--- Log opened Thu Mar 01 00:00:19 2012 | ||
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 00:22 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 00:33 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 00:39 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 01:24 | |
-!- delcoder [~ishaan@101.63.94.119] has joined #shogun | 02:34 | |
-!- delcoder [~ishaan@101.63.94.119] has quit [Quit: Leaving] | 04:22 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 09:01 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 09:57 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 10:08 | |
-!- bais [~bais@n136-p64.kthopen.kth.se] has joined #shogun | 10:57 | |
-!- bais [~bais@n136-p64.kthopen.kth.se] has quit [Client Quit] | 10:58 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 10:59 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 11:16 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Remote host closed the connection] | 11:19 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 11:20 | |
-!- bais [~bais@n136-p64.kthopen.kth.se] has joined #shogun | 11:23 | |
bais | I have just found the CustomKernel class, that should help! | 11:24 |
---|---|---|
-!- bais [~bais@n136-p64.kthopen.kth.se] has quit [Client Quit] | 11:24 | |
-!- Karl-London [57718696@gateway/web/freenode/ip.87.113.134.150] has joined #shogun | 11:40 | |
Karl-London | Hi. I'm a Shogun beginner and I'm trying to run some examples in Octave (modular). | 11:41 |
Karl-London | The problem is, the data directory is missing. Could someone please tell me where the data directory is? Or send me the data? | 11:42 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 11:46 | |
Karl-London | Could someone please send me the data file: label_train_twoclass.dat, fm_train_real.dat and fm_test_real.dat ? | 11:51 |
Karl-London | My email is: ksh309@gmail.com Many thanks for your help! | 11:52 |
n4nd0 | Karl-London: hey! | 11:59 |
n4nd0 | Karl-London: you can download them from the github repo | 11:59 |
n4nd0 | Karl-London: you can find the files you want here https://github.com/shogun-toolbox/shogun-data/tree/master/toy | 12:00 |
n4nd0 | Karl-London: but you may be interested also in getting a local clone of the repo :) | 12:01 |
Karl-London | Got it! That's great. Thank you very much for your help! | 12:04 |
n4nd0 | Karl-London: no problem, you are welcome | 12:04 |
-!- Karl-London [57718696@gateway/web/freenode/ip.87.113.134.150] has quit [Quit: Page closed] | 12:08 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 12:53 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 13:02 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Remote host closed the connection] | 13:17 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 13:17 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 13:19 | |
-!- bais [~bais@n136-p64.kthopen.kth.se] has joined #shogun | 13:51 | |
bais | hi | 13:52 |
bais | i'm having a problem using SVMs in the octave modular interface | 13:52 |
bais | when I try executing the train() method it simply prints "sorry not yet implemented" | 13:55 |
bais | The question at this point might be stupid, but does this mean there is not way of using SVMs with octave right now?! | 13:55 |
bais | should i maybe be using the LibSVM class? | 14:01 |
bais | it seems like the SVM class is just an interface to actual classes like LibSVM.. | 14:16 |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Read error: Operation timed out] | 14:35 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 14:36 | |
sonne|work | bais: yes SVM is just the baseclass | 15:09 |
sonne|work | use any of its child-classes http://www.shogun-toolbox.org/doc/en/current/classshogun_1_1CSVM.html | 15:09 |
sonne|work | wiking: have a look at the updated ideas list - vojtech's libqp has some more detailled descr. now | 15:15 |
wiking | sonne|work: hey! cool i'll check i've had one email exchange with vojtech. he said that he's going to be pretty busy so i'll have to be independent but that's fine by me | 15:16 |
sonne|work | he has a great plan now I think | 15:18 |
sonne|work | just look at it (I recall that this and the latent svm where of interest to you) | 15:19 |
wiking | yep | 15:19 |
wiking | i think | 15:20 |
