Open in new window / Try shogun cloud
--- Log opened Thu Nov 29 00:00:15 2012
wikingthis is funny00:02
wikingn4nd0: i've just found 3 different papers with 3 different author sets on 3 different publication years talking about essentially the same00:02
n4nd0nice!00:03
n4nd0what about?00:03
wikingn4nd0: apparently i've found the solution for this approx00:17
wikingindeed00:18
n4nd0cool00:18
n4nd0what is it?00:18
wiking1/\beta \log \sum exp(\beta * f(x,y,w))00:18
wikinglim \beta -> \inf00:18
wikingwill result in00:18
wikingmax (f(x,y,w))00:19
wikingi mean00:19
wikingi've missed \delta(y,y_i)00:19
wikingbut you know what i mena00:19
wikingmean00:19
n4nd0yeah I kind of get the idea00:19
n4nd0I don't know right now if that is the true answer for the limit00:19
n4nd0but I believe you :)00:19
wikingwell in the paper they even have a proof for it00:20
wikingso yeah i trust them as well00:20
wikingactually i've read it00:21
wikingit makes sense00:21
wikingso anyhow00:21
wikingwe could add a softmax thing for so learning00:21
n4nd0for the so-svm?00:22
wikingyes00:22
n4nd0ok00:22
n4nd0do you get something similar to a CRF then?00:22
wikingyes00:23
wiking\beta = 0 CRF00:23
wiking\beta = inf SSVM00:23
n4nd0interesting00:23
wikingand it's super easy to implement this with BMRM00:24
wikingn4nd0: Entropy and Margin Maximization for Structured Output Learning00:25
wikingcheck this paper00:25
wikingmmm00:36
wikingthis paper is cool00:36
wikingi've just realised that we can do MIL with latent framework :)00:36
n4nd0what's MIL?00:37
wikingmultiple instance learning00:37
wikingso you get n set of (x_1,...,x_n)00:38
wikingand u know that each set has a label y00:38
n4nd0haha it seems that new types of learning don't stop popping out00:38
wikingbut you don't know the exact label for each x_i00:38
wikingand of course it could be that you get a set of x-es00:39
wikingwith label y_i00:39
wikingbut that some of those x-es are not really y_i00:39
wikingWe construct a one-dimensional synthetic dataset which illustrates the de- ficiencies of the MI-SVM. A positive bag consists of p positive instances and 50 - p negative (0 < p <= 50), a negative bag contains 50 negative instances. The individual instances are hard to classify: the positive instances are Gaussian dis- tributed with mean 0.6 whereas the negative instances are Gaussian distributed with mean 0, the variance for both classes is 1.00:40
wikingthis is what they do00:40
n4nd0aham it sounds cool00:41
n4nd0a way to get rid of data outliers maybe00:41
wikinglet see if i can do this00:42
wikingshould be easy to generate this dataset :)00:42
n4nd0yeah00:42
n4nd0time to sleep00:53
n4nd0good night!00:53
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving]00:55
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has joined #shogun01:28
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has quit [Quit: Ex-Chat]02:41
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has joined #shogun02:41
shogun-buildbotbuild #158 of nightly_all is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_all/builds/15803:11
shogun-buildbotbuild #189 of nightly_default is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/18903:23
-!- blackburn [~blackburn@188.168.3.0] has quit [Ping timeout: 246 seconds]04:57
-!- blackburn [~blackburn@37.61.179.230] has joined #shogun05:22
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has quit [Read error: Connection reset by peer]06:31
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has joined #shogun06:32
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has quit [Ping timeout: 250 seconds]07:24
-!- ptizoom [~christian@79-71-89-182.dynamic.dsl.as9105.com] has joined #shogun07:25
-!- blackburn [~blackburn@37.61.179.230] has quit [Quit: Leaving.]08:12
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/228f6ab20ea16dd35d4c0f4056f93ef9507ac04708:26
-shogungit:#shogun- shogun/master 228f6ab Soeren Sonnenburg: drop -fPIC / pthreads on cygwin08:26
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/9417651b7b96268510c64454e720cee08a3a432808:30
-shogungit:#shogun- shogun/master 9417651 Soeren Sonnenburg: fix compile error on cygwin08:30
shogun-buildbotbuild #458 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/458  blamelist: Soeren Sonnenburg <sonne@debian.org>08:35
shogun-buildbotbuild #459 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/459  blamelist: Soeren Sonnenburg <sonne@debian.org>08:36
shogun-buildbotbuild #694 of deb3 - modular_interfaces is complete: Failure [failed compile octave_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/694  blamelist: Soeren Sonnenburg <sonne@debian.org>09:04
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/d517033406b1d90e4fb6400224762a951b843ab909:11
-shogungit:#shogun- shogun/master d517033 Soeren Sonnenburg: make pthreads optional09:11
