--- Log opened Wed Oct 24 00:00:00 2012 | ||
--- Day changed Wed Oct 24 2012 | ||
n4nd0 | atm I am checking the reference Cheng sent to the mailing list | 00:00 |
---|---|---|
n4nd0 | wiking: do you know something about image segmentation using something similar to our so-svm? | 00:00 |
wiking | n4nd0: heheheh | 00:00 |
wiking | n4nd0: yeah i've checked that paper as well | 00:00 |
wiking | i've seen some paper afaik | 00:01 |
wiking | but i have no fucking clue about img segmentation on the other hand | 00:01 |
wiking | i know graph cuts :) | 00:01 |
wiking | and that's it :> | 00:01 |
n4nd0 | hehe ok :) | 00:01 |
wiking | fing help | 00:16 |
wiking | i mean hell | 00:16 |
wiking | who the fuck creates big-endian formated images | 00:16 |
wiking | :D | 00:16 |
wiking | http://researchinprogress.tumblr.com/post/34088637501/fast-vs-exact-solutions | 01:22 |
wiking | :D | 01:22 |
shogun-buildbot | build #125 of nightly_none is complete: Failure [failed compile] Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_none/builds/125 | 03:00 |
shogun-buildbot | build #142 of nightly_default is complete: Failure [failed test] Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/142 | 03:32 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 244 seconds] | 05:21 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 05:57 | |
@sonney2k | n4nd0, up early? | 06:20 |
n4nd0 | sonney2k: yeah! :) | 06:21 |
@sonney2k | n4nd0, why that? | 06:21 |
n4nd0 | sonney2k: I am trying to find out what can I prepare using the SO framework for Christian's talk | 06:22 |
@sonney2k | n4nd0, good excuse :D | 06:24 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 07:03 | |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has quit [Ping timeout: 245 seconds] | 08:22 | |
-!- ptizoom [~christian@85.210.92.173] has joined #shogun | 08:51 | |
-!- blackburn1 [~blackburn@83.234.54.235] has quit [Quit: Leaving.] | 09:57 | |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has joined #shogun | 10:28 | |
wiking | blackburn: any clearcases today? :D | 11:00 |
blackburn | wiking: SURE | 11:00 |
blackburn | my life is empty without CC | 11:00 |
wiking | blackburn: btw how do u like travis' message ?:) specially designed it for u | 11:00 |
wiking | ;)P | 11:00 |
blackburn | yeah I like it | 11:01 |
wiking | ls -la | 11:06 |
wiking | wrong window | 11:06 |
blackburn | good you didn't wrongmessaged sth like 'cd porno/hard/german' | 11:09 |
blackburn | :D | 11:09 |
blackburn | wiking: what are you working on lately? a lot of sparse stuff, where is it used? | 12:17 |
-!- n4nd0 [~nando@n183-p114.kthopen.kth.se] has joined #shogun | 13:10 | |
wiking | blackburn: i'd like to use it | 13:16 |
wiking | blackburn: in my so-svm/latent sosvm implementation | 13:16 |
blackburn | wiking: I just wojnder what is the problem you solve | 13:16 |
blackburn | I mean sparse presentations are not so often used in case of local descriptors or anything like that | 13:16 |
wiking | ah malignant cell detection in xray chest images | 13:16 |
wiking | but it's not about the descriptors | 13:17 |
wiking | it's about the representation of PSI features | 13:17 |
wiking | in so-svm and latent sosvm | 13:17 |
wiking | a lot of times you have like [0,...,0]_n [nnz]_m [0,...,0]_k like spi vectors | 13:18 |
wiking | *psi | 13:18 |
wiking | so you have zero padding before and after the nnz values | 13:18 |
