IRC logs of #shogun for Sunday, 2011-04-10

--- Log opened Sun Apr 10 00:00:36 2011
-!- argetlahmsource [~argetlahm@adsl-99-160-142-89.dsl.euclwi.sbcglobal.net] has left #shogun []00:01
-!- yayo3 [~jakub@ip-78-45-113-245.net.upcbroadband.cz] has quit [Quit: leaving]00:17
-!- aifargonos [~aifargono@46.18.27.35] has left #shogun []00:32
@sonney2kblackburn, would you mind if I go on vacations and you take over the project in the meantime :D00:43
blackburnsonney2k: what do you mean?00:44
@sonney2kblackburn, you seem to mentor the students quite fine.00:44
blackburnahaha00:44
blackburnsonney2k: I hope not getting mentor :))00:45
blackburnsonney2k: anyway, what should I do?00:45
@sonney2kblackburn, all good00:47
@sonney2kit is just nice that you respond - thats all00:47
blackburnsonney2k: ah, joking, see now :)00:48
@bettybooblackburn, rahaha00:48
blackburnsonney2k: EKernelType max is 380 for now, was 300 IIRC :D00:50
blackburnsonney2k: it is 'kernel competition'00:50
alesis-novikHa, that one mail sparked all this00:51
@sonney2ksomeone should send a pointer to unrelated (easy) ml algorithms00:52
@sonney2khmmhh maybe some directory of the Weka toolbox?00:53
blackburnsonney2k: do you really want to continue this?00:53
blackburnsonney2k: with that pace all the GSoC ideas will be complete in next week :D00:55
@sonney2kblackburn, well - it is weekend now and only very few students did submit a patch so far. It is better the other interested ones  just don't run out of 'easy' ideas...00:55
@sonney2kblackburn, heh00:55
@sonney2kI am confident that some of the ideas take > 1 month of full time coding00:56
@sonney2knot just overriding the 'compute' function00:56
blackburnsonney2k: just joking :)00:56
@sonney2kwith some few lines of formula00:56
@sonney2kyes yes :)00:56
@bettyboohihi ;>00:56
blackburnsonney2k: will you have time to evaluate proposals and not only contibutions? :)00:57
@sonney2kblackburn, I am just going through them00:59
@sonney2k13 of 70 scored so far00:59
@sonney2k*sigh*01:00
blackburneh, I see01:01
blackburnsonney2k: is my in 13? :)01:02
@sonney2ksome applications go quick others really take time01:02
@bettybooblackburn, HA01:02
blackburnsonney2k: what's with man with 11 proposals? :) seems like his proposal disappeared01:04
@sonney2kblackburn, one can withdraw a proposal...01:05
blackburnI know, did he?01:05
blackburnwhy I am asking.. whatever :)01:06
@bettybooho? blackburn01:06
blackburnbettyboo: make doc01:06
@bettybooblackburn: i dont know how to make a doc tell me how to make one01:06
blackburnbettyboo: you have to01:06
@bettybooblackburn: over 60 students applied so far - there is now way to make a fair ranking - and we seriously cannot handly more than 10 of you :`-(01:06
blackburnsonney2k: I could mail list of some algorithms to implement tomorrow01:09
blackburnor you should, don't know :)01:09
@sonney2kblackburn, you are the boss :)01:11
blackburnsonney2k: I'm not :)01:11
@bettyboo;> blackburn01:12
@sonney2kbut seriously01:12
@sonney2kanother email with some other easy tasks would help01:12
blackburnjust helping you a bit01:12
@sonney2kand I am thankful01:15
blackburnokay, will send if mind something easy to implement01:15
blackburnsonney2k: 12 forks so far01:16
blackburnand kernels are almost exhausted :)01:17
blackburnsonney2k: it could be a 5-round competition: 1 round - kernels, .... , 5 round - most efficient SVM :D01:19
-!- skydiver [c255a037@gateway/web/freenode/ip.194.85.160.55] has joined #shogun01:21
CIA-31shogun: Shashwat Lal Das master * r7a3ed87 / (4 files in 2 dirs): Added spherical kernel. - http://bit.ly/i1SA8101:38
