MEIN LEBEN! | 21:05 |
| lisitsyn | and in 'return to castle wolfenstein' too ;) | 21:05 |
| besser82 | lisitsyn, :D | 21:05 |
| besser82 | lisitsyn, both games are forbidden in Germany | 21:06 |
| lisitsyn | really? | 21:06 |
| lisitsyn | hahaha | 21:06 |
| lisitsyn | lol | 21:06 |
| besser82 | lisitsyn, but everbody has a copy :-P | 21:06 |
| lisitsyn | why? | 21:06 |
| besser82 | lisitsyn, because of some graphical 3rd Reich symbols | 21:06 |
| besser82 | lisitsyn, like those svatiskas and pictures of Hitler | 21:06 |
| lisitsyn | but they look cool ;) | 21:06 |
| lisitsyn | it has its own aesthetics | 21:07 |
| besser82 | lisitsyn, in Indiana Jones the last Crusade there was a special version for germany too | 21:07 |
| besser82 | lisitsyn, somehow yes, but those symbols are forbidden in germany | 21:08 |
| lisitsyn | I mean it wasn't a cool thing but I don't think restricting every mention of anything | 21:08 |
| lisitsyn | is a good thing to do | 21:08 |
| besser82 | lisitsyn, showing stuff is ~5y prison | 21:08 |
| lisitsyn | like these holocaust denial laws - people should be free to be idiots and believe to anything they want | 21:09 |
| besser82 | lisitsyn, not in germany | 21:09 |
| lisitsyn | besser82: why? | 21:09 |
| lisitsyn | besser82: I know it is that but I mean it would be better to not restrict anything ;) | 21:09 |
| besser82 | lisitsyn, those laws were made by the allied | 21:10 |
| besser82 | lisitsyn, and are still valid | 21:10 |
| besser82 | lisitsyn, like some other crappy law too | 21:10 |
| lisitsyn | besser82: nazi symbols are forbidden here too by the way | 21:11 |
| besser82 | lisitsyn, laws surving stalin, I guess? | 21:11 |
| lisitsyn | besser82: hmm I don't know when was it introduced | 21:11 |
| lisitsyn | besser82: actually stalin's laws were beautiful piece of humanity | 21:12 |
| lisitsyn | but not the real life :D | 21:12 |
| besser82 | lisitsyn, hehe | 21:12 |
| besser82 | lisitsyn, some minor contradiction | 21:12 |
| besser82 | lisitsyn, like in all totalitary systems ;) | 21:13 |
| lisitsyn | besser82: stalin is a hero for like a half of people here ;) | 21:13 |
| besser82 | lisitsyn, shouldn't that be lenin? | 21:13 |
| lisitsyn | besser82: no lenin is like forgotten | 21:13 |
| besser82 | lisitsyn, or Gorbatschow | 21:13 |
| besser82 | lisitsyn, like for people in Ex-GDR | 21:14 |
| lisitsyn | gorbi is hated | 21:14 |
| lisitsyn | besser82: most of the people just hate gorbachev | 21:14 |
| besser82 | lisitsyn, what did he do? | 21:14 |
| lisitsyn | in their minds he destroyed the great country | 21:15 |
| lisitsyn | ;) | 21:15 |
| lisitsyn | besser82: in fact it was on the way to be destroyed anyway | 21:15 |
| besser82 | lisitsyn, yes | 21:15 |
| besser82 | lisitsyn, CCCP was on it's last breathes those days | 21:15 |
| lisitsyn | besser82: same thing for yeltsin | 21:16 |
| lisitsyn | which is totally hated too | 21:16 |
| besser82 | lisitsyn, Boris I can understand ;) | 21:16 |
| lisitsyn | well at least he was a funny guy | 21:16 |
| besser82 | lisitsyn, funny yes, but ... | 21:16 |
| lisitsyn | most of the time being drunk | 21:16 |
| lisitsyn | :D | 21:16 |
| besser82 | lisitsyn, There is some discount vodka named after him in Germany ;) | 21:17 |
| besser82 | lisitsyn, Jeltzin - pure russian vodka :D | 21:17 |
| lisitsyn | besser82: he wasn't the most clever guy but he let quite okish people to rule | 21:17 |
| lisitsyn | so 90s were not that terrible | 21:17 |
| besser82 | lisitsyn, and then came Putin :-P | 21:18 |
| lisitsyn | besser82: yes and he made a few good things and then become a monster | 21:18 |
| besser82 | lisitsyn, yes | 21:18 |