wiking | in the generated psi vector | 13:18 |
blackburn | I see | 13:18 |
wiking | so i want to be able to represent psi both as dense and sparse vectors | 13:19 |
wiking | and for that i have this dotfeatures story | 13:19 |
wiking | but it's a pain in the ass | 13:19 |
blackburn | sonney2k: it seems that SLEP problem is finally solved | 13:32 |
@sonney2k | n4nd0: blackburn shogun-toolbox.org/page/contact/irclogs/ | 14:44 |
blackburn | ha | 14:45 |
n4nd0 | oh your widget application is ready | 14:45 |
n4nd0 | nice! | 14:45 |
@sonney2k | the calendar arrived | 14:45 |
n4nd0 | only 2011 visible so far? | 14:45 |
n4nd0 | sonney2k: any new about the doc sprint, are we applying finally? | 14:47 |
@sonney2k | ugh bug :-! | 14:47 |
@sonney2k | n4nd0: sure ! we are | 14:48 |
n4nd0 | sonney2k: :) | 14:48 |
n4nd0 | so you take care of the application? | 14:48 |
@sonney2k | n4nd0: btw 2012 is in chatlogs but at the very bottom. | 14:50 |
@sonney2k | yeah I have to do it | 14:50 |
@sonney2k | maybe we should reverse the order | 14:51 |
n4nd0 | oh yes 2012 is there | 14:52 |
n4nd0 | I scrolled down too fast :S sorry | 14:52 |
n4nd0 | probably from newest date to oldest is more useful, I think | 14:53 |
@sonney2k | ptizoom: did you checkout git master? | 15:01 |
@sonney2k | n4nd0, now it is reversely ordered http://shogun-toolbox.org/page/contact/irclogs/ | 15:16 |
n4nd0 | wiking: ping | 15:18 |
n4nd0 | wiking: do you happen to know why I get this error ERROR MSK_RES_ERR_SPACE - 'Out of space.' in a problem that I could solve before? | 15:18 |
n4nd0 | wiking: I have just downloaded a new license in case it was that | 15:19 |
n4nd0 | but the error remains | 15:19 |
n4nd0 | sonney2k: nice! | 15:20 |
@sonney2k | blackburn, the website is still confusing to me | 15:29 |
@sonney2k | I think we should move http://shogun-toolbox.org/page/about/contribute | 15:30 |
@sonney2k | to the dev page | 15:30 |
@sonney2k | blackburn, and look here http://shogun-toolbox.org/page/about/ourteam | 15:30 |
@sonney2k | nothing on it | 15:30 |
blackburn | I'll put my drunk photo out there | 15:30 |
@sonney2k | maybe we better remove it | 15:30 |
@sonney2k | this info is under contacts anyways | 15:31 |
blackburn | not really I'd like it to stay | 15:31 |
@sonney2k | blackburn, ok then write some longer content about it | 15:31 |
blackburn | with our photos and things like | 15:32 |
@sonney2k | like list of all developers with coordinates etc | 15:32 |
@sonney2k | please do | 15:32 |
@sonney2k | highly needed | 15:32 |
blackburn | 'ich bin sergej ich liebe das vodka' | 15:32 |
blackburn | not now I am at job :D | 15:32 |
@sonney2k | later then | 15:32 |
* sonney2k is moving contribute over to dev | 15:32 | |
@sonney2k | blackburn, I think the references should be some subpage of about | 15:42 |
@sonney2k | it is not really helpful for users... | 15:42 |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has quit [Ping timeout: 245 seconds] | 15:45 | |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has joined #shogun | 15:46 | |
blackburn | KAWABANGA | 15:46 |
blackburn | BLUE SCREEN OF DEATH | 15:46 |
blackburn | <3 shindows | 15:47 |
blackburn | :D | 15:47 |
n4nd0 | hehe | 15:49 |
@sonney2k | n4nd0, blackburn http://shogun-toolbox.org/page/about/contributions | 15:57 |
@sonney2k | is this better? | 15:57 |
