--- Log opened Wed Apr 10 00:00:26 2013 | ||
blackburn | so is DRPK going to have a party? | 00:01 |
---|---|---|
harpreet | Hi, I was interested in the 'Develop interactive machine learning demos' for GSoc. | 00:14 |
harpreet | Any advice on what sort of work I should do before I apply? | 00:15 |
blackburn | harpreet: this was answered quite a few times on the mailing list - consider starting to contribute with small patches | 00:19 |
harpreet | Since the project is almost entirely concerned with Python, Django, JS, should I submit patches for the core C++ code or something else? | 00:22 |
medeeiip | yesterday I asked about development of a CMakelist. So working under windows is possible.... I did wrote the whole cmake list but the problem is compiler is complaining about a lot of things.... like e.g pthread... | 00:23 |
medeeiip | can pthread be made optional? | 00:25 |
medeeiip | the compiler is complaining about dirent.h too (as that is not available on windows) | 00:26 |
medeeiip | is there any more additional dependencies too ( one was with sys/types.h but I made that inclusion optional using #ifdef _WIN32 .....#else...code) | 00:29 |
blackburn | harpreet: no that's not that required to have C++ patches - any examples using python would work too | 00:33 |
blackburn | medeeiip: yes pthread is optional | 00:33 |
blackburn | medeeiip: that would be very helpful to fix these includes | 00:34 |
medeeiip | can you tell me where and how dirent.h is used ( I mean I can see where, actually how?) | 00:35 |
medeeiip | i mean not specifically | 00:35 |
medeeiip | but to give me an idea | 00:35 |
blackburn | medeeiip: it is used to find files IIRC | 00:36 |
medeeiip | so if i need to replace that dependency I need to do the same thing on windows in a alternative way.... | 00:38 |
blackburn | yes probably | 00:38 |
medeeiip | is there any available list about what feature is optional and what is required? | 00:41 |
blackburn | feature of? | 00:41 |
medeeiip | i mean optional components like (pthread as you said).... specifically what library is required and what is optional? | 00:43 |
blackburn | ahh no I don't think we have a list | 00:44 |
blackburn | basically it should compile always | 00:44 |
medeeiip | it is directly complaining about pthread..... | 00:44 |
blackburn | medeeiip: all pthreads code should be guarded with preprocessor | 00:45 |
blackburn | so that's a bug | 00:45 |
medeeiip | no no.... hold a sec .... as confg.h generation was not automated i created a config.h where I used #define HAVE_PTHREAD 1 | 00:47 |
medeeiip | let me check if it works without it. | 00:47 |
medeeiip | an idiot i am....:D | 00:49 |
blackburn | medeeiip: yeah HAVE_PTHREAD should not be set if you don't have it | 00:51 |
medeeiip | what is lbfgsfloatval_t datatype? | 01:05 |
medeeiip | in optimization_libfgs.cpp | 01:05 |
blackburn | medeeiip: what kind of answer do you expect? ;) | 01:08 |
medeeiip | from what library it is? | 01:09 |
blackburn | lbfgs | 01:09 |
medeeiip | 'n that is? | 01:09 |
medeeiip | BFGS method? | 01:09 |
medeeiip | lib for that? | 01:09 |
blackburn | limited memory broyden fletcher goldfarb shanno | 01:09 |
blackburn | yes | 01:09 |
blackburn | medeeiip: it is incldued to the sources | 01:11 |
medeeiip | but lbfgsfloatval_t is not defined in lbfgs.h | 01:13 |
blackburn | medeeiip: where is it used? | 01:16 |
medeeiip | shogun\examples\undocumented\libshogun\optimization_lbfgs.cpp | 01:18 |
medeeiip | example though..... | 01:18 |
blackburn | hmm | 01:21 |
blackburn | it might be it wasn't properly updated | 01:21 |
blackburn | have to sleep now | 01:23 |
medeeiip | it's night at ur country | 01:23 |
medeeiip | ??????? | 01:23 |
blackburn | medeeiip: yes deep night slowly becoming a morning | 01:24 |
medeeiip | same in india...... but | 01:24 |
medeeiip | u 'r from germany ? | 01:24 |
medeeiip | isn't it? | 01:25 |
blackburn | no, russia | 01:25 |
blackburn | it is sonney2k who is from berlin | 01:25 |
medeeiip | i see..... | 01:26 |
medeeiip | btw good nite | 01:26 |
medeeiip | i too need a sleep badly....... tomorrow i've college | 01:26 |
blackburn | nite | 01:26 |
-!- medeeiip [~medeeiip@103.2.132.232] has quit [Quit: Leaving] | 01:32 | |
-!- FSCV [~FSCV@187.210.54.166] has quit [Quit: Leaving] | 02:09 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has joined #shogun | 03:15 | |
-!- phd [lvoronova1@77-120-201-195.sumy.volia.net] has quit [Ping timeout: 248 seconds] | 03:39 | |
-!- foulwall_ [~foulwall@li379-21.members.linode.com] has joined #shogun | 03:44 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has quit [Read error: Connection reset by peer] | 03:44 | |
-!- gsomix [~Miranda@188.168.14.137] has quit [Ping timeout: 245 seconds] | 04:51 | |
-!- blackburn [~blackburn@188.168.14.137] has quit [Ping timeout: 245 seconds] | 04:51 | |
-!- blackburn [~blackburn@83.234.169.3] has joined #shogun | 05:07 | |
-!- rishabh [~rishabh@202.78.172.162] has joined #shogun | 05:34 | |
-!- abinash [75ef5e6e@gateway/web/freenode/ip.117.239.94.110] has joined #shogun | 06:56 | |
-!- gzhd79 [~zeller@mnhm-4d011972.pool.mediaWays.net] has joined #shogun | 06:59 | |
-!- gzhd79 [~zeller@mnhm-4d011972.pool.mediaWays.net] has quit [Client Quit] | 07:03 | |
-!- foulwall_ [~foulwall@li379-21.members.linode.com] has quit [Ping timeout: 252 seconds] | 07:33 | |
-!- harpreet [~ceo@202.3.77.220] has quit [Read error: Connection reset by peer] | 07:59 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has joined #shogun | 07:59 | |
@sonney2k | blackburn, muhaaaahaa you do boosting now :D | 08:00 |
blackburn | sonney2k: yeah | 08:12 |
blackburn | sonney2k: is the code crappy? | 08:13 |
@sonney2k | blackburn, no idea - but this reminds me of my multiboost endeavour | 08:14 |
@sonney2k | and I guess their code is massive - they have won a couple of competitions with that | 08:14 |
blackburn | sonney2k: I do not know what to answer to last message | 08:14 |
blackburn | :D | 08:14 |
@sonney2k | gtg brb | 08:15 |
-!- bogdanc [~bogdan@p16.eregie.pub.ro] has joined #shogun | 08:22 | |
sonne|work | blackburn: haha | 08:24 |
blackburn | sonne|work: I am scared by your laugh | 08:25 |
blackburn | :D | 08:25 |
sonne|work | *MUUUUUAUAUAAAAAAHHHH* | 08:25 |
blackburn | sonne|work: is adaboost-reg really that cool? | 08:30 |
blackburn | I am quite surprised they got back to 2001 paper | 08:30 |
sonne|work | blackburn: it certainly is a reasonable method | 08:31 |
sonne|work | like any other method of course | 08:31 |
blackburn | sonne|work: what I do not get - do they have C++ impl of their adaboost? | 08:33 |
sonne|work | blackburn: "they"? | 08:34 |
blackburn | sonne|work: or him - whatever :) | 08:34 |
blackburn | sonne|work: he said 'they' are limited with time | 08:34 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 08:35 | |
-!- abinash [75ef5e6e@gateway/web/freenode/ip.117.239.94.110] has quit [Ping timeout: 245 seconds] | 08:35 | |
sonne|work | blackburn: I don't know whom you mean | 08:36 |
blackburn | sonne|work: roni | 08:36 |
sonne|work | ahh so not the multiboost guys | 08:36 |
sonne|work | well adaboost needs early stopping (or it will overfit) | 08:37 |
sonne|work | or some kind of regularization | 08:37 |
sonne|work | and of course it sucks to have too many baselearners in the end | 08:37 |
sonne|work | (slow when using it) | 08:37 |
n4nd0 | adaboost doesn't overfit (or so said my professor always) | 08:37 |
sonne|work | haha | 08:37 |
sonne|work | then he has never used it | 08:37 |
n4nd0 | she* | 08:38 |
blackburn | hmm how can it *not* overfit? | 08:38 |
n4nd0 | mm I am not completely sure what was her claim | 08:38 |
n4nd0 | let me check slides | 08:39 |
blackburn | n4nd0: I just can't see any 'rule' that makes it not overfitting algorithm | 08:39 |
blackburn | I mean how many weak classifiers do we need? | 08:39 |
blackburn | 50, 100? | 08:40 |
sonne|work | n4nd0: gunnar back in the days wrote a paper about it improving the situation | 08:40 |
n4nd0 | blackburn: what do you want to use it for? | 08:40 |
sonne|work | back in 1999 or so | 08:40 |
blackburn | n4nd0: I was stupid enough last night to suggest to help with integrating some adaboost reg | 08:40 |
n4nd0 | blackburn: multiboost code to shogun? | 08:41 |
blackburn | n4nd0: no, adaboost reg by gunnar | 08:41 |
sonne|work | blackburn: I guess adaboost* is easy | 08:41 |
sonne|work | but you need the framework for baselearners | 08:41 |
sonne|work | so basically any shogun classifier could be baselearner | 08:42 |
blackburn | sonne|work: what kind of framework? | 08:42 |
sonne|work | blackburn: CBaseLearner ? | 08:42 |
blackburn | sonne|work: you mean some ensemble classifier | 08:43 |
n4nd0 | sonne|work: if any shogun classifier could be a baselearner, why not CMachine? | 08:43 |
blackburn | that outputs this weighted sum? | 08:43 |
blackburn | that sounds easy - the only thing is how to learn these classifiers and their weights right? | 08:43 |