CIA-31shogun: Shashwat Lal Das master * rdd967ce / (2 files): Fixed constructor in spherical kernel - http://bit.ly/hxrtBg01:38
CIA-31shogun: Shashwat Lal Das master * r662d190 / (src/libshogun/kernel/Kernel.cpp src/modular/Kernel.i): Add K_SPHERICAL to kernel files - http://bit.ly/eqk5En01:38
CIA-31shogun: Shashwat Lal Das master * r54a2581 / (2 files): Modify copyrights in sphericalkernel - http://bit.ly/erEKD201:38
CIA-31shogun: Shashwat Lal Das master * re49b406 / src/libshogun/kernel/SphericalKernel.h : Fixed get_name in SphericalKernel - http://bit.ly/gTHDMc01:38
dvevresonney2k: hehe... thank you for the extreme patience :)01:39
@bettyboo;>01:39
@sonney2kdvevre, you intend to do the vw - sg integration right?01:43
dvevresonney2k: yeah01:44
@sonney2kI was just reading your proposal.01:44
dvevre*shivers*01:44
@sonney2kwell 10 more mentors have to read it too01:44
dvevreso you have any advice for me?01:45
dvevrei mean on stuff i can start working on01:45
@sonney2kdvevre, it would be nice if you could do something already in that direction01:45
-!- epps [~epps@unaffiliated/epps] has quit [Ping timeout: 258 seconds]01:46
blackburneh, it is becoming not fair! :)01:46
@bettyboo;D01:46
@sonney2kI mean for example a streaming feature class (yet very simple) and show some proof of concept code01:46
@sonney2kblackburn, you have a totally different project so you both are not competing anyways01:46
dvevresonney2k: sounds good.01:46
blackburnsonney2k: competing for slots :)01:47
@bettyboo;>01:47
dvevrei was thinking of doing that anyways, but blackburn got us involved in a totally different direction :P01:47
@sonney2kfor example you could do the streamingfature class (assuming some simple things) and then make shogun's sgd use it or so01:47
dvevresonney2k: yeah.. i was going to start implementing it assuming file based input first01:47
@sonney2kor integrate one i/o method of vw in there01:48
@sonney2kexactly01:48
@sonney2kgood plan01:48
dvevresonney2k: can you suggest a way for me to get in touch with the vw people?01:48
dvevreor can i ask doubts abt vw on this channel itself?01:48
-!- epps [~epps@unaffiliated/epps] has joined #shogun01:49
blackburnsonney2k: btw, will you tell us when slots number will be known?01:49
-!- ameerkat [~ameerkat@184-98-140-155.phnx.qwest.net] has quit [Ping timeout: 260 seconds]01:49
@sonney2kdvevre, ask john when you have specific questions ... he seemed to like your proposal btw.01:50
@sonney2kblackburn, sure - but that is a long way still01:50
dvevresonney2k: thanks. i'm reading up some things which he commented about..01:50
blackburnsonney2k: 12-14, April in calendar for 'slots adjusting'01:51
blackburnwill it take some more time?01:52
@sonney2khttp://www.google-melange.com/document/show/gsoc_program/google/gsoc2011/timeline01:53
@sonney2ksays april 25 / april 22 is a conflict resolution01:53
blackburnhttp://www.google-melange.com/gsoc/events/google/gsoc201101:53
blackburnMentor Slots are Announced and adjusted as necessaryДата12 – 14 апр 201101:53
blackburnsorry for russian date :)01:53
@sonney2kbut yes you are right01:53
@sonney2kthen we have to hurry to have mentors for everyone...01:54
blackburnI have one :D but he isn't asking in irc for some reason :)01:54
blackburnanswering*01:55
blackburnsonney2k: btw, have you read my proposal? :)01:56
@sonney2kdvevre, anyway thanks for the patch01:56
@sonney2kblackburn, yes01:56
@sonney2k:D01:56
blackburnhow it is? :)01:57
@sonney2kI like it - but I won't be your mentor - so...01:57
