--- Log opened Sun Apr 28 00:00:29 2013 | ||
NP | Hey! | 00:01 |
---|---|---|
-!- NP [~NP@61.12.19.99] has left #shogun [] | 00:03 | |
-!- mikhailBelous [~towelenee@109.205.249.161] has quit [] | 00:07 | |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has quit [Ping timeout: 245 seconds] | 00:36 | |
-!- HeikoS [~heiko@90.219.232.242] has quit [Quit: Leaving.] | 01:20 | |
-!- dvalcarce [5b757556@gateway/web/freenode/ip.91.117.117.86] has quit [Quit: Page closed] | 01:21 | |
shogun-buildbot | build #373 of nightly_default is complete: Failure [failed test] Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/373 | 04:07 |
-!- kaushalb09 [~kaushalb0@cpe-74-77-131-130.buffalo.res.rr.com] has joined #shogun | 04:32 | |
kaushalb09 | hi,my name is kaushal.I have a doubt regarding issue # 863 | 04:36 |
kaushalb09 | Add sparse inverse covariance graphical example | 04:36 |
kaushalb09 | can i generate my own data or can i use the toy data provided | 04:37 |
kaushalb09 | hi , i am a GSoC applicant. is anyone there to help | 05:32 |
-!- kaushalb09 [~kaushalb0@cpe-74-77-131-130.buffalo.res.rr.com] has quit [Ping timeout: 246 seconds] | 07:01 | |
-!- kaushalb09 [~kaushalb0@cpe-74-77-131-130.buffalo.res.rr.com] has joined #shogun | 07:15 | |
-!- sids_aquarius [~sidi_@li400-124.members.linode.com] has joined #shogun | 08:37 | |
-!- gsomix [~Miranda@83.149.21.112] has joined #shogun | 08:58 | |
@wiking | sonney2k: try to turn on python_modular interface on fbsd bot... although at the current state of shogun it will fail for sure :( | 09:42 |
-!- gsomix [~Miranda@83.149.21.112] has quit [Ping timeout: 240 seconds] | 09:45 | |
-!- kaushalb09 [~kaushalb0@cpe-74-77-131-130.buffalo.res.rr.com] has quit [Ping timeout: 272 seconds] | 09:52 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 11:03 | |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has joined #shogun | 11:08 | |
-!- vivekn [318aa50a@gateway/web/freenode/ip.49.138.165.10] has joined #shogun | 11:16 | |
-!- sijin [~smuxi@144.214.222.109] has joined #shogun | 11:26 | |
vgorbati | lisitsyn: hi, there? | 11:48 |
@lisitsyn | vgorbati: yes | 11:48 |
vgorbati | lisitsyn: about t-SNE interface task, do I remember the steps correctly: 1. Add include in modular/Converter_includes.i . 2. Add rename in modular/Converter.i. 3. Add some graphical example in examples/undocumented/python_modular/graphical? | 11:50 |
@lisitsyn | vgorbati: better not graphical but simple in examples/undocumented/python_modular first | 11:51 |
-!- gsomix [~Miranda@83.149.21.112] has joined #shogun | 11:52 | |
vivekn | Hi, I am interested in implementing the Hashing Trick for GSoC, as I have used it for reducing memory usage with n-grams in one of my previous projects on sentiment classification. What are the specific features the shogun project is expecting for this? | 11:53 |
vgorbati | lisitsyn: ok. Something like converter_factoranalysis_modular.py (with FactorAnalysis replaced by t-SNE class) should work, shouldn't it? | 11:54 |
@lisitsyn | vgorbati: exactly like that | 11:54 |
vivekn | How can I start contributing and are there any entrance tasks pertaining to this idea? | 11:57 |
@lisitsyn | vivekn: no there is no particular task for this idea so feel free to contribute with anything you like | 11:59 |
deerishi | @lisitsyn :i am having some problem with the integration test | 12:02 |
deerishi | https://github.com/deerishi/KernelPCA-test-check/commit/743d3d9bc373aa36b0ea4bb7065ed36ba8706a93 | 12:03 |
deerishi | could you check it? | 12:03 |
@lisitsyn | deerishi: let me check | 12:03 |
@lisitsyn | deerishi: looks like exactly like heiko said | 12:04 |
@lisitsyn | the name of the function should be the same with the filename | 12:04 |
deerishi | which function are we talking about? | 12:04 |
deerishi | def generator? | 12:05 |
deerishi | or def get_fname? | 12:05 |
@lisitsyn | deerishi: the function in the preprocessor_kpca.py | 12:05 |
-!- vivekn [318aa50a@gateway/web/freenode/ip.49.138.165.10] has quit [Quit: Page closed] | 12:07 | |
deerishi | i renames the function in preprocessor_kpca as the same, but it still gives the same error | 12:10 |
deerishi | @lisitsyn any comments? | 12:11 |
@lisitsyn | deerishi: I haven't seen the file | 12:12 |
deerishi | i had sent a pull request https://github.com/shogun-toolbox/shogun/pull/1014 | 12:14 |
@lisitsyn | deerishi: the function and the file have different names here | 12:16 |
deerishi | this is the old pr.i tried renaming the function in the file preprocessor_kpca just now | 12:17 |
@lisitsyn | then it should work please carefully check they are the same | 12:18 |
deerishi | one more thing,should i add the name preprocessor_kpca.py to the backlist of the file generator.py | 12:20 |
deerishi | ? | 12:20 |
@lisitsyn | deerishi: only if it fails :) | 12:22 |
deerishi | i added that,which removed the error,but the 2 text files being created are broken | 12:25 |
-!- deerishi_ [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 12:28 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 12:32 | |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has quit [Ping timeout: 245 seconds] | 12:38 | |
