--- Log opened Mon Apr 23 00:00:19 2012 | ||
gsomix | it's time to sleep | 00:03 |
---|---|---|
gsomix | good night, guys | 00:03 |
n4nd0 | good night gsomix | 00:04 |
-!- puffin444 [230bf329@gateway/web/freenode/ip.35.11.243.41] has quit [Quit: Page closed] | 00:06 | |
-!- blackburn_ [d5578adc@gateway/web/freenode/ip.213.87.138.220] has quit [Ping timeout: 245 seconds] | 00:41 | |
harshit_ | good night guys. | 00:56 |
-!- harshit_ [~harshit@182.68.43.52] has quit [Remote host closed the connection] | 00:57 | |
-!- av3ngr [~av3ngr@42.62.206.121] has quit [Ping timeout: 240 seconds] | 00:59 | |
-!- av3ngr [~av3ngr@42.62.206.121] has joined #shogun | 01:00 | |
-!- PhilTillet [~Philippe@vir78-1-82-232-38-145.fbx.proxad.net] has quit [Read error: Connection reset by peer] | 01:10 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 01:13 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 02:23 | |
-!- av3ngr [~av3ngr@42.62.206.121] has quit [Ping timeout: 260 seconds] | 02:43 | |
-!- av3ngr [~av3ngr@42.62.206.121] has joined #shogun | 02:45 | |
-!- pluskid [~pluskid@173.254.214.60] has joined #shogun | 03:42 | |
-!- av3ngr [~av3ngr@42.62.206.121] has quit [Ping timeout: 260 seconds] | 04:02 | |
-!- xiangwang [~chatzilla@159.226.60.224] has joined #shogun | 04:04 | |
-!- xiangwang [~chatzilla@159.226.60.224] has quit [Client Quit] | 04:05 | |
CIA-64 | shogun: Soeren Sonnenburg master * r719f903 / examples/undocumented/ruby_modular/modelselection_parameter_tree_modular.rb : disable printing of combinations - http://git.io/jAv3ew | 06:16 |
-!- Netsplit *.net <-> *.split quits: pluskid | 06:57 | |
-!- Netsplit over, joins: pluskid | 07:14 | |
shogun-buildbot | build #507 of ruby_modular is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/ruby_modular/builds/507 | 07:30 |
-!- vikram360 [~vikram360@117.192.178.207] has quit [Ping timeout: 250 seconds] | 07:37 | |
-!- av3ngr [~av3ngr@122.110.219.42] has joined #shogun | 07:54 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 08:26 | |
-!- vikram360 [~vikram360@117.192.190.229] has joined #shogun | 08:46 | |
-!- vikram360 [~vikram360@117.192.190.229] has quit [Ping timeout: 246 seconds] | 08:51 | |
-!- vikram360 [~vikram360@117.192.190.169] has joined #shogun | 08:52 | |
sonne|work | pluskid: I was just reading your patch thx! | 09:10 |
sonne|work | pluskid: to get the idea - you introduce multiclassstrategy that can be attached to any multiclass classifier and will generate configurations to be learned by some binary classifier / same with evaluation | 09:16 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 09:17 | |
pluskid | sonne|work, yes | 09:18 |
sonne|work | pluskid: that makes a lot of sense - only thing that might be confusing to users is that we have *true* multiclass machines | 09:19 |
pluskid | sonne|work, the motivation is that blackburn suggest me to write a tree-structure training strategy | 09:19 |
sonne|work | like KNN that will directly do everything | 09:19 |
pluskid | sonne|work, yes, like many MulticlassSVM | 09:19 |
pluskid | they handle training directly | 09:20 |
sonne|work | I think we have only one multiclass svm - GMNP | 09:20 |
sonne|work | libsvmmulticlass is also doing OvO | 09:20 |
pluskid | maybe LaRank? | 09:20 |
sonne|work | yes | 09:20 |
sonne|work | true | 09:20 |
-!- emrecelikten [~Anubis@176.40.232.233] has joined #shogun | 09:20 | |
pluskid | I think either way (the old and the new) of handling multiclass strategy is OK, if we only have OvO and OvR | 09:21 |
pluskid | but if we want to introduce more, this way is more convenient | 09:21 |
pluskid | especially when we want to attach some more data to the strategy | 09:21 |
pluskid | for example, in a tree-structure training, we need the user to define a tree and attach that to the strategy object | 09:22 |
sonne|work | pluskid: no your way is much better | 09:23 |
sonne|work | I mean you should read the ECOC JMLR paper / code | 09:23 |
sonne|work | there are so many strategies to do so it is much nicer the new way | 09:23 |
pluskid | hmm, you are right | 09:23 |
sonne|work | pluskid: only thing I don't know is how to make clear in the class hierarchy that we have true multiclass machines | 09:24 |
sonne|work | not requiring any of this | 09:24 |
pluskid | yes, and training and applying can also be separated | 09:24 |
sonne|work | yeah only training is true multiclass for many 'true' multiclass machines | 09:25 |
pluskid | some machines use their own training, but use the generic strategy for applying | 09:25 |
sonne|work | yeah | 09:25 |
sonne|work | not clear how to handle that yet | 09:25 |
sonne|work | pluskid: anyway the best way to proceed is to implement other ECOC schemes from the paper - things will become clear then | 09:27 |
sonne|work | pluskid: btw I added some (minor) comments to your patch - please address them and then it is ok to be merged | 09:27 |
pluskid | sonne|work, yes, practicing is the best teacher :p | 09:27 |
pluskid | OK, I'll look at them | 09:28 |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 09:28 | |
sonne|work | pluskid: thanks and please continue the great work! | 09:32 |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has joined #shogun | 09:32 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has quit [Changing host] | 09:32 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 09:32 | |
pluskid | sonne|work, thanks, I'll try my best :) | 09:32 |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has joined #shogun | 10:00 | |
-!- emrecelikten [~Anubis@176.40.232.233] has quit [Quit: Leaving.] | 10:20 | |
blackburn | sonne|work: multiclassliblinear and multiclassocas are true too | 10:24 |
pluskid | haha, my laptop is *just* capable of compiling shogun, if I opened other big prog (like Matlab) simutaneously, the compiler will complain out-of-memory | 10:26 |
blackburn | pluskid: oh you are here ;) | 10:27 |
blackburn | you did great work | 10:27 |
blackburn | I haven't been thinking about it actually | 10:28 |
pluskid | blackburn, thanks, I'm making some modifications according to sonney2k's feedback | 10:28 |
-!- emrecelikten [~Anubis@176.40.232.233] has joined #shogun | 10:29 | |
-!- pluskid [~pluskid@173.254.214.60] has quit [Ping timeout: 260 seconds] | 10:34 | |
-!- pluskid [~pluskid@111.120.43.103] has joined #shogun | 10:52 | |
pluskid | sonne|work, blackburn I updated the pull request | 10:52 |
blackburn | yes thanks | 10:52 |
blackburn | pluskid: have you seen ECOC training code in scikits learn? | 10:54 |
pluskid | blackburn, no | 10:54 |
blackburn | pluskid: https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/multiclass.py | 10:54 |
blackburn | looks pretty easy actually | 10:55 |
pluskid | hmm | 10:56 |
pluskid | I think I understand the basics | 10:57 |
pluskid | after reading its comments | 10:57 |
blackburn | pluskid: yes rather easy - I am also having some long discussion with v. franc about training ECOC svm in true multiclass fashion | 10:58 |
pluskid | what does that mean? "in true multiclass fashion" | 10:59 |
blackburn | pluskid: in one optimization problem | 11:00 |
blackburn | implies better normalization | 11:00 |
pluskid | OK | 11:00 |
blackburn | pluskid: I mean in basic ECOC with given K bits | 11:01 |
blackburn | you would need to construct K classifiers | 11:01 |
pluskid | yes | 11:01 |
blackburn | right? | 11:01 |
pluskid | so you train the K classifiers in one time? | 11:01 |
blackburn | I've been thinking about SVM (looks like SO at some place) | 11:01 |
blackburn | actually I'd like to construct K classifier for each class | 11:02 |
blackburn | however K classifiers at the same time would work too may be | 11:02 |
pluskid | K classifier for each class? | 11:02 |
blackburn | hmm yes if I didn't get crazy | 11:03 |
pluskid | I didn't quite get you, so e.g. we have C class, K bits, then we have CK classifiers | 11:03 |
pluskid | with a test point x, how do we classify it? | 11:04 |
blackburn | max_{c\in C} \sum_{k\in K} <w_c^k, x> | 11:05 |
blackburn | does it make sense for you? | 11:05 |
blackburn | I am unsure already | 11:05 |
blackburn | :D | 11:05 |
pluskid | let me think | 11:06 |
blackburn | it is not ECOC :D | 11:06 |
blackburn | daaamn I have to think about it | 11:06 |
pluskid | no, this has nothing to do with ECOC | 11:07 |
pluskid | ordinary one-vs-rest linear machine, haha | 11:07 |
pluskid | move the sum inside the inner product | 11:07 |
blackburn | not ovr | 11:07 |
blackburn | composition of K OvR - useless | 11:08 |
pluskid | hmm, but you can move the sum into the inner product, then \sum_{k\in K}w_c^k can be treated as a single w_c | 11:09 |
blackburn | ah right | 11:09 |
blackburn | pluskid: no, decision rule is wrong | 11:10 |
pluskid | blackburn, why? | 11:10 |
blackburn | should be | 11:10 |
pluskid | I don't understand | 11:11 |
blackburn | should be different I mean | 11:11 |
blackburn | :D | 11:11 |
blackburn | pluskid: nevermind this time | 11:11 |
pluskid | :) | 11:11 |
CIA-64 | shogun: Soeren Sonnenburg master * r1ccb9e4 / (21 files in 5 dirs): | 11:17 |