dvevresonney2k: i'm hoping to keep them coming :)01:57
@bettyboo;D dvevre01:57
@sonney2kalright01:57
@sonney2kI need to sleep01:57
blackburnsonney2k: Chris will, I hope :)01:57
@bettybooyep!01:57
@sonney2kl8r01:58
blackburnme too01:58
dvevrebbye guys01:58
blackburnbye01:58
@bettybooblackburn, cu01:58
@sonney2khave to get up in 5 hrs01:58
@sonney2kl8r01:58
-!- blackburn [~qdrgsm@188.122.237.148] has quit [Quit: Leaving.]02:05
-!- skydiver [c255a037@gateway/web/freenode/ip.194.85.160.55] has quit [Quit: Page closed]02:12
-!- lionelc_ [adfdb883@gateway/web/freenode/ip.173.253.184.131] has joined #shogun02:16
-!- ameerkat [~ameerkat@184-98-140-155.phnx.qwest.net] has joined #shogun02:27
-!- ameerkat [~ameerkat@184-98-140-155.phnx.qwest.net] has quit [Ping timeout: 258 seconds]02:53
-!- Ryaether [~Ryaether@50-80-170-245.client.mchsi.com] has quit [Remote host closed the connection]02:57
-!- Ryaether [~Ryaether@50-80-170-245.client.mchsi.com] has joined #shogun02:57
-!- Mengyun [~dorothyk@S0106001a803d4363.vc.shawcable.net] has joined #shogun05:03
serialhexis there anybody here?05:04
serialhex...i guess not :P05:05
-!- epps [~epps@unaffiliated/epps] has quit [Ping timeout: 240 seconds]05:22
Ziyuanmorning05:36
ZiyuanHmm...I plan to use boost.math to implement bessel kernel05:37
-!- siddharth [~siddharth@117.211.88.150] has quit [Ping timeout: 276 seconds]05:37
-!- siddharth [~siddharth@117.211.88.150] has joined #shogun05:54
MengyunHi sonney2k, I saw your comment on my application for mloss/mldata.org integration ( for a minor patch like a  facebook like button..) I just find mldata already have a share button. Are you referring to like buttons for every data/task/method ?07:07
serialhexMengyun: i dont think he's here yet, so you may want to e-mail him that Q07:09
serialhexit's ~7am in berlin07:09
-!- lionelc_ is now known as lionelc707:15
serialhexfree book on neural networks: http://page.mi.fu-berlin.de/rojas/neural/index.html.html  <307:17
Mengyunserialhex, yep, Thanks07:17
serialhexnp :D07:18
* serialhex loves free knowlege07:18
-!- Ziyuan [~Ziyuan@116.22.50.70] has quit []07:24
-!- ameerkat [~ameerkat@184-98-140-155.phnx.qwest.net] has joined #shogun07:43
-!- lionelc7 [adfdb883@gateway/web/freenode/ip.173.253.184.131] has quit [Quit: Page closed]08:36
-!- Mengyun [~dorothyk@S0106001a803d4363.vc.shawcable.net] has quit [Read error: Connection reset by peer]08:47
-!- aifargonos [~aifargono@46.18.27.35] has joined #shogun08:53
-!- akhil_ [75d35896@gateway/web/freenode/ip.117.211.88.150] has joined #shogun09:09
-!- sploving [~root@124.16.139.196] has joined #shogun09:23
splovinghello all09:24
@bettyboosploving, morning09:24
splovingmorning bettyboo09:24
@bettyboosploving: I mean I created it and it doesn't importing09:24
@bettyboosploving, moin09:24
-!- yayo3 [~jakub@ip-78-45-113-245.net.upcbroadband.cz] has joined #shogun09:25
yayo3morning09:25
@bettyboomoin09:25
@sonney2kmorning all - I am not available for the next several hours though09:25
@sonney2kso talk to you in the evening hopefully09:26
splovingok09:27
-!- akhil_ [75d35896@gateway/web/freenode/ip.117.211.88.150] has quit [Quit: Page closed]09:43
-!- ameerkat [~ameerkat@184-98-140-155.phnx.qwest.net] has quit [Ping timeout: 248 seconds]09:50
-!- epps [~epps@84.18.157.200] has joined #shogun09:55
-!- epps [~epps@84.18.157.200] has quit [Changing host]09:55
-!- epps [~epps@unaffiliated/epps] has joined #shogun09:55
yayo3####ing merge conflicts10:04
yayo3I hate them10:04
-!- Tanmoy [75d35896@gateway/web/freenode/ip.117.211.88.150] has quit [Ping timeout: 253 seconds]10:49