-!- iglesiasg [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 12:39 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 12:40 | |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has joined #shogun | 13:00 | |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has quit [Ping timeout: 245 seconds] | 13:14 | |
deerishi_ | @lisitsyn : i have tried renaming the the function as the filename,but it still is giving an error.only if i write the functioname in the backlist,i do not get any error.but then the text files that are created are broken | 13:15 |
deerishi_ | what could be the problem? | 13:18 |
deerishi_ | https://github.com/shogun-toolbox/shogun/pull/1024 | 13:19 |
-!- gsomix [~Miranda@83.149.21.112] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org] | 13:38 | |
deerishi_ | ? | 14:19 |
deerishi_ | ? | 14:19 |
@iglesiasg | deerishi_: what is the error you get? | 14:33 |
deerishi_ | @iglesiasg : https://github.com/deerishi/KernelPCA-test-check/commit/743d3d9bc373aa36b0ea4bb7065ed36ba8706a93 | 14:34 |
@iglesiasg | deerishi_: have you tried tracking the line where this fails? | 14:37 |
deerishi_ | yes, | 14:37 |
deerishi_ | in the generator.py | 14:37 |
@iglesiasg | Heiko said that ir worked in his machine after renaming thoug | 14:37 |
@iglesiasg | though* | 14:37 |
deerishi_ | i tried that too. | 14:37 |
deerishi_ | i submited a pr after changing the function name | 14:38 |
@sonney2k | shogun-buildbot, force build 'cyg1 - libshogun' | 15:16 |
shogun-buildbot | build #713 forced | 15:16 |
shogun-buildbot | I'll give a shout when the build finishes | 15:16 |
-!- gsomix [~Miranda@37.61.181.11] has joined #shogun | 15:22 | |
gsomix | hi | 15:22 |
-!- sids_aquarius [~sidi_@li400-124.members.linode.com] has quit [Remote host closed the connection] | 15:27 | |
shogun-buildbot | build #713 of cyg1 - libshogun is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/713 | 15:27 |
-!- alibezz [~alibezz@177.182.164.8] has joined #shogun | 15:28 | |
alibezz | hey yall | 15:29 |
@iglesiasg | hi alibezz | 15:29 |
alibezz | I'm interested in working as an intern at GSoC 2013 | 15:29 |
alibezz | just sent my introduction to the mail list yesterday (Aline Bessa) | 15:29 |
@iglesiasg | alibezz: yeah, I read your mails :) welcome around here | 15:29 |
alibezz | iglesiasg: thanks! | 15:30 |
-!- sids_aquarius [~sidi_@14.139.122.114] has joined #shogun | 15:30 | |
@iglesiasg | alibezz: I am also applying for a GSoC project | 15:30 |
alibezz | iglesiasg: Is it necessary to solve a bug to be able to participate? | 15:30 |
@iglesiasg | alibezz: yes. You should submit a patch; that can be a bug fix, a new feature,... | 15:31 |
alibezz | iglesiasg: ok, any ideas about how I can find a good yet simple one? | 15:31 |
@iglesiasg | alibezz: there are some tasks in github issues | 15:31 |
@iglesiasg | alibezz: see for the ones labelled as entrance | 15:32 |
alibezz | iglesiasg: thanks! i have a paper deadline for Thursday, but will do my best to accomplish that. =) | 15:32 |
@iglesiasg | alibezz: cool, also you are free to submit patches after the application deadline | 15:32 |
@iglesiasg | IMHO, the more contributions you have, the stronger applicant you are | 15:33 |
alibezz | iglesiasg: chill, thanks for the advice. but it is important to have something **before** the application deadline, right? it's a pity I just became aware of GSoC late this year. | 15:33 |
@iglesiasg | alibezz: mmm I don't think it is specially important to have something **before** | 15:34 |
alibezz | iglesiasg: I'll try to figure it out. ;) thanks, iglesiasg! | 15:34 |
@iglesiasg | it is important to have it before the mentors make their decision | 15:34 |
@iglesiasg | alibezz: you are welcome :) | 15:35 |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has joined #shogun | 15:40 | |
alibezz | iglesiasg: I'm gonna work on this entrance here: https://github.com/shogun-toolbox/shogun/issues/948 :) | 16:10 |
alibezz | ...or this one https://github.com/shogun-toolbox/shogun/issues/873 | 16:12 |
alibezz | but actually, if I could find an entrance related to the dimred project http://shogun-toolbox.org/page/Events/gsoc2013_ideas#dimred, that'd be better | 16:13 |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has quit [Ping timeout: 245 seconds] | 16:21 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 16:36 | |
-!- kaushalb09 [~kaushalb0@cpe-74-77-131-130.buffalo.res.rr.com] has joined #shogun | 16:52 | |
kaushalb09 | Hi ,I am a GSoC applicant. I have few doubts regarding issue# 863. Is anyone there | 16:56 |
-!- gsomix [~Miranda@37.61.181.11] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org] | 16:58 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds] | 17:40 | |
-!- gsomix [~Miranda@185.6.239.174] has joined #shogun | 17:47 | |
gsomix | sonney2k: ping | 17:47 |
gsomix | there is no "pong" for me... | 17:57 |
@iglesiasg | alibezz: sure, just pick the one you like the most | 17:58 |
alibezz | iglesiasg: are you aware of any that is specifically related to dimensionality reduction? | 17:58 |
@iglesiasg | alibezz: mmm not really | 17:58 |
alibezz | iglesiasg: k ;) | 17:59 |
@iglesiasg | alibezz: I know some DR methods that we are missing and the matlab DR toolbox has | 17:59 |
@iglesiasg | such as maximum variance unfolding IIRC | 17:59 |
alibezz | iglesiasg: maybe that could be a starting point | 17:59 |
alibezz | iglesiasg: like, integrating these. | 17:59 |