CIA-64 | shogun: Merge pull request #479 from pluskid/multiclass-strategy | 11:17 |
CIA-64 | shogun: Multiclass strategy refactoring (+14 more commits...) - http://git.io/Lf2ifA | 11:17 |
sonne|work | pluskid: thanks for the quick updates! | 11:17 |
pluskid | sonne|work, :) | 11:17 |
pluskid | hope build-bot is happy with my patches, though I have already checked locally | 11:18 |
sonne|work | pluskid: I hope so too :D | 11:18 |
sonne|work | blackburn: how many exams left? | 11:18 |
blackburn | sonne|work: 3 | 11:18 |
blackburn | sonne|work: but they are not related with optics | 11:18 |
blackburn | so I am available | 11:19 |
blackburn | :D | 11:19 |
sonne|work | haha | 11:19 |
blackburn | sonne|work: do you think I should ask SLEP guys to GPL their code under their copyright? | 11:20 |
sonne|work | blackburn: I think we should do the intrusive renaming changes rather soon | 11:20 |
blackburn | sonne|work: which ones? | 11:20 |
sonne|work | blackburn: like moving around multiclass classifiers | 11:20 |
sonne|work | and multitask / domainadaptation stuff | 11:20 |
blackburn | sonne|work: I still think multitask normalizers are better suited to be in transfer/multitask.. | 11:21 |
sonne|work | then the only remaining intrusive stuff for this year is IMHO sgvector | 11:21 |
sonne|work | maybe but the should be totally hidden from the user | 11:22 |
blackburn | sonne|work: but what to rename? | 11:24 |
-!- pluskid [~pluskid@111.120.43.103] has left #shogun ["for supper"] | 11:26 | |
-!- pluskid [~pluskid@111.120.43.103] has joined #shogun | 11:27 | |
shogun-buildbot | build #726 of cmdline_static is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cmdline_static/builds/726 blamelist: pluskid@gmail.com | 11:33 |
sonne|work | blackburn: rename includes moving things around :) | 11:34 |
blackburn | right | 11:35 |
sonne|work | blackburn: so? | 11:35 |
blackburn | sonne|work: so what? | 11:36 |
blackburn | actually you are right we need to move knn, etc to multiclass | 11:36 |
shogun-buildbot | build #704 of r_static is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/r_static/builds/704 blamelist: pluskid@gmail.com | 11:36 |
sonne|work | I am saying better do all the rename / move now than later when we run on full steam | 11:37 |
blackburn | ah yeah I'll do something about it | 11:39 |
shogun-buildbot | build #705 of octave_static is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/octave_static/builds/705 blamelist: pluskid@gmail.com | 11:41 |
shogun-buildbot | build #692 of python_static is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/python_static/builds/692 blamelist: pluskid@gmail.com | 11:45 |
n4nd0 | see you later people! | 11:46 |
blackburn | see yopu | 11:47 |
blackburn | you | 11:47 |
blackburn | hehe | 11:47 |
n4nd0 | :D | 11:47 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 11:48 | |
shogun-buildbot | build #727 of cmdline_static is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cmdline_static/builds/727 | 11:49 |
shogun-buildbot | build #705 of r_static is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/r_static/builds/705 | 11:54 |
shogun-buildbot | build #706 of octave_static is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/octave_static/builds/706 | 11:59 |
shogun-buildbot | build #693 of python_static is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/python_static/builds/693 | 12:04 |
shogun-buildbot | build #517 of java_modular is complete: Failure [failed compile] Build details are at http://www.shogun-toolbox.org/buildbot/builders/java_modular/builds/517 blamelist: pluskid@gmail.com | 12:04 |
shogun-buildbot | build #540 of octave_modular is complete: Failure [failed compile] Build details are at http://www.shogun-toolbox.org/buildbot/builders/octave_modular/builds/540 blamelist: pluskid@gmail.com | 12:05 |
shogun-buildbot | build #528 of csharp_modular is complete: Failure [failed compile] Build details are at http://www.shogun-toolbox.org/buildbot/builders/csharp_modular/builds/528 blamelist: pluskid@gmail.com | 12:06 |
shogun-buildbot | build #516 of lua_modular is complete: Failure [failed compile] Build details are at http://www.shogun-toolbox.org/buildbot/builders/lua_modular/builds/516 blamelist: pluskid@gmail.com | 12:07 |
shogun-buildbot | build #513 of python_modular is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/python_modular/builds/513 blamelist: pluskid@gmail.com | 12:14 |
shogun-buildbot | build #508 of ruby_modular is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/ruby_modular/builds/508 blamelist: pluskid@gmail.com | 12:20 |
shogun-buildbot | build #512 of r_modular is complete: Failure [failed compile] Build details are at http://www.shogun-toolbox.org/buildbot/builders/r_modular/builds/512 blamelist: pluskid@gmail.com | 12:21 |
shogun-buildbot | build #518 of java_modular is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/java_modular/builds/518 | 12:31 |
-!- pluskid [~pluskid@111.120.43.103] has quit [Ping timeout: 246 seconds] | 12:35 | |
-!- pluskid [~pluskid@173.254.214.60] has joined #shogun | 12:35 | |
shogun-buildbot | build #541 of octave_modular is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/octave_modular/builds/541 | 12:39 |
-!- PhilTillet [~Philippe@vir78-1-82-232-38-145.fbx.proxad.net] has joined #shogun | 12:41 | |
PhilTillet | hello everybodyh | 12:44 |
shogun-buildbot | build #529 of csharp_modular is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/csharp_modular/builds/529 | 12:46 |
shogun-buildbot | build #517 of lua_modular is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/lua_modular/builds/517 | 12:53 |
gsomix | >> Welcome again to the Machine learning class. | 12:59 |
gsomix | huh | 12:59 |
gsomix | hi all | 12:59 |
PhilTillet | hi gsomix | 12:59 |
PhilTillet | :p | 12:59 |
PhilTillet | gsomix, yes, a lot of coursera courses start today | 13:04 |
PhilTillet | :p | 13:04 |
shogun-buildbot | build #509 of ruby_modular is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/ruby_modular/builds/509 | 13:06 |
shogun-buildbot | build #513 of r_modular is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/r_modular/builds/513 | 13:13 |
blackburn | happiness again | 13:22 |
sonne|work | blackburn: look at python_modular | 13:27 |
blackburn | sonne|work: yes | 13:30 |
blackburn | I'll fix once I get home | 13:30 |
blackburn | sonne|work: I think that's ok to patch true multiclass machines | 13:35 |
blackburn | to avoid this error we can add trained flag there | 13:35 |
blackburn | so once we train machine we store number of classes | 13:35 |
sonne|work | blackburn: just think of doing x-validation with Mc machines | 13:36 |
blackburn | what is the problem? | 13:36 |
sonne|work | for some subsets you observer x for some x-1 classes or less | 13:36 |
sonne|work | same problem with sparse features btw | 13:36 |
sonne|work | there could be 1000 dims or 100 | 13:36 |
wiking | hey | 13:36 |
blackburn | ?? | 13:36 |
blackburn | that is the reason to train on different number of classes? | 13:36 |
blackburn | no sense for me at all | 13:37 |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Remote host closed the connection] | 13:37 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 13:37 | |
sonne|work | blackburn: you said you want to *infer* number of classes based on labels | 13:37 |
sonne|work | that is dangerous | 13:38 |
blackburn | yes train labels should contain all possible classes | 13:38 |
sonne|work | I think specifying nr classes is better | 13:38 |
sonne|work | blackburn: why should they? | 13:38 |
sonne|work | if you do x-val you might not have some classes | 13:38 |
blackburn | bad x-val | 13:38 |
sonne|work | not bad xval | 13:38 |
blackburn | but you will get wrong results! | 13:39 |
blackburn | sonne|work: how can you ignore some classes? | 13:40 |
blackburn | so if you skip class C you will have zero w_c, what makes you able to use it? | 13:42 |
blackburn | wiking: hey | 13:43 |
pluskid | I support specifying nr classes when constructing a classifier | 13:43 |
pluskid | just a matter of taste, however | 13:43 |
blackburn | pluskid: both ways are ok | 13:43 |
blackburn | however crossvalidation with class ignores makes no sense to me | 13:44 |
blackburn | things would be totally wrong if you skip some classes | 13:44 |
pluskid | I mean it would be easier to implement the code if we are given nr classes at first | 13:45 |
pluskid | then no need for a flag like m_trained or similar | 13:45 |
blackburn | pluskid: where is it set? | 13:45 |
pluskid | no in current code | 13:45 |
pluskid | but I think it might be needed | 13:45 |
blackburn | pluskid: yes that's I am asking | 13:45 |
blackburn | on which step? | 13:46 |
blackburn | ah | 13:46 |
blackburn | not in current? | 13:46 |
pluskid | for example, you do not know the number of machines, until it is trained | 13:46 |
blackburn | how does it work now | 13:46 |
blackburn | ? | 13:46 |
pluskid | for example, in MulticlassMachine | 13:47 |
pluskid | get_num_machines() | 13:47 |
pluskid | will return "wrong" value, until the machine is actually trained | 13:47 |
