--- Log opened Thu Sep 06 00:00:17 2012 | ||
shogun-buildbot | build #89 of nightly_default is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/89 | 03:39 |
---|---|---|
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 07:17 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 07:48 | |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has joined #shogun | 08:07 | |
sonne|work | blackburn: we forgot heiko in NEWS! | 08:48 |
-!- uricamic [~uricamic@2001:718:2:1634:c1c6:c21e:4fd:ba07] has joined #shogun | 08:51 | |
-!- uricamic1 [~uricamic@2001:718:2:1634:c1c6:c21e:4fd:ba07] has joined #shogun | 08:55 | |
-!- uricamic [~uricamic@2001:718:2:1634:c1c6:c21e:4fd:ba07] has quit [Read error: Connection reset by peer] | 08:55 | |
blackburn | sonne|work: oh well | 08:56 |
sonne|work | we better mention him in 2.0.1 :D | 08:56 |
blackburn | sonne|work: 9 critical level tickets here | 08:59 |
blackburn | I start to hate that | 08:59 |
sonne|work | http://freecode.com/projects/shogun/date_metrics | 09:04 |
blackburn | sonne|work: LLTSA author? heheh | 09:51 |
sonne|work | nice right :) | 10:21 |
sonne|work | blackburn: I thought this is the motivation you need. | 10:22 |
blackburn | yeah kind of | 10:22 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 10:35 | |
blackburn | sonne|work: and what is number of downloads? | 10:36 |
sonne|work | no idea | 10:37 |
sonne|work | we don't monitor this | 10:37 |
sonne|work | usually between 100 and 300 unique visitors per days (has the website) | 10:37 |
blackburn | sonne|work: we've got no mentions on any website this time :( | 10:42 |
sonne|work | blackburn: we didn't write any blog post yet... | 10:43 |
sonne|work | blackburn: and we don't have nice figures - so it is hard for me to blog this time | 10:45 |
-!- cronor [~cronor@fb.ml.tu-berlin.de] has joined #shogun | 10:47 | |
blackburn | sonne|work: I have free 3 days from tomorrow may be I'll come with something | 10:55 |
blackburn | free from that damned | 10:55 |
blackburn | support | 10:55 |
blackburn | n4nd0: can HMM recognize strategy of flipping a coin? | 11:07 |
blackburn | i.e. if we learn it on | 11:07 |
blackburn | 0101010101 | 11:07 |
blackburn | or | 11:07 |
blackburn | 11111100000 | 11:07 |
blackburn | or 000111000111000 | 11:08 |
blackburn | see what I mean? | 11:08 |
n4nd0 | blackburn: do you always use the same coin? | 11:08 |
n4nd0 | ? | 11:08 |
blackburn | n4nd0: oh, wait | 11:09 |
blackburn | I think it is not a fair coin :D | 11:09 |
n4nd0 | I mean if the probability of tail and head is always the same | 11:09 |
blackburn | what I meant is can it learn to recognize that? | 11:09 |
blackburn | 3 zeros, 3 ones | 11:09 |
n4nd0 | is the prob. of tail and head always the same? | 11:09 |
blackburn | 4 zeros, 4 ones | 11:09 |
blackburn | etc | 11:09 |
blackburn | let it be not a coin anymore | 11:10 |
n4nd0 | ?? | 11:10 |
blackburn | :D | 11:10 |
n4nd0 | haha ok... | 11:10 |
blackburn | forget what I said before, from the beginining | 11:10 |
blackburn | I show you one or two fingers | 11:10 |
blackburn | with some strategy I know | 11:10 |
blackburn | can you learn a HMM that will adapt to my strategy? | 11:10 |
n4nd0 | yes | 11:11 |
n4nd0 | there is a couple of details though | 11:11 |
n4nd0 | the thing would be like | 11:11 |
n4nd0 | sometimes you follow one strategy where the prob. of 1 finger is x and of two fingers is 1-x | 11:11 |
n4nd0 | other times you follow another strategy where the prob. ... | 11:12 |
n4nd0 | you may change between strategies | 11:12 |
blackburn | I am just thinking of graphical example of HMMs | 11:12 |
n4nd0 | the only thing I can see is the fingers you show me | 11:12 |
n4nd0 | that can be modelled easily with an HMM | 11:12 |
n4nd0 | blackburn: do you mean to put some nice related to HMMs in the webpage or in graphical examples? | 11:14 |
blackburn | n4nd0: webpage/blog post | 11:14 |
n4nd0 | aham ok | 11:14 |
n4nd0 | they use HMMs in bioinformatics quite a bit | 11:15 |
n4nd0 | and they have nice plots | 11:15 |
n4nd0 | maybe sonne|work can point out something useful for this | 11:15 |
blackburn | n4nd0: yeah but we would need 4 state model | 11:15 |
blackburn | or we end with limited thymine and adenine alphabet :D | 11:16 |
blackburn | n4nd0: btw do you know why 4 aminoacids are there? | 11:16 |
n4nd0 | I don't know anything about aminoacids | 11:17 |
sonne|work | blackburn: what is wrong with the loaded cube example (that we already have)? | 11:19 |
blackburn | sonne|work: ? | 11:19 |
sonne|work | I created an example where 3 dices (all loaded) are drawn | 11:19 |
blackburn | n4nd0: a little more than 3 is the optimal base for a numeral system | 11:20 |