@iglesiasg | alibezz: in my opinion it could be indeed | 17:59 |
@iglesiasg | alibezz: lisitsyn (mentor of the project) may be around and give his opinion about this | 18:00 |
alibezz | iglesiasg: k, i'm currently finishing my formal application. once i'm done, can i ping you to discuss it in further detail? | 18:00 |
alibezz | iglesiasg: oh cool, i'll ping him | 18:00 |
alibezz | iglesiasg: which project are you applying to? | 18:01 |
alibezz | iglesiasg: just out of curiosity | 18:01 |
-!- deerishi_ [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Quit: Page closed] | 18:03 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 18:03 | |
@iglesiasg | alibezz: structured learning | 18:03 |
alibezz | iglesiasg: chill. were you at NIPS last year? | 18:04 |
@lisitsyn | re | 18:19 |
@lisitsyn | alibezz: answering your mail: the sooner the better I think | 18:21 |
alibezz | lisitsyn: you mean the patch? | 18:22 |
@lisitsyn | yes | 18:22 |
@lisitsyn | we will have a pressure to decide rather fast after the deadline | 18:22 |
-!- mikhailBelous [~towelenee@109.205.249.161] has joined #shogun | 18:22 | |
@lisitsyn | alibezz: just fyi you currently have one competitor for the dimension reduction idea :) | 18:23 |
alibezz | chill. iglesiasg suggested me to try integrating some DM matlab algorithms to the shogun framework. do you think this is simple enough giving the time constraints? I'd like to send something until Thursday or Friday. Thing is: I have a conference deadline as well, so my time will be rather small. | 18:23 |
mikhailBelous | Hello, I would like to know, do you need some students who intrested in machine learning and have a huge c++ optimization expirience. Is some mentor for this task now online? | 18:24 |
alibezz | lisitsyn: I'm also interested in the mldata project, given that I have some background with accessibility | 18:24 |
@lisitsyn | mikhailBelous: huge is like? | 18:24 |
alibezz | lisitsyn: but seriously, if you could suggest me a task that you think that can be accomplished until Friday? | 18:25 |
alibezz | lisitsyn: no "if you" haha just "could..." | 18:25 |
@lisitsyn | alibezz: I'll come with something and let you know | 18:25 |
alibezz | lisitsyn: thank you VERY much! i am really interested in this project! | 18:26 |
deerishi | @lisitsyn :the integration test passed :) | 18:27 |
@lisitsyn | deerishi: nice | 18:27 |
-!- gsomix [~Miranda@185.6.239.174] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org] | 18:28 | |
@lisitsyn | mikhailBelous: gsoc is mainly about searching long-term contributors - that's why we are pushing students to contribute | 18:29 |
deerishi | turns out there was a broken file in the folder | 18:29 |
-!- sids_aquarius_ [~sidi_@li400-124.members.linode.com] has joined #shogun | 18:31 | |
-!- sids_aquarius [~sidi_@14.139.122.114] has quit [Read error: Connection reset by peer] | 18:31 | |
@lisitsyn | mikhailBelous: so if you have such optimization skills just optimize something in shogun ;) | 18:31 |
alibezz | lisitsyn: is there anything in the Issues entry at github related to the DimRed project? | 18:35 |
alibezz | lisitsyn: I was thinking of starting today ;) | 18:36 |
@lisitsyn | alibezz: I am thinking if some of them are still left | 18:36 |
alibezz | lisitsyn: yay #figerscrossed | 18:36 |
@iglesiasg | alibezz: nah, no NIPS nor any other conference/journal for me yet :( | 18:37 |
@iglesiasg | alibezz: what about you? | 18:37 |
alibezz | alibezz: I've been in NIPS once. but not because of a paper. haha I submitted a project to Women in Machine Learning and it got approved. It was co-located with NIPS so I went to both | 18:38 |
alibezz | iglesiasg: DUH i wrote alibezz instead of your name. haha | 18:38 |
@iglesiasg | alibezz: nice! | 18:38 |
@iglesiasg | yeah | 18:38 |
@iglesiasg | lisitsyn: you see? I am not the only one around here talking to himself now :D | 18:39 |
@lisitsyn | iglesiasg: haha | 18:39 |
@lisitsyn | nice | 18:39 |
alibezz | lolz | 18:39 |
@lisitsyn | alibezz: okay what about sammon mapping? | 18:39 |
@lisitsyn | I can describe it quite detailed | 18:39 |
alibezz | lisitsyn: woot go for it please ;) | 18:40 |
@lisitsyn | alibezz: matlab code is available | 18:40 |
alibezz | lisitsyn: where's this issue? the sammon mapping one? on github? | 18:40 |
@lisitsyn | alibezz: writing it | 18:40 |
@iglesiasg | alibezz: just out of curiosity, what's about the paper you are preparing for this upcoming conference? | 18:44 |
alibezz | iglesiasg: according to the field of behavioral economics, everytime one looks at a list of items and is supposed to choose one of them, it does not depend only on the individual properties of each one of the items, but also on the universe of presented alternatives. so i model this mutual influence items exert on each other as pairwise factors, and individual qualities as individual factors. one can see it as a markov network, and then perform MAP inferenc | 18:48 |
@iglesiasg | alibezz: it sounds interesting | 18:49 |
alibezz | iglesiasg: thx! | 18:49 |
@iglesiasg | alibezz: do you have to perform parameter learning for the MRF too or are they given by some expert knowledge? | 18:50 |