-!- Ryaether [~Ryaether@50-80-170-245.client.mchsi.com] has left #shogun []11:01
-!- siddharth [~siddharth@117.211.88.150] has quit [Remote host closed the connection]11:50
-!- siddharth [~siddharth@117.211.88.150] has joined #shogun11:52
-!- heiko [~heiko@134.91.53.124] has joined #shogun12:13
-!- heiko [~heiko@134.91.53.124] has left #shogun []12:15
-!- blackburn [~qdrgsm@188.122.237.148] has joined #shogun12:19
-!- heiko [~heiko@134.91.53.124] has joined #shogun12:24
-!- heiko [~heiko@134.91.53.124] has left #shogun []12:30
-!- akhil_ [75d35896@gateway/web/freenode/ip.117.211.88.150] has joined #shogun13:27
-!- siddharth [~siddharth@117.211.88.150] has quit [Read error: Connection reset by peer]13:44
-!- siddharth [~siddharth@117.211.88.150] has joined #shogun13:48
-!- yayo3 [~jakub@ip-78-45-113-245.net.upcbroadband.cz] has quit [Quit: leaving]13:50
-!- akhil_ [75d35896@gateway/web/freenode/ip.117.211.88.150] has quit [Ping timeout: 252 seconds]14:39
-!- heiko [~heiko@134.91.53.124] has joined #shogun14:47
-!- heiko [~heiko@134.91.53.124] has left #shogun []14:48
-!- blackburn [~qdrgsm@188.122.237.148] has quit [Quit: Leaving.]15:01
-!- blackburn [~qdrgsm@188.122.237.148] has joined #shogun15:03
splovinghello sonney2k, I write a small lua one-dim array typemap file. The code is : http://pastebin.com/wk9GNzpU.15:33
splovingwhen you have time, please take a look at it.15:33
-!- sploving [~root@124.16.139.196] has left #shogun []15:39
-!- epps [~epps@unaffiliated/epps] has quit [Quit: Leaving]16:02
siddharthblackburn, I am not able to pull request16:18
siddharthits saying There was a problem with your request.16:18
siddharthUsually this means that your user authenticity token is incorrect. This can happen when you visit a page with a form, then login in a separate window, and then return to resubmit the form. Here's what you can try next:16:18
siddharthi logged out and then logged in...no change :|16:18
blackburnsiddharth: did you set up all these things with rsa?16:19
siddharthyes16:19
siddharthi am able to fork it16:19
siddharthi am able to push also16:19
blackburnand git config --global github.token ...?16:19
-!- josip [~josip@unaffiliated/josip] has quit [Remote host closed the connection]16:19
siddharthyes16:19
blackburnhmm. strange16:19
siddharthi have mailed them...but they have not replied :|16:21
blackburnsiddharth: have you added a SSH key to your github profile?16:24
siddharthyes the publickey right16:24
blackburnseems that there is some problem with it16:24
blackburnyou could re-generate it and delete then add it to your profile16:25
siddharthhmm let me try16:25
blackburnmaybe it could help :)16:26
@bettyboo:> blackburn16:26
siddharthblackburn, did you leave passphrase empty?16:30
blackburnnope16:30
blackburnjust enter there some word you will remember16:30
blackburnsiddharth: how it is? working now?16:39
siddharthnope...irritated :|16:40
siddharthI am behind a proxy but all m using transconnect for it16:41
siddhartheverything is working fine except for pull request16:42
* siddharth thinks of deleting account and making new one :|16:42
blackburnstrange..16:44
-!- Tanmoy [75d35896@gateway/web/freenode/ip.117.211.88.150] has joined #shogun16:53
-!- dvevre [b49531e3@gateway/web/freenode/ip.180.149.49.227] has quit [Ping timeout: 252 seconds]17:00
-!- dvevre [b49531e3@gateway/web/freenode/ip.180.149.49.227] has joined #shogun17:41
-!- Tanmoy [75d35896@gateway/web/freenode/ip.117.211.88.150] has quit [Ping timeout: 252 seconds]17:53
-!- Netsplit *.net <-> *.split quits: serialhex, cwidmer, blackburn18:09