shogun-buildbotbuild #460 of cyg1 - libshogun is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/460  blamelist: Soeren Sonnenburg <sonne@debian.org>09:26
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/c333f44e165a880497253af95d049831c852cb5209:29
-shogungit:#shogun- shogun/master c333f44 Soeren Sonnenburg: fix error: cannot convert 'shogun::SGMatrix<double>' to 'const float64_t*' in assignment...09:29
shogun-buildbotbuild #695 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/69509:32
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has joined #shogun09:37
shogun-buildbotbuild #461 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/461  blamelist: Soeren Sonnenburg <sonne@debian.org>09:39
blackburnsomeone has stolen a compiler :(09:41
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/fab840bb26e039e7ae8a703ab4fd643ba31e3c3609:49
-shogungit:#shogun- shogun/master fab840b Soeren Sonnenburg: remove leftin function call09:49
sonne|workblackburn: cleaned up your desk?09:49
blackburnsonne|work: haha I am cleaning up my tasks09:49
shogun-buildbotbuild #696 of deb3 - modular_interfaces is complete: Failure [failed compile octave_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/696  blamelist: Soeren Sonnenburg <sonne@debian.org>09:50
sonne|workit is important that they shine once you leave :)09:50
blackburnwell I still have 3 tickets in bugtracking09:50
blackburnsonne|work: and hey as I don't use anything but computer I have nothing on my desk09:51
sonne|workhmmhh buildslave #4 seems to have some issues09:53
blackburngood to have slaves in the modern world we live09:53
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun09:55
blackburnokay I described one problem through phase of moon09:55
blackburnn4nd0: aha! catch you09:56
n4nd0blackburn: hello hello :)09:56
shogun-buildbotbuild #462 of cyg1 - libshogun is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/46209:59
shogun-buildbotbuild #463 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/463  blamelist: Soeren Sonnenburg <sonne@debian.org>09:59
-!- cavalier- [~cavalier-@70.118.110.73] has quit [Read error: Connection reset by peer]10:02
-!- cavalier- [cavalier-@70.118.110.73] has joined #shogun10:02
wikingsonne|work: i'm getting this warning with the HEAD of the master branch: ../shogun/lib/SGSparseVector.h:26:27: note: previous use is here10:05
wikingtemplate <class T> struct SGSparseVectorEntry ^10:05
wiking../shogun/io/File.h:23:21: note: did you mean struct here?10:05
wikingtemplate <class ST> class SGSparseVectorEntry;10:05
shogun-buildbotbuild #698 of deb3 - modular_interfaces is complete: Failure [failed compile octave_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/698  blamelist: Soeren Sonnenburg <sonne@debian.org>10:23
shogun-buildbotbuild #697 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/69710:28
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/a85550094ec860e020803d5a7b5540fd285e307910:31
-shogungit:#shogun- shogun/master a855500 Soeren Sonnenburg: fix compile error when pthreads are not available10:31
sonne|workprevious?10:33
sonne|workwhat?10:33
wikingyes10:34
wikingsonne|work: i mean in SGSparseVector you declare SGSparseVectorEntry as struct10:34
wikingbut then for example in File.h you refere to it as a class10:35
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/8a5274f3ea1a32182a5e6c6af08fc012a48df8c910:35
-shogungit:#shogun- shogun/master 8a5274f Soeren Sonnenburg: SGSparseVectorEntry is in fact a struct10:35
wikingvoila10:35
sonne|workwiking: does it build now?10:45
wikingit was building before10:46
wikingjust flooding with warnings about this10:46
wikingyeah better thnx10:47
sonne|workk10:48
shogun-buildbotbuild #464 of cyg1 - libshogun is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/464  blamelist: Soeren Sonnenburg <sonne@debian.org>10:50
wikingok i thnk i have an awesome example coming up for shogun10:52
wikingit's even graphical10:52
wikingso you can post it to the webpage10:52
blackburndoes it contain tits?10:53
wikingblackburn: :D10:54
wikingsonne|work: mmm i think there's a way to connect buildbots as well to the pull request hook in github10:54
sonne|workwiking: ?10:55
wikingsonne|work: well i've just seen this10:55
sonne|workit is10:55
wikingsonne|work: https://github.com/Itseez/opencv/pull/18110:55
sonne|workahh10:55
sonne|workok so you mean even before commit10:55
wikingsonne|work: and that link there for good to merge10:55
wikingis not coming from travis10:55
wikingit's from a buildbot10:55
wikingsonne|work: indeed, so it's like travis10:56
wikingsonne|work: ah ok i think i know how it works10:58
wikingsonne|work: you need to define in service hooks10:58
wikinga webhook url10:58
wikingas it says10:58
wiking"We'll hit these URLs with POST requests when you push to us, passing along information about the push. More information can be found in the Post-Receive Guide."10:58