pluskid | We need to know the state of the machine to see whether some "knowledge" is already available | 13:48 |
pluskid | here the number of classes | 13:48 |
pluskid | the same problem happens in MulticlassStrategy | 13:48 |
pluskid | MulticlassStrategy::get_num_machines(int32_t num_classes) | 13:48 |
pluskid | one has to pass num_classes as an argument when calling the function | 13:49 |
pluskid | but if we know num_classes from the begining, there's no need for this argument | 13:49 |
pluskid | anyway, I think it is just a matter of taste, coz the current solution also works | 13:49 |
blackburn | pluskid: I think we can assume no ties in labels while training | 13:50 |
blackburn | however it looks like it would be better to store number of classes on training | 13:50 |
blackburn | and obtain it while applying | 13:50 |
blackburn | this way you can be sure it works with ties in test set | 13:51 |
wiking | blackburn: ho :) | 13:51 |
blackburn | wiking: does it make sense to you to skip some classes in train set? | 13:52 |
blackburn | (that are not skipped in test one) | 13:52 |
wiking | ?? | 13:52 |
pluskid | blackburn, I'm OK with this solution | 13:52 |
wiking | blackburn: why would one do that? | 13:52 |
blackburn | wiking: x-val | 13:52 |
pluskid | though still prefer specifying nr classes in advance :p (my taste...) | 13:53 |
blackburn | pluskid: in advance? | 13:53 |
pluskid | I mean when constructing the classifier | 13:53 |
blackburn | when training or constructing? | 13:53 |
blackburn | pluskid: problem is that if you have ties in labels | 13:54 |
blackburn | like test set contains only 1,2, 4 instances | 13:54 |
blackburn | of classes 1,2,4 | 13:54 |
blackburn | I mean | 13:54 |
blackburn | so get_num_classes would return 3 | 13:54 |
blackburn | and you will run only 3 of 4 machines | 13:55 |
blackburn | skipping last one - it looks incorrect | 13:55 |
pluskid | I mean in the constructor | 13:58 |
pluskid | before training | 13:58 |
pluskid | what do you mean by "ties in labels"? | 13:58 |
blackburn | 1,2,4 | 13:58 |
blackburn | no 3 | 13:58 |
blackburn | pluskid: but what if you set different labels? | 13:58 |
pluskid | but the number of labels won't change, will it? | 13:59 |
pluskid | number of classes | 13:59 |
blackburn | why not? | 13:59 |
blackburn | you can set features, kernel, labels - everything | 14:00 |
pluskid | You mean 1,2,4 or 2,3,4, or 1,2,3, etc? | 14:00 |
blackburn | just some missing class | 14:00 |
pluskid | OK | 14:00 |
pluskid | so we need set_num_classes, haha | 14:00 |
blackburn | pluskid: yes we need to save num of classes on training | 14:00 |
blackburn | we can't rely only on constructor - set_labels() would make no sense then | 14:01 |
pluskid | blackburn, OK, that makes sense | 14:01 |
blackburn | however this requires to set number of classes in strategy | 14:02 |
blackburn | in every true svm as well | 14:02 |
blackburn | true multiclass* | 14:02 |
pluskid | yes, that's the problem | 14:02 |
blackburn | not really | 14:02 |
blackburn | oneliner | 14:02 |
sonne|work | blackburn: I think that is the same issue like having a linear classifier operating on sparse features - sometimes not the last dim in any of the sparse feature vectors is non-zero so you need to specify length of w of the sparse classifier or so | 14:03 |
blackburn | sonne|work: are you still sure train set can contain a few classes? | 14:03 |
blackburn | (not all classes) | 14:04 |
sonne|work | sure | 14:04 |
blackburn | sonne|work: and how would you classify class X if you haven't trained a classifier for that? | 14:05 |
sonne|work | it will never predict class X but operate correctly otherwise | 14:06 |
blackburn | sonne|work: it broke other class' results | 14:06 |
blackburn | break* | 14:06 |
sonne|work | why? | 14:06 |
blackburn | sonne|work: hmm I assume it would be w=0 | 14:06 |
sonne|work | exactly | 14:07 |
blackburn | sonne|work: it can override some other class scores | 14:07 |
blackburn | ah damn | 14:07 |
sonne|work | ? | 14:07 |
blackburn | 0 always | 14:07 |
blackburn | hah | 14:07 |
blackburn | okay sorry | 14:07 |
blackburn | <0,x>=0 | 14:07 |
blackburn | sonne|work: it complicates things pretty much | 14:08 |
pluskid | I think if the samples for a class is missing, the problem is not with predicting, but with training. You need a one-class algorithm to train a "classifier" :D | 14:09 |
blackburn | pluskid: I am not even sure that would work with current algorithms | 14:09 |
blackburn | like gmnp/etc | 14:09 |
pluskid | however, if this happens, it should be an error | 14:10 |
pluskid | so if we are specified nr classes, we might be able to detect this kind of error | 14:10 |
pluskid | be back in a while | 14:10 |
blackburn | we need map there | 14:11 |
blackburn | to avoid any errors | 14:11 |
blackburn | we can't keep things like that anymore | 14:11 |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has quit [Quit: Page closed] | 14:57 | |
-!- gsomix [~gsomix@188.168.5.255] has quit [Ping timeout: 265 seconds] | 15:00 | |
-!- PhilTillet [~Philippe@vir78-1-82-232-38-145.fbx.proxad.net] has quit [Read error: Connection reset by peer] | 15:09 | |
-!- harshit_ [~harshit@182.68.43.52] has joined #shogun | 15:27 | |
-!- ted__ [2e1fd566@gateway/web/freenode/ip.46.31.213.102] has joined #shogun | 15:28 | |
ted__ | hi all | 15:28 |
ted__ | what type of Features do I need to use (c++) if I want to train multiclass svm using string kernel with float array samples of different size ? | 15:30 |
ted__ | thanks for help | 15:30 |
ted__ | and which string kernel is suitable for samples of different sizes ? | 15:31 |
-!- PhilTillet [~Philippe@vir78-1-82-232-38-145.fbx.proxad.net] has joined #shogun | 15:40 | |
-!- gsomix [~gsomix@83.234.54.70] has joined #shogun | 15:50 | |
ted__ | RE: hi there, what type of Features do I need to use (c++) if I want to train multiclass svm using string kernel with float array samples of different size ? and which string kernel is suitable for samples of different sizes ? Thanks in advance. T. | 15:54 |
harshit_ | ted__, hi ,I havn't had much experience with strings in shogun but i think you need to use CStringFeatures<char> as features . | 16:04 |
harshit_ | ted__, have a look at examples/documented/libshogun/features_copy_subset_string_features | 16:04 |
harshit_ | ted__, Also wait for a while till other members are online [blackburn, soeren] they will help you better | 16:05 |
-!- blackburn [~qdrgsm@188.168.2.179] has joined #shogun | 16:18 | |
PhilTillet | 4hours40 :S | 16:20 |
harshit_ | blackburn, hi , I think ted__ needs ur help :) | 16:20 |
blackburn | harshit_: yes I just read logs | 16:21 |
-!- vikram360 [~vikram360@117.192.190.169] has quit [Ping timeout: 265 seconds] | 16:21 | |
harshit_ | PhilTillet, really nervous moments :/ | 16:21 |
sonne|work | ted__: StringCharFeatures ? | 16:23 |
PhilTillet | harshit_, let's just get our body frozen for 5 hours | 16:26 |
blackburn | pluskid: I don't like dependency of multiclassmachine and multiclassstrategy | 16:30 |
harshit_ | PhilTillet, haha .. | 16:30 |
-!- pluskid [~pluskid@173.254.214.60] has quit [Ping timeout: 246 seconds] | 16:33 | |
harshit_ | blackburn, I think there might me some issue with svmlin.cpp present in shogun, I have compared the results of original svmlin and that of shogun's, they are close but not exactly similar . | 16:39 |
harshit_ | blackburn, have a look : http://snipt.org/uhjhh3 | 16:40 |
blackburn | are all parameters equal? | 16:40 |
harshit_ | blackburn, In svmlin original library we can set lambda so I set it to 1 | 16:41 |
ted__ | sonne|work, harshit_ : sorry, was away | 16:41 |
gsomix | harshit_, I think I'll drink vodka in the 17-18UTC and go to bed, aha | 16:41 |
harshit_ | blackburn, but in shogun internaly lambda = 1/2c so I make value of C as 0.5 | 16:41 |
blackburn | harshit_: but other params? are there any? | 16:42 |
harshit_ | blackburn, only other parameter is epsilon which is 1e-5 in both | 16:42 |
ted__ | Let's resume: I have samples of different sizes as vectors of float , I am looking for best shogun feature structure and kernel to train multiclass svm. | 16:43 |
ted__ | I'll take a look at StringCharFeatures | 16:43 |
ted__ | what about kernels ? | 16:43 |
blackburn | ted__: vectors of float? | 16:44 |
ted__ | yes | 16:44 |
blackburn | of variable size? | 16:44 |
harshit_ | gsomix, I think vodka is for blackburn, we should go for tequila ! | 16:44 |
ted__ | y | 16:44 |
blackburn | ted__: non-trivial.. how can you compare it? | 16:44 |
ted__ | I know, I'm guessing stringkernel could do the work as they implement smthg like dtw distance , | 16:45 |
ted__ | ? | 16:45 |
gsomix | harshit_, huh, I have never tried tequila. | 16:46 |
ted__ | It can be vectors of int but sizes are different. | 16:46 |
harshit_ | g2g, blackburn i'll do some more debugging and will let you know about the status ! | 16:46 |
blackburn | harshit_: yeah thanks | 16:46 |
blackburn | ted__: yes makes sense but they are for more 'finite' features like char | 16:47 |