-!- Netsplit over, joins: blackburn, serialhex, cwidmer18:13
-!- Mengyun [~dorothyk@S0106001a803d4363.vc.shawcable.net] has joined #shogun18:52
-!- aifargonos [~aifargono@46.18.27.35] has left #shogun []19:01
-!- Bautista [4f91c37b@gateway/web/freenode/ip.79.145.195.123] has joined #shogun19:02
-!- Bautista [4f91c37b@gateway/web/freenode/ip.79.145.195.123] has quit [Quit: Page closed]19:23
-!- ameerkat [~ameerkat@184-98-140-155.phnx.qwest.net] has joined #shogun19:27
-!- bran_ [458cb0fe@gateway/web/freenode/ip.69.140.176.254] has joined #shogun19:32
bran_Is there any way to disable all printing to the console, including the copyright info?19:32
bran_I'm thinking about using this on Hadoop streaming with performs is communication over stdin/err/out, which makes it impossible to use with Shogun as it is19:35
bran_if anyone knows how please email me at bwhite (at) cs umd edu19:36
-!- bran_ [458cb0fe@gateway/web/freenode/ip.69.140.176.254] has quit [Quit: Page closed]19:38
-!- serialhex [~quassel@99-101-149-136.lightspeed.wepbfl.sbcglobal.net] has quit [Remote host closed the connection]19:43
-!- serialhex [~quassel@99-101-149-136.lightspeed.wepbfl.sbcglobal.net] has joined #shogun19:44
-!- alesis-novik [~alesis@188.74.87.84] has quit [Quit: I'll be Bach]19:54
-!- skydiver [c255a037@gateway/web/freenode/ip.194.85.160.55] has joined #shogun20:11
-!- Bautista [4f91c37b@gateway/web/freenode/ip.79.145.195.123] has joined #shogun21:17
Bautistahello everybody!21:17
blackburnhi21:17
@bettyboomorning21:17
Bautistasee, I'm kind of newbie in the whole git world21:18
Bautistai need step by step instructions21:18
Bautista(even more detailed than in the mail)21:18
blackburnyou can find it in the mailing list21:18
Bautistahehe21:19
Bautistathe thing is, i've already implemented a new distance21:19
blackburnjust fork repo, commit to your fork and do pull request21:20
Bautistaokk so how is the commit exactly performed? (sorry for the inconveniences)21:21
-!- blackburn [~qdrgsm@188.122.237.148] has quit [Ping timeout: 260 seconds]21:25
-!- Bautista [4f91c37b@gateway/web/freenode/ip.79.145.195.123] has quit [Ping timeout: 252 seconds]21:29
-!- blackburn [~qdrgsm@188.122.237.148] has joined #shogun21:30
* sonney2k is back21:37
blackburnmany issues are waiting only for you, Soeren! :D21:38
@bettyboo;> blackburn21:38
serialhexyeah, i've been catching up on whats going on in the chat, and it seemd like a BUNCH of people are having various problems!21:39
serialhexblackburn: you're still awake??  what are you, a vampire?? :P21:40
@bettyboo;D21:40
blackburnserialhex: why I am? it is only 23-40 here21:41
serialhex??  then my clock is wrong, it says it's 2:42am...21:41
serialhexomsk right?21:42
blackburnoh, omsk is far away from here :)21:42
@bettybooblackburn,21:42
blackburnsamara, or moscow, same time21:43
serialhexahh, ok21:43
serialhex...interesting, apparently it lists +- moscow time, not gmt for some reason :-/21:44
blackburnomsk is in siberia21:44
serialhexok... thats better :D21:44
@bettyboo;>21:44
serialhexyeah, i was like 'wtf, i'm up that late sometimes... but he seems to be up that late all the time' :P21:44
* serialhex sets moscow clock to moscow time :P21:45
-!- blackburn1 [~qdrgsm@188.122.237.148] has joined #shogun21:46
blackburn1disconnected for some reason I don't know :)21:46
@bettyboo;>21:46
-!- blackburn [~qdrgsm@188.122.237.148] has quit [Read error: Connection reset by peer]21:46
serialhexhmm.. it hasnt shown that you've left21:46
serialhex...until just now21:46
serialhexanyway, i got ya on the right time now :D21:46