sonne|work | n4nd0: yeah CMachine is what you will likely need to use though regression methods etc don't fit | 08:43 |
n4nd0 | aham true | 08:44 |
sonne|work | blackburn: err no - you have a fixed parameter setting for base learners | 08:44 |
sonne|work | blackburn: so you just call train(data) | 08:44 |
sonne|work | and it is not so clear to me how to reweight data | 08:44 |
blackburn | sonne|work: ah so we need weighting support here | 08:45 |
n4nd0 | found it in slides directly | 08:45 |
n4nd0 | Beauty of AdaBoost | 08:46 |
n4nd0 | among other things | 08:46 |
n4nd0 | Test Errot: Asmptotes - no over-fitting observed. It continues to decrease after training error vanishes. | 08:46 |
blackburn | n4nd0: sounds like the best method | 08:46 |
blackburn | n4nd0: that's crazy | 08:46 |
n4nd0 | Asymptotes* | 08:46 |
blackburn | n4nd0: I am sure it will increase as weighted sum of a lot of floats gets not really stable with time | 08:47 |
n4nd0 | blackburn: IIRC the weights are normalized | 08:47 |
n4nd0 | not sure if that helps in what you are referring to | 08:48 |
sonne|work | n4nd0: read the abstract of http://books.google.com/books?id=bMuzXPzlkG0C&pg=PA564&lpg=PA564&dq=gunnar+ratsch+adaboost*&source=bl&ots=MvrfwEEJIj&sig=G72a-SjKPxM6VIA0Xj_RdPtDYJc&hl=en&sa=X&ei=2QplUbe3KMm7hAePioBI&ved=0CC8Q6AEwAA | 08:48 |
sonne|work | n4nd0: no overfitting if no noise | 08:48 |
sonne|work | but w/ noise (which is always present) | 08:48 |
sonne|work | you have it | 08:48 |
sonne|work | that is gunnars '99 paper | 08:48 |
sonne|work | adaboost reg | 08:48 |
n4nd0 | sonne|work: aham ok, I don't really know how to interpret the noise here | 08:48 |
sonne|work | n4nd0: well any measured data is noise | 08:49 |
n4nd0 | the course I am talking about was really focused on using Adaboost for face detection | 08:49 |
n4nd0 | with this Haar features | 08:49 |
sonne|work | yeah noisy | 08:49 |
sonne|work | as hell | 08:49 |
blackburn | modern method | 08:49 |
blackburn | :D | 08:49 |
sonne|work | svm is from 1997 :P | 08:49 |
n4nd0 | mmm I don't really understand then why it is said like that in the slides | 08:50 |
blackburn | sonne|work: yeah that's why svm is 'outdated' comparing to deep learning things | 08:50 |
sonne|work | and neural nets from decades ago :P | 08:50 |
blackburn | sonne|work: svm is from 1970 actually | 08:50 |
blackburn | :D | 08:50 |
sonne|work | even more outdated then | 08:50 |
sonne|work | deep learning is just hip again since people had SVMs (with guaranteed convergence) for long enough | 08:51 |
sonne|work | blackburn: theory of SVM maybe but not the applications | 08:51 |
blackburn | sonne|work: no it is pretty different | 08:51 |
blackburn | sonne|work: learning features is important | 08:51 |
-!- gzhd79 [~zeller@embln.embl.de] has joined #shogun | 08:52 | |
sonne|work | blackburn: so? | 08:52 |
blackburn | sonne|work: I can't see SVM learning anything on features so it is quite different I'd say | 08:52 |
sonne|work | it is like what NN's have always done | 08:52 |
sonne|work | but since NN's have local minima | 08:52 |
sonne|work | they are extremely hard to control | 08:52 |
blackburn | sonne|work: when working on images it gets important | 08:52 |
sonne|work | the right features are important for any task | 08:53 |
shogun-buildbot | build #355 of nightly_default is complete: Failure [failed test] Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/355 | 08:53 |
n4nd0 | blackburn: aren't there ways to use large margin SVM-like learning to choose features? | 08:53 |
blackburn | sonne|work: yes but we should not engineer it | 08:53 |
blackburn | n4nd0: just *choose* | 08:53 |
n4nd0 | on the meantime test break... :S | 08:53 |
blackburn | n4nd0: deep learning is about constructing a few levels from low-level (pixels) | 08:54 |
sonne|work | blackburn: yeah sure that is the dream - but there still is a long way to go | 08:54 |
blackburn | sonne|work: yes but it is already making some steps to automagic features | 08:54 |
sonne|work | blackburn: with NNs you choose certain architectures, ways of training etc that have a massive impact on what features it learns | 08:54 |
blackburn | sonne|work: that's true | 08:54 |
sonne|work | blackburn: well then use MKL with all kinds of kernels | 08:55 |
sonne|work | it also learns some features | 08:55 |
blackburn | sonne|work: but SVM do not consider it at all | 08:55 |
sonne|work | but sucks badly | 08:55 |
sonne|work | I don't agree | 08:55 |
sonne|work | SVM learns important features in feature space | 08:55 |
sonne|work | but you limit the features by choosing a kernel | 08:55 |
n4nd0 | sonne|work: why does the approach MKL with lot of kernels suck? | 08:55 |
sonne|work | and yes you have a hard time to express these features | 08:56 |
blackburn | sonne|work: feature selection is totally different from feature extraction | 08:56 |
sonne|work | n4nd0: it usually gives worse performance than just adding kernels | 08:56 |
blackburn | sonne|work: I tend to believe we should *extract* features not select | 08:56 |
sonne|work | I don't | 08:56 |
sonne|work | I think the more features - the better | 08:56 |
blackburn | sonne|work: yes, but again back to images | 08:57 |
blackburn | sonne|work: if we just select pixels | 08:57 |
blackburn | it always sucks | 08:58 |
blackburn | but if we have some high-level on it (like sift) | 08:58 |
sonne|work | blackburn: sure | 08:59 |
sonne|work | so they use convolutional NN's | 08:59 |
sonne|work | others suck too | 08:59 |
blackburn | sonne|work: yes convolution is a good base to transform pixels to something | 08:59 |
blackburn | sonne|work: it is engineered to do convolution - that's bad | 09:00 |
sonne|work | blackburn: believe - a ML method is as good as the person who trains it | 09:00 |
sonne|work | there are soo many tricks involved in everything | 09:01 |
sonne|work | there usually is no general rule | 09:01 |
sonne|work | saying that methodA is better than methodB | 09:01 |
sonne|work | some methods are a more natural choice for certain tasks of course but that's it | 09:01 |
blackburn | sonne|work: I am speaking not about methods but the approach | 09:01 |
shogun-buildbot | build #1009 of deb3 - modular_interfaces is complete: Failure [failed test python_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1009 blamelist: Soeren Sonnenburg <sonne@debian.org> | 09:02 |
blackburn | learning some method on engineered features is (in general) worse | 09:03 |
blackburn | than learning features and then method | 09:03 |
sonne|work | blackburn: to the contrary | 09:03 |
sonne|work | learning features & method together is usually worse | 09:04 |
sonne|work | humans are usually much better in engineering features | 09:04 |
blackburn | sonne|work: depends | 09:04 |
blackburn | sonne|work: with images (it is the only field I know) it is not - sparse features, dictionary learning, convolution stuff - these are better | 09:05 |
blackburn | sonne|work: e.g there is no heuristic in fitting HOG cell sizes | 09:06 |
blackburn | it is just stupid model selection | 09:06 |
sonne|work | blackburn: yes but if you then add prior knowledge and force the thing to focus on reasonable features | 09:07 |
sonne|work | you will get even better results | 09:07 |
blackburn | bah I am still at home should be at job | 09:08 |
blackburn | :D | 09:08 |
sonne|work | http://yann.lecun.com/exdb/mnist/ | 09:09 |
sonne|work | blackburn: and I am at work but hmm... ok working now | 09:09 |
sonne|work | cu! | 09:09 |
blackburn | sonne|work: you see | 09:09 |
blackburn | everything from the top is about extracting | 09:09 |
sonne|work | yes and lots of is about learning | 09:09 |
sonne|work | and lots of is about extracting & learning | 09:09 |
sonne|work | and limiting learning of features | 09:10 |
blackburn | sonne|work: ensembling + feature extraction + learning | 09:10 |
sonne|work | k-NN with 0.63! | 09:10 |
sonne|work | beats halv of the NN | 09:11 |
sonne|work | s | 09:11 |
blackburn | sonne|work: very well engineered distance I guess | 09:11 |
blackburn | shape context matching should be something that matches these curves in a smart way | 09:11 |
sonne|work | actually even 0.52 | 09:11 |
sonne|work | better than any VM | 09:11 |
sonne|work | committee of 35 conv. net, 1-20-P-40-P-150-10 [elastic distortions] width normalization 0.23 | 09:12 |
sonne|work | haha | 09:12 |
blackburn | sonne|work: it must be pretty slow | 09:12 |
sonne|work | and they do distortions etc | 09:13 |
sonne|work | the usual invariance tricks | 09:13 |
blackburn | sonne|work: yes most of top methods do it with distortions | 09:13 |
sonne|work | rotation/shifting lala | 09:13 |
-!- blackburn [~blackburn@83.234.169.3] has quit [Quit: Leaving.] | 09:14 | |
-!- medeeiip [~medeeiip@103.18.168.49] has joined #shogun | 09:28 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 09:29 | |
-!- gzhd79 [~zeller@embln.embl.de] has left #shogun [] | 09:32 | |