| lisitsyn | basically starting from 2004-2006 he has been doing crazy things | 21:18 |
| besser82 | lisitsyn, then went to make some serious $$$$$$$$$$$$$$$ and left over to Medwedwew | 21:19 |
| lisitsyn | besser82: extra oil prices made people quite rich in 2001-2004 | 21:19 |
| lisitsyn | most of the people bought cars these days etc | 21:19 |
| lisitsyn | it was really good years economics-wise | 21:19 |
| besser82 | lisitsyn, but now most people are poor again :( | 21:19 |
| lisitsyn | besser82: there are crazy amounts of money still | 21:20 |
| lisitsyn | actually just go get it | 21:20 |
| besser82 | lisitsyn, lets-a-go !!!!! | 21:20 |
| lisitsyn | :D | 21:20 |
| besser82 | lisitsyn, tell me where to grab :-P | 21:20 |
| lisitsyn | besser82: no really, goverment still has a lot of money | 21:20 |
| besser82 | lisitsyn, and some very few others | 21:21 |
| lisitsyn | besser82: we do a project for russian railways | 21:21 |
| besser82 | lisitsyn, which one? | 21:21 |
| lisitsyn | well some scheduling | 21:21 |
| lisitsyn | and we get quite $$$ | 21:21 |
| lisitsyn | for nothing serious | 21:21 |
| besser82 | lisitsyn, then go out and grab some more ;) | 21:22 |
| lisitsyn | they have millions of $ flying around | 21:22 |
| lisitsyn | besser82: they just care to grab some % of that and redirect the work | 21:22 |
| lisitsyn | and still in the end (2-3 step of redirection) | 21:22 |
| lisitsyn | it is quite good money | 21:22 |
| lisitsyn | if I wasn't on salary I'd get rich with such things :D | 21:23 |
| besser82 | lisitsyn, then put your foot in the door and save me a place in that chain ;) | 21:23 |
| besser82 | lisitsyn, and then we redirect the work to some indians | 21:23 |
| lisitsyn | besser82: that's quite easy - you just talk to right people | 21:24 |
| lisitsyn | :D | 21:24 |
| besser82 | lisitsyn, like 75% us / 25% them :D | 21:24 |
| lisitsyn | besser82: I can say they are mostly located in moscow :D | 21:24 |
| besser82 | lisitsyn, then start talking ;) | 21:24 |
| lisitsyn | no other places to make deals in russia | 21:24 |
| besser82 | lisitsyn, don't you live in Moscow | 21:24 |
| lisitsyn | besser82: no, samara | 21:24 |
| besser82 | lisitsyn, is it a great distance? | 21:25 |
| lisitsyn | besser82: 1K km | 21:25 |
| lisitsyn | 1.15 by plane ;) | 21:25 |
| besser82 | lisitsyn, ok | 21:25 |
| besser82 | lisitsyn, so need to go | 21:25 |
| besser82 | lisitsyn, so I need to go there ;) | 21:25 |
| lisitsyn | besser82: really all the money is in moscwo | 21:26 |
| lisitsyn | quite easy thing :D | 21:26 |
| besser82 | lisitsyn, but i don't speak tha ??????? :( | 21:26 |
| lisitsyn | I suspect you have less centric distribution of money flow | 21:26 |
| besser82 | lisitsyn, actually yes | 21:26 |
| besser82 | lisitsyn, in Germany it is more shattered | 21:27 |
| lisitsyn | as far as I know munich is kind of more rich city than berlin | 21:27 |
| besser82 | lisitsyn, yes | 21:27 |
| besser82 | lisitsyn, and more expensive by far | 21:27 |
| lisitsyn | that's impossible for russia - everything is in moscow | 21:27 |
| lisitsyn | :D | 21:27 |
| besser82 | lisitsyn, 2-room, ghetto-style flat starts @ 1K5EUR without heating | 21:28 |
| lisitsyn | that's crazy | 21:28 |
| besser82 | lisitsyn, there are more people than flats ;) | 21:28 |
| lisitsyn | besser82: however quite similar to moscow prices | 21:28 |
| besser82 | lisitsyn, let's go moscow and grab some $$$, then | 21:29 |
| lisitsyn | besser82: I pay ~625$ here | 21:29 |
| besser82 | lisitsyn, for what? | 21:29 |
| lisitsyn | 1 room, ~70 m2 | 21:29 |
| besser82 | lisitsyn, that's normal price like in Germany | 21:29 |
| lisitsyn | besser82: quite above average here | 21:30 |
