--- Log opened Sun Jun 10 00:00:41 2012 | ||
n4nd0 | blackburn: ? | 00:01 |
---|---|---|
blackburn | :D | 00:01 |
blackburn | need matlab to test slep :) | 00:01 |
n4nd0 | blackburn: aham | 00:10 |
n4nd0 | do you need a copy of matlab? | 00:10 |
blackburn | no I have one | 00:11 |
n4nd0 | all right then :) | 00:11 |
blackburn | and I dooo nooot have crack for that | 00:11 |
blackburn | :D | 00:11 |
n4nd0 | ooh I see | 00:11 |
blackburn | sonney2k: still there? | 00:12 |
blackburn | I am searching for a conference where I can present shogun :D | 00:12 |
n4nd0 | nice | 00:13 |
n4nd0 | in Russia, or you don't mind where? | 00:13 |
blackburn | better far away :) | 00:14 |
n4nd0 | blackburn: just let me know if you want some help with it | 00:14 |
n4nd0 | I would love to participate in any way | 00:14 |
blackburn | sure | 00:14 |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 01:53 | |
-!- emrecelikten [~Anubis@176.40.238.218] has quit [Quit: Leaving.] | 02:21 | |
n4nd0 | good night guys | 02:34 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 02:36 | |
-!- blackburn [~blackburn@31.28.59.65] has quit [Ping timeout: 260 seconds] | 03:00 | |
-!- blackburn [~blackburn@31.28.59.65] has joined #shogun | 03:09 | |
CIA-9 | shogun: Sergey Lisitsyn master * rbc2cf3e / (6 files in 3 dirs): Added sparse inverse covariance estimator - http://git.io/TYOcFQ | 04:32 |
-!- blackburn [~blackburn@31.28.59.65] has quit [Ping timeout: 265 seconds] | 05:21 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 07:26 | |
-!- wiking_ [~wiking@huwico/staff/wiking] has joined #shogun | 07:54 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Read error: No route to host] | 07:54 | |
-!- wiking_ is now known as wiking | 07:54 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 08:10 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 08:15 | |
-!- blackburn [~blackburn@31.28.59.65] has joined #shogun | 10:38 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 10:47 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 10:54 | |
-!- blackburn [~blackburn@31.28.59.65] has quit [Ping timeout: 245 seconds] | 11:25 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 12:02 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 240 seconds] | 12:16 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has joined #shogun | 12:34 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has quit [Changing host] | 12:34 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 12:34 | |
-!- pluskid [~pluskid@111.120.62.134] has joined #shogun | 13:36 | |
-!- emrecelikten [~Anubis@176.40.238.218] has joined #shogun | 13:54 | |
-!- gsomix [~gsomix@80.234.31.247] has joined #shogun | 14:06 | |
gsomix | hi all | 14:06 |
emrecelikten | gsomix: Hello gsomix | 14:06 |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 260 seconds] | 14:24 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 252 seconds] | 14:38 | |
-!- blackburn [~blackburn@109.226.76.136] has joined #shogun | 14:40 | |
-!- pluskid [~pluskid@111.120.62.134] has quit [Quit: Leaving] | 14:41 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 14:44 | |
-!- blackburn [~blackburn@109.226.76.136] has left #shogun [] | 15:50 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 246 seconds] | 15:56 | |
-!- puffin444 [472e31fb@gateway/web/freenode/ip.71.46.49.251] has joined #shogun | 16:37 | |
-!- puffin444 [472e31fb@gateway/web/freenode/ip.71.46.49.251] has quit [Quit: Page closed] | 16:58 | |
-!- puffin444 [472e31fb@gateway/web/freenode/ip.71.46.49.251] has joined #shogun | 16:58 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has joined #shogun | 17:01 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has quit [Changing host] | 17:01 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 17:01 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 17:24 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 244 seconds] | 18:27 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has joined #shogun | 18:28 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has quit [Changing host] | 18:28 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 18:28 | |