-!- blackburn [~lisitsin@mxs.kg.ru] has joined #shogun | 09:38 | |
blackburn | sonne|work: did you like that new mentor? | 09:42 |
sonne|work | blackburn: 25minutes! | 09:42 |
blackburn | I had to reject him though | 09:42 |
blackburn | sonne|work: what 25 minutes? | 09:42 |
sonne|work | blackburn: yeah cool guy | 09:42 |
sonne|work | it took you to get to work | 09:42 |
blackburn | sonne|work: hah yes with no jams it is pretty fast | 09:43 |
blackburn | sonne|work: 1.5 km actually | 09:43 |
sonne|work | blackburn: we certainly need more mentors like him | 09:44 |
blackburn | sonne|work: it takes more time to enter the building and get to that 13th floor | 09:44 |
blackburn | than moving by bus | 09:44 |
sonne|work | though it is a bit sad that he didn't speak japanese | 09:44 |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds] | 09:44 | |
blackburn | sonne|work: he has other cool skills | 09:45 |
blackburn | sonne|work: he speaks machine code at least | 09:46 |
sonne|work | brainfuck? | 09:48 |
sonne|work | topcoder - within top 10? | 09:48 |
sonne|work | we need HEROS | 09:49 |
blackburn | sonne|work: I don't mind a fluent machine code speaker in team | 09:49 |
sonne|work | true | 09:49 |
sonne|work | kind of a matrix operator | 09:49 |
sonne|work | we should equip ourselves with phone booths to exit the matrix | 09:50 |
sonne|work | and some reasonable mobiles | 09:50 |
blackburn | sonne|work: matrix operator is a must have | 09:50 |
sonne|work | but only one that can cope with a few liters of vodka per day | 09:51 |
blackburn | sonne|work: mind releasing 2.1.1 once I get finished with tapkee? | 09:56 |
sonne|work | you have to ask our release manager | 09:56 |
sonne|work | I think it has to be 2.2 btw | 09:56 |
blackburn | sonne|work: ahhh yes | 09:56 |
sonne|work | we changed some api's | 09:57 |
blackburn | sonne|work: no 2.1.1 | 09:57 |
blackburn | wherE? | 09:57 |
-!- bogdanc [~bogdan@p16.eregie.pub.ro] has quit [Quit: Konversation terminated!] | 10:06 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 10:10 | |
-!- rishabh [~rishabh@202.78.172.162] has quit [Remote host closed the connection] | 10:21 | |
-!- Yanglittle [deb20af1@gateway/web/freenode/ip.222.178.10.241] has joined #shogun | 10:40 | |
Yanglittle | Hi, I have a question, how should I do when I want to combine the pre-computed kernel ? I didn't find the examples in the shogun. | 10:42 |
sonne|work | Yanglittle: just add the kernel like any other kernel | 10:44 |
sonne|work | you just don't need any features... | 10:47 |
Yanglittle | this is the usage of add_kernel: sg('add_kernel', weight, kernel-specific parameters) the third parameter is replaced by pre-computed kernel matrix? | 10:50 |
blackburn | sonne|work: your biggest mistake was to formulate that demos idea :D | 10:51 |
sonne|work | ohh static interfaces | 10:52 |
sonne|work | Yanglittle: please use a modular interface | 10:52 |
blackburn | we have somewhat 5 6 7 may be 8? | 10:52 |
-!- n4nd0 [~test@n133-p134.kthopen.kth.se] has joined #shogun | 10:52 | |
sonne|work | blackburn: in worst case we create one project for interactive *on the web* and one for shogun's examples | 10:53 |
sonne|work | blackburn: hmmhh this adaboost code must be magic... | 10:54 |
blackburn | sonne|work: that would push some ML idea out.. | 10:54 |
sonne|work | I still don't know how you get weights for examples to work w/ shogun | 10:54 |
sonne|work | but hey | 10:54 |
blackburn | sonne|work: I LOVE MAGIC CODE | 10:54 |
blackburn | :D | 10:54 |
sonne|work | blackburn: why should it push out an ML idea? | 10:55 |
blackburn | sonne|work: # slots? | 10:55 |
sonne|work | well last year we gave one back | 10:55 |
sonne|work | so... | 10:55 |
blackburn | sonne|work: true | 10:55 |
sonne|work | blackburn: look at http://codingplayground.blogspot.nl/2009/03/adaboost-improve-your-performance.html | 11:05 |
sonne|work | he also has the orginal adaboost algorithm in there | 11:05 |
blackburn | hahah funny code | 11:06 |
sonne|work | I didn't know that http://en.wikipedia.org/wiki/Godel_prize | 11:06 |
sonne|work | the adaboost authors got that prize | 11:06 |
sonne|work | but for real boosting use http://www.multiboost.org/ | 11:06 |
blackburn | sonne|work: in two words - what is the problem with integrating multiboost? | 11:07 |
sonne|work | blackburn: it is huuuge | 11:08 |
blackburn | sonne|work: and one more important Q - why multiboost doesn't have adaboost reg | 11:08 |
sonne|work | it has everytthing invented like we do but different ... | 11:08 |
sonne|work | it has adaboost MH | 11:08 |
blackburn | sonne|work: what is MH? | 11:08 |
sonne|work | www.lsi.upc.edu/~lluism/cursos/docSoria00/add2.ps.gz | 11:09 |
sonne|work | seems like for multilable | 11:09 |
sonne|work | label | 11:09 |
sonne|work | and w/ confidences | 11:10 |
sonne|work | but no regularization!? | 11:10 |
blackburn | I mean if adaboost reg is kind of cool - why didn't they implement it | 11:10 |
blackburn | that seems to be strange at least | 11:11 |
sonne|work | I guess early stopping is good enough | 11:13 |
-!- foulwall [~foulwall@li379-21.members.linode.com] has quit [Remote host closed the connection] | 11:20 | |
-!- gsomix [~Miranda@83.234.169.3] has joined #shogun | 11:30 | |
gsomix | hi guys | 11:30 |
n4nd0 | hey gsomix! | 11:30 |
medeeiip | hello everyone I's working to build the shogun on windows.... seems almost I've to entirely rewrite math.h using _win32 preprocessor | 11:40 |
medeeiip | but currently i'm stuck with this error: | 11:40 |
medeeiip | use of undefined type 'shogun::shogun::CStringFeatures<ST>' | 11:40 |
medeeiip | 1> with | 11:40 |
medeeiip | 1> [ | 11:40 |
medeeiip | 1> ST=uint16_t | 11:40 |
medeeiip | 1> ] | 11:40 |
n4nd0 | ups rewriting code... that must be touch | 11:41 |
medeeiip | from src\shogun/distributions/HMM.h(582) | 11:41 |
n4nd0 | medeeiip: how are you trying to build it? using cygwin? | 11:41 |
medeeiip | no native vs compiler...... | 11:41 |
n4nd0 | mmm I am not entirely sure what you mean | 11:41 |
blackburn | medeeiip: you probably didn't define a thing that enables that type | 11:41 |
n4nd0 | in any case I think that the easiest way to get shogun working in windows is with cygwin, algthough I have no experience | 11:42 |
medeeiip | it's not about working.... I've enthu of porting shogun to windows... | 11:43 |
n4nd0 | ah ok I see | 11:43 |
medeeiip | about that error: | 11:43 |
medeeiip | there is a def : | 11:44 |
medeeiip | CStringFeatures<uint16_t>* p_observations; | 11:45 |
medeeiip | one more thing i did.... | 11:45 |
blackburn | medeeiip: -DUSE_UINT16 and so on | 11:45 |
medeeiip | implmented mmanwin32 | 11:45 |
-!- 18VAAX8BZ [~lvoronova@77-120-201-195.sumy.volia.net] has joined #shogun | 11:46 | |
-!- phd [~lvoronova@77-120-201-195.sumy.volia.net] has joined #shogun | 11:46 | |
-!- 18VAAX8BZ [~lvoronova@77-120-201-195.sumy.volia.net] has left #shogun [] | 11:46 | |
medeeiip | blackburn: I didn't understand | 11:46 |
medeeiip | what u just said | 11:46 |
medeeiip | -DUSE_UINT16 where? | 11:47 |
blackburn | medeeiip: you should have USE_UINT16 define to have features for uint16_t enabled | 11:47 |
medeeiip | actually i used a #ifdef _WIN32 and included stdint.h | 11:48 |
medeeiip | where uint16_t is available | 11:48 |
medeeiip | I also implemented mman-win32 from https://code.google.com/p/mman-win32/ | 11:52 |
medeeiip | so memory management shouldn't be a problem..... | 11:52 |
blackburn | medeeiip: hmm wait it is not necessary to have that define | 11:53 |
blackburn | it is relevant for interfaces but not for anything else it seems | 11:53 |
-!- heiko1 [~heiko@nat-183-237.internal.eduroam.ucl.ac.uk] has joined #shogun | 11:56 | |
blackburn | n4nd0: thanks for answering these messages | 11:58 |
n4nd0 | blackburn: :) it's nothing | 11:59 |
medeeiip | blackburn: any hack for that....... | 12:02 |
-!- arp [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 12:06 | |
arp | hey.shogun does not work in ubuntu 64 bit rite? | 12:07 |
n4nd0 | arp: Hi, it does work. | 12:07 |
n4nd0 | what problem do you have? | 12:08 |
arp | i was installing it via the ubutu software center,it jept on failing,saying that it failed to download some packages | 12:09 |
arp | or some packages could not be downloaded | 12:09 |
n4nd0 | Aham, I assumed you were installing from source, sorry. | 12:09 |
n4nd0 | I didn't actually know you can install it via de software center :) | 12:10 |
n4nd0 | arp: What package in particular did you try to install? I have just tried with libshogun11 and it worked fine here. | 12:12 |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 12:12 | |
Yanglittle | It can't find the module when I run "from shogun.Kernel import GaussianKernel" and I can't find them on my PC.. | 12:13 |
n4nd0 | Yanglittle: I guess you built shogun for python_modular. Did you make install after building as well? | 12:14 |
-!- arp [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Quit: Page closed] | 12:15 | |