| lisitsyn | besser82: though I decided to take it due to larger volume and 18th floor ;) | 21:30 |
| besser82 | lisitsyn, 18th floor !?!? | 21:31 |
| lisitsyn | besser82: yeah | 21:31 |
| besser82 | lisitsyn, who wants to live there? | 21:31 |
| lisitsyn | besser82: I ;) | 21:31 |
| lisitsyn | besser82: why not? | 21:31 |
| besser82 | lisitsyn, It's da much too high | 21:31 |
| lisitsyn | besser82: http://instagram.com/p/cedWoGTfWN/ | 21:31 |
| lisitsyn | that's what I see from da window when it is summer | 21:32 |
| besser82 | lisitsyn, looks like somewhere in an industrial complex.... | 21:33 |
| lisitsyn | besser82: no typical russian city | 21:34 |
| lisitsyn | besser82: the closest building is university :D | 21:34 |
| besser82 | lisitsyn, other countries, other circumstances ;) | 21:34 |
| lisitsyn | besser82: I'd say it is kind of new centre of the city actually | 21:35 |
| besser82 | lisitsyn, you can imagine me house like some in a typical uptown-suburb | 21:35 |
| lisitsyn | besser82: I like that kind of houses | 21:35 |
| besser82 | lisitsyn, wanna come to germany? | 21:36 |
| besser82 | lisitsyn, have some flat in spare ;) | 21:36 |
| lisitsyn | besser82: not decided where I want to come :) | 21:36 |
| lisitsyn | besser82: I am 'almost married' ;) | 21:36 |
| besser82 | lisitsyn, good for getting visa :-P | 21:37 |
| besser82 | lisitsyn, esp. USA | 21:37 |
| besser82 | lisitsyn, :D | 21:37 |
| lisitsyn | they didn't give it to me though | 21:37 |
| besser82 | lisitsyn, then go get married and you'll get ;) | 21:37 |
| lisitsyn | yes but a bit later :D | 21:37 |
| besser82 | lisitsyn, for germany you just need grandpa with german shepheard :-P | 21:38 |
| lisitsyn | besser82: I am neither jew nor related to germany in any means ;) | 21:39 |
| lisitsyn | so no options but typisch process :) | 21:39 |
| besser82 | lisitsyn, yes, like I said | 21:40 |
| besser82 | lisitsyn, you need to tell you granpa had a german shepheard ;) | 21:40 |
| besser82 | lisitsyn, BITTE SEHR, IHRE PAPIERE :D | 21:40 |
| lisitsyn | besser82: I am not sure I understand what is shepheard ;) | 21:42 |
| besser82 | lisitsyn, it's a dog ;) | 21:49 |
| lisitsyn | besser82: ah then I get it right | 21:49 |
| -!- travis-ci [~travis-ci@ec2-54-235-33-212.compute-1.amazonaws.com] has joined #shogun | 21:50 | |
| 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/16080333 | 21:50 |
| -!- travis-ci [~travis-ci@ec2-54-235-33-212.compute-1.amazonaws.com] has left #shogun [] | 21:50 | |
| besser82 | lisitsyn, ^^^ Heiko hat's kaputt gemacht ;) | 21:51 |
| besser82 | lisitsyn, let's get back to some more serious stuff, like "How we can make SHOGUN rule the world" ;) | 21:53 |
| lisitsyn | besser82: shogun kaputt gemacht | 21:53 |
| besser82 | lisitsyn, yes ;) | 21:53 |
| besser82 | lisitsyn, how do we want to start making shogun better? | 21:55 |
| besser82 | lisitsyn, liking doing plugins and all? | 21:55 |
| lisitsyn | besser82: 1) I get some rest 2) I start doing some shit | 21:55 |
| lisitsyn | :D | 21:55 |
| lisitsyn | besser82: we should start experimenting with plugins | 21:55 |
| lisitsyn | cause I have barely any idea how to develop that so far | 21:55 |
| besser82 | lisitsyn, did you read the link in my email? | 21:55 |
| lisitsyn | not yet | 21:56 |
| lisitsyn | good you remind | 21:56 |
| besser82 | lisitsyn, there is some idea about how to get some very basic plugin"server" | 21:56 |
| besser82 | lisitsyn, like loading and registering symbols | 21:56 |
| besser82 | lisitsyn, some ground to start from | 21:57 |
| besser82 | lisitsyn, wiking had some idea about using shared_ptr for CSGObjects | 21:57 |
| besser82 | lisitsyn, and derived | 21:57 |