wiking | although it's a bit much but there could be a great way to do them both | 15:20 |
wiking | i mean creating nice interface for libqp within shogun | 15:20 |
wiking | and use libqp for the latent svm | 15:20 |
wiking | that'd be great | 15:20 |
wiking | since now some of the latent svm 'implementations' use mosek | 15:21 |
wiking | hmm that reminds me i should clean up my latest patch and submit a pull request ;P | 15:30 |
bais | sonne|work: yes thanks I am finally making my way through the documentation in a more ordered way :P Although, I cannot find some form of description of the various classes, like what would be the difference between using LightSVM and SVMLib, and such.. | 15:36 |
sonne|work | yeah that is missing | 15:37 |
sonne|work | ^ bais | 15:37 |
sonne|work | some more documentation is certainly needed in particular when people don't want to google 'libsvm' 'svmlight' etc | 15:37 |
sonne|work | wiking: well it depends if you intend to sleep | 15:38 |
sonne|work | and of course if you want to stick around after gsoc ... in the end gsoc is just 3 months | 15:38 |
wiking | sonne|work: hahaha well it's part of my phd | 15:39 |
wiking | soooo | 15:39 |
wiking | i mean i'm not doing this for gsoc | 15:39 |
wiking | it's for myself and my phd | 15:39 |
sonne|work | well then just start the work :D | 15:39 |
wiking | yep | 15:39 |
wiking | i'd just like to talk about it with somebody | 15:40 |
wiking | so that you guys get something in a form you'd like to see it | 15:40 |
wiking | so everybody wins :P | 15:40 |
sonne|work | I havent read the latentsvm papers but I can very well imagine what it does and that the resulting problem is non-convex | 15:41 |
wiking | yep | 15:41 |
sonne|work | but in the end it is just an SVM :) | 15:41 |
wiking | currently that's why i'm trying out other optimizers | 15:41 |
wiking | but yep.. essentially it's an sum :) | 15:42 |
wiking | *svm | 15:42 |
sonne|work | wiking: what about http://opencv.itseez.com/trunk/modules/objdetect/doc/latent_svm.html | 15:42 |
wiking | lesse | 15:43 |
wiking | yeah | 15:43 |
wiking | i know the opencv based one | 15:43 |
wiking | it's very very specific | 15:43 |
sonne|work | wiking: do you want to do that for linear svms? | 15:47 |
sonne|work | I mean linear latent svm | 15:47 |
sonne|work | (I just had a look at the paper) | 15:47 |
wiking | well afaik the latent svm that joachim's student did was linear... he haven't made the kernel extension of it... | 15:48 |
sonne|work | kernel would be too slow | 15:48 |
sonne|work | so then we need hog and lbp features and then live is good :) | 15:48 |
bais | sonne|work: oh ok i didnt figure out they were external references to the project, i thought they were internal stuff like "KernelMachine" or such | 15:51 |
sonne|work | bais: yeah we use other people's code if we can... | 15:52 |
bais | ok well, thanks | 15:53 |
-!- bais [~bais@n136-p64.kthopen.kth.se] has quit [Quit: leaving] | 15:53 | |
sonne|work | wiking: certainly quite some effort to get things to work... | 15:54 |
sonne|work | can be as hard a general structured output learning (HM-SVM) | 15:54 |
wiking | nyiha | 15:55 |
wiking | i haven't even knew about HM-SVM | 15:55 |
wiking | let me check what's that thingy :) | 15:55 |
wiking | apart from being a hidden markov -svm :))) | 15:55 |
sonne|work | it is frequently used in vision/text/bioinformatics to segment things | 15:55 |
sonne|work | e.g. dna into genes/non-genes | 15:55 |
wiking | mmm | 15:56 |
wiking | shogun has it? :) | 15:56 |
wiking | or it's only implemented in hm-svm toolbox | 15:57 |
sonne|work | well shogun has the argmax but no more | 16:01 |
sonne|work | one of the projects is to get hm-svm toolbox back into shogun | 16:02 |
sonne|work | with some nice design though | 16:02 |
sonne|work | code quality is not so good in that part... | 16:02 |
sonne|work | ^wiking | 16:11 |
wiking | mmm | 16:12 |
wiking | i'll give it a look | 16:12 |
sonne|work | wiking: well, choose what you are most interesting in / what does motivate you to do most :) | 16:33 |
-!- delcoder [~ishaan@115.242.61.210] has joined #shogun | 16:42 | |
-!- delcoder [~ishaan@115.242.61.210] has quit [Quit: Leaving] | 17:14 | |
-!- blackburn [~qdrgsm@188.168.128.87] has joined #shogun | 17:46 | |