harshit_ | gsomix, you should, maybe today !! | 16:47 |
ted__ | blackburn: and I dont want to use simple bag-of-words representation | 16:47 |
blackburn | ted__: what is the nature of your features? | 16:47 |
ted__ | blackburn: [w1,w2,w3,..] with w_iEUR[0..N] N could be 10, 100, 1000, or 10000 | 16:48 |
ted__ | and size of the "string" is not defined | 16:48 |
blackburn | ted__: have you any idea of similarility function? you can create custom kernel actually | 16:49 |
harshit_ | ted__, what I think is that you should first use string to manipulate the words and then prepare its corresponding VSM model which you can store in a normal CSimpleFeatures | 16:49 |
-!- pluskid [~pluskid@li381-56.members.linode.com] has joined #shogun | 16:50 | |
ted__ | harshit_: VSM ? | 16:50 |
harshit_ | vector space model | 16:50 |
harshit_ | in which you have words as features and documents as examples/datapoints | 16:51 |
-!- harshit_ [~harshit@182.68.43.52] has left #shogun ["Leaving"] | 16:52 | |
ted__ | I thought that "string kernel" could handle features of different size ? | 16:52 |
blackburn | ted__: yes if they are contained with chars | 16:53 |
ted__ | then clustered by [0,255] ? | 16:54 |
blackburn | yes would work | 16:54 |
ted__ | Then it could be same algo for string of cluster indices (indices btw 0 and 1000..) | 16:55 |
ted__ | Well, which shogun features/cluster could I overload if I want to do that ? | 16:56 |
blackburn | ted__: could you please clarify what exactly you need? | 16:56 |
ted__ | Well, which shogun features/kernel could I overload if I want to do that ? | 16:57 |
blackburn | repeating sentence is not a clarification I'd say :D | 16:57 |
ted__ | blackburn: sry, I just type cluster and meant kernel . | 16:58 |
blackburn | aha | 16:58 |
blackburn | ted__: you don't really need to overload features for that | 16:59 |
ted__ | blackburn: So. I have a set of indices vector of different size. Let's say Indices EUR [0,100]. Each vector is associated with a label. let's say we have 10 labels. | 16:59 |
-!- pluskid [~pluskid@li381-56.members.linode.com] has quit [Quit: Leaving] | 17:00 | |
ted__ | blackburn: I want to train multiclass svm of this data. Problem is that features have different size. | 17:00 |
-!- pluskid [~pluskid@li381-56.members.linode.com] has joined #shogun | 17:00 | |
ted__ | blackburn: indeed I have read about string kernel apply to string of char. I want to do the same with my problem. | 17:01 |
blackburn | ted__: you may try to construct StringFeatures<char> with your features | 17:01 |
blackburn | and use any kernel | 17:01 |
blackburn | string kernel I mean | 17:01 |
ted__ | ok. And if I have more than 256 indices ? | 17:02 |
blackburn | ted__: you have vecotr x = [x1,x2,...] where xi \in [0,255]? | 17:03 |
blackburn | vector* | 17:03 |
blackburn | what do you mean more that 256 indices? wider range of xi values? | 17:03 |
ted__ | blackburn: I have vector x = [x1,x2, ..] where xi \in [0,1000] | 17:05 |
-!- vikram360 [~vikram360@117.192.190.169] has joined #shogun | 17:05 | |
blackburn | ted__: ok and what is the similarity measure you want to use? | 17:06 |
blackburn | I am asking because string kernel are mainly used by bioinformaticians to estimate similarities between some ACTCGCT and TCGCGA - not a big alphabet | 17:07 |
ted__ | I know | 17:08 |
@sonney2k | wiking,gsomix,pluskid,blackburn, do you have some homepage I could link to? | 17:08 |
blackburn | I haven't | 17:08 |
wiking | sonney2k: mmm not really :( | 17:08 |
wiking | but i'm planning to keep a weekly posting on maeth.com | 17:08 |
pluskid | sonney2k: http://pluskid.org | 17:08 |
wiking | but currently it's personal rampage :) | 17:08 |
gsomix | shogun-buildbot, nope | 17:09 |
gsomix | sonney2k, nope | 17:09 |
gsomix | huh | 17:09 |
ted__ | Dunno exactly yet what type of similarity measure. | 17:09 |
ted__ | Maybe for the beginning one you used with your already implemented string kernels which support features of diff size | 17:10 |
ted__ | LocalAlignmentString ? | 17:10 |
ted__ | Any advice ? :p | 17:10 |
blackburn | ted__: it is based on alignment - I am not sure it would work for your features | 17:11 |
blackburn | ted__: where these features come from? | 17:11 |
gsomix | pluskid, cool pictures :] | 17:12 |
ted__ | spatial arrangement of my vectors are important: features extracted from video sequence, then I would like to keep the temporal info (=as in ACTCGCT) | 17:13 |
pluskid | gsomix: you like them? :D | 17:13 |
gsomix | pluskid, yep | 17:13 |
pluskid | gsomix: thanks! haha, I like drawing | 17:14 |
blackburn | ted__: really non-trivial thing you do | 17:14 |
ted__ | I know but it seems to be a good way to keep temporal information which is lost in a bag-of-frame algo. | 17:22 |
ted__ | maybe I could try with a mapping to [0,255] and see the results | 17:22 |
-!- pluskid [~pluskid@li381-56.members.linode.com] has quit [Quit: Konversation terminated!] | 17:23 | |
ted__ | thanks for helping , I anythg else come in mind, I stay here for a couple hours more | 17:23 |
-!- pluskid [~pluskid@li381-56.members.linode.com] has joined #shogun | 17:23 | |
blackburn | pluskid: you forgot to ref multiclass strategy ;) | 17:24 |
pluskid | blackburn, oh, my fault! | 17:25 |
blackburn | pluskid: I fixed | 17:25 |
pluskid | could you please help me to fix this? | 17:25 |
blackburn | will commit in a min | 17:25 |
pluskid | thanks! | 17:25 |
blackburn | pluskid: I am currently working on getting examples to work | 17:25 |
blackburn | because of that enum -> class change | 17:25 |
ted__ | blackburn: do you have an example using string kernel and multiclass svm ? | 17:26 |
pluskid | blackburn, I see, I fixed libshogun, but it seems I forgot python_modular | 17:26 |
pluskid | blackburn, thanks again! | 17:26 |
-!- av3ngr [~av3ngr@122.110.219.42] has quit [Ping timeout: 260 seconds] | 17:26 | |
blackburn | ted_: in C++? | 17:26 |
ted__ | blackburn: yes please, with features of diff size | 17:27 |
blackburn | ted__: check examples/undocumented/libshogun/ for some examples | 17:30 |
blackburn | I don't think we have any complex C++ example of string features and kernel learning | 17:30 |
ted__ | arf | 17:31 |
ted__ | blackburn: I will try EAlphabet with RAW BYTE - [0-255] | 17:32 |
pluskid | sonney2k: why did you collected the time zones for all the candidates? Is there any online conferences? | 17:32 |
blackburn | pluskid: yes for kickoff meeting | 17:32 |
pluskid | when? | 17:32 |
ted__ | I already checked those examples, was hoping you had somthg more specific on your own machines. | 17:32 |
blackburn | ted__: I am sorry, I haven't worked with such features (yet) | 17:33 |
ted__ | blackburn: thanks anyway for all those info. Maybe anyone else have ? | 17:34 |
blackburn | ted__: yes sonney2k has some experience with strings but he is pretty busy | 17:34 |
blackburn | pluskid: got error :( | 17:35 |
blackburn | File "classifier_multiclassmachine_modular.py", line 25, in classifier_multiclassmachine_modular | 17:35 |
blackburn | mc_classifier.train() | 17:35 |
blackburn | SystemError: [ERROR] Not a two class labeling - no positively labeled examples found | 17:35 |
pluskid | blackburn: leave that for me | 17:37 |
blackburn | pluskid: ok I'll push some fixes | 17:37 |
pluskid | blackburn: thanks! | 17:37 |
CIA-64 | shogun: Sergey Lisitsyn master * reeaa056 / (5 files in 3 dirs): A bunch of fixes for multiclass machine - http://git.io/jfjX2g | 17:38 |
pluskid | blackburn: BTW, what is the kickoff meeting for? | 17:38 |
blackburn | pluskid: hmm I think we need to discuss our plans | 17:39 |
blackburn | it is already late there you are.. | 17:39 |
pluskid | yes | 17:39 |
blackburn | pluskid: will you be available at 19 UTC? | 17:39 |
pluskid | I mean, we don't need to be there when Google announce the result, do we? | 17:39 |
blackburn | yes that's not required | 17:39 |
pluskid | OK, cool | 17:40 |
pluskid | if it's 15:40 UTC now | 17:40 |
pluskid | I don't think I will be available at 19 UTC | 17:40 |
pluskid | unless required | 17:40 |
pluskid | have to sleep :p | 17:40 |
blackburn | pluskid: when will you wake up? | 17:41 |
pluskid | let me calculate | 17:41 |
pluskid | around UTC 00:30 maybe | 17:41 |
blackburn | there is a chance I will not be sleeping yet | 17:41 |
blackburn | a lot of things to do today | 17:42 |
blackburn | :) | 17:42 |
pluskid | a regular sleeping time is good for a programmer! | 17:42 |
pluskid | to live longer! :D | 17:42 |
blackburn | pluskid: it would be painful to not complete these things today :) | 17:43 |
blackburn | pluskid: okay just read logs once you wake up | 17:43 |
pluskid | blackburn: hmm, then wish you finish them soon | 17:43 |
pluskid | sure | 17:43 |
ted__ | sonney2k: I am looking for C++ example of string features (with different size) and (multiclass) kernel learning. I know you are pretty busy, I you have time to share an example it would be great. Thank you in advance. | 17:52 |
ted__ | If* | 17:52 |
-!- blackburn [~qdrgsm@188.168.2.179] has quit [Ping timeout: 272 seconds] | 17:52 | |