| @sonney2k | besser82, lisitsyn are you guys enjoying the trip? | 21:58 |
| besser82 | sonney2k, which trip? | 21:58 |
| lisitsyn | sonney2k: what trip? | 21:58 |
| @sonney2k | seems like it is not cold enough to kill the fancy mushrooms | 21:58 |
| lisitsyn | besser82: is he trying to offence us? :D | 21:59 |
| besser82 | lisitsyn, looks like | 21:59 |
| @sonney2k | you have all the dictator power of stalin and hitler combined | 21:59 |
| @sonney2k | I am more than afraid! | 21:59 |
| lisitsyn | I don't get this language :D | 21:59 |
| lisitsyn | I feel unsafe | 22:00 |
| besser82 | sonney2k, we don't want to harm anybody ;) | 22:00 |
| @sonney2k | lisitsyn, what is wrong with java modular javac: file not found: /home/travis/build/shogun-toolbox/shogun/build/src/interfaces/java_modular/*.java ?! | 22:00 |
| lisitsyn | besser82: I see auto_ptr here | 22:00 |
| @sonney2k | lisitsyn, did someone remove java stuff? | 22:00 |
| lisitsyn | auto_ptr is a NO-GO | 22:00 |
| lisitsyn | :D | 22:00 |
| lisitsyn | sonney2k: it wasn't me | 22:01 |
| lisitsyn | ;) | 22:01 |
| besser82 | lisitsyn, that is just an example | 22:01 |
| besser82 | lisitsyn, we need to mod that ;) | 22:02 |
| lisitsyn | besser82: okay well I get it | 22:02 |
| lisitsyn | besser82: this technique is known for me | 22:02 |
| besser82 | lisitsyn, auto_ptr is one posibility amoung much and some few better ones ;) | 22:02 |
| lisitsyn | shared_ptr for the win! | 22:02 |
| besser82 | lisitsyn, +1 ;) | 22:03 |
| lisitsyn | besser82: auto_ptr is dangerous with its strange copy behaviour | 22:03 |
| lisitsyn | besser82: the main problem I see is | 22:04 |
| besser82 | lisitsyn, which | 22:04 |
| lisitsyn | how to resolve some plugin in python | 22:04 |
| lisitsyn | i.e. I wrote a plugin | 22:04 |
| lisitsyn | how can I resolve it w/o the need to recompile shogun with its support | 22:04 |
| lisitsyn | shogun.mypluginclassifier | 22:04 |
| lisitsyn | this won't work | 22:04 |
| besser82 | lisitsyn, like building a c++-base pluging which provides the interpreter | 22:04 |
| lisitsyn | I mean basic shogun doesn't export plugin names | 22:05 |
| besser82 | lisitsyn, we don't need | 22:05 |
| lisitsyn | well but you need to get that shogun.myclassifier | 22:05 |
| lisitsyn | how do you do that? | 22:05 |
| besser82 | lisitsyn, we need some cheap metadata around with the plugins | 22:05 |
| lisitsyn | yes it is clear | 22:06 |
| lisitsyn | but still | 22:06 |
| besser82 | lisitsyn, datatypes and object shall get unified | 22:06 |
| lisitsyn | for our user | 22:06 |
| lisitsyn | how does he call new plugin? | 22:06 |
| lisitsyn | shogun.load('mystuff')? | 22:06 |
| lisitsyn | or what? | 22:06 |
| besser82 | lisitsyn, sth. like that | 22:06 |
| besser82 | lisitsyn, and that plugin should have some init() to provide ptr to all stuff inside | 22:07 |
| lisitsyn | well factory | 22:07 |
| lisitsyn | it should provide a factory | 22:07 |
| besser82 | lisitsyn, the plugin? | 22:09 |
| lisitsyn | yeah so you can create new classifier | 22:09 |
| besser82 | lisitsyn, classifier like? | 22:10 |
| lisitsyn | besser82: well you loaded a module | 22:10 |
| besser82 | lisitsyn, yes | 22:10 |
| lisitsyn | and then use its factory to create new instance | 22:10 |
| lisitsyn | of classifier it supports | 22:10 |
| lisitsyn | or whatever it does | 22:11 |
| besser82 | lisitsyn, that should be handled by some kind of plugin server.... | 22:11 |
| lisitsyn | why? | 22:11 |
| lisitsyn | you should be able to create multiple instances of that | 22:11 |
| besser82 | lisitsyn, yes | 22:11 |
| lisitsyn | so it should be explicit | 22:11 |
| besser82 | lisitsyn, but handled through plugin-server | 22:12 |