alibezz | iglesiasg: actually, all factors are either spit out from a individual-item recommender learner, such as SVD++, and all pairwise factors are computed in a rather naive way so far, via maximum likelihood. I didn't go too far with the MRF anyways because it was too costly. I just show some experiments that illustrate that and show how the bayesian approximation is way faster, and also rather good, in practice. | 18:52 |
-!- sids_aquarius_ is now known as sids_aquarius | 18:53 | |
-!- gsomix [~Miranda@188.168.3.32] has joined #shogun | 18:53 | |
@iglesiasg | alibezz: I see | 18:53 |
@lisitsyn | alibezz: https://github.com/shogun-toolbox/shogun/issues/1025 | 18:56 |
alibezz | lisitsyn: chill, thx. i'll check it out asap. | 18:56 |
@lisitsyn | oops forgot one step | 18:56 |
@iglesiasg | lisitsyn: oh that's a nice description | 18:57 |
@iglesiasg | directly to the game :) | 18:57 |
@lisitsyn | iglesiasg: I start to understand nobody give a damn about my code :D | 18:58 |
@iglesiasg | lisitsyn: c'mon, I don't think so | 18:59 |
@lisitsyn | alibezz: I tried to make it detailed as you are out of time | 19:03 |
@lisitsyn | but ask if something is not clear | 19:03 |
@lisitsyn | alibezz: iglesiasg should be able to answer some questions too I think | 19:03 |
@lisitsyn | lala you are out touch I am out of time lala | 19:03 |
@lisitsyn | iglesiasg: know that song? | 19:04 |
@lisitsyn | from the vice city :D | 19:04 |
alibezz | lisitsyn: thank you so much for being so understanding! in South America, this time of the year is always full of things... we don't actually don't have summer vacation in school here at this point of time because... uh... it's not summer! haha | 19:04 |
@iglesiasg | lisitsyn: mmm I don't think I have heard it | 19:04 |
alibezz | lisitsyn: iglesiasg: you guys are fun hahaha chill | 19:04 |
@lisitsyn | iglesiasg: hall and oates | 19:04 |
alibezz | iglesiasg: lisitsyn: yuppppp gotta dance like crazy with these guys. | 19:05 |
@lisitsyn | http://www.youtube.com/watch?feature=player_detailpage&v=s_8KR-n2fBQ iglesiasg | 19:05 |
@iglesiasg | lisitsyn: dafuq | 19:05 |
@lisitsyn | iglesiasg: I actually do like it more than some of ehm modern music | 19:06 |
@iglesiasg | heh | 19:07 |
alibezz | iglesiasg: lisitsyn: the description of the issue is mad detailed yall. thanks!!!! | 19:09 |
alibezz | iglesiasg: lisitsyn: if you guys like music, you might enjoy this one. just guessing... hahaha http://www.youtube.com/watch?feature=player_embedded&v=T5Cp55MvX54 | 19:11 |
@lisitsyn | get lucky is nice yes | 19:11 |
alibezz | iglesiasg: lisitsyn: one has to try ;) | 19:11 |
alibezz | lisitsyn: but this is a chill cover for rainy days or so. ha | 19:11 |
@lisitsyn | yeah listening | 19:11 |
@iglesiasg | lisitsyn: maybe you can ask alibezz about music in Portuguese if you are still into it :) | 19:12 |
@lisitsyn | haha yes | 19:12 |
@lisitsyn | not sure what should I ask though | 19:12 |
alibezz | lisitsyn: hell yea. I'm a big music nerd actually. hahaha and i love the music of my country, especially non-cliche shit | 19:12 |
@iglesiasg | alibezz: the song is nice idd | 19:12 |
@lisitsyn | alibezz: I am a big fan of bossa like antonio carlos jobim | 19:13 |
@lisitsyn | seu jorge, etc | 19:13 |
alibezz | lisitsyn: hey do you know Astrud Gilberto? | 19:13 |
@lisitsyn | of course | 19:13 |
-!- sids_aquarius [~sidi_@li400-124.members.linode.com] has left #shogun ["Off to code | sleep"] | 19:14 | |
@lisitsyn | caetano veloso | 19:14 |
@lisitsyn | to name a few | 19:14 |
@lisitsyn | :D | 19:14 |
@lisitsyn | alibezz: this thing rocks http://www.youtube.com/watch?feature=player_detailpage&v=Q7xZ3cBPxIE | 19:14 |
alibezz | tim maia? erasmo carlos? | 19:14 |
@lisitsyn | I have seen 'tim maia' name somewhere in brazil drum'n'bass collections | 19:15 |
@lisitsyn | no idea about erasmo carlos though | 19:15 |
alibezz | gonna think of something good to show ya. bout to eat something now. it's lunch time here. lemme know where you guys are from! | 19:20 |
deerishi | @lisitsyn : how do go about merging?now that the integration test has passed | 19:20 |
@lisitsyn | deerishi: let Heiko merge it | 19:20 |
@lisitsyn | alibezz: russia | 19:20 |
@iglesiasg | alibezz: Spain, hidden in Sweden | 19:21 |
-!- kaushalb09 [~kaushalb0@cpe-74-77-131-130.buffalo.res.rr.com] has quit [Ping timeout: 260 seconds] | 19:23 | |
-!- chintak [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 19:23 | |
-!- chintak [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Client Quit] | 19:23 | |
* gsomix missed all fun. Again. :( | 19:26 | |
-!- kaushalb09 [~kaushalb0@cpe-74-77-131-130.buffalo.res.rr.com] has joined #shogun | 19:36 | |
-!- HeikoS [~heiko@90.219.232.242] has joined #shogun | 19:36 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 19:36 | |
@lisitsyn | HeikoS: hey could you please take a look at deerishi PR? | 19:41 |
@HeikoS | lisitsyn: yes, hwy? | 19:41 |
@HeikoS | why | 19:41 |
deerishi | i 'll will send the revised pr in a few minutes | 19:41 |
@HeikoS | deerishi: ah hi | 19:41 |
deerishi | integration test passes | 19:41 |
@HeikoS | I just sent you an email | 19:41 |
@HeikoS | did you manage to create a new test file? | 19:41 |
alibezz | lisitsyn: iglesiasg: Could you please update the sammon mapping issue assigning it to me? | 19:42 |
@lisitsyn | alibezz: just write you will take care | 19:42 |