-!- harshit_ [~harshit@182.68.43.52] has joined #shogun | 17:57 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 17:57 | |
gsomix | n4nd0, yo | 17:58 |
n4nd0 | gsomix: hey! | 17:58 |
gsomix | n4nd0, how are you? | 17:58 |
n4nd0 | gsomix: I am fine, what about you? | 17:58 |
n4nd0 | and hey to all people :) | 17:59 |
gsomix | n4nd0, fine. I'm waiting for 19UTC. | 18:00 |
n4nd0 | gsomix: I think we all are! | 18:01 |
n4nd0 | gsomix: are you nervous? | 18:01 |
PhilTillet | gsomix, join the club :p | 18:01 |
n4nd0 | I am getting more and more nervous :O | 18:01 |
gsomix | n4nd0, a little. | 18:01 |
pluskid | boys, keep relax, since no matter what we do *within the several hours*, we can not change the (currently unknown) result :p | 18:03 |
PhilTillet | :p | 18:08 |
pluskid | good night all, hope to see good news when I wake up | 18:09 |
-!- pluskid [~pluskid@li381-56.members.linode.com] has quit [Quit: Leaving] | 18:09 | |
n4nd0 | pluskid: good night and good luck! | 18:10 |
ted__ | sonney2k: Is there any example of using string features with Alphabet [0,255] ? | 18:23 |
wiking | :>>> | 18:36 |
PhilTillet | =>> | 18:36 |
@sonney2k | gsomix, dont' wait code :) | 18:52 |
gsomix | sonney2k, moin. what do you mean? | 18:54 |
@sonney2k | gsomix, SGVector is waiting for you :) | 18:55 |
gsomix | sonney2k, ah, I know. I'm doing now a small patches for wiking and blackburn. | 18:56 |
CIA-64 | shogun: Soeren Sonnenburg master * r9cd129d / src/shogun/multiclass/MulticlassOneVsRestStrategy.cpp : | 18:56 |
CIA-64 | shogun: Merge pull request #481 from pluskid/multiclass-fix | 18:56 |
CIA-64 | shogun: Fix a bug in OvR strategy: inc counter in wrong point - http://git.io/CEHNVw | 18:56 |
CIA-64 | shogun: Chiyuan Zhang master * r0814c83 / src/shogun/multiclass/MulticlassOneVsRestStrategy.cpp : Fix a bug in OvR strategy: inc counter in wrong point - http://git.io/B-fS_Q | 18:56 |
@sonney2k | gsomix, SGVector is more important I think - lets hope blackburn knows that | 18:57 |
wiking | gsomix: \o/ | 18:57 |
-!- blackburn [5bde8018@gateway/web/freenode/ip.91.222.128.24] has joined #shogun | 19:06 | |
blackburn | sonney2k: yes but it needs clarification | 19:08 |
blackburn | hey guys less than two hours left | 19:10 |
PhilTillet | yes ^^' | 19:11 |
-!- wiking_ [~wiking@huwico/staff/wiking] has joined #shogun | 19:12 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 246 seconds] | 19:15 | |
-!- wiking_ is now known as wiking | 19:15 | |
@sonney2k | blackburn, clarification in which respect? | 19:15 |
blackburn | sonney2k: explain to gsomix more what to do | 19:17 |
@sonney2k | gsomix, I don't know what to explain actually ... is there anything in particular that is unclear? | 19:18 |
blackburn | ref/unref? | 19:18 |
@sonney2k | I mean just continue to use SGVector& in arguments | 19:18 |
@sonney2k | blackburn, but the SGVector& stuff is not yet complete or gsomix? | 19:19 |
blackburn | ah this transition should be pretty clear | 19:19 |
gsomix | huh, I had planned to discuss it tomorrow. :) | 19:19 |
gsomix | but ok | 19:19 |
blackburn | sonney2k: ok then | 19:19 |
@sonney2k | gsomix, so please explain :) | 19:20 |
gsomix | sonney2k, hmm, I already did almost work for SGVector -> SGVector& in arguments, isn't it? | 19:21 |
gsomix | I am still wondering how to return SGVector from methods. | 19:22 |
@sonney2k | gsomix, SGVector& except if you return a new sgvectro | 19:22 |
@sonney2k | SGVector | 19:22 |
@sonney2k | then just ^ | 19:22 |
@sonney2k | gsomix, ok? | 19:23 |
gsomix | sonney2k, it's clear for me. But I worry about how it is then transferred to ref/unref. | 19:24 |
-!- harshit_ [~harshit@182.68.43.52] has quit [Quit: Leaving] | 19:24 | |
gsomix | sonney2k, ok, I'll do it in the next few days. | 19:25 |
@sonney2k | gsomix, we do the ref/unref business after this is done | 19:25 |
@sonney2k | gsomix, note that this will likely cause some errors that we cannot fix before we have ref/unref stuff | 19:26 |
@sonney2k | gsomix, if you want to know about ref/unref stuff - let us just copy the functions we have in SGObject for now | 19:26 |
gsomix | sonney2k, well I think step-by-step strategy is correct. | 19:27 |
@sonney2k | later on we should investigate whether we add another base class to derive SGVector / SGObject from | 19:27 |
@sonney2k | gsomix, btw, the same task then for SGMatrix/SGStringList/SGSparseVector ... | 19:27 |
@sonney2k | SGNDArray .. you name | 19:27 |
@sonney2k | it | 19:27 |
n4nd0 | sonney2k: a question, why to return SGVector and not a reference? | 19:32 |
n4nd0 | I mean what you said to gsomix ^ | 19:32 |
@sonney2k | n4nd0, if he allocates a new vector within a function (as local variable) - he cannot return a reference to it | 19:33 |
@sonney2k | since the reference will be invalid | 19:33 |
@sonney2k | (object de-allocated when the function returns) | 19:34 |
n4nd0 | wouldn't it better to create the vector to return in the heap? | 19:34 |
gsomix | blackburn, check a git. | 19:34 |
gsomix | *github | 19:34 |
n4nd0 | some stuff ... = new SGVector<...more stuff | 19:35 |
blackburn | gsomix: nice | 19:35 |
gsomix | n4nd0, I think it's not good at now (without ref/unref). | 19:37 |
n4nd0 | sonney2k: well actually SGVector is kind of light, a pointer and a couple of more things so probably the return by value shouldn't be much worse than returning a reference | 19:37 |
n4nd0 | gsomix: do you prefer return by value as well? | 19:37 |
CIA-64 | shogun: Soeren Sonnenburg master * r3007eb5 / src/shogun/io/StreamingAsciiFile.cpp : | 19:37 |
CIA-64 | shogun: pure whitespace changes | 19:37 |
CIA-64 | shogun: - convert file to use tabsize 4 - http://git.io/Ud_lvg | 19:37 |
CIA-64 | shogun: Soeren Sonnenburg master * r098c0cc / src/shogun/lib/v_array.h : | 19:37 |
CIA-64 | shogun: delete v_array's begin on exit | 19:37 |
CIA-64 | shogun: This hopefully fixes the memory leak observed in | 19:37 |
CIA-64 | shogun: https://github.com/shogun-toolbox/shogun/pull/401 - http://git.io/gfTGeA | 19:37 |
gsomix | n4nd0, it is a necessity, I mean | 19:38 |
n4nd0 | gsomix: I wondered if return by reference was an option | 19:43 |
@sonney2k | n4nd0, yes the object is alloc'd on heap however just *once* since RVO is used https://en.wikipedia.org/wiki/Return_value_optimization | 19:44 |
@sonney2k | n4nd0, there is no option | 19:44 |
@sonney2k | n4nd0, see http://stackoverflow.com/questions/4643713/c-returning-reference-to-local-variable and the first answer | 19:45 |
n4nd0 | sonney2k: ok, thank you | 19:45 |
blackburn | sonney2k: any preliminary date for kickoff? | 19:47 |
@sonney2k | n4nd0, I tried to fix that leak in streaming features - can you check if it still occurs? | 19:48 |
@sonney2k | blackburn, some people have ECML deadline tomorrow morning so not before that :D | 19:48 |
@sonney2k | maybe wednesday or some other day this week | 19:49 |
blackburn | sonney2k: yeah chris has | 19:49 |
@sonney2k | yeah he is using shogun :) | 19:49 |
blackburn | what is the relation? | 19:49 |
n4nd0 | sonney2k: I'll check it | 19:52 |
CIA-64 | shogun: Evgeniy Andreev master * r15afb6f / (14 files in 5 dirs): knn and conjugate to multiclass - http://git.io/XbLKXg | 19:59 |
CIA-64 | shogun: Soeren Sonnenburg master * rc46d1c0 / (14 files in 5 dirs): | 19:59 |
CIA-64 | shogun: Merge pull request #482 from gsomix/to_multiclass | 19:59 |
CIA-64 | shogun: move KNN and ConjugateIndex to shogun/multiclass - http://git.io/uSr2zQ | 19:59 |
n4nd0 | I think qda could me moved to shogun/multiclass too ^ | 20:00 |
n4nd0 | be moved, sorry | 20:00 |
-!- harshit_ [~harshit@182.68.43.52] has joined #shogun | 20:01 | |
@sonney2k | n4nd0, send a pull request | 20:02 |
@sonney2k | n4nd0, btw didn't you also convert LDA to multiclass? | 20:02 |
@sonney2k | blackburn, that reminds me - didn't we want to have CMulticlassLabels? | 20:03 |
n4nd0 | sonney2k: no, I didn't do anything to LDA | 20:03 |
n4nd0 | sonney2k: once, even before QDA was done, it was commented though | 20:03 |
blackburn | sonney2k: yes | 20:04 |
blackburn | sonney2k: we will do it once | 20:04 |
@sonney2k | n4nd0, too bad - that's really something we should have if we have multiclass enabled QDA | 20:05 |
n4nd0 | sonney2k: just for completeness or is there something else? | 20:06 |
@sonney2k | n4nd0, multiclass LDA is certainly much more robust than QDA :) | 20:07 |
@sonney2k | so it is really a very useful baseline method | 20:07 |
@sonney2k | much faster than SVM on low-dim data sets and even optimal when you can assume gaussian distribs | 20:07 |
blackburn | sonney2k: btw have you seen that SLEP supports some logistic regression even? | 20:09 |
@sonney2k | SLEP? | 20:09 |
blackburn | yes that package I am going to integrate | 20:10 |
@sonney2k | but we have logistic regression in liblinear already?! | 20:10 |
blackburn | sonney2k: why not to provide alternative? | 20:11 |
@sonney2k | I just don't understand your excitement | 20:11 |
blackburn | sonney2k: it is library containing a lot of L1 and L1/L2 things | 20:12 |
n4nd0 | sonney2k: I quite don't understand why QDA may fail somewhere LDA successes | 20:14 |