serialhexoh, and instead of "why i am?"  it's "why am i?" blackburn121:48
blackburn1hehe, sorry21:48
@bettyboohe!?! :> blackburn121:48
blackburn1I see now21:48
serialhexno problem21:48
serialhexyou'll get it21:48
-!- blackburn1 is now known as blackburn21:49
serialhexas far as i understand it, english is one of the hardest languages to learn as a second language, and russian is even harder... :P21:50
@bettyboohihi? serialhex21:50
serialhexhihi bettyboo21:50
@bettybooserialhex: stack traceback:21:50
CIA-31shogun: Sergey Bartunov master * r6a2967b / (5 files in 2 dirs): SplineKernel added - http://bit.ly/gzaygV21:50
serialhexwhy are you doing a stack traceback on me bettyboo??21:50
@bettybooserialhex: i am not sure if this has been the case before you added gsoc21:50
CIA-31shogun: Soeren Sonnenburg master * re8cd282 / (5 files in 2 dirs):21:50
CIA-31shogun: Merge branch 'master' of https://github.com/sbos/shogun21:50
CIA-31shogun: Resolve conflicts in:21:50
CIA-31shogun:  src/libshogun/kernel/Kernel.cpp21:50
CIA-31shogun:  src/libshogun/kernel/Kernel.h21:50
CIA-31shogun:  src/modular/Kernel.i - http://bit.ly/dPrtS821:51
blackburnserialhex: don't think so, german is a bit difficult too, may be even more21:51
serialhexblackburn: i dont know, all i understand is english, and from what i understand russian has a bunch of tenses and stuff to get right... i'm just relating what i know21:52
serialhexeither way, you're doing well!21:52
blackburnserialhex: thank you :)21:52
blackburnthere is 17 forks so far, seems that ~30% will seriously commit something, it will simplify your task, sonney2k :)21:54
blackburnthere are, there are21:54
serialhexdamn... 17 forks!!!21:56
* serialhex gets to work doing cool stuff!!21:57
blackburnIIRC it was ~5-6 before start of our work21:57
serialhexyeah i know :-/21:57
CIA-31shogun: Soeren Sonnenburg master * r572762e / (3 files):22:04
CIA-31shogun: fix the latex code in some of the newer kernel documentation22:04
CIA-31shogun: ams math is not avilable, so commands like text, lVert etc won't work - http://bit.ly/g2ApIw22:04
-!- Gagan [~brocolli@199.48.241.32] has joined #shogun22:10
CIA-31shogun: Soeren Sonnenburg master * r05a1d7d / (7 files in 3 dirs):22:13
CIA-31shogun: Merge branch 'master' of https://github.com/diver-in-sky/shogun22:13
CIA-31shogun: Fix merge conflicts in (stemming from changed kernel enums)22:13
CIA-31shogun:  src/libshogun/kernel/Kernel.cpp22:13
CIA-31shogun:  src/libshogun/kernel/Kernel.h - http://bit.ly/fiejPo22:13
-!- siddharth [~siddharth@117.211.88.150] has quit [Remote host closed the connection]22:15
serialhexjust a note:  i've found it's easier to sync with shogun's master repository by 'git pull upstream master' && 'git push' than it is to use the fork queue22:16
serialhex...if anyone is paying attention & is interested :P22:16
@sonney2kblackburn, impressive...22:21
blackburnsonney2k: what?22:22
@sonney2kthe 17 forks22:22
* sonney2k is slow to respond - going though his email and trying to apply patches one by one22:22
blackburnsonney2k: that's it, in same time your task to choose <10 will not became simple at all because there will be >10 contributors :)22:25
@bettyboolol!! ;>22:25
serialhexi couldnt have said it better than bettyboo!!22:26
@bettybooserialhex: they are basically solving the dual form22:26
@sonney2kblackburn, we need new criteria then - but the quality of the patches etc are already an indication22:26
blackburnsonney2k: at least you have proposals :)22:27
-!- yayo3 [~jakub@ip-78-45-113-245.net.upcbroadband.cz] has joined #shogun22:27