Yanglittle | three steps I all executed. | 12:16 |
n4nd0 | Yanglittle: is it possible that you have more than one python version in your system? | 12:17 |
n4nd0 | it could be that shogun was built against one of the versions and you are trying to run it using another | 12:17 |
-!- ZeThomas [~thomaspr@2a02:2c40:100:a000:a81c:41e5:277a:c326] has joined #shogun | 12:18 | |
Yanglittle | i installed two versions ago. now just version 2.7 | 12:20 |
Yanglittle | i will rebuild it | 12:20 |
n4nd0 | make sure the version of python detected by shogun is the same one you want to use | 12:21 |
n4nd0 | detected by shogun = detected by the configure script | 12:21 |
Yanglittle | i will, thanks. | 12:22 |
ZeThomas | hey I get an error while compiling shogun2.1 | 12:23 |
n4nd0 | ZeThomas: and the error is? | 12:23 |
ZeThomas | http://pastebin.com/GYeSxjGR | 12:24 |
ZeThomas | lapack | 12:24 |
ZeThomas | i just installed openblas (and removed atlas) | 12:24 |
ZeThomas | and wanted to recompile | 12:24 |
ZeThomas | numpy 1.7 has been recompiled successfully just now | 12:24 |
n4nd0 | mm then maybe it is imcopatible with openblas? | 12:25 |
ZeThomas | I don't know, is it? | 12:25 |
n4nd0 | I didn't know about it, it might be | 12:26 |
ZeThomas | bugger | 12:27 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 12:28 | |
-!- arp [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 12:28 | |
n4nd0 | ZeThomas: excuse me? | 12:28 |
ZeThomas | n4nd0: I should watch my language, sorry | 12:29 |
-!- blackburn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.] | 12:33 | |
-!- foulwall [~foulwall@2001:da8:215:6901:69a5:d6ae:aaf0:6ef1] has joined #shogun | 12:34 | |
-!- blackburn [~lisitsin@mxs.kg.ru] has joined #shogun | 12:36 | |
blackburn | ha! | 12:37 |
-!- heiko1 [~heiko@nat-183-237.internal.eduroam.ucl.ac.uk] has quit [Quit: Leaving.] | 12:42 | |
-!- medeeiip [~medeeiip@103.18.168.49] has quit [Quit: Leaving] | 12:49 | |
-!- abinashpanda [75ef5e6e@gateway/web/freenode/ip.117.239.94.110] has joined #shogun | 12:59 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 13:04 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 13:05 | |
-!- abinashpanda [75ef5e6e@gateway/web/freenode/ip.117.239.94.110] has quit [Ping timeout: 245 seconds] | 13:07 | |
-!- shrey [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has joined #shogun | 13:11 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 13:15 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 13:16 | |
ZeThomas | blackburn: do you know if shogun supports openblas? | 13:18 |
blackburn | ZeThomas: should be possible I think | 13:24 |
ZeThomas | i compiled it, and successfully numpy with it | 13:25 |
ZeThomas | but shogun gives me errors | 13:25 |
ZeThomas | http://pastebin.com/GYeSxjGR | 13:25 |
blackburn | ZeThomas: ehmmm 311 looks strange to me :) | 13:28 |
blackburn | why 'I' is treated as complex | 13:28 |
blackburn | ZeThomas: what is the compiler? | 13:28 |
ZeThomas | gcc | 13:29 |
ZeThomas | 4.7 | 13:30 |
blackburn | ZeThomas: could you please try to put some char('I') or so? | 13:30 |
blackburn | 311 and 347 of src/shogun/mathematics/lapack.cpp | 13:31 |
ZeThomas | so line 311 becomes: char I = char('I') ? | 13:33 |
blackburn | yes | 13:34 |
ZeThomas | ok | 13:34 |
ZeThomas | done, I'm trying to recompile now | 13:36 |
* ZeThomas goes for coffee while shogun compiles | 13:38 | |
blackburn | ZeThomas: you may use make -j4 to use 4 cores | 13:39 |
ZeThomas | oh, that's handy :) | 13:39 |
sonne|work | and ccache! | 13:40 |
blackburn | yes and ccache | 13:40 |
-!- n4nd0_ [~test@n133-p134.kthopen.kth.se] has joined #shogun | 13:42 | |
-!- n4nd0 [~test@n133-p134.kthopen.kth.se] has quit [Quit: Reconnecting] | 13:42 | |
ZeThomas | so "make -j8 -ccache" ? | 13:42 |
ZeThomas | blackburn: still the same error | 13:42 |
blackburn | ZeThomas: no, ccache is a wrapper on top of compiler | 13:44 |
blackburn | not a key | 13:44 |
blackburn | ZeThomas: hmm let me think | 13:44 |
-!- n4nd0_ is now known as n4nd0 | 13:44 | |
n4nd0 | ZeThomas: no, ccache make -j8 | 13:44 |
-!- n4nd0 [~test@n133-p134.kthopen.kth.se] has quit [Client Quit] | 13:44 | |
-!- n4nd0_ [~test@n133-p134.kthopen.kth.se] has joined #shogun | 13:45 | |
-!- n4nd0_ [~test@n133-p134.kthopen.kth.se] has quit [Client Quit] | 13:45 | |
-!- n4nd0 [~test@n133-p134.kthopen.kth.se] has joined #shogun | 13:45 | |
ZeThomas | thanks n4nd0 | 13:45 |
blackburn | n4nd0: ?? would that work? | 13:46 |
sonne|work | IIRC you need to install ccache and put a certain dir in $PATH | 13:46 |
sonne|work | such that it calls ccache when gcc is called | 13:47 |
blackburn | and replace CXX and CC with ccache g++ | 13:47 |
n4nd0 | blackburn: do you normally use it as make --ccache? | 13:47 |
n4nd0 | blackburn: I always do ccache make blah blah | 13:47 |
blackburn | n4nd0: so does that work with ccache make? | 13:47 |
blackburn | bah.. | 13:47 |
n4nd0 | hehe I hope so | 13:47 |
blackburn | sonne|work: any idea why 'I' is treated as complex? | 13:48 |
blackburn | ZeThomas: okay other guess | 13:48 |
blackburn | '\I' maybe? | 13:48 |
sonne|work | what is going on? | 14:03 |
sonne|work | blackburn: my suggestion is to use char range = 'I'; | 14:06 |
sonne|work | then it should work | 14:06 |
sonne|work | blackburn: I think it treats I as a complex nr | 14:06 |
-!- arp [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 14:10 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds] | 14:11 | |
ZeThomas | yes that seems the case, so is this an OpenBLAS specific thing then? | 14:12 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 14:15 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 14:17 | |
ZeThomas | blackburn, sonne|work, this seems to have done the trick: I refactored every occurence of [varname] I to range | 14:17 |
ZeThomas | however it gives errors for SGMatrix.cpp (http://pastebin.com/KV8Yzi9Y), I suspect of the same kind | 14:18 |
sonne|work | SGMatrix<int8_t> I(size, size); | 14:19 |
sonne|work | yes | 14:19 |
sonne|work | could you please call this identity there? | 14:20 |
sonne|work | ZeThomas: and please send us a git pull request against the develop branch | 14:20 |
ZeThomas | sonne|work: I'd be delighted, but you might want to talk me through it, I'm a git noob... | 14:23 |
sonne|work | n4nd0: could you please help ZeThomas? | 14:25 |
n4nd0 | back again | 14:27 |
sonne|work | n4nd0: ^ please help him to set up git if you have time - thanks! | 14:28 |
n4nd0 | sure | 14:28 |
n4nd0 | ZeThomas: so how is it going | 14:28 |
n4nd0 | although I have not tried yet myself git-flow, blackburn or someone else can tell us if we need some help about it though | 14:30 |
ZeThomas | ok n4nd0, I git cloned git://github.com/shogun-toolbox/shogun.git | 14:30 |
blackburn | yeah sure | 14:30 |
ZeThomas | and made the changes | 14:30 |
n4nd0 | ZeThomas: ok then let's start again | 14:31 |
blackburn | ZeThomas: I am afraid you would need to transfer your changs to your fork | 14:31 |
blackburn | ZeThomas: is it ubuntu/debian you are running? | 14:31 |
ZeThomas | blackburn: yes | 14:31 |
n4nd0 | because you fork the main repository and clone that one | 14:31 |
blackburn | ZeThomas: please install git flow (sudo apt-get install git-flow) | 14:31 |
n4nd0 | blackburn: I guess however it can enough enough changing the remotes? | 14:31 |
blackburn | n4nd0: ha good point | 14:31 |
ZeThomas | blackburn: done | 14:32 |
blackburn | alright | 14:33 |
blackburn | now you have to 'git flow init' | 14:33 |
ZeThomas | ok: Branch name for production releases: [master] | 14:34 |
ZeThomas | ? | 14:34 |
n4nd0 | ZeThomas: give me a second | 14:34 |
n4nd0 | this was already explained in the mailing list | 14:34 |
blackburn | ZeThomas: yes that's default | 14:34 |
n4nd0 | I can send you a link | 14:34 |
blackburn | ZeThomas: next release is develop | 14:35 |
ZeThomas | n4nd0: please do | 14:35 |
ZeThomas | sonne|work: it still fails to compile: http://pastebin.com/G25Fyx0g | 14:36 |
n4nd0 | ZeThomas: http://news.gmane.org/gmane.comp.ai.machine-learning.shogun | 14:36 |
n4nd0 | Crtl+f workflow | 14:36 |
n4nd0 | mail by Sergey on April 4th | 14:36 |
sonne|work | ZeThomas: argh - please rename the varialbe I in line 106 to index | 14:37 |
sonne|work | too | 14:37 |
-!- n4nd0 [~test@n133-p134.kthopen.kth.se] has quit [Ping timeout: 246 seconds] | 14:41 | |
ZeThomas | sonne|work: in which file? | 14:42 |
ZeThomas | blackburn: ok git flow is set up | 14:42 |
sonne|work | shogun/structure/libbmrm.h:106 | 14:45 |
sonne|work | I guess there too shogun/lib/external/libqp.h:61 | 14:45 |
-!- votjak [~votjak@host-46-241-3-209.bbcustomer.zsttk.net] has joined #shogun | 14:48 | |
-!- kpn [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 14:48 | |
-!- n4nd0 [~test@n133-p134.kthopen.kth.se] has joined #shogun | 14:49 | |
n4nd0 | got connection troubles | 14:49 |
blackburn | ZeThomas: once it compiles do 'git flow support complex_literals' | 14:49 |