-!- blackburn [~qdrgsm@188.168.128.87] has quit [Quit: Leaving.] | 18:04 | |
-!- blackburn [~qdrgsm@188.168.128.177] has joined #shogun | 18:10 | |
-!- n4nd0 [82ede325@gateway/web/freenode/ip.130.237.227.37] has joined #shogun | 18:22 | |
n4nd0 | blackburn: how is it going? | 18:28 |
blackburn | n4nd0: hey, fine, what about you? | 18:28 |
n4nd0 | blackburn: fine too | 18:29 |
n4nd0 | blackburn: I am taking a look to the gsoc ideas at the webpage | 18:29 |
blackburn | and what do you think about it? | 18:29 |
n4nd0 | I think they are really interesting in general | 18:30 |
n4nd0 | I find quite appealing the one of structured output | 18:31 |
n4nd0 | but being honest, it may be a bit out of scope for my knowledge right now | 18:31 |
n4nd0 | a bit too out of scope :P | 18:31 |
blackburn | well I didn't know anything about dimreduction last year when I did a lot of dimreduction :) | 18:31 |
n4nd0 | oh | 18:32 |
n4nd0 | I thought you had done sth with it before maybe in a math course or the like | 18:33 |
n4nd0 | you are working in multitask right? | 18:34 |
blackburn | will be working on MT, yeah | 18:34 |
n4nd0 | the first one about writing tutorials and documentation looks nice too | 18:37 |
n4nd0 | is that a priority for shogun? | 18:37 |
blackburn | well probably most interesting and important idea, yeah | 18:38 |
n4nd0 | yeah, that one looks good definetely | 18:39 |
n4nd0 | it can be a good way to learn the fundamentals so later (post-gsoc) one can continue with more advanced stuff | 18:40 |
n4nd0 | blackburn: anyhow, do you have any other idea do work on in the meantime? | 18:42 |
blackburn | well.. not really | 18:47 |
blackburn | let me think about it | 18:47 |
blackburn | n4nd0: maybe QDA? | 19:01 |
n4nd0 | blackburn: it sounds almost like LDA :P | 19:02 |
n4nd0 | I guess they must be related to each other | 19:02 |
blackburn | yes, but no code for QDA here :) | 19:02 |
n4nd0 | I will take a look to it then | 19:03 |
n4nd0 | do you know something it is applied for? | 19:03 |
blackburn | not really, but we are general-purpose ML lib | 19:06 |
blackburn | :) | 19:06 |
n4nd0 | I just wondered :P | 19:07 |
blackburn | may be spectral clustering | 19:07 |
blackburn | btw spectral clustering is a very simple task | 19:11 |
blackburn | one need to just combine ready KMeans and LaplacianEigenmaps | 19:11 |
n4nd0 | is it used for dimensionality reduction? | 19:11 |
blackburn | spectral clusterin is used for clustering, but it uses laplacian eigenmaps | 19:12 |
blackburn | simply it is kmeans in laplacian eigenmaps domain | 19:13 |
-!- blackburn [~qdrgsm@188.168.128.177] has quit [Quit: Leaving.] | 19:30 | |
-!- n4nd0 [82ede325@gateway/web/freenode/ip.130.237.227.37] has quit [Quit: Page closed] | 19:50 | |
-!- delcoder [~ishaan@101.63.49.239] has joined #shogun | 21:00 | |
CIA-64 | shogun: Soeren Sonnenburg master * r013f6d2 / src/Makefile : fix make linecount - http://git.io/lNMR3Q | 21:17 |
CIA-64 | shogun: Soeren Sonnenburg master * radec872 / examples/undocumented/python_modular/graphical/interactive_clustering_demo.py : mark also cluster centers - http://git.io/9dUOdw | 21:17 |
CIA-64 | shogun: Soeren Sonnenburg master * re2a16a6 / src/shogun/clustering/KMeans.cpp : whitespace fix - http://git.io/J7hFgw | 21:17 |
shogun-buildbot | build #509 of octave_static is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/octave_static/builds/509 blamelist: sonne@debian.org | 21:46 |
-!- blackburn [~qdrgsm@188.168.14.46] has joined #shogun | 21:59 | |
-!- delcoder [~ishaan@101.63.49.239] has quit [Ping timeout: 244 seconds] | 22:02 | |
shogun-buildbot | build #510 of octave_static is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/octave_static/builds/510 | 22:04 |
blackburn | sonney2k: you found time to do interactive demos ;) | 22:05 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 22:25 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 22:43 | |
-!- blackburn [~qdrgsm@188.168.14.46] has quit [Ping timeout: 252 seconds] | 22:43 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 22:53 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 23:36 | |
--- Log closed Fri Mar 02 00:00:19 2012 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!