blackburnsonney2k: working on gaussian naive bayes now, is it necessary at all?22:31
@sonney2kyou mean as classifier? for real valued inputs?22:32
-!- Gagan [~brocolli@199.48.241.32] has quit [Ping timeout: 240 seconds]22:32
blackburnsonney2k: yeap22:33
@sonney2kwe don't have that - so another good low hanging fruit I guess :)22:34
@sonney2k(we only have that for discrete inputs via HMMs)22:35
blackburnsonney2k: okay, seems that I will finish it tomorrow22:35
@sonney2knice22:37
serialhexdammit blackburn!!  why do you have to be such an overachiever!!! :D22:38
blackburnserialhex: do you think so? just doing some things I can :)22:39
blackburnat least I'm really interested now and not only because GSoC22:39
serialhexi'm just messing with ya, but you do seem to be doing a lot more than everyone else22:39
blackburnI did only 2 kernels and some improvement of CKNN :)22:41
@bettyboo;>22:41
serialhexhey, do you know if you'd need gaussian random numbers??  i was thinking of implementing that if it'll be used22:42
* serialhex wonders why it hasnt been implemented if one would need it, but it may have been overlooked22:44
blackburndouble rand_normal(double mean, double dev) {22:44
blackburn    double x = 1.0*rand()/RAND_MAX, y=1.0*rand()/RAND_MAX;22:44
blackburn    double z = sqrt(-log(x))*cos(2*M_PI*y);22:44
blackburn    return sqrt(dev)*z+mean;22:44
blackburn}22:44
blackburnhehe22:44
blackburnor more fast bux-mueller transformation: http://pastebin.com/x5nSUc8b22:45
blackburnbox*, sorry22:45
serialhexyeah... thats what i was going to implement...22:45
blackburnso do it22:46
serialhextho the ziggurat alg says it's faster: http://en.wikipedia.org/wiki/Ziggurat_algorithm22:46
serialhexyeah, i guess it couldnt hurt :D22:46
blackburnI guess, you will have to spend much more time for ziggurat22:46
serialhexyeah, maybe later cause it looks like a pain in the ass!22:47
yayo3hmmm, Makefile bug22:50
-!- skydiver [c255a037@gateway/web/freenode/ip.194.85.160.55] has quit [Quit: Page closed]22:53
yayo3making a patch for adding one character seems so wasteful22:56
blackburnI had one removing '_' yesterday, named it 'Commit the Great' :)22:56
@bettyboo;> blackburn22:56
yayo3well, I have one pull request open already so it needs to wait I guess22:59
blackburnyou could commit, and that small patch will be in your pull request, I guess it isn't bad23:00
yayo3well it's Mac OS specific and noone uses that enymore :)\23:02
yayo3ok, I'll take a look at it23:02
@sonney2kyayo3, are you referring to this https://github.com/shogun-toolbox/shogun/pull/7 ?23:03
@sonney2kIf so it needs some attention from your side ...23:04
yayo3sonney2k: well, I fixed the stuff (I think, but I tested it using python modular and it seemd to work) and added a new kernel. Do I need to resubmit it or something?23:05
@sonney2kblackburn, I guess you can answer :)23:06
serialhexsonney, the float64 & 32 random number generators, are they [min, max] or (min, max)?? I'm thinking the first, but i dont want to be wrong...23:06
blackburnsonney2k: it is all about serialization? :)23:07
blackburnis it23:07
@sonney2kblackburn, I meant does he need to re-issue a pull request or ...?23:07
blackburnsonney2k: ah, he only could commit and it will be in pull request23:08
@sonney2kyayo3, ^^ so did you git push?23:08
blackburnhe didn't23:10
CIA-31shogun: Shashwat Lal Das master * r01a33a7 / src/libshogun/kernel/SphericalKernel.h : Add doc for parameter in SphericalKernel - http://bit.ly/hyz5iM23:10
@sonney2kyayo3, please do then!23:11
@sonney2kthough I guess a git rebase is necessary by now23:11