blackburn | complex_literals is the name of branch to be created | 14:49 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 14:55 | |
ZeThomas | blackburn: Unknown subcommand: 'complex_literals' | 14:56 |
ZeThomas | sonne|work: it compiles now, thanks. Is this an OpenBLAS specific thing? | 14:56 |
-!- kpn [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 14:57 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 14:57 | |
n4nd0 | ZeThomas: my guess is that there is a global in OpenBlas called I | 14:57 |
ZeThomas | n4nd0: oh, and in c++, global names take precedence over local ones? or in no case can you have doubles? | 14:58 |
n4nd0 | ZeThomas: with doubles you mean two variables with the same name? | 14:58 |
ZeThomas | n4nd0: yes | 14:58 |
ZeThomas | as you can do in python e.g. | 14:59 |
n4nd0 | I am not sure | 14:59 |
n4nd0 | ZeThomas: what could it be otherwise? | 14:59 |
blackburn | ZeThomas: oops sorry | 14:59 |
blackburn | ZeThomas: git flow support start complex_literals | 14:59 |
ZeThomas | blackburn: Missing argument <base> | 15:00 |
n4nd0 | blackburn: why do you think this error with the I happened? | 15:00 |
blackburn | n4nd0: no idea but your explanation looks valid | 15:01 |
ZeThomas | blackburn: I add develop at the end? | 15:01 |
blackburn | ZeThomas: hmm yeah try that | 15:01 |
blackburn | ZeThomas: sorry we changed our model to git flow may be a week ago | 15:01 |
blackburn | :) | 15:01 |
-!- medeeiip [~medeeiip@103.18.168.49] has joined #shogun | 15:01 | |
ZeThomas | ok: You are now on branch 'support/complex_literals' | 15:01 |
blackburn | ZeThomas: okay so call 'git status' and check the files you made changes in | 15:02 |
blackburn | 'git add [filename]' each of them | 15:02 |
ZeThomas | blackburn: ok, done | 15:07 |
blackburn | ZeThomas: git commit -m 'your message here' | 15:08 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 15:08 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 15:09 | |
ZeThomas | blackburn, ok, what's next? | 15:11 |
blackburn | ZeThomas: tricky part | 15:12 |
blackburn | ZeThomas: git remote rename origin upstream | 15:12 |
blackburn | ZeThomas: https://github.com/shogun-toolbox/shogun press the fork button | 15:13 |
ZeThomas | blackburn: error: Could not rename config section 'remote.origin' to 'remote.upstream' | 15:13 |
blackburn | ZeThomas: hmm okay | 15:14 |
blackburn | ZeThomas: git remote add upstream https://github.com/shogun-toolbox/shogun.git | 15:14 |
-!- foulwall [~foulwall@2001:da8:215:6901:69a5:d6ae:aaf0:6ef1] has quit [Remote host closed the connection] | 15:17 | |
ZeThomas | blackburn: ok | 15:18 |
blackburn | ZeThomas: done? | 15:18 |
ZeThomas | yes | 15:18 |
ZeThomas | so now? | 15:19 |
blackburn | ZeThomas: git remote set-url [ssh url to your fork] | 15:20 |
blackburn | argh | 15:20 |
blackburn | ZeThomas: git remote set-url origin [ssh url to your fork] | 15:21 |
blackburn | ZeThomas: you may get that link on the page of your fork | 15:21 |
ZeThomas | blackburn: fatal: No such remote 'origin' | 15:25 |
ZeThomas | git remote add origin https://github.com/shogun-toolbox/shogun.git ? | 15:26 |
n4nd0 | ZeThomas: origin should point to your fork, not to the main one | 15:27 |
n4nd0 | ZeThomas: it should be something like ...ZeThomas/shogun.git | 15:27 |
n4nd0 | where ZeThomas is your username in github | 15:27 |
ZeThomas | oh, I'm so bad at git... | 15:28 |
ZeThomas | so git remote add origin https://github.com/tp199911/shogun.git ? | 15:28 |
n4nd0 | it is hard at the beginning, I suggest you to check gitbook and some videocasts | 15:28 |
n4nd0 | ZeThomas: did you already fork in github? | 15:29 |
ZeThomas | yes, that link is my fork | 15:29 |
blackburn | ZeThomas: better use ssh url | 15:29 |
blackburn | however https would work too | 15:29 |
n4nd0 | blackburn: faster with ssh? | 15:30 |
blackburn | n4nd0: requires no authorization | 15:30 |
ZeThomas | ok, so I do: git remote add origin git@github.com:tp199911/shogun.git ? | 15:30 |
blackburn | ZeThomas: yes | 15:30 |
n4nd0 | blackburn: oh yeah, sure | 15:30 |
ZeThomas | blackburn: ok, that's done | 15:31 |
blackburn | ZeThomas: ok near to ready! | 15:31 |
blackburn | ZeThomas: try to do git push origin | 15:31 |
ZeThomas | Permission denied (publickey). fatal: The remote end hung up unexpectedly... | 15:32 |
n4nd0 | ZeThomas: you have to put your publich ssh key in github I think | 15:33 |
blackburn | ZeThomas: just to check could you please git remote -v? | 15:34 |
n4nd0 | https://help.github.com/articles/generating-ssh-keys | 15:34 |
blackburn | and paste the output | 15:34 |
ZeThomas | http://pastebin.com/yQ7Mae73 | 15:35 |
blackburn | ZeThomas: alright | 15:36 |
blackburn | ZeThomas: please follow the link n4nd0 put | 15:36 |
blackburn | and set up a public key | 15:36 |
ZeThomas | I'm on it | 15:36 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 15:37 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 15:38 | |
ZeThomas | blackburn: I get this after the push | 15:39 |
ZeThomas | http://pastebin.com/w4cRiMri | 15:39 |
blackburn | ZeThomas: alright | 15:40 |
blackburn | no problem | 15:40 |
blackburn | ZeThomas: git checkout master | 15:40 |
blackburn | git pull --rebase origin master | 15:40 |
blackburn | git checkout develop | 15:40 |
blackburn | git pull --rebase origin develop | 15:40 |
blackburn | git checkout support/complex_literals | 15:40 |
blackburn | git rebase develop | 15:41 |
blackburn | git push origin support/complex_literals | 15:41 |
blackburn | something like this I think | 15:41 |
ZeThomas | after git pull origin master I get: Aborting\n could not detach HEAD | 15:42 |
-!- gaurang [~quassel@triband-mum-59.182.146.169.mtnl.net.in] has joined #shogun | 15:43 | |
blackburn | ZeThomas: what is the branch you are on? | 15:46 |
blackburn | git status | 15:46 |
ZeThomas | on branch master | 15:47 |
ZeThomas | the files I added have disappeared though :D | 15:49 |
blackburn | ZeThomas: you commited them right? | 15:49 |
ZeThomas | yes | 15:50 |
ZeThomas | well, I guess | 15:50 |
blackburn | so that's ok they are on a different branch | 15:50 |
-!- foulwall [~foulwall@2001:da8:215:6901:d1f1:befd:256d:419] has joined #shogun | 15:51 | |
blackburn | ZeThomas: I am confused with thar error | 15:51 |
blackburn | that* | 15:51 |
ZeThomas | blackburn: you and I both... | 15:52 |
blackburn | ZeThomas: okay lets get back to that support/complex_literals branch | 15:52 |
blackburn | git checkout complex_literals | 15:52 |
blackburn | git flow support finish complex_literals | 15:53 |
-!- foulwall [~foulwall@2001:da8:215:6901:d1f1:befd:256d:419] has quit [Ping timeout: 245 seconds] | 15:55 | |
ZeThomas | Unknown subcommand: 'finish' | 15:56 |
blackburn | ZeThomas: argghhhh | 15:57 |
blackburn | my bad | 15:57 |
blackburn | ZeThomas: git branch -m support/complex_literals feature/complex_literals | 15:58 |
-!- yefuneh [3edb8730@gateway/web/freenode/ip.62.219.135.48] has joined #shogun | 15:59 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 15:59 | |
yefuneh | Can someone point me to a **simple** example using shogun in lua? | 16:00 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:00 | |
blackburn | yefuneh: https://github.com/shogun-toolbox/shogun/blob/master/examples/undocumented/lua_modular/classifier_libsvm_minimal_modular.lua something like that | 16:00 |
ZeThomas | blackburn: done | 16:00 |
yefuneh | no, too complicated | 16:01 |
blackburn | ZeThomas: git feature finish feature/complex_literals | 16:01 |
-!- gsomix [~Miranda@83.234.169.3] has quit [Read error: Connection reset by peer] | 16:02 | |
ZeThomas | blackburn: I suppose you mean git flow feature finish feature/complex_literals | 16:02 |
ZeThomas | blackburn: No branch matches prefix 'feature/complex_literals' | 16:02 |
yefuneh | blackburn when looking at svmlight for example, you just run svm_learn and svm_classify, is there something that just shows how to take a training file create a model and then run that model against the test data? | 16:03 |
blackburn | ZeThomas: ohhhh | 16:03 |
blackburn | ZeThomas: was it renamed? | 16:03 |
blackburn | yefuneh: no probably there is no such example | 16:04 |
n4nd0 | yefuneh: https://github.com/shogun-toolbox/shogun/blob/master/examples/undocumented/lua_modular/classifier_averaged_perceptron_modular.lua | 16:04 |
n4nd0 | yefuneh: this looks simple enough | 16:04 |
-!- heiko1 [~heiko@nat-184-136.internal.eduroam.ucl.ac.uk] has joined #shogun | 16:05 | |
blackburn | n4nd0: could you please back me up with ZeThomas? have to get to work | 16:05 |
n4nd0 | blackburn: sure | 16:05 |
n4nd0 | blackburn: but not idea about git flow yet :) | 16:06 |
blackburn | n4nd0: it is just about branches I think | 16:07 |
yefuneh | n4nd0 - thanks. What is load_labels for? My test and training data are in a format like: -1 1:5 2:7 ... where -1 is the label | 16:07 |
n4nd0 | yefuneh: wild guess - load labels is to take the labels from a file and put them in a variable | 16:09 |
yefuneh | n4nd0 thanks | 16:10 |