n4nd0 | I thought of QDA as a generalization, within class covariances need not to be equal, so I assumed it would give better results | 20:15 |
n4nd0 | even if teh covariances are the same for all the classes, case where LDA would perform fine too | 20:15 |
-!- blackburn [5bde8018@gateway/web/freenode/ip.91.222.128.24] has quit [Ping timeout: 245 seconds] | 20:16 | |
@sonney2k | n4nd0, estimating more parameters is difficult - so for noisy / not a lot of data LDA will work better | 20:25 |
n4nd0 | sonney2k: ok, intuituively I understand too that QDA may be more prone to overfitting | 20:26 |
* sonney2k is bored | 20:31 | |
gsomix | 27min :] | 20:32 |
n4nd0 | sonney2k: double free or corruption in muticlasslinearmachine example, pretty ugly stack trace :S | 20:33 |
@sonney2k | n4nd0, paste :) | 20:33 |
n4nd0 | http://snipt.org/uhjja8 | 20:35 |
-!- nickon [~noneedtok@kotnet-147.kulnet.kuleuven.be] has joined #shogun | 20:36 | |
-!- vikram360 [~vikram360@117.192.190.169] has quit [Ping timeout: 260 seconds] | 20:39 | |
-!- blackburn [~qdrgsm@188.168.2.179] has joined #shogun | 20:41 | |
n4nd0 | sonney2k: gdb trace is pretty cool too | 20:41 |
n4nd0 | full of | 20:41 |
n4nd0 | [New Thread 0xb6345b70 (LWP 22769)] | 20:41 |
n4nd0 | [Thread 0xb6345b70 (LWP 22769) exited] | 20:41 |
blackburn | 15 in room | 20:41 |
blackburn | oh | 20:41 |
n4nd0 | blackburn: do you mean 15 people there? | 20:41 |
blackburn | yeap | 20:42 |
n4nd0 | blackburn: oooh! you have followers man | 20:42 |
blackburn | followers? | 20:42 |
n4nd0 | yeah, I don't know :P | 20:42 |
n4nd0 | I thought you were all waiting for GSoC's answer | 20:43 |
PhilTillet | n4nd0, GSoC answer? What are you talking about? :p | 20:43 |
wiking | hahah the crowd gets excited ... ;) | 20:43 |
PhilTillet | what is GSoC? :o | 20:43 |
wiking | i bet gsoc site will be flooded | 20:43 |
PhilTillet | like last year | 20:43 |
PhilTillet | but as far as I remember the result is also send by mail | 20:44 |
wiking | hehehe i wonder why don't they do a caching | 20:44 |
wiking | and just do a simple precompiled html and that's it | 20:44 |
* emrecelikten starts biting nails | 20:44 | |
n4nd0 | sonney2k: oh wait, maybe I was handling wrong SG_UNREF in my example | 20:45 |
blackburn | are you fskin' ready? ;) | 20:49 |
PhilTillet | f5 spree | 20:50 |
blackburn | no way to see that even after 5 minutes of deadline | 20:51 |
n4nd0 | haha | 20:52 |
PhilTillet | yes I know | 20:52 |
PhilTillet | :p | 20:52 |
PhilTillet | at first not all the names are loaded anyway | 20:52 |
harshit_ | Is it gonna be same as the organizations names were released ? | 20:53 |
PhilTillet | no | 20:53 |
PhilTillet | in principle there's also an e-mail | 20:53 |
blackburn | email is pretty late | 20:54 |
PhilTillet | last year I had the e-mail 5mins before the deadline | 20:54 |
PhilTillet | :p | 20:54 |
PhilTillet | but I guess it was some bug | 20:55 |
@sonney2k | n4nd0, can you be more specific? | 20:55 |
@sonney2k | or attach a gdb bt | 20:55 |
@sonney2k | I cannot make sense of that | 20:55 |
n4nd0 | sonney2k: I think the double free was caused by SG_UNREF I am doing and shouldn't be done | 20:56 |
@sonney2k | which one? | 20:56 |
n4nd0 | in any case I think that to use more SG_UNREF shouldn't cause that | 20:56 |
n4nd0 | in the example classifier_multiclasslinearmachine | 20:57 |
n4nd0 | the one with which I detected those memory leaks | 20:57 |
emrecelikten | The emails are going out | 20:57 |
n4nd0 | well good luck to all guys! | 21:00 |
blackburn | ;) | 21:00 |
harshit_ | good luck guys | 21:00 |
PhilTillet | good luck :) | 21:01 |
blackburn | sonney2k: you are free to announce I think ;) | 21:01 |
harshit_ | site is down :( | 21:02 |
blackburn | hehe | 21:02 |
wiking | "Error retrieving data: please refresh the list or the whole page to try again" | 21:02 |
wiking | :> | 21:02 |
wiking | yeey for google :) | 21:02 |
n4nd0 | :) | 21:02 |
emrecelikten | I'm in too from another project | 21:03 |
emrecelikten | Best of luck to you guys | 21:03 |
emrecelikten | :) | 21:03 |
n4nd0 | emrecelikten: congrats! | 21:03 |
harshit_ | emrecelikten, congrats ! | 21:04 |
emrecelikten | Thanks guys :) | 21:04 |
blackburn | any of you received email? | 21:04 |
wiking | - | 21:05 |
emrecelikten | I did | 21:05 |
harshit_ | no | 21:05 |
PhilTillet | I didn't | 21:05 |
n4nd0 | - | 21:05 |
blackburn | heheh melange is really SLOW | 21:05 |
PhilTillet | yes | 21:06 |
emrecelikten | The servers experience something like delta(t) * e^(-at) now I guess | 21:06 |
PhilTillet | it loaded about 500 students (no shogun :p) and then crashed | 21:06 |
harshit_ | PhilTillet, same here ! | 21:06 |
PhilTillet | same again | 21:09 |
PhilTillet | after 400, crash! | 21:09 |
PhilTillet | XD | 21:09 |
wiking | heheh | 21:12 |
wiking | they have only the first 100 loaded | 21:12 |
blackburn | oh mailing list will be full of crazy stuff again | 21:13 |
n4nd0 | :D | 21:13 |
-!- karlnapf [~heiko@host86-180-223-101.range86-180.btcentralplus.com] has joined #shogun | 21:13 | |
harshit_ | blackburn, why ? | 21:14 |
karlnapf | hi all | 21:14 |
PhilTillet | hi karlnapf :p | 21:15 |
blackburn | harshit_: gsoc mailing list is full of stuff every year | 21:15 |
karlnapf | PhilTillet, hi there | 21:15 |
blackburn | karlnapf: hey | 21:15 |
karlnapf | blackburn, hi | 21:15 |
blackburn | karlnapf: received email? ;) | 21:15 |
karlnapf | blackburn, no not yet, but just turned on computer | 21:16 |
karlnapf | have been vomiting all day :( | 21:16 |
blackburn | are you ok? | 21:16 |
karlnapf | not really | 21:16 |
-!- vikram360 [~vikram360@117.192.162.41] has joined #shogun | 21:16 | |
blackburn | I hope you will recover soon | 21:17 |
karlnapf | blackburn, thanks | 21:17 |
blackburn | I can recall last year Soeren disappeared on the announcement too hehe | 21:17 |
n4nd0 | karlnapf: yeah, take care ;) | 21:17 |
karlnapf | n4nd0, thanks | 21:18 |
emrecelikten | It might a common thing, because my mentor disappeared too | 21:18 |
karlnapf | I only got a mail fro carol | 21:18 |
blackburn | karlnapf: which mail? | 21:19 |
karlnapf | Hi GSoC Students, | 21:19 |
karlnapf | As with every year around this time, we are about to add new students to this mailing list in a.... | 21:19 |
blackburn | ahaha | 21:19 |
blackburn | yeah | 21:19 |
karlnapf | the melange site seems to be down :) | 21:19 |
blackburn | yes tends to | 21:19 |
PhilTillet | 774 loaded and no shogun o_o | 21:20 |
wiking | the first 500 does not contain shogun related proj | 21:20 |
wiking | PhilTillet: :D | 21:20 |
blackburn | no shogun at all! | 21:20 |
blackburn | lol | 21:20 |
emrecelikten | Yeah | 21:20 |
wiking | well there were 4k applications | 21:20 |
wiking | i don't know how many of them are going to get accepted | 21:21 |
n4nd0 | PhilTillet: you can load quite a lot man, are you using GPU in the browser too? :P | 21:21 |
blackburn | wiking: ~1.1K | 21:21 |
PhilTillet | n4nd0, XD | 21:21 |
PhilTillet | but it stopped at 774 | 21:21 |
harshit_ | I have a different case all together, cant even open melange site :( | 21:21 |
PhilTillet | no more loading | 21:21 |
n4nd0 | harshit_: I am mostly there too | 21:22 |
@sonney2k | which site are you all trying to load? | 21:22 |
PhilTillet | http://www.google-melange.com/gsoc/projects/list/google/gsoc2012 | 21:22 |
shogun-buildbot | build #516 of python_modular is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/python_modular/builds/516 | 21:22 |
harshit_ | ^ | 21:22 |
blackburn | it should be a nice DDoS attack - to tell gsoc students accepted project are announced on site X | 21:24 |
PhilTillet | lol | 21:25 |
n4nd0 | :D | 21:25 |
PhilTillet | GSoC accepted projects are on universalmusic.com | 21:25 |
wiking | :> | 21:25 |
harshit_ | lol | 21:25 |
PhilTillet | once again, stopped at 774 | 21:25 |
PhilTillet | and no shogun! | 21:26 |
blackburn | I have 100 still | 21:26 |
emrecelikten | I'm trying too, for you guys :) | 21:26 |
wiking | PhilTillet: i guess they are inserting lines by lines | 21:26 |
-!- puffin444 [230bf329@gateway/web/freenode/ip.35.11.243.41] has joined #shogun | 21:27 | |
PhilTillet | wiking, well, seems like i've loaded everything though | 21:27 |
wiking | PhilTillet: yeah 774 and it gets stuck | 21:27 |
PhilTillet | I think we'll have to wait some additional time | 21:27 |
blackburn | lets have some vodka instead | 21:27 |
blackburn | who's in? | 21:28 |
karlnapf | lol, not me :) | 21:28 |
PhilTillet | I don't drink :p | 21:28 |
blackburn | PhilTillet: bad way | 21:28 |
PhilTillet | :) | 21:29 |
puffin444 | Well this is quite suspensful | 21:30 |
PhilTillet | it's beyond suspens | 21:31 |
PhilTillet | ^^ | 21:31 |
karlnapf | hey guys, i will check back later in an hour or so | 21:31 |
puffin444 | Yeah it's sort of torturous :) | 21:31 |
puffin444 | I wonder how google updates these. Do you think my dashboard/email would tell me first? | 21:32 |
blackburn | puffin444: IIRC that list is first | 21:32 |
wiking | 774 is the magic number :D | 21:33 |
n4nd0 | 400 here | 21:33 |
PhilTillet | again 774 | 21:33 |
PhilTillet | :p | 21:33 |