yayo3I see five commits in total in the pull request23:12
yayo3so I need to do more things to push them though, or the code is wrong?23:13
blackburnyayo3: you hadn't push your last commit as seen on https://github.com/yayo3/shogun23:13
@sonney2kyayo3, so push push - I'd like to apply it.23:20
yayo3blackburn: I didn't? My last commit is from "about 10 hours ago" and I don't have any more code. I added it to the pull request.23:20
blackburnyayo3: yeap, but I thougth you mind to commit another one23:21
yayo3oh. I see.23:21
@sonney2kyayo3, you also need to update to git master / rebase or merge. There are some conflicts now.23:22
yayo3ok.23:22
blackburngit fetch upstream23:22
blackburngit merge master/upstream23:22
blackburnIIRC23:22
yayo3let me guess. someone named some kernel the same way I did hours ago23:23
blackburnupstream/master23:23
yayo3numbered I mean23:23
@sonney2kyayo3, exactly.23:24
blackburnoh, no, seems that rebase23:25
blackburni'm tangled :)23:26
@sonney2kblackburn, seems that rebase what?23:26
blackburnthat he should rebase, I'll check it on my fork in a sec23:27
yayo3when does Makefile.template get used?23:27
yayo3I thought Makefile gets generated from it, but clean clone gets working Makefile already23:27
yayo3anyway the problem is that configure generates configure-SOMETHING-CRAZY files23:29
yayo3make clean is supposed to clean them up on botched build, but it fails since they are actually directories on Mac OS23:30
@sonney2kyayo3, the .template is used to genreate makefiles under e.g. libshogun/ etc23:31
@sonney2kyayo3, being a bit more specific / showing the error msg. would help.23:32
blackburnsonney2k: yes, git fetch upstream + git merge upstream/master is sufficient to remove unrelated commits from pull request23:32
yayo3I added a -r on the rm that's supposed to do that, but maybe there should be checking whether we're on mac os or something, as rm -rf running wild might get ugly23:32
@sonney2kyayo3, adding -r sounds way to dangerous23:33
@sonney2kjust name the files that are there23:34
@sonney2kor dirs23:34
yayo3it's a Mac OS bundle23:34
@sonney2kthen we have to figure out which test creates them23:34
yayo3anyway I resolved the merge conflict and updated the pull request (rm -rf not included)23:34
yayo3ok, the file now is called "configure-21124-3819.o.dSYM"23:37
serialhexsonney2k, blackburn, tell me what you think: https://gist.github.com/91275323:39
yayo3directory I mean23:39
serialhex...when you have a chance :D23:40
yayo3and somehow it's not mentioned in configure.log at all23:40
yayo3ok, it's a Mac OS gcc feature. I'll take a look at it.23:45
serialhexyayo3: i'd hardly call problems a 'feature'... unless you're in windows :P23:46
dvevreserialhex: nice :)23:46
@bettybooyeah! ;D23:46
yayo3serialhex: it's not actually a problem. they just introduced new type of debug symbols in external files that are actually directories23:46
serialhexi was hoping someone would like my joke! :P23:47
@bettybooyeah23:47
serialhexahh.. ok23:47
yayo3serialhex: only it's the default now.23:47
CIA-31shogun: Jakub Jirku master * r33c7e58 / (11 files in 3 dirs):23:52
CIA-31shogun: Merge remote-tracking branch 'upstream/master'23:52
CIA-31shogun: Conflicts:23:52
CIA-31shogun:  src/libshogun/kernel/Kernel.cpp23:52
CIA-31shogun:  src/libshogun/kernel/Kernel.h - http://bit.ly/grnbdX23:52
yayo3squeee!23:53
yayo3anyway see ya tomorrow23:58
-!- yayo3 [~jakub@ip-78-45-113-245.net.upcbroadband.cz] has left #shogun []23:58
--- Log closed Mon Apr 11 00:00:19 2011

Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!