wikingsonne|work: and it seems there should be a module for this to be handled in buildbot10:59
wikinghttp://trac.buildbot.net/ticket/218611:00
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has quit [Ping timeout: 264 seconds]11:00
wikingalthoguh i couldnt' find any docs about this11:01
wikingsonne|work: http://buildbot.net/buildbot/docs/0.8.3/reference/buildbot.status.web.hooks.github-module.html11:01
shogun-buildbotbuild #465 of cyg1 - libshogun is complete: Failure [failed compile]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/465  blamelist: Soeren Sonnenburg <sonne@debian.org>11:04
sonne|workwiking: not so clear to me how to set this up - I mean we don't want to mix commits with committed stuff11:06
sonne|workblackburn: cyg1 is still failing with your last commit :/11:07
blackburnja11:08
shogun-buildbotbuild #699 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/69911:15
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has joined #shogun11:15
shogun-buildbotbuild #700 of deb3 - modular_interfaces is complete: Failure [failed compile octave_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/700  blamelist: Soeren Sonnenburg <sonne@debian.org>11:19
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving]12:37
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has quit [Quit: Page closed]15:55
--- Log closed Thu Nov 29 18:10:38 2012
--- Log opened Thu Nov 29 18:10:47 2012
-!- shogun-toolbox [~shogun@7nn.de] has joined #shogun18:10
-!- Irssi: #shogun: Total of 9 nicks [1 ops, 0 halfops, 0 voices, 8 normal]18:10
-!- Irssi: Join to #shogun was synced in 6 secs18:10
wikingmmm it's weird19:47
wikingcan't i make a 1 dim DenseFeatures via numpy19:48
wiking?19:48
wikingi mean features dim = 119:48
wikingit seems so :(19:52
wikingsonne|work: ping?19:52
@sonney2kwiking, wassup?20:24
@sonney2kwiking, [[1,2,3]]20:25
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 246 seconds]20:29
-!- blackburn [~blackburn@188.168.3.33] has joined #shogun20:57
-!- wiking [~wiking@info2k1.hu] has joined #shogun21:21
wikingmmm21:21
wikingi wonder what was happening here21:21
wiking:)21:21
wikingsonney2k: yeah in the meanwhile i've found out that i needed to transpose that vector21:22
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun22:01
-!- blackburn [~blackburn@188.168.3.33] has quit [Quit: Leaving.]22:02
@sonney2kwiking, I finally understand why the buildbot is failing on cygwin22:04
@sonney2kwe use a reserved word22:04
@sonney2k__from! nothing should be __'t22:05
wiking??22:07
wikinglolface22:07
wikingyeah that's uricamic's  coe22:07
wikingcode22:07
-shogungit:#shogun- [shogun] sonney2k pushed 1 new commit to master: https://github.com/shogun-toolbox/shogun/commit/1f7a6cca7f7bb58616eb0f2b9f884617c49b19f222:08
-shogungit:#shogun- shogun/master 1f7a6cc Soeren Sonnenburg: fixup cygwin build -> never use __reserved __keywords!22:08
wikinghahah i bet blackburn woulld love this nameing: MILFeature ;)22:08
@sonney2kwiking, no blackburn introduced some convenience constructor +       TMultipleCPinfo(uint32_t __from, uint32_t _N) : _from(__from), N(_N) {  }22:09
@sonney2kand that kills it22:09
@sonney2ktook me ages to find out :/22:09
wiking:(22:10
wikingyeah i would have never guessed that __* is reserved in cygwin's gcc22:10
@sonney2kwiking, btw I wrote some example for dynarray and likely found the issue22:10
wikingsonney2k: great!22:10
@sonney2kwe memset(0) the newly allocated memory22:10
wikingsonney2k: yeah i've seen that22:10
@sonney2kthat doesn't work with objects of course22:10
wikingbut it's only memsetting the new area no?22:10
@sonney2kkaboom22:10
@sonney2ksure but that would be and SGVector sitting on that new area22:11
@sonney2kmemset it22:11
@sonney2kand then at some point it gets destructed22:11
@sonney2kI can easily fix this - only question is if sth relies on the memset 0 behavior22:11
wikingwho's using DynArray?22:12
@sonney2klots of stuff22:12
wikingah well then yeah good luck with debugging it out22:12
wiking:(22:12
wikingbtw something like this would be fun as well22:13
wikinghttp://lxr.linux.no/+trees22:13
@sonney2kgood thing with the new malloc templates22:13
@sonney2kis though that one can fix it22:13
@sonney2kwiking, I actually think that this might not be a problem at all - let me commit and see what fails22:15
wikingheheh glhf22:15
wiking;)22:15
wikingwtf: TypeError: not a numpy vector of appropriate type22:29
shogun-buildbotbuild #466 of cyg1 - libshogun is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/46622:30
wikingsonney2k: seems u fixed it! \o/22:37
shogun-buildbotbuild #701 of deb3 - modular_interfaces is complete: Failure [failed compile octave_modular]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/701  blamelist: Soeren Sonnenburg <sonne@debian.org>22:42
--- Log closed Fri Nov 30 00:00:15 2012