n4nd0 | yefuneh: if you labels are in that format you can either code your own functions to parse the data in a similar way to how it is done in the load.lua script | 16:10 |
n4nd0 | yefuneh: or you can create new data files following the convention of the files used in the example and re-use the load functions | 16:10 |
ZeThomas | n4nd0, me to I have to get to work... can we do this some other time? | 16:11 |
n4nd0 | ZeThomas: sure | 16:11 |
n4nd0 | just come around some time | 16:11 |
ZeThomas | ok, cheers | 16:11 |
ZeThomas | cu, thanks for the help | 16:11 |
yefuneh | n4nd0 - I'll have to look at the example files. Based on the examples it seems much more complicated then running svmlight/libsvm directly on the data | 16:11 |
-!- ZeThomas [~thomaspr@2a02:2c40:100:a000:a81c:41e5:277a:c326] has quit [Quit: Leaving] | 16:11 | |
-!- phd [~lvoronova@77-120-201-195.sumy.volia.net] has quit [Ping timeout: 252 seconds] | 16:13 | |
-!- frl [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 16:15 | |
-!- phd [~lvoronova@77-120-201-195.sumy.volia.net] has joined #shogun | 16:15 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has joined #shogun | 16:17 | |
-!- yefuneh [3edb8730@gateway/web/freenode/ip.62.219.135.48] has quit [Quit: Page closed] | 16:17 | |
-!- frl_ [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 16:17 | |
-!- n4nd0 [~test@n133-p134.kthopen.kth.se] has quit [Ping timeout: 255 seconds] | 16:18 | |
frl_ | in the issue "Kernel PCA test/check" . we have to implement th ekernel pca algorithm in python without using the shogun libraries rite? | 16:18 |
frl_ | m i correct? | 16:18 |
frl_ | ? | 16:18 |
frl_ | ? | 16:19 |
-!- frl [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 16:19 | |
frl_ | ne suggestions? | 16:19 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:19 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:20 | |
frl_ | ? | 16:21 |
frl_ | ? | 16:21 |
heiko1 | frl_: see email | 16:27 |
-!- ppletscher [~ppletsche@84-75-165-165.dclient.hispeed.ch] has quit [Quit: ppletscher] | 16:36 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:39 | |
-!- ppletscher [~ppletsche@84-75-165-165.dclient.hispeed.ch] has joined #shogun | 16:40 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:40 | |
-!- medeeiip__ [~medeeiip@103.18.168.49] has joined #shogun | 16:47 | |
-!- medeeiip [~medeeiip@103.18.168.49] has quit [Ping timeout: 245 seconds] | 16:49 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has quit [Remote host closed the connection] | 16:49 | |
-!- ppletscher [~ppletsche@84-75-165-165.dclient.hispeed.ch] has quit [Quit: ppletscher] | 16:56 | |
@sonney2k | heiko1, https://docs.google.com/spreadsheet/ccc?key=0Aunb9cCVAP6NdDVBMzY1TjdPcmx4ei1EeUZNNGtKUHc&usp=sharing | 17:03 |
@sonney2k | blackburn, ^ | 17:03 |
@sonney2k | I am just creating some python script to generate a page from that which we can put on the homepage | 17:03 |
heiko1 | sonney2k: ok | 17:04 |
heiko1 | I dont find that overly usefull though | 17:04 |
heiko1 | I think we should rather have a list of what we can do, which is a bit more detailed | 17:04 |
heiko1 | like "features" | 17:04 |
heiko1 | since this matrix oversimplifies somewhat | 17:04 |
frl_ | what does "integration test for kernel pca means? | 17:04 |
-!- frl_ [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Quit: Page closed] | 17:05 | |
blackburn | ha | 17:05 |
blackburn | he left no chance to answer | 17:05 |
heiko1 | frl_: this saves the results of the example to a file once and then everytime the example is run it is compared against the old result. This way we can avoid that algorithms change wihtout us noticing | 17:05 |
-!- kipp [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 17:05 | |
heiko1 | going for coffee .... | 17:06 |
@sonney2k | heiko1, blackburn - here it is https://dl.dropbox.com/u/31480471/matrix.png | 17:06 |
kipp | what does integration test for kernal pca mean?does it means that the kernel matrix is valid? | 17:07 |
kipp | m i correct? | 17:07 |
blackburn | kipp: what is a valid kernel matrix? | 17:08 |
blackburn | kipp: that was for your probably? | 17:10 |
blackburn | (7:05:25 PM) heiko1: frl_: this saves the results of the example to a file once and then everytime the example is run it is compared against the old result. This way we can avoid that algorithms change wihtout us noticing | 17:10 |
kipp | blackburn: kernel matrix should be postive definate | 17:10 |
blackburn | kipp: yeah but it would be a test for kernel not pca, right? ;) | 17:10 |
kipp | yes ,yes.sorry typed pca by mistake | 17:11 |
blackburn | kipp: integration tests are usually checking if results are staying the same | 17:11 |
blackburn | so it would be just project something and check if it is the same as it was before | 17:11 |
kipp | so its like a accuracy checker | 17:11 |
blackburn | kipp: some sort of | 17:12 |
kipp | k,so its like checking ,if the data has become linearly separable or not | 17:17 |
blackburn | kipp: no the data stays so the output should stay too | 17:17 |
kipp | as in the output data is same as the input data after kernel pca | 17:18 |
blackburn | kipp: no as output data is the same after some code is changed | 17:20 |
blackburn | naywhayare: how many SFINAE experts so far? ;) | 17:20 |
-!- kipp [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 17:29 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 17:29 | |
-!- phd [~lvoronova@77-120-201-195.sumy.volia.net] has quit [Ping timeout: 264 seconds] | 17:29 | |
-!- shrey [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has quit [Ping timeout: 245 seconds] | 17:30 | |
-!- Yanglittle [deb20af1@gateway/web/freenode/ip.222.178.10.241] has quit [Quit: Page closed] | 17:30 | |
-!- shrey [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has joined #shogun | 17:34 | |
-!- blackburn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.] | 17:36 | |
-!- shrey [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has quit [Ping timeout: 245 seconds] | 17:41 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 17:42 | |
-!- gaurang [~quassel@triband-mum-59.182.146.169.mtnl.net.in] has quit [Read error: Connection reset by peer] | 17:42 | |
-!- phd [lvoronova1@77-120-201-195.sumy.volia.net] has joined #shogun | 17:47 | |
naywhayare | blackburn: no SFINAE experts but a large number of first-year undergraduates seem interested in the "9/10 difficulty" (as I listed it) project on improving the abstract tree traversals mlpack has... | 18:03 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:03 | |
naywhayare | and when I say "no, really, I promise, enjoy your summer, you can't learn this stuff in two weeks", they seem to end up even more interested | 18:03 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:05 | |
-!- phd [lvoronova1@77-120-201-195.sumy.volia.net] has quit [Ping timeout: 260 seconds] | 18:08 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:13 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:15 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:17 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:19 | |
-!- phd [~lvoronova@77-120-201-195.sumy.volia.net] has joined #shogun | 18:19 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:21 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:23 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 18:24 | |
deerishi | i know Kernal PCA ,but I am still unclear about what do we need to do the kernalpca issue https://github.com/shogun-toolbox/shogun/issues/940 | 18:25 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:25 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:27 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:29 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:31 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 18:31 | |
heiko1 | deerishi: the point is to write a nice example to illustrate how the algorithm works | 18:32 |
heiko1 | some easy data, then you do your kernel PCA, and see how it changes the data | 18:32 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:34 | |
deerishi | heikol: thank you | 18:34 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:35 | |
-!- shaumik [d2d43ae7@gateway/web/freenode/ip.210.212.58.231] has joined #shogun | 18:36 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:38 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:39 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:42 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:43 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 18:44 | |
-!- shaumik [d2d43ae7@gateway/web/freenode/ip.210.212.58.231] has quit [Quit: Page closed] | 18:45 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:46 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:47 | |
-!- blackburn [~blackburn@188.168.2.135] has joined #shogun | 18:47 | |
blackburn | heiko1: | 18:49 |
blackburn | one thing on kpca | 18:49 |
heiko1 | blackburn: | 18:49 |