@sonney2k | (as in more useful?) | 15:57 |
@sonney2k | question is where should the 'we participate in gsoc2012' from http://shogun-toolbox.org/page/about/information go? | 15:58 |
blackburn | sonney2k: I am ok with place it is now | 15:59 |
blackburn | it is a good advertisement :D | 15:59 |
@sonney2k | no it is basically hidden too deep | 16:00 |
@sonney2k | hard to find... | 16:00 |
n4nd0 | sonney2k: maybe we should include the HMSVM toolbox http://mloss.org/software/view/250/ in acknowledgements? | 16:01 |
@sonney2k | n4nd0, please do it | 16:02 |
@sonney2k | actually also the GP toolbox | 16:02 |
@sonney2k | still a long way to go... | 16:02 |
@sonney2k | the dev page becomes useful | 16:02 |
@sonney2k | btw, I realize we have no explicit download page | 16:03 |
@sonney2k | maybe we should rename news to something else | 16:03 |
n4nd0 | sonney2k: shogun-toolbox.org/admin to login and after edit? | 16:05 |
ptizoom | sonney2k: I have not rebased for a while.... | 16:06 |
@sonney2k | ptizoom, well don't rebase but just check it out | 16:07 |
ptizoom | sonney2k: do I need to? | 16:07 |
blackburn | ich mochte schlafen | 16:10 |
blackburn | !! | 16:10 |
ptizoom | ihren Sie erm?den | 16:16 |
@sonney2k | n4nd0, admin/ sure | 16:34 |
n4nd0 | sonney2k: contributions updated | 16:34 |
@sonney2k | ptizoom, please test a clean git master and then try ./configure --interfaces=perl_modular | 16:34 |
n4nd0 | sonney2k: yeah ... I was missing the slash at the end, blackburn pointed it out :) | 16:34 |
ptizoom | sonney2k: I see... | 16:35 |
n4nd0 | blackburn: what about the chef port? | 16:35 |
@sonney2k | n4nd0, could you please add http://www.gaussianprocess.org/gpml/code | 16:35 |
@sonney2k | are GP stuff is based on theirs | 16:35 |
n4nd0 | I just closed the window but yes I can do that | 16:36 |
@sonney2k | please do | 16:37 |
@sonney2k | ptizoom, we should get that to work in master | 16:37 |
@sonney2k | then we can add examples etc from your stuff | 16:37 |
n4nd0 | sonney2k: done | 16:38 |
n4nd0 | I am adding the matlab DR toolbox too | 16:40 |
ptizoom | sonney2k: I have not much to test thy perl_modular, excepted my humoungus batch I put in tests/integration/pdl_modular | 16:40 |
@sonney2k | ptizoom, well test if it compiles! | 16:41 |
@sonney2k | n4nd0, thx | 16:41 |
@sonney2k | ptizoom, then some example - if that works we can continue | 16:41 |
blackburn | n4nd0: chef is a long run project sure | 16:43 |
blackburn | we are in design phase | 16:43 |
blackburn | we've got to estimate impact and regressions | 16:43 |
blackburn | i.e. how much money we can earn from that :D | 16:44 |
n4nd0 | :D | 16:44 |
n4nd0 | need a break, see you later! | 16:46 |
-!- n4nd0 [~nando@n183-p114.kthopen.kth.se] has quit [Quit: leaving] | 16:46 | |
-!- heiko [~heiko@host86-176-125-223.range86-176.btcentralplus.com] has joined #shogun | 16:59 | |
-!- heiko [~heiko@host86-176-125-223.range86-176.btcentralplus.com] has left #shogun [] | 16:59 | |
ptizoom | sonney2k:...Checking for bzip2 ... yes | 17:19 |
ptizoom | Checking for lzma ... no | 17:19 |
ptizoom | Checking for Posix thread spinlock support ... yes | 17:19 |
ptizoom | Checking for libshogun ... yes (internal in source tree) | 17:19 |
ptizoom | Checking for Perl ... 5.14.2 | 17:19 |
ptizoom | Checking for Perl ExtUtils::Embed ... yes | 17:19 |
ptizoom | Checking for PDL ... yes | 17:19 |
