| --- Log opened Wed Jan 25 00:00:03 2012 | ||
| --- Day changed Wed Jan 25 2012 | ||
| @sonney2k | trees | 00:00 |
|---|---|---|
| @sonney2k | not everything but everything needed :) | 00:00 |
| @sonney2k | blackburn, I totally forgot to ask - did your exam go well? | 00:01 |
| blackburn | sonney2k: I consider shogun as 'core', but didn't formulated what is it to be core | 00:01 |
| @sonney2k | the optical illusion one I mean :D | 00:01 |
| blackburn | yeah, got 4/5 | 00:01 |
| @sonney2k | perfect | 00:01 |
| blackburn | I am happy to get 4 for this shit hah | 00:01 |
| blackburn | wait a min I would show you presentation | 00:02 |
| @sonney2k | blackburn, that is a nice page http://scikit-learn.org/stable/auto_examples/index.html | 00:02 |
| blackburn | http://dl.dropbox.com/u/10139213/share/05-%D0%B8%D1%82%D0%B5%D1%80%D0%B0%D1%86%D0%B8%D0%BE%D0%BD%D0%BD%D1%8B%D0%B9.pdf | 00:02 |
| blackburn | yes, I like these guys | 00:02 |
| blackburn | they rock | 00:02 |
| @sonney2k | sth like that for shogun would be really great for gsoc and fun for a student | 00:02 |
| blackburn | but they dimreduction sucks haha | 00:03 |
| blackburn | their | 00:03 |
| @sonney2k | blackburn, heh - nice :) | 00:03 |
| blackburn | sonney2k: how can one memorize this? | 00:03 |
| @sonney2k | by looking it up | 00:04 |
| blackburn | sonney2k: 28 topics totally, it is one of them | 00:04 |
| @sonney2k | if we manage to get in to say 2-3 years of gsoc we really have a wide base of all the basic features one usually needs in shogun | 00:05 |
| @sonney2k | even now I am using it for a lot of things | 00:06 |
| blackburn | sonney2k: about sklearn dimreduction | 00:06 |
| @sonney2k | I have to sleep now! | 00:06 |
| blackburn | I know a way (2 lines fix) how to speed up LLE | 00:07 |
| blackburn | but didn't tell it to them yet | 00:07 |
| @sonney2k | blackburn, well their SVMs certainly are no match to shogun stuff | 00:07 |
| blackburn | do you think it is fair? :) | 00:07 |
| @sonney2k | either | 00:07 |
| @sonney2k | as you want | 00:07 |
| @sonney2k | did they fix anything in shogun? | 00:07 |
| blackburn | not really :) | 00:07 |
| @sonney2k | http://jmlr.csail.mit.edu/papers/volume12/pedregosa11a/pedregosa11a.pdf | 00:08 |
| @sonney2k | have you read? | 00:08 |
| @sonney2k | table 1 lists some experiments for very tiny data sets | 00:08 |
| blackburn | sonney2k: glanced over, is there anything interesting? | 00:08 |
| @sonney2k | showing that shogun is slower | 00:08 |
| blackburn | sonney2k: well did you see my measures? | 00:09 |
| @sonney2k | but that is all in the 1 second range | 00:09 |
| @sonney2k | not sure how they can mark 5.2 seconds vs. 5.63 seconds and then 5.2 secs as the clear winner | 00:09 |
| blackburn | 19.43 vs 473.34 | 00:10 |
| blackburn | seconds | 00:10 |
| blackburn | for 10000 3d vectors | 00:10 |
| @sonney2k | why is our PCA slower? | 00:10 |
| blackburn | no idea | 00:10 |
| blackburn | well it is not really optimal | 00:11 |
| blackburn | sonney2k: what did they measure? | 00:12 |
| @sonney2k | whatever - I don't even look at data sets of size ~5000 nowadays | 00:13 |
| @sonney2k | nothing below 100k examples :) | 00:13 |
| blackburn | sonney2k: hah that's funny I care too much with measures | 00:13 |
| blackburn | trying to make them fair.. | 00:13 |
| @sonney2k | I guess hardly any toolkit except shogun works for that | 00:13 |
| @sonney2k | I wonder why reviewers didn't criticize this table | 00:14 |
| @sonney2k | but hey | 00:14 |
| blackburn | sonney2k: okay I would BLAS the PCA just now | 00:14 |