blackburn | heiko1: tapkee supports kpca, may be it makes sense to change the code to use it | 18:50 |
heiko1 | blackburn: yes that might be good | 18:50 |
heiko1 | could you update the task? | 18:50 |
blackburn | heiko1: yeah will do | 18:50 |
heiko1 | or maybe add another one for that | 18:50 |
heiko1 | since the example doesnt depend on the backend | 18:50 |
blackburn | yes | 18:50 |
blackburn | that's better | 18:50 |
heiko1 | cool | 18:50 |
blackburn | naywhayare: I see | 18:51 |
heiko1 | blackburn: could we go into the git issue again? :) | 18:52 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:52 | |
blackburn | heiko1: what is the issue? | 18:52 |
heiko1 | still no idea how to commit ;) | 18:52 |
heiko1 | git flow | 18:52 |
heiko1 | remember? | 18:52 |
blackburn | oh shh | 18:52 |
blackburn | sure | 18:52 |
blackburn | heiko1: could you please remind me the status we got into? | 18:52 |
heiko1 | git flow init | 18:53 |
heiko1 | Which branch should be used for bringing forth production releases? | 18:53 |
heiko1 | - master | 18:53 |
heiko1 | Branch name for production releases: [master] | 18:53 |
heiko1 | I cannot select development | 18:53 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:53 | |
blackburn | heiko1: yes because it is not here | 18:53 |
heiko1 | exactly | 18:53 |
blackburn | probably | 18:53 |
blackburn | heiko1: is it your fork? | 18:54 |
heiko1 | yes | 18:54 |
blackburn | heiko1: lets try | 18:54 |
blackburn | git checkout -b develop upstream/develop | 18:54 |
heiko1 | fatal: git checkout: updating paths is incompatible with switching branches. | 18:55 |
heiko1 | Did you intend to checkout 'upstream/develop' which can not be resolved as commit? | 18:55 |
blackburn | hmm | 18:55 |
blackburn | git branch develop upstream/develop? | 18:55 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:56 | |
blackburn | heiko1: wait | 18:56 |
blackburn | heiko1: lets do that | 18:56 |
blackburn | git fetch upstream | 18:56 |
blackburn | it should load all branches | 18:56 |
heiko1 | argh | 18:56 |
heiko1 | yes | 18:56 |
heiko1 | of course | 18:56 |
blackburn | heiko1: but you'd still need to create your branch too | 18:56 |
heiko1 | remember when you said git pull --rebase includes git fetch upstream? | 18:57 |
blackburn | heiko1: yes | 18:57 |
blackburn | pull is fetch + merge | 18:57 |
blackburn | or fetch + rebase | 18:57 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:57 | |
heiko1 | okay hzow to fetch the branch now? | 18:57 |
blackburn | heiko1: does git branch -va | 18:57 |
blackburn | show you remote branch | 18:57 |
blackburn | of develop? | 18:57 |
heiko1 | yes | 18:58 |
blackburn | heiko1: alright | 18:58 |
blackburn | heiko1: just create develop | 18:58 |
blackburn | ahh | 18:58 |
blackburn | yes | 18:58 |
blackburn | :D | 18:58 |
blackburn | heiko1: I think you need to create a branch based on upstream/develop | 18:59 |
heiko1 | is there a difference between dev and develop? | 18:59 |
-!- kakashi_ [~kakashi_@nltk/kakashi] has quit [Ping timeout: 252 seconds] | 18:59 | |
blackburn | heiko1: dev should be deleted | 18:59 |
heiko1 | ok | 18:59 |
blackburn | heiko1: it is deleted but git is distributed so it stayed in your copy | 19:00 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:00 | |
heiko1 | blackburn: ok so now I created develop | 19:00 |
blackburn | alright | 19:00 |
blackburn | heiko1: git push origin develop | 19:00 |
heiko1 | how to update it? | 19:00 |
blackburn | git rebase upstream/develop | 19:00 |
heiko1 | ok pushed | 19:01 |
-!- kakashi_ [~kakashi_@nltk/kakashi] has joined #shogun | 19:01 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:01 | |
heiko1 | and now? | 19:01 |
blackburn | heiko1: git flow init | 19:01 |
heiko1 | blackburn: how to update the develop branch? | 19:01 |
blackburn | heiko1: git gets easier once you get the basic idea that branches are homeomorphic endofunctors mapping submanifolds.. | 19:02 |
blackburn | :D | 19:02 |
blackburn | heiko1: what do you mean by update? | 19:02 |
heiko1 | haha :) | 19:02 |
heiko1 | nevermind | 19:02 |
heiko1 | Which branch should be used for integration of the "next release"? | 19:02 |
blackburn | next release is develop | 19:02 |
heiko1 | and production release? | 19:02 |
heiko1 | Ill check your mail | 19:02 |
blackburn | heiko1: master | 19:03 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:04 | |
blackburn | heiko1: everything is default | 19:05 |
heiko1 | ok worked | 19:05 |
blackburn | except versiontag | 19:05 |
heiko1 | so now I modify develop right? | 19:05 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:05 | |
blackburn | nope | 19:05 |
blackburn | now you | 19:05 |
blackburn | git flow feature start my_awesome_feature | 19:05 |
blackburn | git commit ... | 19:05 |
heiko1 | ok | 19:05 |
blackburn | git flow feature finish my_awesome_feature | 19:05 |
heiko1 | what does this? merge it into develop | 19:06 |
heiko1 | ? | 19:06 |
blackburn | heiko1: finish merges and deletes the branch | 19:06 |
heiko1 | I see | 19:06 |
heiko1 | and then I push develop? | 19:06 |
blackburn | yes | 19:06 |
heiko1 | whats the command for that? | 19:06 |
heiko1 | just push? | 19:06 |
blackburn | heiko1: but actually I have absolutely no idea how to handle that better | 19:06 |
blackburn | heiko1: yes | 19:06 |
blackburn | heiko1: I'd suggest you to finally commit in the main repository | 19:06 |
heiko1 | at which point? | 19:07 |
blackburn | PRs are additional headache here | 19:07 |
heiko1 | what do you mean by that? | 19:07 |
heiko1 | a I see | 19:07 |
heiko1 | no Ill do PR | 19:07 |
heiko1 | better for me | 19:07 |
blackburn | heiko1: just merge it directly | 19:07 |
heiko1 | I tend to mess up repos | 19:07 |
blackburn | why? | 19:07 |
heiko1 | fucked up mine about 100 times | 19:07 |
blackburn | hmm | 19:07 |
heiko1 | so safer for me to work on the fork | 19:07 |
heiko1 | also we have a history | 19:07 |
heiko1 | think thats better | 19:07 |
heiko1 | and travis | 19:08 |
heiko1 | and I can see the diff in the PR view | 19:08 |
blackburn | heiko1: yeah that has pros and conses | 19:08 |
heiko1 | I know, but Id rather stay on the safe side here | 19:08 |
heiko1 | thanks for the help! | 19:08 |
blackburn | heiko1: but actually it would be better so consider moving some day ;) | 19:08 |
blackburn | heiko1: the thing is that we do group commits this way | 19:09 |
blackburn | so develop could look like a bunch of merges | 19:09 |
heiko1 | thats not bad right? | 19:09 |
blackburn | ideally it looks like | 19:09 |
blackburn | merge feature a | 19:09 |
blackburn | merge feature b | 19:09 |
blackburn | master is even more compressive | 19:09 |
blackburn | it contains only tags | 19:10 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:10 | |
blackburn | heiko1: homeomorphic endofunctors! keep that in mind | 19:10 |
heiko1 | blackburn: yeah, whatever :D | 19:10 |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 19:10 | |
blackburn | heiko1: git flow stuff is just a script | 19:11 |
blackburn | that creates branches and deletes it | 19:11 |
heiko1 | I know | 19:11 |
blackburn | heiko1: just hides a few commands | 19:11 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:11 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:14 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:15 | |
-!- votjak [~votjak@host-46-241-3-209.bbcustomer.zsttk.net] has quit [Quit: Leaving] | 19:16 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 19:17 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:18 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:19 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:24 | |
-!- bogdanc [~bogdan@p16.eregie.pub.ro] has joined #shogun | 19:25 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:25 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:28 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:29 | |
-!- ppletscher [~ppletsche@84-75-165-165.dclient.hispeed.ch] has joined #shogun | 19:30 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Read error: Connection reset by peer] | 19:34 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:35 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:38 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:39 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:43 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:44 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:48 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:49 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:52 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:53 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 19:55 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:56 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:57 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:02 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 20:04 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:07 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:11 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:12 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:16 | |