ptizoom | Checking for SWIG ... 2.0.7 | 17:19 |
ptizoom | ... | 17:19 |
ptizoom | so far so good... | 17:19 |
@sonney2k | ptizoom, compile! | 17:21 |
ptizoom | I have only a...celeron 2GHz (chipest...) | 17:24 |
ptizoom | cheapest... | 17:24 |
ptizoom | you have warnings... | 17:25 |
ptizoom | for ShiftDataGenerator.cpp.o | 17:26 |
ptizoom | i am disappointed | 17:28 |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has quit [Quit: Page closed] | 17:30 | |
ptizoom | yet other warnings in sg_print_functions.cpp.o | 17:34 |
ptizoom | aint working modshogun_wrap.cxx: In function 'pdl* if_piddle(SV*)': | 17:37 |
ptizoom | modshogun_wrap.cxx:3568:22: error: 'PDL' was not declared in this scope | 17:37 |
ptizoom | ... | 17:37 |
ptizoom | it shall had been defined like Core* PDL; in modshogun_wrap.cxx | 17:48 |
ptizoom | like in src/interfaces/modular/SGBase.i, | 17:49 |
ptizoom | patched ...with #if defined (SWIGPERL) //&& HAVE_PDL | 17:51 |
ptizoom | + | 17:51 |
ptizoom | +%header %{ | 17:51 |
ptizoom | + SV* CoreSV; | 17:51 |
ptizoom | + Core* PDL; | 17:51 |
ptizoom | +%} | 17:51 |
ptizoom | + | 17:51 |
ptizoom | +%init %{ | 17:51 |
ptizoom | + //PTZ120930 boot PDL, load PDL stuff??? and define a PDL | 17:51 |
ptizoom | + //pl_require('PDL'); | 17:51 |
ptizoom | + //check Core.xs //Core* PDL_p = pdl__Core_get_Core(); | 17:51 |
ptizoom | + //PDL_COMMENT("Get pointer to structure of core shared C routines") | 17:51 |
ptizoom | + //PDL_COMMENT("make sure PDL::Core is loaded") | 17:51 |
ptizoom | + | 17:51 |
ptizoom | + perl_require_pv("PDL::Core"); | 17:51 |
ptizoom | + CoreSV = perl_get_sv("PDL::SHARE",FALSE); | 17:51 |
ptizoom | + // PDL_COMMENT("SV* value") | 17:51 |
ptizoom | + if (CoreSV == NULL) | 17:51 |
ptizoom | + Perl_croak(aTHX_ "Can't load PDL::Core module"); | 17:51 |
ptizoom | + PDL = INT2PTR(Core*, SvIV( CoreSV )); | 17:51 |
ptizoom | + // PDL_COMMENT("Core* value") | 17:51 |
ptizoom | + | 17:51 |
ptizoom | ... | 17:51 |
ptizoom | sonney2k: have you got more patches in you bag or do you want me to carry on from there? | 17:53 |
ptizoom | sonney2k:by the way all these //, #if 0 /**/ shall be zapped on the release branch! | 17:57 |
-!- Netsplit *.net <-> *.split quits: wiking | 18:08 | |
-!- wiking [~wiking@info2k1.hu] has joined #shogun | 18:08 | |
-!- wiking [~wiking@info2k1.hu] has quit [Changing host] | 18:09 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 18:09 | |
wiking | mmm splitting...? | 18:09 |
wiking | sonney2k: ping | 18:28 |
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has quit [Ping timeout: 245 seconds] | 18:39 | |
-!- blackburn [~blackburn@83.234.54.200] has joined #shogun | 18:59 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 19:38 | |
-!- zxtx [~zv@guest-wireless-nup-198-147-123-236.usc.edu] has joined #shogun | 19:48 | |
blackburn | n4nd0: got any ideas for example? | 20:02 |
n4nd0 | blackburn: yep | 20:02 |
n4nd0 | I am working hard on it since this afternoon | 20:02 |
blackburn | can I help you? | 20:03 |
n4nd0 | let's see if I get sth cool in time | 20:03 |
n4nd0 | mmm I don't know | 20:03 |
n4nd0 | brb in a minute | 20:03 |
n4nd0 | blackburn: I tell you the idea | 20:05 |
n4nd0 | so in this pattern recognition course I finished last week we developed an isolated word recognizer using HMMs | 20:06 |
n4nd0 | let's say you want to recognize numbers from 1-5 | 20:06 |