| @sonney2k | I really need to sleep now | 00:14 |
| blackburn | or no | 00:14 |
| blackburn | would better do the same | 00:14 |
| @sonney2k | it is blas'd IIRC? | 00:14 |
| @sonney2k | the same? | 00:14 |
| blackburn | sonney2k: sub-mean ops could be blased too | 00:14 |
| blackburn | sonney2k: yeah, would better sleep too | 00:14 |
| blackburn | sonney2k: I'm going to submit my paper in 2 days | 00:15 |
| @sonney2k | anything that fits in memory is reasonably fast | 00:16 |
| @sonney2k | doesn't matter if it takes 1.5 or 1.0 seconds :) | 00:16 |
| @sonney2k | well I am really off now | 00:16 |
| @sonney2k | cu | 00:16 |
| blackburn | sonney2k: well scikits multiplies two 10000 * 10000 matrices when embedding 10000 vectors | 00:17 |
| blackburn | (we don't) :D | 00:17 |
| blackburn | okay see you | 00:17 |
| @sonney2k | http://packages.python.org/milk/benchmarks.html | 00:17 |
| blackburn | so it matters sometimes | 00:17 |
| @sonney2k | look at this milk :) | 00:17 |
| * sonney2k is typing while asleep | 00:17 | |
| blackburn | milk is pretty fast and for kmeans and SVM learning it is the fastest system. | 00:17 |
| blackburn | :D | 00:17 |
| blackburn | :D | 00:17 |
| blackburn | :D | 00:18 |
| blackburn | LOL | 00:18 |
| blackburn | omg lol | 00:18 |
| blackburn | sleep well :) | 00:19 |
| -!- blackburn [~qdrgsm@31.28.63.56] has quit [Quit: Leaving.] | 00:54 | |
| -!- Ram108 [~amma@14.96.38.9] has joined #shogun | 07:22 | |
| -!- Ram108 [~amma@14.96.38.9] has quit [Ping timeout: 240 seconds] | 08:38 | |
| -!- blackburn [~qdrgsm@31.28.63.56] has joined #shogun | 08:57 | |
| -!- jekintrivedi [~jekin@27.4.211.159] has joined #shogun | 09:01 | |
| -!- jekintrivedi [~jekin@27.4.211.159] has left #shogun [] | 09:02 | |
| -!- jekintrivedi [~jekin@27.4.211.159] has joined #shogun | 09:03 | |
| -!- jekintrivedi [~jekin@27.4.211.159] has quit [Read error: Connection reset by peer] | 09:32 | |
| -!- blackburn [~qdrgsm@31.28.63.56] has quit [Quit: Leaving.] | 09:59 | |
| -!- blackburn [~qdrgsm@31.28.63.56] has joined #shogun | 10:00 | |
| blackburn | surprising truth, sckikits is faster on toy data ;( | 15:40 |
| -!- Ram108 [~amma@219.64.146.186] has joined #shogun | 16:14 | |
| -!- Netsplit *.net <-> *.split quits: Ram108 | 16:34 | |
| -!- Netsplit *.net <-> *.split quits: naywhayare, blackburn | 16:34 | |
| -!- Netsplit *.net <-> *.split quits: @sonney2k, sonne|work | 16:34 | |
| -!- Netsplit *.net <-> *.split quits: shogun-buildbot | 16:34 | |
| -!- Netsplit *.net <-> *.split quits: CIA-10 | 16:34 | |
| -!- Netsplit over, joins: CIA-10, shogun-buildbot, naywhayare, @sonney2k, Ram108, sonne|work | 16:55 | |
| -!- blackburn [~qdrgsm@31.28.63.56] has joined #shogun | 17:06 | |
| -!- Ram108 [~amma@219.64.146.186] has quit [Ping timeout: 248 seconds] | 18:39 | |
| -!- Ram108 [~amma@14.96.105.129] has joined #shogun | 19:08 | |
| blackburn | sonney2k: around? | 19:19 |
| -!- cwidmer [8ca3fe9d@gateway/web/freenode/ip.140.163.254.157] has joined #shogun | 19:33 | |
| blackburn | sonney2k: I would need your time today :) | 19:57 |
| -!- blackburn [~qdrgsm@31.28.63.56] has quit [Quit: Leaving.] | 21:07 | |
| -!- blackburn [~qdrgsm@31.28.63.56] has joined #shogun | 21:11 | |
| shogun-buildbot | build #115 of nightly_default is complete: Failure [failed test_1] Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/115 | 21:52 |
| -!- Ram108 [~amma@14.96.105.129] has quit [Ping timeout: 245 seconds] | 23:22 | |
| --- Log closed Thu Jan 26 00:00:19 2012 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!