-!- wiking_ [~wiking@68.68.31.85.customer.bluemilenetworks.com] has joined #shogun | 19:06 | |
-!- wiking_ [~wiking@68.68.31.85.customer.bluemilenetworks.com] has quit [Changing host] | 19:06 | |
-!- wiking_ [~wiking@huwico/staff/wiking] has joined #shogun | 19:06 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 252 seconds] | 19:08 | |
-!- wiking_ is now known as wiking | 19:08 | |
-!- blackburn [d55784cd@gateway/web/freenode/ip.213.87.132.205] has joined #shogun | 19:37 | |
gsomix | sonney2k, around? | 20:34 |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 260 seconds] | 20:35 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has joined #shogun | 20:36 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has quit [Changing host] | 20:36 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 20:36 | |
@sonney2k | gsomix, yes - whats up? | 20:39 |
gsomix | sonney2k, machines | 20:39 |
gsomix | can you look at director machines? | 20:39 |
-!- blackburn [d55784cd@gateway/web/freenode/ip.213.87.132.205] has quit [Ping timeout: 245 seconds] | 20:45 | |
-!- blackburn [~blackburn@188.168.3.9] has joined #shogun | 20:49 | |
@sonney2k | gsomix, looks fine - now we only need some example | 20:57 |
gsomix | sonney2k, have any ideas? | 20:58 |
blackburn | I see one problem here | 21:05 |
blackburn | train_machine is a) protected b) have default parameter | 21:05 |
@sonney2k | gsomix, you could do parzen windows with gaussian kernel (equation 55 in https://www.cs.utah.edu/~suyash/Dissertation_html/node11.html ) | 21:07 |
@sonney2k | for the directorkernelmachine | 21:07 |
blackburn | haha parzen finally | 21:07 |
blackburn | or with Epanechnikov kernel! | 21:08 |
@sonney2k | sounds like a disease | 21:08 |
blackburn | sonney2k: no typical kernel in atypical kernel mean :) | 21:09 |
blackburn | statistic kernel I mean | 21:10 |
blackburn | is there anyone to describe me what is advantage having sparse inverse covariance estimator? | 21:11 |
@sonney2k | gsomix, and for linearmachine - you could implement a linear perceptron algorithm is here (page 2 on top https://www.cs.cmu.edu/~avrim/ML09/lect0126.pdf) | 21:11 |
gsomix | ok | 21:12 |
@sonney2k | gsomix, but for the very first test just do something as simple as using alpha=-1 for negatively labeled examples and +1 otherwise for the directorkernelmachine | 21:13 |
gsomix | sonney2k, ok. what about dotfeatures? | 21:15 |
@sonney2k | and some random w vector in linear machines (b=0) | 21:18 |
@sonney2k | gsomix, for directordotfeatures - best example would be to just implement all the required operations for some numpy double matrix | 21:19 |
-!- blackburn [~blackburn@188.168.3.9] has quit [Ping timeout: 244 seconds] | 21:19 | |
@sonney2k | gsomix, then you can run the code once using your director features and once using shogun's densefeatures | 21:19 |
@sonney2k | and result should be the same | 21:19 |
@sonney2k | gsomix, I started a directordotfeatures example - but it is not totally there yet | 21:20 |
@sonney2k | gsomix, it might need some new operators in SGVector ( overload * / - + etc) | 21:21 |
@sonney2k | += / -= ... | 21:21 |
@sonney2k | gsomix, got disconnected | 21:32 |
@sonney2k | not sure if you've seen this | 21:32 |
@sonney2k | <sonney2k> gsomix, then you can run the code once using your director features and once using shogun's densefeatures | 21:32 |
@sonney2k | <sonney2k> and result should be the same | 21:32 |
@sonney2k | <sonney2k> gsomix, I started a directordotfeatures example - but it is not totally there yet | 21:32 |
@sonney2k | <sonney2k> gsomix, it might need some new operators in SGVector ( overload * / - + etc) | 21:32 |
@sonney2k | <sonney2k> += / -= ... | 21:32 |
gsomix | sonney2k, aha | 21:33 |
@sonney2k | gsomix, first line was gsomix, for directordotfeatures - best example would be to just implement all the required operations for some numpy double matrix | 21:33 |