n4nd0 | then you train 5 HMMs, each HMM models a number (the word) | 20:06 |
n4nd0 | for new examples, you just run an algorithm in every HMM and recognize with the HMM that gives highest score | 20:07 |
n4nd0 | I aim at doing exactly the same thing, but training HM-SVMs instead of HMMs | 20:07 |
blackburn | can you describe example of data? | 20:07 |
blackburn | so it is a text? | 20:07 |
n4nd0 | no | 20:07 |
n4nd0 | it is speech | 20:07 |
blackburn | image? | 20:07 |
blackburn | ahh | 20:07 |
n4nd0 | you record your voice saying 'one' | 20:08 |
n4nd0 | saying 'two' | 20:08 |
n4nd0 | etc | 20:08 |
n4nd0 | do you see what I mean? | 20:08 |
blackburn | yes probably | 20:08 |
n4nd0 | I just need to implement a new StateModel | 20:09 |
n4nd0 | so far we've got in shogun the twoState | 20:09 |
n4nd0 | for speech recognition you use other kind of state models | 20:09 |
blackburn | I see | 20:09 |
n4nd0 | blackburn: do you think it sounds ok? | 20:11 |
n4nd0 | I think it is something I should be able to do for tomorrow and prepare a couple of slides with fancy pictures | 20:12 |
-!- n4nd0 is now known as n4nd0_away | 20:17 | |
blackburn | would be nice | 20:37 |
-!- zxtx [~zv@guest-wireless-nup-198-147-123-236.usc.edu] has quit [Ping timeout: 272 seconds] | 20:55 | |
-!- n4nd0_away is now known as n4nd0 | 20:56 | |
@sonney2k | ptizoom: is that stuff already in Sgbase? | 20:57 |
ptizoom | yes | 21:33 |
ptizoom | sonney2k: in my "pull-request" branch...! | 21:34 |
@sonney2k | ptizoom, ahh in SGBase.i | 21:47 |
@sonney2k | hmmhh quite cluttered - lemme try | 21:47 |
n4nd0 | sonney2k: can I ask you a question regarding PLiFs? | 22:20 |
@sonney2k | I have to enter the airplane now | 22:21 |
n4nd0 | ups ok | 22:21 |
@sonney2k | cu | 22:21 |
n4nd0 | good flight then | 22:21 |
shogun-buildbot | build #579 of deb3 - modular_interfaces is complete: Failure [failed compile python_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/579 blamelist: Soeren Sonnenburg <sonne@debian.org> | 22:28 |
shogun-buildbot | build #470 of deb2 - static_interfaces is complete: Failure [failed test libshogun] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb2%20-%20static_interfaces/builds/470 blamelist: Soeren Sonnenburg <sonne@debian.org> | 22:31 |
shogun-buildbot | build #464 of bsd1 - libshogun is complete: Failure [failed test] Build details are at http://www.shogun-toolbox.org/buildbot/builders/bsd1%20-%20libshogun/builds/464 blamelist: Soeren Sonnenburg <sonne@debian.org> | 22:33 |
shogun-buildbot | build #11 of ubu1 - libshogun is complete: Failure [failed test] Build details are at http://www.shogun-toolbox.org/buildbot/builders/ubu1%20-%20libshogun/builds/11 blamelist: Soeren Sonnenburg <sonne@debian.org> | 22:43 |
shogun-buildbot | build #61 of rpm1 - libshogun is complete: Failure [failed test] Build details are at http://www.shogun-toolbox.org/buildbot/builders/rpm1%20-%20libshogun/builds/61 blamelist: Soeren Sonnenburg <sonne@debian.org> | 22:46 |
-!- zxtx [~zv@guest-wireless-nup-198-147-121-195.usc.edu] has joined #shogun | 22:48 | |
ptizoom | sonney2k:the perl stuff is quite an Entwurf, and needs nice reduction... when it works | 23:04 |
-!- zxtx [~zv@guest-wireless-nup-198-147-121-195.usc.edu] has quit [Ping timeout: 244 seconds] | 23:19 | |
--- Log closed Thu Oct 25 00:00:17 2012 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!