| lisitsyn | 'I need liblinear SVM instance' | 22:12 |
| lisitsyn | besser82: the worst thing here is that we need to use a lot of strings | 22:12 |
| lisitsyn | to name modules, to resolve them during runtime | 22:12 |
| besser82 | lisitsyn, pluginserver.request('instance of type') | 22:13 |
| besser82 | lisitsyn, which then returns ptr to that new instace | 22:13 |
| lisitsyn | I think it is shogun.create('LibLinearSVM') or whatever | 22:13 |
| lisitsyn | or even | 22:13 |
| lisitsyn | shogun.createClassifier(..) | 22:13 |
| lisitsyn | cause you need it explicit in python | 22:14 |
| besser82 | lisitsyn, how ever we name the child ;) | 22:14 |
| lisitsyn | otherwise people won't able to cast it back to classifier | 22:14 |
| besser82 | lisitsyn, unfied object /data- types ??? | 22:14 |
| lisitsyn | besser82: how unified? | 22:15 |
| lisitsyn | is it up to sgobject or whatever it is called? | 22:15 |
| besser82 | lisitsyn, like having only one type of each thing | 22:15 |
| besser82 | lisitsyn, Things deriving from CSGObj should be unified eg | 22:15 |
| lisitsyn | ehmmm so kernel and classifier have the same set of methods? | 22:16 |
| besser82 | lisitsyn, posibly | 22:16 |
| besser82 | lisitsyn, lemme check code | 22:16 |
| lisitsyn | besser82: it could be a problem | 22:16 |
| besser82 | lisitsyn, why? | 22:16 |
| lisitsyn | well what is apply for kernel? | 22:16 |
| lisitsyn | or what is train for distance | 22:17 |
| besser82 | lisitsyn, empty stub, that does nothing? | 22:18 |
| lisitsyn | besser82: sounds bad | 22:18 |
| lisitsyn | I mean too many methods that do nothing but exist there for a reason | 22:18 |
| besser82 | lisitsyn, how to resolve otherwise? | 22:19 |
| besser82 | sonney2k, are you still around?!? | 22:19 |
| lisitsyn | besser82: create_classifier, create_processor etc | 22:19 |
| besser82 | lisitsyn, andd how to annouce / declare the availability? | 22:20 |
| lisitsyn | la excepcione ! | 22:20 |
| lisitsyn | throw new FuckYouException | 22:21 |
| lisitsyn | :D | 22:21 |
| shogun-buildbot | build #150 of clang34 - undefined behaviour analysis is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/150 blamelist: Saurabh7 <saurabh.mahindre@gmail.com> | 22:22 |
| * besser82 throws an instace of FuckYouException: (What?) MyAss!!! | 22:22 | |
| besser82 | lisitsyn, ????? | 22:22 |
| lisitsyn | besser82: okay availability of what? | 22:23 |
| @sonney2k | besser82, yes | 22:23 |
| besser82 | sonney2k, what's wrong with java? | 22:23 |
| @sonney2k | besser82, travis is unhappy | 22:23 |
| @sonney2k | I am waiting for the buildbot to see if this is just a heisenbug | 22:24 |
| besser82 | lisitsyn, about which methods / functions are inside the new avail class | 22:24 |
| @sonney2k | besser82, actually all good on the buildbot | 22:24 |
| besser82 | sonney2k, looks like a glitch to me | 22:25 |
| besser82 | lisitsyn, without needing predecl or headers? | 22:25 |
| lisitsyn | besser82: if we speak about plugins we have no headers | 22:25 |
| @sonney2k | lisitsyn, I only see the problem with registering sth new to non-interpreted interfaces (java, csharp, ...) | 22:26 |
| lisitsyn | sonney2k: yes but that's terribly major proble | 22:26 |
| @sonney2k | lisitsyn, why? everything has to be available *at compile time* | 22:27 |
| @sonney2k | that is normal with such $langs | 22:27 |
| lisitsyn | sonney2k: how can plugin be available at compile time? | 22:27 |
| besser82 | lisitsyn, they don't need but some decl needs to be | 22:28 |
| lisitsyn | then we have to put every and each plugin to shogun itself | 22:28 |
| lisitsyn | I mean at least their exported names | 22:28 |