alibezz | ok! | 19:42 |
@lisitsyn | I can't assign it to arbitrary person | 19:42 |
@HeikoS | deerishi: so if the generator worked, try the tester.py with the same argument, if this passes, commit the test data and the data version in the development repository | 19:42 |
@lisitsyn | only shogun-toolbox members are available there unfortunately | 19:42 |
deerishi | @HeikoS : yes, 2 text files were created as you said | 19:42 |
@HeikoS | deerishi: nice, so add them to the commit | 19:43 |
@HeikoS | and also add "data" in the main repo | 19:43 |
alibezz | lisitsyn: no worries | 19:43 |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 19:49 | |
mikhailBelous | Can you suggest any library that should be optimized? | 19:49 |
@iglesiasg | mikhailBelous: what do you mean? | 19:50 |
mikhailBelous | I wanna to do some contribution in optimization field, can you suggested something | 19:51 |
@iglesiasg | within Shogun, or are you talking in general? | 19:52 |
mikhailBelous | Shogun | 19:52 |
@iglesiasg | ah ok so math optimization I guess you are talking about | 19:52 |
mikhailBelous | Or worktime optimization | 19:53 |
@iglesiasg | mikhailBelous: I implemented a method that you can find in structure/PrimalMosekSOSVM | 19:53 |
deerishi | @HeikoS : i am having some problem in pushing the repo | 19:53 |
@iglesiasg | mikhailBelous: which needs Mosek | 19:53 |
@iglesiasg | mikhailBelous: it would be nice if we could drop this requirement by adding a QP solver with box contraints | 19:54 |
@HeikoS | lisitsyn: how does it work if someone who is not core dev wants to push to data? | 19:54 |
deerishi | i could not clone the repository,so i downloaded the as zip folder and installed shogun | 19:54 |
@iglesiasg | mikhailBelous: I can let you know more about the QP it needs to solve if you are interested | 19:54 |
deerishi | now how do i push the entire thing into a particular branch of the repository i had forked | 19:54 |
mikhailBelous | It will be nice | 19:54 |
@iglesiasg | mikhailBelous: I wrote a bit about it here http://iglesiashogun.wordpress.com/2012/06/04/second-weekly-report-gsoc-2012/ | 19:55 |
@lisitsyn | HeikoS: PR | 19:57 |
@lisitsyn | as usual | 19:57 |
@lisitsyn | :) | 19:57 |
@HeikoS | lisitsyn: so push to the forked data repo? | 19:57 |
@HeikoS | and then PR the data? | 19:57 |
@lisitsyn | yes | 19:57 |
@iglesiasg | mikhailBelous: in that post there is some application dependent information on structured learning | 19:57 |
@lisitsyn | the data PR should be merged first HeikoS | 19:57 |
@HeikoS | well thats easy, deerishi, see above | 19:57 |
@iglesiasg | mikhailBelous: but the idea is basically to get a solver like http://docs.mosek.com/6.0/capi/node007.html#250330792 | 19:57 |
@HeikoS | lisitsyn: thanks for the tip! | 19:57 |
@HeikoS | indeed thats a good idea .) | 19:57 |
@iglesiasg | mikhailBelous: however I have just used that as a black box, I have no idea how complex can it be to implement that. It might even a better idea to find some open source code able to do that and integrate it | 19:58 |
@lisitsyn | mikhailBelous: you may just learn any linear machine and benchmark how it computes the output on the test data | 20:01 |
@lisitsyn | and make it faster if you have any ideas | 20:01 |
@lisitsyn | without the loss of flexibility and crazy refactoring | 20:01 |
mikhailBelous | It looks simplier than qp realization | 20:02 |
@iglesiasg | mikhailBelous: It may be a good idea to start off with something simpler ;) | 20:03 |
@lisitsyn | I wouldn't expect it is feasible to come up with something much better than libqp in a reasonable time | 20:04 |
deerishi | @HeikoS: will do that ,and let you know soon | 20:04 |
mikhailBelous | But I think that QP can be solved by simple gradient method. But I may be wrong | 20:04 |
@iglesiasg | lisitsyn: the problem with libqp and this problem are the constraints | 20:06 |
@iglesiasg | for box constraints, it only handles an equlity contrainst | 20:06 |
@lisitsyn | all these libqp, mosek etc have been under development for a long long time | 20:06 |
@iglesiasg | while this problem needs inequalities | 20:06 |
@iglesiasg | IRRC | 20:06 |
@lisitsyn | so no that's not simple at all | 20:07 |
@iglesiasg | indeed | 20:07 |
@HeikoS | deerishi: nice, Ill have a look tomorrow though since got some friends now visiting, see you! | 20:09 |
mikhailBelous | We have convex function on compact space, why it would be solved by gradient method? | 20:09 |
-!- HeikoS [~heiko@90.219.232.242] has quit [Quit: Leaving.] | 20:09 | |
mikhailBelous | *would not | 20:09 |
@lisitsyn | mikhailBelous: they are all gradient based but they are not simple at all | 20:11 |
mikhailBelous | Okay, let me suggest this solution. | 20:11 |
@lisitsyn | it matters a lot how general the problem you solve is | 20:12 |
@lisitsyn | libqp is quite general, liblinear optimizer is less general | 20:13 |
mikhailBelous | Do you or do you not have a solution for this problem(math not code)? | 20:13 |
@lisitsyn | mikhailBelous: which problem? | 20:14 |
mikhailBelous | Quadratic optimization with inequalities boundaries | 20:15 |
@lisitsyn | in libqp this method is implemented if you are interested: http://www.recognition.mccme.ru/pub/papers/SVM/keerthi00convergence.pdf | 20:15 |