puffin444 | How many in total is it this year? 1000? | 21:33 |
PhilTillet | 774 + 9 | 21:33 |
PhilTillet | XD | 21:33 |
wiking | :)) | 21:33 |
blackburn | puffin444: should be ~1.1K again I think | 21:34 |
harshit_ | 774 , no shogun :( | 21:34 |
emrecelikten | Yep, I'm stuck at 774 too | 21:34 |
puffin444 | I suspect they haven't released all of them at the same time for some reason. | 21:35 |
puffin444 | They did the same thing with the sponsoring organizations | 21:35 |
puffin444 | Some didn't pop up until days after the release data :o | 21:35 |
blackburn | puffin444: no with orgs the reason was different | 21:35 |
blackburn | organizations had to complete their profiles | 21:35 |
PhilTillet | I think it's just a bug of theirs | 21:35 |
blackburn | not a bug | 21:36 |
PhilTillet | a feature? | 21:36 |
blackburn | with orgs - yes | 21:36 |
PhilTillet | i was talking about the students :p | 21:37 |
blackburn | as for students no idea what is going on hehe | 21:37 |
PhilTillet | maybe it's a feature too | 21:37 |
PhilTillet | :p | 21:37 |
PhilTillet | 786! | 21:37 |
PhilTillet | the number is evolving | 21:38 |
PhilTillet | XD | 21:38 |
blackburn | terribly slow thing | 21:39 |
puffin444 | Did anyone receive that "none" email a few days back? | 21:39 |
blackburn | I didn't | 21:39 |
PhilTillet | I didn't either | 21:39 |
emrecelikten | I did | 21:40 |
emrecelikten | It was a positive one | 21:40 |
blackburn | 811 loaded | 21:40 |
blackburn | no shogun ;) | 21:40 |
PhilTillet | haha blackburn, new record! | 21:40 |
PhilTillet | i'm refreshing again | 21:40 |
PhilTillet | maybe i'll get a better score | 21:40 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: Lost terminal] | 21:40 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 21:41 | |
puffin444 | emrecelikten: what do you mean be "positive"? | 21:41 |
puffin444 | *by* | 21:41 |
wiking | 814 | 21:42 |
wiking | :) | 21:42 |
PhilTillet | 827 :p | 21:42 |
PhilTillet | no shogun though | 21:42 |
puffin444 | Do you know how many slots shogun had this year? | 21:42 |
emrecelikten | puffin444: There were two types of emails as far as I know. The "thank you" ones that corresponded to negative ones and the "congratulations" ones. Some guy claimed that he got "pick one of the projects" email | 21:42 |
emrecelikten | 836 | 21:43 |
puffin444 | This happened a few days ago? | 21:44 |
emrecelikten | puffin444: Yes, Friday I think | 21:44 |
wiking | 846 | 21:44 |
PhilTillet | 855 | 21:45 |
PhilTillet | :p | 21:45 |
blackburn | hehe it becomes faster | 21:45 |
emrecelikten | It's like an auction | 21:45 |
@sonney2k | http://google-opensource.blogspot.de/2012/04/students-announced-for-google-summer-of.html | 21:45 |
@sonney2k | look 1212 is the number | 21:45 |
PhilTillet | yep | 21:45 |
wiking | ok so 1212 :> | 21:45 |
blackburn | big # | 21:46 |
puffin444 | I just got a rejection email from another organization | 21:46 |
wiking | fyi: 5 500 * 1 212 = 6 666 000 | 21:46 |
wiking | ;) | 21:46 |
PhilTillet | 943 | 21:47 |
wiking | funny number... i wonder whether it was intentional ;) | 21:47 |
PhilTillet | no shogun | 21:47 |
PhilTillet | :p | 21:47 |
blackburn | wiking: in russia it would be 300-400 - you have to stole 60% :D | 21:47 |
emrecelikten | 950 here | 21:47 |
emrecelikten | No shogun | 21:47 |
wiking | blackburn: :DDD | 21:48 |
wiking | 955 | 21:48 |
PhilTillet | 982 :p | 21:48 |
emrecelikten | You guys seems to be the last :P | 21:48 |
blackburn | 978 | 21:48 |
* wiking refreshing :) | 21:49 | |
puffin444 | Maybe I'll never see the decision. I'll just die from a heart attack first :P | 21:49 |
PhilTillet | same for me | 21:49 |
emrecelikten | This is cruel | 21:49 |
n4nd0 | emrecelikten: yeah it is | 21:50 |
PhilTillet | 999! | 21:50 |
PhilTillet | no shogun | 21:50 |
PhilTillet | :p | 21:50 |
emrecelikten | PhilTillet: Same | 21:50 |
harshit_ | 1000 , no sign of shogun :( | 21:50 |
wiking | 1000 | 21:50 |
blackburn | we are elite | 21:50 |
PhilTillet | XD | 21:51 |
wiking | ok | 21:51 |
wiking | they are just at R | 21:51 |
wiking | so soon | 21:51 |
wiking | 1007 | 21:51 |
PhilTillet | how do you know? XD | 21:51 |
PhilTillet | i've got some T loaded | 21:51 |
PhilTillet | and it will be hell cause Shogun's list will progressively appear | 21:51 |
PhilTillet | :p | 21:51 |
wiking | PhilTillet: yeah | 21:51 |
wiking | PhilTillet: but then now suddenly between 900-1000 | 21:51 |
wiking | new P and R projects got in | 21:51 |
blackburn | 1049 | 21:51 |
wiking | ;) | 21:51 |
blackburn | no shogun | 21:51 |
emrecelikten | n4nd0: You're in | 21:52 |
emrecelikten | If you're Fernando Garcia | 21:52 |
n4nd0 | emrecelikten: really? | 21:52 |
wiking | yeeey | 21:52 |
emrecelikten | Yep | 21:52 |
blackburn | hehe congrats | 21:52 |
puffin444 | Congrats! | 21:52 |
wiking | congrats n4nd0 | 21:52 |
emrecelikten | pluskid in | 21:52 |
n4nd0 | well Iglesias should be in between | 21:52 |
karlnapf | nice :), congrats | 21:52 |
emrecelikten | Sergey Lisitsyn in | 21:52 |
blackburn | hehe | 21:52 |
emrecelikten | Michal in | 21:52 |
n4nd0 | :D | 21:52 |
wiking | plsukid is in | 21:52 |
n4nd0 | I wan to see that guys! | 21:52 |
emrecelikten | Jacob Walker in | 21:52 |
puffin444 | YES | 21:52 |
emrecelikten | vigsterkr in | 21:52 |
wiking | \o/ | 21:52 |
puffin444 | Oh wow! | 21:52 |
n4nd0 | still 1024 here | 21:53 |
PhilTillet | sergey in | 21:53 |
emrecelikten | Just filter it ;) | 21:53 |
n4nd0 | maaaail | 21:53 |
wiking | got a mail :D | 21:53 |
n4nd0 | fuck I was not selected for multiclass :O | 21:53 |
karlnapf | just got the mail :) | 21:53 |
n4nd0 | :P | 21:53 |
blackburn | hehe got the mail | 21:53 |
wiking | n4nd0: heheheh i'm more happy that you'll do that | 21:54 |
n4nd0 | GOT SO! | 21:54 |
emrecelikten | Congratulations to you guys! | 21:54 |
PhilTillet | got a negative mail | 21:54 |
PhilTillet | :( | 21:54 |
blackburn | PhilTillet: I am sorry | 21:54 |
harshit_ | -ve :( | 21:54 |
wiking | PhilTillet: :(((( | 21:54 |
puffin444 | PhilTillet: Sorry PhilTillet :( :( | 21:54 |
emrecelikten | PhilTillet: Next year mate :) I was rejected last year too, I know how it feels | 21:54 |
n4nd0 | PhilTillet: sorry man | 21:54 |
PhilTillet | emrecelikten, I was rejected last year too :p | 21:54 |
wiking | n4nd0: when are you going to start the SO? | 21:55 |
emrecelikten | PhilTillet: I was being rejected for two years :P | 21:55 |
n4nd0 | n4nd0: asap :) | 21:55 |
n4nd0 | ups, again talking to me | 21:55 |
n4nd0 | haha | 21:55 |
wiking | n4nd0: hahahah | 21:55 |
wiking | n4nd0: i'll need it | 21:55 |
wiking | asap :D | 21:56 |
n4nd0 | wiking: cool, we can probably exchange some thoughts about it :) | 21:56 |
wiking | yeah sure | 21:56 |
wiking | are you going to use libqp? | 21:56 |
blackburn | PhilTillet: okay actually we are just not ready for gpuing.. | 21:56 |
n4nd0 | wiking: yes, that was the first option I thought of | 21:57 |
wiking | aha | 21:57 |
wiking | so then you are the third one | 21:57 |
PhilTillet | blackburn, why so? | 21:57 |
wiking | since i'm going to use it and michal as well | 21:57 |
blackburn | harshit_: I am very sorry - the problem is that Soeren is not capable to mentor two students | 21:57 |
wiking | so i've started to talk with sonney2k about how to refactor some stuff | 21:57 |
blackburn | PhilTillet: no real mentor for that | 21:57 |
blackburn | and no base for that actually | 21:57 |
wiking | n4nd0: i'll push this week something so your comments are more than welcome | 21:58 |
PhilTillet | hmmm ok | 21:58 |
n4nd0 | wiking: thank you | 21:58 |
harshit_ | blackburn, can understand :| | 21:58 |
blackburn | I really hope you are not upset with anything | 21:58 |
n4nd0 | wiking: I'll start to be much more active again after Saturday | 21:58 |
n4nd0 | damned exams | 21:58 |
PhilTillet | blackburn, I'm not ^^ | 21:58 |
wiking | n4nd0: no worries | 21:58 |
blackburn | PhilTillet: that's nice :) | 21:59 |
blackburn | PhilTillet: we are still welcome your contributions | 22:00 |
blackburn | harshit_: your contributions are welcome too I hope you are not very frustrated! | 22:01 |
PhilTillet | blackburn, I'll probably do the project on my own :) that's why I sent only one application, there's no point in working in something i'm not fully interested in ; | 22:02 |
PhilTillet | :) | 22:02 |
gsomix | hey, congrats guys | 22:02 |
harshit_ | blackburn, no, actually a lil. but still can work on NN for shogun , and my research | 22:02 |
@sonney2k | PhilTillet, harshit_ let me explain a bit | 22:03 |
blackburn | you all have done really a lot of work | 22:03 |
harshit_ | blackburn, but summer could be better with gsoc | 22:03 |
@sonney2k | I am/was the only mentor for your projects and the one we selected gsomix for | 22:03 |
@sonney2k | I would have loved to take you both too but I simply cannot handle >1 student | 22:04 |
@sonney2k | so we decided for the task we needed most - and that is some restructuring / accessibility cleanups | 22:04 |
@sonney2k | there was nothing wrong with your proposal or involvement or anything | 22:04 |
PhilTillet | sonney2k, we know :) | 22:05 |