-!- aegis1 [aegis1@123.237.52.165] has joined #shogun | 20:17 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:17 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:21 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:22 | |
-!- gsomix [~Miranda@188.168.2.135] has joined #shogun | 20:25 | |
gsomix | hi | 20:25 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:25 | |
n4nd0 | hey gsomix | 20:26 |
n4nd0 | what is your project search going? | 20:26 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:26 | |
n4nd0 | how* | 20:27 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:30 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:31 | |
gsomix | n4nd0: in progress... any ideas? I think I can implement typemaps for model selection (ought to have done a long time ago >__<), but it's not enough. | 20:33 |
-!- ppletscher [~ppletsche@84-75-165-165.dclient.hispeed.ch] has quit [Quit: ppletscher] | 20:33 | |
gsomix | sonney2k: there? good evening | 20:34 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 264 seconds] | 20:35 | |
-!- aegis1 [aegis1@123.237.52.165] has quit [Ping timeout: 255 seconds] | 20:35 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:36 | |
@sonney2k | gsomix, hey there | 20:37 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:38 | |
gsomix | sonney2k: how are you? I want to continue the last conversation. what's wrong with R modular interface now? | 20:38 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:40 | |
-!- phd [~lvoronova@77-120-201-195.sumy.volia.net] has quit [Ping timeout: 264 seconds] | 20:41 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:42 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:46 | |
@sonney2k | gsomix, it is crashing | 20:46 |
@sonney2k | gsomix, IDK if you recall | 20:47 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:47 | |
@sonney2k | but whenever some object is transferred from C++ -> python | 20:47 |
@sonney2k | python needs to keep somehow track of the object and potentially destroy / deallocate it | 20:48 |
@sonney2k | this is done via reference counts | 20:48 |
@sonney2k | so we register 2 functions | 20:48 |
@sonney2k | called ref() | 20:48 |
@sonney2k | and unref() | 20:48 |
@sonney2k | that take care of this from python | 20:48 |
@sonney2k | for R modular these functions seem not to work | 20:48 |
@sonney2k | at least we just get crashes when these are enabled | 20:49 |
gsomix | sonney2k: got it | 20:50 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:53 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:54 | |
gsomix | sonney2k: are there any else interfaces related issues? model selection, R, lua, static interfaces... hm? | 20:56 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:58 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:00 | |
@sonney2k | gsomix, yeah well blackburn had this nice model selection idea | 21:02 |
gsomix | sonney2k: already know :) | 21:03 |
@sonney2k | gsomix, it is not a fulltime project though but just something to warm up LD | 21:03 |
@sonney2k | :) | 21:03 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:03 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:05 | |
@sonney2k | gsomix, otherwise interface wise we have some even harder problem - our wrapper code size is to big. we would need some way of splitting up the swig generated wrapper | 21:05 |
-!- heiko1 [~heiko@nat-184-136.internal.eduroam.ucl.ac.uk] has left #shogun [] | 21:05 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 21:08 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:10 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:11 | |
gsomix | sonney2k: not clear. should this give a profit (less time to recompile or so)? | 21:12 |
blackburn | gsomix: sonney2k: I will try to come up with idea on all these generic programming things tomorrow | 21:13 |
@sonney2k | gsomix, yes drastically reduced compile time | 21:16 |
@sonney2k | gsomix, and much less memory requirements | 21:16 |
blackburn | https://github.com/shogun-toolbox/shogun/issues/969 we have standartizer! | 21:16 |
@sonney2k | instead of 3-4GB just few 100MB | 21:16 |
-!- n4nd0_ [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 21:17 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 276 seconds] | 21:17 | |
-!- n4nd0_ is now known as n4nd0 | 21:17 | |
blackburn | sonney2k: hmm he is going to standartize all libraries on github! | 21:17 |
@sonney2k | n4nd0, I managed to https://docs.google.com/spreadsheet/ccc?key=0Aunb9cCVAP6NdDVBMzY1TjdPcmx4ei1EeUZNNGtKUHc&hl=en#gid=0 | 21:19 |
@sonney2k | convert this to html with some python script | 21:19 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:19 | |
n4nd0 | sonney2k: I saw the matrix before, it looks good I think | 21:19 |
@sonney2k | https://dl.dropbox.com/u/31480471/matrix.png | 21:19 |
gsomix | sonney2k: ok, thanks a lot. there is a lot to think about | 21:19 |
@sonney2k | n4nd0, so where do we put it | 21:20 |
n4nd0 | sonney2k: let met see | 21:20 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:20 | |
n4nd0 | sonney2k: somewhere inside about, what do you think? | 21:20 |
n4nd0 | isn't already something similar on the page? about > related projects | 21:21 |
n4nd0 | what about there? | 21:21 |
n4nd0 | maybe a subpage | 21:21 |
n4nd0 | sub-subpage | 21:21 |
blackburn | sonney2k: I am surprised with that guy - he opens issues in various repos on that | 21:22 |
blackburn | I just do not really get the motivation beneath | 21:22 |
@sonney2k | blackburn, yeah but it is cool | 21:23 |
@sonney2k | staying within standard (to the extent possible) will keep us out of trouble | 21:23 |
blackburn | sonney2k: yeah and out of native compiling on window$ | 21:24 |
@sonney2k | n4nd0, it is not just related projects though but also feature comparison | 21:24 |
blackburn | :D | 21:24 |
@sonney2k | blackburn, why? | 21:24 |
blackburn | sonney2k: VS heavily violates everything I know | 21:24 |
@sonney2k | ahh that | 21:25 |
n4nd0 | sonney2k: btw why does it appear shogun last updates 2010?? | 21:25 |
@sonney2k | n4nd0, because that is when we updated the .csv | 21:25 |
@sonney2k | err google doc | 21:25 |
@sonney2k | blackburn, I don't like having a UUID in there though | 21:26 |
blackburn | sonney2k: UUID is ok but ugly | 21:26 |
@sonney2k | then everything should be autogenerated | 21:26 |
@sonney2k | yeah because of ugliness | 21:26 |
blackburn | sonney2k: I see no actual reason to do that | 21:26 |
n4nd0 | sonney2k: I would go for either about > information > feature comparison or about > related projects > feature comparison | 21:27 |
blackburn | chances STL get a header named SG_CUSTOM_KERNEL_H_ are ehmm low | 21:27 |
n4nd0 | sonney2k: I think it would be nice though to have a link to it or sth where one can see part of the matrix png from the homepage | 21:27 |
@sonney2k | n4nd0, agreed | 21:28 |
@sonney2k | but it needs updating too | 21:28 |
n4nd0 | sonney2k: what needs updating? | 21:28 |
@sonney2k | n4nd0, that was pre shogun 1.0 IIRC | 21:28 |
@sonney2k | we have like 100'd more kernels | 21:28 |
@sonney2k | and so might have other projects | 21:28 |
@sonney2k | and other features of course | 21:29 |
n4nd0 | aaah ok I thought the google docs you have sent around today was up-to-date | 21:29 |
-!- shrey [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has joined #shogun | 21:29 | |
@sonney2k | blackburn, well let him do the work no? | 21:29 |
blackburn | sonney2k: he is not going to do that I think | 21:30 |
blackburn | be back later | 21:30 |
-!- phd [~lvoronova@77-120-201-195.sumy.volia.net] has joined #shogun | 21:49 | |
gsomix | time to install linux | 21:50 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 21:51 | |
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has quit [Ping timeout: 252 seconds] | 21:51 | |
-!- gsomix [~Miranda@188.168.2.135] has quit [Read error: Connection reset by peer] | 21:58 | |
-!- medeeiip [~medeeiip@103.18.168.49] has joined #shogun | 22:44 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Ping timeout: 245 seconds] | 22:45 | |
-!- medeeiip__ [~medeeiip@103.18.168.49] has quit [Ping timeout: 252 seconds] | 22:46 | |
-!- bogdanc [~bogdan@p16.eregie.pub.ro] has quit [Quit: Konversation terminated!] | 22:55 | |
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has joined #shogun | 22:56 | |
-!- ahcorde [553324b1@gateway/web/freenode/ip.85.51.36.177] has joined #shogun | 23:16 | |
shrey | Hello everyone. | 23:20 |
shrey | I am new to machine learning.Can someone point me in the right direction so that i can start working on the binary classification problem?links to reading materials/demos would be helpful. | 23:21 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 23:22 | |
--- Log closed Thu Apr 11 00:00:27 2013 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!