@lisitsyn | I am not sure what the method the mosek stuff implements | 20:18 |
@iglesiasg | me neither actually | 20:19 |
@lisitsyn | it depends on the formulation | 20:19 |
@lisitsyn | iglesiasg: is the qp problem in SO the same as in basic svm? | 20:19 |
@iglesiasg | lisitsyn: no, I don't think so. Let me ensure it | 20:20 |
@iglesiasg | lisitsyn: this version achieved with the cutting plane problem it is I think | 20:21 |
@iglesiasg | lisitsyn: the problem with SO is that you have a pretty large number of constraints, using a cutting plane algorithm you select the active ones and then you need to solve a QP | 20:22 |
@iglesiasg | this QP is pretty much like Crammer-Singer with the margin scaled | 20:23 |
@lisitsyn | oh yes yes | 20:24 |
@lisitsyn | C-S is the thing that made SO | 20:24 |
@lisitsyn | :) | 20:24 |
@iglesiasg | :) | 20:25 |
@lisitsyn | mikhailBelous: what is the function which gradient you would use there btw? | 20:27 |
mikhailBelous | To target function | 20:32 |
@lisitsyn | but this doesn't handle the constraints right? | 20:33 |
@lisitsyn | I mean we have to use the lagrange function there as we have constraints | 20:33 |
mikhailBelous | We go along gradient vector until we reach boundary | 20:33 |
mikhailBelous | What is lagrange fucnction when we use inequalities? | 20:34 |
@lisitsyn | mikhailBelous: please check the paper I send to you - QP is usually formulated through its dual form | 20:37 |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds] | 20:49 | |
-!- vivekn [65d75db0@gateway/web/freenode/ip.101.215.93.176] has joined #shogun | 20:49 | |
-!- mikhailBelous [~towelenee@109.205.249.161] has quit [Ping timeout: 260 seconds] | 20:52 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 20:55 | |
alibezz | lisitsyn: I just submitted my GSoC proposal. If you could take a look at it when you have time and tell me if it's ok, I'd be glad!! | 20:56 |
alibezz | lisitsyn: http://www.google-melange.com/gsoc/proposal/review/google/gsoc2013/alibezz/1 | 20:56 |
-!- mikhailBelous [~towelenee@109.205.249.161] has joined #shogun | 20:57 | |
@iglesiasg | alibezz: I think that if you want to apply for two different ideas you need to submit two proposals | 20:57 |
@iglesiasg | even if it is for the same org | 20:57 |
@iglesiasg | just mentioning because I saw this Idea 1 and Idea 2, sorry if you already knew that :) | 20:58 |
alibezz | iglesiasg: I think I'll just apply for the DimRed project. I'm much more interested in this one! | 20:58 |
alibezz | iglesiasg: ok, I'll fix this later | 20:58 |
alibezz | ;) | 20:58 |
alibezz | later today. | 20:58 |
mikhailBelous | Please, give me an advise how to watch project in some using linux? | 20:58 |
@iglesiasg | mikhailBelous: what do you mean? | 20:58 |
mikhailBelous | I need IDE in order to read code, what do you use in linux? | 20:59 |
gsomix | mikhailBelous: vim | 21:00 |
@iglesiasg | mikhailBelous: ah ok. I am not an IDE guy, I use terminal and vim as editor | 21:00 |
@iglesiasg | mikhailBelous: eclipse of code::blocks might be a choice | 21:01 |
mikhailBelous | And how do you find function declaration and implementation? | 21:03 |
@iglesiasg | mikhailBelous: I grep all the way :D | 21:03 |
@lisitsyn | mikhailBelous: ctags + vim supports that quite ok | 21:04 |
@lisitsyn | eclipse supports that natively so feel free to use it | 21:04 |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has joined #shogun | 21:05 | |
alibezz | iglesiasg: lisitsyn: fixed! this is the current version of my application http://www.google-melange.com/gsoc/proposal/review/google/gsoc2013/alibezz/1 if you guys have any suggestions, please let me know | 21:07 |
@lisitsyn | alibezz: bahh you need a new notebook ;) | 21:09 |
@iglesiasg | hehe | 21:10 |
alibezz | lisitsyn: seriously? well, I can always ssh a better machine in my university | 21:10 |
@lisitsyn | alibezz: just kidding | 21:10 |
@lisitsyn | I mean it is pretty old :) | 21:10 |
alibezz | lisitsyn: lolz YEAH IT IS hahaha my newer one broke. | 21:10 |
@iglesiasg | too bad | 21:11 |
alibezz | lisitsyn: and now i myself am super low on cash... hahaha | 21:11 |
alibezz | lisitsyn: gonna buy a new one, maybe a mac-os, when I make some dough. ;) | 21:11 |
vgorbati | lisitsyn: hi there. So I implemented all the steps, also run generator.py on the created examle (it succeeded). But, when I run tests, my new example gives an 'EXCEPTION No module named modshogun', what is kind a strange, regarding that all other tests work ok. Any suggestions what this can be related to? | 21:12 |
@lisitsyn | vgorbati: hey | 21:12 |
@lisitsyn | not really | 21:12 |
gsomix | alibezz: interesting proposal :) | 21:15 |
@lisitsyn | alibezz: vgorbati: fight! :D | 21:16 |
@lisitsyn | you are the competitors | 21:16 |
@lisitsyn | *mortal kombat theme* | 21:16 |
alibezz | hahahahah well i might be losing already because I just arrived. | 21:16 |
@iglesiasg | so epic one | 21:16 |
@lisitsyn | alibezz: you submitted the proposal first though | 21:17 |
alibezz | well, no I'm gonna work on my paper, and then on the issue! ;) | 21:18 |
alibezz | see yall later! | 21:18 |
@lisitsyn | see you | 21:18 |
@iglesiasg | bye bye | 21:18 |
-!- alibezz [~alibezz@177.182.164.8] has left #shogun [] | 21:18 | |
@lisitsyn | iglesiasg: vgorbati: a puzzle! | 21:19 |