@sonney2k | gsomix, does that make sense? | 21:33 |
gsomix | sonney2k, yep. I think I'll start with dotfeatures. | 21:34 |
@sonney2k | gsomix, can you fix the return 0 -> return false in your pull request | 21:42 |
@sonney2k | I would then merge these | 21:42 |
gsomix | ah ^__^'' | 21:42 |
gsomix | sonney2k, done | 21:45 |
CIA-9 | shogun: Evgeniy Andreev master * rb01c399 / src/shogun/machine/DirectorKernelMachine.h : added DirectorKernelMachine - http://git.io/KAkBrA | 21:49 |
CIA-9 | shogun: Evgeniy Andreev master * r0df5d4d / src/shogun/machine/DirectorLinearMachine.h : added DirectorLinearMachine - http://git.io/s7NjUQ | 21:49 |
CIA-9 | shogun: Evgeniy Andreev master * r1914e86 / (3 files in 2 dirs): new classes added to modular - http://git.io/-os58w | 21:49 |
CIA-9 | shogun: Evgeniy Andreev master * r60a7f70 / (2 files): fixed returns in train_function - http://git.io/19v1oQ | 21:49 |
CIA-9 | shogun: Soeren Sonnenburg master * r51c3b36 / (5 files in 2 dirs): Merge pull request #578 from gsomix/director_classes - http://git.io/SKuf0w | 21:49 |
@sonney2k | gsomix, btw please use git commit --amend next time | 21:51 |
@sonney2k | then push --force | 21:52 |
gsomix | sonney2k, mmm? why? | 21:52 |
@sonney2k | gsomix, to avoid the useless fix | 21:58 |
gsomix | a, ok | 21:58 |
@sonney2k | so one just has one OK commit | 21:58 |
n4nd0 | sonney2k: hey! can I double check with you an issue about the cutting plane algorithm we are using for SO?? | 22:03 |
@sonney2k | n4nd0, ask | 22:08 |
n4nd0 | sonney2k: http://www.cs.cornell.edu/people/tj/publications/finley_joachims_08a.pdf take a look here to second page | 22:08 |
n4nd0 | the algorithm, line 10 | 22:08 |
n4nd0 | and the gist here https://gist.github.com/2634487 | 22:08 |
n4nd0 | line 41 | 22:09 |
n4nd0 | in the paper it seems that they solve the QP each time they add a new constraint | 22:09 |
n4nd0 | inside the for loop | 22:09 |
n4nd0 | while in the gist it's outside the for loop | 22:09 |
n4nd0 | I wonder if both are equivalent | 22:10 |
n4nd0 | I am checking in some HMSVM code that Nico sent me and it seems to be done outside the for loop as well, like in the gist | 22:12 |
n4nd0 | this may be the right solution then | 22:12 |
@sonney2k | n4nd0, both will converge to the same solution | 22:17 |
@sonney2k | n4nd0, only difference is speed | 22:17 |
n4nd0 | ok | 22:17 |
@sonney2k | n4nd0, and that really depends on the optimizer speed / argmax speed | 22:17 |
@sonney2k | n4nd0, but it is clear - if you have zero costs for doing the optimization inside the loop then do it :) | 22:18 |
@sonney2k | n4nd0, and actually with cplex at least ( I guess mosek too) one can use the dual simplex method and 'hotstarts' | 22:19 |
n4nd0 | n4nd0: I am still debugging, I think I have a bug somewhere since the weight vector learned is all zeros | 22:19 |
n4nd0 | and I think that must be an error even if the dataset is so simply as the one I am using | 22:19 |
n4nd0 | oh I am back to black ... talking to myself ^ :D | 22:20 |
@sonney2k | n4nd0, well the trivial solution might mean that your example is too hard :) | 22:43 |
n4nd0 | sonney2k: :) I think there's no way for that to happen, I have 17 training examples of two classes, data distributed as Gaussians | 22:44 |
@sonney2k | n4nd0, why so many? | 22:51 |
@sonney2k | why not 2 examples - one in each class? | 22:51 |
n4nd0 | sonney2k: ok, I am going to try with that | 22:52 |
n4nd0 | it seems funny though | 22:53 |
@sonney2k | n4nd0, rule #1 try with the simplest possible case first | 22:55 |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 23:01 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has joined #shogun | 23:03 | |
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has quit [Changing host] | 23:03 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 23:03 | |
-!- gsomix [~gsomix@80.234.31.247] has quit [Ping timeout: 240 seconds] | 23:59 | |
--- Log closed Mon Jun 11 00:00:41 2012 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!