| besser82 | lisitsyn, or we just use unified stuff ;) | 22:28 |
| lisitsyn | unified interface is the way | 22:29 |
| lisitsyn | but it is not unified like one possible class | 22:29 |
| lisitsyn | rather a few possible classes | 22:29 |
| besser82 | lisitsyn, as i said | 22:29 |
| besser82 | lisitsyn, a handfull of generic stuff and all is fine ;) | 22:29 |
| besser82 | sonney2k, did you read those full backlog? | 22:31 |
| besser82 | sonney2k, like the beginning of my chat with lisitsyn ? | 22:31 |
| besser82 | sonney2k, what's your oppionion about that?!? | 22:31 |
| @sonney2k | besser82, yeah but no idea what you have in your minds | 22:32 |
| besser82 | sonney2k, mein kopf ist so klar, wie der korn darin :-P | 22:32 |
| besser82 | sonney2k, features, features, features | 22:33 |
| lisitsyn | haha we have to write specification | 22:33 |
| lisitsyn | we all know what do we want to get | 22:33 |
| besser82 | sonney2k, like pluginize everything and make stuff possible like with e.g. rubygems | 22:34 |
| besser82 | sonney2k, shogun-samurai install fancy-extension-for-shogun ;) | 22:34 |
| besser82 | lisitsyn, spec like? | 22:34 |
| lisitsyn | besser82: describe anything | 22:34 |
| lisitsyn | before coding | 22:35 |
| besser82 | lisitsyn, ahh kk like ISO :D | 22:35 |
| @sonney2k | besser82, bakufu is the name | 22:35 |
| besser82 | sonney2k, that was the name i wanted to reserve for the "core" stuff ;) | 22:35 |
| lisitsyn | besser82: yes first ISO certified machine learning toolbox :D | 22:35 |
| lisitsyn | can we speak docs | 22:36 |
| lisitsyn | ? | 22:36 |
| lisitsyn | besser82: sonney2k: are you ok with describing it in gdoc? | 22:36 |
| besser82 | sonney2k, like having shogun the interface, bakufu the "core" plugins maintained by us and samurai for community / gem stuff ;) | 22:36 |
| @sonney2k | lisitsyn, or github wiki! | 22:37 |
| besser82 | sonney2k, +1 ;) | 22:37 |
| shogun-buildbot | build #148 of clang34 - thread analysis is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/148 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Saurabh7 <saurabh.mahindre@gmail.com> | 22:38 |
| lisitsyn | sonney2k: besser82: it is less realtime but ok | 22:39 |
| lisitsyn | https://github.com/shogun-toolbox/shogun/wiki/Shogun-splitting | 22:39 |
| lisitsyn | how does it handle multiple changes? | 22:40 |
| lisitsyn | are they overwritten? | 22:40 |
| @sonney2k | lisitsyn, realtime is etherpad | 22:42 |
| @sonney2k | e.g. http://beta.etherpad.org/ | 22:43 |
| lisitsyn | sonney2k: google docs is as well? | 22:43 |
| lisitsyn | I trust google docs more :D | 22:44 |
| lisitsyn | especially due | 22:44 |
| lisitsyn | to ' Pad contents are destroyed periodically' | 22:44 |
| besser82 | lisitsyn, sonney2k which one now? gh-wiki, gdocs or ethercrap? | 22:49 |
| lisitsyn | no idea :) | 22:50 |
| @sonney2k | I am for gh-wiki | 22:52 |
| @sonney2k | it is public -> better | 22:52 |
| lisitsyn | okay | 22:52 |
| besser82 | lisitsyn, sonney2k me is fine with either | 22:57 |
| shogun-buildbot | build #165 of clang34 - static analysis is complete: Failure [failed analyse] Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/165 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Saurabh7 <saurabh.mahindre@gmail.com> | 23:21 |
| shogun-buildbot | build #151 of clang34 - undefined behaviour analysis is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/151 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 23:32 |
| -!- HeikoS [~heiko@90.211.150.15] has joined #shogun | 23:58 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 23:58 | |
| --- Log closed Sun Dec 29 00:00:27 2013 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!