@lisitsyn | https://github.com/shogun-toolbox/shogun/blob/develop/src/shogun/lib/tapkee/traits/callbacks_traits.hpp | 21:19 |
@lisitsyn | what is the C++ idiom here in lines 20-31? | 21:19 |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 21:20 | |
@iglesiasg | haha no idea -- let me see if I come up with anything to say :D | 21:20 |
@iglesiasg | the dummy pattern | 21:20 |
@lisitsyn | haha | 21:21 |
@lisitsyn | no | 21:21 |
@iglesiasg | lisitsyn: so what is it? | 21:26 |
@lisitsyn | iglesiasg: sfinae | 21:26 |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Quit: Page closed] | 21:40 | |
vgorbati | lisitsyn: my mind is wrecked:) The purpose of this code is to determine whether a type T has a 'dummy' inside it, isn't it? | 21:40 |
@lisitsyn | vgorbati: yes | 21:40 |
vgorbati | lisitsyn: that is why I do not like C++) | 21:41 |
@lisitsyn | why? | 21:41 |
vgorbati | lisitsyn: there are too much possibilities to shoot yourself in the foot:) | 21:41 |
@lisitsyn | it is important that it is compile-time | 21:41 |
@lisitsyn | TBH I am not in the trend of C++ haters :D | 21:43 |
* sonney2k qualifies to be a java hater | 21:43 | |
@lisitsyn | hahah | 21:43 |
@lisitsyn | well C++ sucks when you don't get it | 21:43 |
@lisitsyn | that's for sure | 21:43 |
@lisitsyn | sucks much more than everything else probably | 21:44 |
gsomix | sonney2k: hey! | 21:44 |
@iglesiasg | we are all matlab haters, aren't we? :D | 21:44 |
gsomix | iglesiasg: nope | 21:44 |
gsomix | :3 | 21:44 |
@iglesiasg | gsomix: but.. but, python! | 21:44 |
vgorbati | lisitsyn: maybe it is like a very sophisticated tool, but personally I prefer understandable tools | 21:45 |
gsomix | iglesiasg: there is not python in my lab. | 21:45 |
gsomix | *no | 21:45 |
@lisitsyn | vgorbati: what is understandable for you? | 21:45 |
@iglesiasg | gsomix: wow so the pay for matlab licenses but don't download python? funny | 21:46 |
vgorbati | lisitsyn: hard to explain:) python is understandable tool, pure C is understandable:) | 21:47 |
@lisitsyn | iglesiasg: pay? I don't think so :D | 21:47 |
vgorbati | lisitsyn: the only thing that I miss in pure C is templates | 21:47 |
@iglesiasg | hehe | 21:47 |
@lisitsyn | vgorbati: what else is left? classes/ | 21:48 |
gsomix | iglesiasg: so, such is the objective reality | 21:48 |
gsomix | lisitsyn: tsssss | 21:48 |
@sonney2k | gsomix, hey | 21:48 |
gsomix | sonney2k: yay! glad to see you | 21:48 |
gsomix | sonney2k: have a minute? | 21:48 |
gsomix | or second... | 21:49 |
vgorbati | lisitsyn: personally I do not see a big difference between calling myVector.sort() and sort(myVector). | 21:49 |
@lisitsyn | vgorbati: for big projects it matters a lot | 21:49 |
gsomix | lisitsyn: vgorbati: http://www.youtube.com/watch?v=I6knCdiUBz4 | 21:50 |
@sonney2k | gsomix, ask | 21:50 |
vgorbati | lisitsyn: maybe, maybe:) btw, have you seen the fake Stroustrup interview?:) | 21:50 |
gsomix | sonney2k: https://dl.dropboxusercontent.com/u/19029407/ideas%20for%20gsoc%202013.txt | 21:50 |
@lisitsyn | vgorbati: which? | 21:50 |
vgorbati | lisitsyn: http://www.snopes.com/computer/program/stroustrup.asp this one:) | 21:51 |
vgorbati | lisitsyn: even though it is fake, it is kind a interesting:)_ | 21:51 |
@sonney2k | gsomix, all certainly valid problems | 21:51 |
@sonney2k | gsomix, but might be too difficult | 21:51 |
@sonney2k | gsomix, e.g. r-modular is not sth I think you can fix nor the swig -> multiple modules | 21:52 |
@lisitsyn | hah vgorbati I do not get the point of hating C++ a lot | 21:52 |
@lisitsyn | I always hear things like 'I do not like pointers' | 21:53 |
@lisitsyn | or 'it is very complex' | 21:53 |
@lisitsyn | but nobody forces to use it, right? :D | 21:53 |
-!- mikhailBelous [~towelenee@109.205.249.161] has quit [] | 21:54 | |
-!- vivekn [65d75db0@gateway/web/freenode/ip.101.215.93.176] has quit [Ping timeout: 245 seconds] | 21:54 | |
vgorbati | lisitsyn: well, I do not actually hate it, I just think that some thing could be done better, that's all:) Like in python (and when you will need the speed - just implement this part in C and use it as a linked library) | 21:55 |
@lisitsyn | vgorbati: do you really have any improvement to C++? | 21:55 |
@lisitsyn | some things do feel wrong (like syntax) | 21:56 |
vgorbati | lisitsyn: you mean python vs C++? | 21:56 |
gsomix | sonney2k: sad. are there another ideas for project? | 21:56 |
@lisitsyn | but I do not have any suggestion at all | 21:56 |
vgorbati | lisitsyn: actually yes, I guess I am mainly concerned about syntax - after python I always fill that I write too much code when in C++) | 21:58 |
@sonney2k | vgorbati, I totally agree | 21:59 |
@sonney2k | gsomix, well we need file i/o stuff | 21:59 |
@sonney2k | not sure if this is of interest to you | 21:59 |
@lisitsyn | well that's not the intrinsic property of C++ | 21:59 |
@sonney2k | lisitsyn, c'mon it is | 21:59 |
@lisitsyn | sonney2k: example? | 22:00 |
@sonney2k | python is like 10 characters and c++ would be 1000 lines | 22:00 |
@lisitsyn | if you avoid libraries or what? | 22:00 |
gsomix | sonney2k: http://shogun-toolbox.org/page/Events/gsoc2013_ideas#fastio this? | 22:00 |
@sonney2k | gsomix, yes | 22:00 |
@sonney2k | lisitsyn, sure I am exaggerating but its like it | 22:01 |