@sonney2k | it really is about being at our capacity limits | 22:05 |
PhilTillet | sonney2k, nothing prevents us from contributing without being paid I guess :) | 22:05 |
harshit_ | sonney2k, no worries :) | 22:05 |
@sonney2k | and there is nothing I could do about it so late | 22:05 |
@sonney2k | in the end I *never* expected that we get 9 slots | 22:05 |
@sonney2k | if I knew that beforehand I would have asked a lot more people :) | 22:06 |
PhilTillet | :p | 22:06 |
@sonney2k | anyways whole wikimedia foundation has 8 (or 9?) slots | 22:06 |
puffin444 | Wow | 22:06 |
@sonney2k | so just to give you some impression how many we received | 22:06 |
blackburn | we keep receiving strong gsoc support :) | 22:07 |
@sonney2k | PhilTillet, harshit_ of course we welcome your contributions | 22:07 |
harshit_ | sonney2k, sure ..! | 22:07 |
@sonney2k | karlnapf, do you have a homepage I can link to | 22:08 |
@sonney2k | puffin444, ^ ? | 22:08 |
@sonney2k | and does any of the students here blog? I can put you on shogun planet if you want... | 22:09 |
puffin444 | sonney2k: Not at the current moment. I might be able to make one if you need one though. | 22:09 |
blackburn | sonney2k: do you think introduction @ mailing list is a good idea? | 22:09 |
@sonney2k | not required - I just wanted to give it some traffic - hehe | 22:09 |
@sonney2k | blackburn, I think so | 22:09 |
n4nd0 | sonney2k: I want to take advantage of gsoc and start doing it | 22:09 |
blackburn | hey guys | 22:10 |
@sonney2k | blackburn, I totally forgot how we did it last year | 22:10 |
blackburn | sonney2k: just some introduction mail nothing special I think | 22:10 |
@sonney2k | blackburn: every student had to send some short email introducing himself + project right? | 22:10 |
blackburn | yeap | 22:10 |
blackburn | n4nd0: puffin444: wiking: karlnapf: gsomix: could you please do this ^? | 22:10 |
blackburn | karlnapf: however you are well-known | 22:10 |
@sonney2k | but still | 22:11 |
blackburn | yes there is a need to introduce your project | 22:11 |
@sonney2k | you can look up in the mailinglist archive how these things look like | 22:11 |
n4nd0 | all right | 22:11 |
blackburn | jsut some short introduction | 22:11 |
puffin444 | blackburn: Yeah sure. | 22:11 |
@sonney2k | the other more annoying job is to send a weekly update (from the official beginning to end of gsoc) | 22:11 |
blackburn | and some feature plan would be nice | 22:11 |
@sonney2k | so another short email | 22:12 |
@sonney2k | *short* | 22:12 |
@sonney2k | cannot be said often enough | 22:12 |
blackburn | attach pics! | 22:12 |
blackburn | :) | 22:12 |
wiking | blackburn: on shogun mailing list? | 22:13 |
blackburn | wiking: right | 22:13 |
@sonney2k | yeah | 22:13 |
wiking | alrighty i'll do it tomo morning | 22:13 |
@sonney2k | wiking, thanks | 22:13 |
wiking | nw | 22:13 |
karlnapf | sonney2k, yes I just created a hp, but its not completely ready, will send you the link this week | 22:13 |
karlnapf | blackburn, I will send an introducionary mail :) | 22:14 |
@sonney2k | karlnapf, your old blog is dead too it seems | 22:14 |
blackburn | sonney2k: do you understand things will go even more crazy next days? | 22:14 |
blackburn | I'll send too | 22:14 |
@sonney2k | blackburn, why should they? | 22:14 |
blackburn | sonney2k: once active development begins it would become crazy I am sure :) | 22:14 |
karlnapf | sonney2k, yes i took it offline to replace it by something nicer | 22:15 |
* wiking wonders about the average number of users on #shogun after a month ;) | 22:15 | |
n4nd0 | :D | 22:15 |
@sonney2k | wiking, let me guess 8 :D | 22:15 |
wiking | sonney2k: + the bots! | 22:15 |
blackburn | n4nd0: your task is really important and you will definitely collaborate with wiking and michal! | 22:16 |
n4nd0 | blackburn: sounds very exciting | 22:16 |
wiking | yeah i'll most def will need a SO asap :>> | 22:16 |
@sonney2k | n4nd0, that is why you were chosen ... we have high trust in you | 22:16 |
wiking | n4nd0: SO WORK!!! | 22:16 |
n4nd0 | wiking: :D | 22:16 |
n4nd0 | sonney2k: thank you!! | 22:16 |
blackburn | however I have high trust in pluskid too ;) | 22:16 |
wiking | n4nd0: but anyhow when u start do define the new labels class let me know | 22:17 |
n4nd0 | wiking: ok | 22:17 |
@sonney2k | well actually we know that you all can do the task - that is why you got selected ;) | 22:17 |
blackburn | okay congrats you guys | 22:18 |
blackburn | I have to get back to my shity deals | 22:18 |
n4nd0 | blackburn: congrats to you too man | 22:18 |
@sonney2k | blackburn, vodka? | 22:18 |
@sonney2k | dealer? | 22:18 |
blackburn | sonney2k: not yet | 22:18 |
blackburn | sonney2k: yes I deal vodka | 22:18 |
wiking | blackburn: :DDD what's your poison ?:) | 22:18 |
@sonney2k | vodka blackburn the true soul of shogun :D | 22:19 |
blackburn | wiking: my poison is water | 22:19 |
n4nd0 | :D | 22:20 |
gsomix | blackburn, `prostavlyaisya'. :] | 22:20 |
blackburn | gsomix: you should | 22:20 |
blackburn | hehe | 22:20 |
gsomix | why? | 22:20 |
puffin444 | So just to clarify: An introductory email about me and my project tomorrow on the Shogun Mailing List, and possibly a link to a homepage? | 22:21 |
gsomix | blackburn, you should in the past year. so where? :] | 22:21 |
blackburn | puffin444: yes | 22:21 |
blackburn | gsomix: heh ok | 22:22 |
puffin444 | ok. | 22:22 |
blackburn | okay guys congrats again and see you later | 22:23 |
gsomix | I should review accepted projects. | 22:23 |
gsomix | it's very interesting | 22:23 |
blackburn | all 1212? | 22:23 |
blackburn | will take a night | 22:23 |
-!- blackburn [~qdrgsm@188.168.2.179] has quit [Quit: Leaving.] | 22:23 | |
gsomix | blackburn, aha | 22:23 |
gsomix | wow, `Category Theory Module' for SymPy. | 22:27 |
emrecelikten | gsomix: Nice | 22:29 |
n4nd0 | see you later guys | 22:29 |
emrecelikten | n4nd0: Bye! | 22:29 |
shogun-buildbot | build #731 of cmdline_static is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cmdline_static/builds/731 blamelist: sonne@debian.org, gsomix@gmail.com | 22:29 |
gsomix | :( | 22:30 |
puffin444 | ok guys, you'll be hearing from me tomorrow. See you later. | 22:32 |
gsomix | shogun-buildbot, aww, I forget about it. | 22:32 |
shogun-buildbot | build #709 of r_static is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/r_static/builds/709 blamelist: sonne@debian.org, gsomix@gmail.com | 22:33 |
-!- puffin444 [230bf329@gateway/web/freenode/ip.35.11.243.41] has left #shogun [] | 22:34 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 246 seconds] | 22:34 | |
wiking | shogun-buildbot what's up :) | 22:36 |
gsomix | wiking, my bag | 22:36 |
shogun-buildbot | build #710 of octave_static is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/octave_static/builds/710 blamelist: sonne@debian.org, gsomix@gmail.com | 22:37 |
@sonney2k | gsomix, what goes wrong? | 22:41 |
shogun-buildbot | build #697 of python_static is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/python_static/builds/697 blamelist: sonne@debian.org, gsomix@gmail.com | 22:41 |
gsomix | sonney2k, I forget about libshogun tests. | 22:42 |
@sonney2k | gsomix, these are not even tests but just examples | 22:42 |
gsomix | but I do not know what with classifier_multiclasslinearmachine | 22:42 |
gsomix | sonney2k, aha | 22:42 |
@sonney2k | they are supposed to be run as tests later but we didn't do it yet | 22:43 |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 22:54 | |
gsomix | sonney2k, I corrected conjugate example. check github | 22:58 |
-!- harshit_ [~harshit@182.68.43.52] has quit [Ping timeout: 244 seconds] | 22:58 | |
-!- PhilTillet [~Philippe@vir78-1-82-232-38-145.fbx.proxad.net] has left #shogun ["Leaving"] | 22:58 | |
CIA-64 | shogun: Evgeniy Andreev master * r51e0223 / examples/undocumented/libshogun/classifier_conjugateindex.cpp : small fix in example - http://git.io/Ci5UwQ | 22:59 |
CIA-64 | shogun: Soeren Sonnenburg master * r6ab1607 / examples/undocumented/libshogun/classifier_conjugateindex.cpp : | 22:59 |
CIA-64 | shogun: Merge pull request #483 from gsomix/to_multiclass | 22:59 |
CIA-64 | shogun: Small fix in example - http://git.io/Y19esw | 22:59 |
gsomix | sonney2k, I need to go. Thanks a lot for your choose. | 23:01 |
gsomix | good night guys | 23:01 |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has joined #shogun | 23:04 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has quit [Changing host] | 23:04 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 23:04 | |
-!- harshit_ [~harshit@182.68.43.52] has joined #shogun | 23:10 | |
-!- harshit_ [~harshit@182.68.43.52] has quit [Client Quit] | 23:15 | |
@sonney2k | nite everyone | 23:19 |
-!- karlnapf [~heiko@host86-180-223-101.range86-180.btcentralplus.com] has quit [Quit: Leaving.] | 23:26 | |
-!- av3ngr [~av3ngr@106.70.3.222] has joined #shogun | 23:50 | |
-!- nickon [~noneedtok@kotnet-147.kulnet.kuleuven.be] has quit [Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )] | 23:54 | |
--- Log closed Tue Apr 24 00:00:19 2012 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!