@lisitsyn | sonney2k: usually people just do not exploit capabilities and write C+- instead of C++ | 22:02 |
gsomix | sonney2k: hm, nice project | 22:02 |
@lisitsyn | sonney2k: for example people keep writing inner loops to say compute vector norm | 22:05 |
@lisitsyn | it is not only eigen that provides such things but blitz++ and armadillo and etc | 22:05 |
vgorbati | lisitsyn: I guess, I've just lost one point as a GSoC applicant :) | 22:07 |
@lisitsyn | sonney2k: furthermore you are assuming stl is wrong - but it does reduce the # of LoC | 22:07 |
@lisitsyn | vgorbati: no, why? | 22:08 |
vgorbati | lisitsyn: because I do not like C++ as much as you do:) just kidding) | 22:08 |
@lisitsyn | vgorbati: am I stalin or what? :D | 22:09 |
@lisitsyn | vgorbati: I do like all the languages they are all cool | 22:09 |
gsomix | sonney2k: I think it's interesting idea. should/can I add something in it from my list of ideas? | 22:11 |
@iglesiasg | ok, one more application to the batch :) | 22:11 |
vgorbati | lisitsyn: let's stop the holywar on this point:) | 22:11 |
@lisitsyn | haha | 22:11 |
@lisitsyn | sonney2k: we with iglesiasg do need a quick idea for some conference paper :D | 22:15 |
@iglesiasg | lisitsyn: coffin is structured svm is the way :) | 22:19 |
@lisitsyn | I do not get how it can help it though | 22:19 |
@lisitsyn | iglesiasg: http://cs409823.vk.me/v409823955/1836/D_Ps3YSxsak.jpg new photo from the city of accident black holes | 22:21 |
vgorbati | lisitsyn: ok, it looks like I've managed to finish the task - the test passes. Is it ok if I submit a PR now? | 22:22 |
@lisitsyn | vgorbati: yes | 22:22 |
@lisitsyn | vgorbati: btw t-sne has local minima so I'd blacklist it for now | 22:23 |
@iglesiasg | lisitsyn: wtf | 22:23 |
vgorbati | lisitsyn: but then we need a totally different test:) | 22:24 |
@iglesiasg | lisitsyn: you could help in lot of stuff for it. You understand coffin, you are a coding master, etc :) | 22:24 |
@lisitsyn | vgorbati: yes - all converters are blacklisted IIRC | 22:24 |
@lisitsyn | vgorbati: it is only PCA that produces the same result and even it can vary a little | 22:25 |
vgorbati | lisitsyn: :D yes, I guess some tests are 'blacklisted' in interesting manner - they call 'converter.apply(features)' and then 'return features' - that's how the result will be similar every time:) | 22:27 |
vgorbati | lisitsyn: but at list they are checking, whether the apply method does not call an exception | 22:27 |
vgorbati | lisitsyn: is it ok if my test will be 'black-listed' in a similar manner? | 22:28 |
@lisitsyn | vgorbati: just blacklist it for real | 22:28 |
vgorbati | lisitsyn: ok | 22:29 |
@lisitsyn | however | 22:29 |
@lisitsyn | no idea | 22:29 |
@lisitsyn | rather do a PR now | 22:29 |
@lisitsyn | :) | 22:29 |
@lisitsyn | but yes, probably blacklisting it from the very beginning is the way | 22:29 |
vgorbati | lisitsyn: ok, but currently test works as I described - it return original features - that is how it passes | 22:29 |
@lisitsyn | vgorbati: no, it produces the same result then | 22:30 |
@lisitsyn | returned features are serialized with generator.py | 22:30 |
@lisitsyn | then de-serialized and checked | 22:30 |
vgorbati | lisitsyn: yes, and the test passes (if only the apply method throws an exception) | 22:31 |
@lisitsyn | if it was producing the different result | 22:31 |
@lisitsyn | it would fail | 22:31 |
vgorbati | lisitsyn: does not throw* | 22:31 |
vgorbati | lisitsyn: the thing is, that calling 'converter.apply(features)' does not change the 'features' itself - it returns the embedding. So, 'return features' will just return original data every time - that's why the test won't fail | 22:33 |
vgorbati | lisitsyn: anyway, I got it - I will add it to blacklist and send a PR | 22:34 |
@lisitsyn | ahhh | 22:34 |
@lisitsyn | got it | 22:34 |
@lisitsyn | yes you right | 22:34 |
@lisitsyn | vgorbati: it came from the old time when it did change the feautres | 22:35 |
vgorbati | lisitsyn: so this tests should probably be refactored? | 22:35 |
@lisitsyn | well they won't start working as tests anyway | 22:36 |
@lisitsyn | vgorbati: even for say LLE | 22:36 |
@lisitsyn | all the features in the embedded space | 22:37 |
@lisitsyn | can be both multiplied by +1 or -1 | 22:37 |
@lisitsyn | and it will be correct | 22:37 |
@lisitsyn | I mean sign of the eigenvector doesn't matter | 22:37 |
vgorbati | lisitsyn: I meant that they should be gradually refactored so that they actually do return an embedding instead of original data + that they will be tested 'the right way' | 22:38 |
@lisitsyn | yes but sounds like a long way | 22:39 |
@lisitsyn | and it actually could become a unit test instead of integration test :D | 22:39 |
vgorbati | lisitsyn: So the integration test can only test for equality of results, right? | 22:39 |
@lisitsyn | no but it is supposed to do something like that | 22:40 |
vgorbati | lisitsyn: ok, I got to go now - will submit a blacklisted test tomorrow | 22:40 |
@lisitsyn | alright | 22:40 |
vgorbati | see you | 22:41 |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has quit [Quit: Page closed] | 22:41 | |
@lisitsyn | 10 proposals so far | 22:41 |
@iglesiasg | good number | 22:53 |
--- Log closed Mon Apr 29 00:00:29 2013 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!