sonne|work | the HMM is used to learn the distribution that one can draw from each dice and when which dice is used | 11:20 |
blackburn | sonne|work: I've never seen that example :D | 11:20 |
blackburn | where? | 11:20 |
sonne|work | octave_static/graphical/hmm.m | 11:21 |
sonne|work | example is probably 8 years old :D | 11:21 |
blackburn | sonne|work: well, there is nothing wrong with it except I've never seen that working :) | 11:26 |
blackburn | I'll run it once I get home | 11:27 |
sonne|work | yeah old code in static if's | 11:27 |
blackburn | that would take a few hours today, switching city | 11:27 |
blackburn | I wish my mom gave me a teleport | 11:27 |
-!- cronor [~cronor@fb.ml.tu-berlin.de] has quit [Quit: cronor] | 11:58 | |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has quit [Quit: Page closed] | 14:20 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 14:42 | |
-!- uricamic1 [~uricamic@2001:718:2:1634:c1c6:c21e:4fd:ba07] has left #shogun [] | 14:43 | |
-!- dan_ [2eda6d52@gateway/web/freenode/ip.46.218.109.82] has joined #shogun | 15:08 | |
dan_ | Hi, I am trying to install shogun with MOSEK and CPLEX support but ./configure fails to find those lib | 15:09 |
dan_ | How can I add "/usr/local/Aimms/3.12.6/Solvers" to shogun links ? thats the path where libmosek.so, libcplex.so are. Thanks. | 15:10 |
sonne|work | dan_: ./configure --libs | 15:11 |
dan_ | thanks but MOSEK and CPLEX support are still not found .. | 15:15 |
dan_ | Is there a configure output example with CPLEX support ? | 15:49 |
-!- hoijui [~hoijui@141.23.65.251] has joined #shogun | 15:52 | |
sonne|work | dan_: sry I don't have either... | 15:53 |
dan_ | thanks anyway | 15:53 |
-!- dan_ [2eda6d52@gateway/web/freenode/ip.46.218.109.82] has quit [Quit: Page closed] | 15:53 | |
-!- jogin [86600232@gateway/web/freenode/ip.134.96.2.50] has joined #shogun | 16:32 | |
-!- jogin [86600232@gateway/web/freenode/ip.134.96.2.50] has quit [Client Quit] | 16:34 | |
-!- jogin [86600232@gateway/web/freenode/ip.134.96.2.50] has joined #shogun | 16:35 | |
jogin | Hello, I have some data for training and afterwards testing by python interface. As I see in shogun examples, I need my data in *.dat format. And that's my question - how can I create *.dat format? Can I find somewhere a sample? | 17:01 |
@sonney2k | jogin, no just load your data with python and pass a numpy matrix | 17:13 |
@sonney2k | (and use the python_modular interface!!) | 17:13 |
-!- ChanServ changed the topic of #shogun to: Shogun Machine Learning Toolbox | Version 2.0.0 just released | When asking questions please be patient or use the mailing list. We always answer but it might take a while. | Channel logs http://bit.ly/eQDk8Q | 17:16 | |
-!- heiko [~heiko@82.132.139.20] has joined #shogun | 17:16 | |
-!- heiko [~heiko@82.132.139.20] has left #shogun [] | 17:16 | |
jogin | @sonney2k: So, which object class to use for loading the data with Python? | 17:32 |
-!- jogin [86600232@gateway/web/freenode/ip.134.96.2.50] has quit [Quit: Page closed] | 18:05 | |
-!- blackburn [~blackburn@109.226.119.190] has joined #shogun | 18:31 | |
-!- hoijui [~hoijui@141.23.65.251] has quit [Ping timeout: 245 seconds] | 19:31 | |
@sonney2k | blackburn, alive? | 19:45 |
blackburn | sonney2k: yeap | 19:45 |
blackburn | sonney2k: what's up? wanna make me fixing that crash? :) | 19:46 |
@sonney2k | blackburn, yes! | 19:56 |
@sonney2k | do it! | 19:56 |
-!- hoijui [~hoijui@141.23.65.251] has joined #shogun | 19:59 | |
blackburn | okayy let me pack my code for gsoc now and I will check | 20:00 |
blackburn | locality preserving projections right? | 20:03 |
@sonney2k | blackburn, yes - fails when arpack is not available | 20:04 |
blackburn | sonney2k: ahh okay | 20:05 |
blackburn | easy tehn | 20:05 |
blackburn | sonney2k: so you will visit NY? | 20:05 |
@sonney2k | blackburn, not easy - it does sth in the atlas call | 20:11 |
blackburn | sonney2k: hah | 20:11 |
@sonney2k | ...which fails | 20:11 |
blackburn | sonney2k: still - easy | 20:12 |
@sonney2k | then please do it | 20:12 |
blackburn | doing | 20:12 |
@sonney2k | it is blocking me doing things on the new bb | 20:12 |
blackburn | oh blocker issue | 20:12 |
blackburn | I fscking love it | 20:12 |
blackburn | :D | 20:12 |
blackburn | QA always says the issue I work on blocks everything | 20:13 |
blackburn | we can't start nuclear war! | 20:13 |
@sonney2k | luckily you slow down the process :) | 20:16 |
@sonney2k | so war can wait :D | 20:17 |
@sonney2k | blackburn, I just figured out why I got tons of errors on the web site | 20:17 |
blackburn | sonney2k: why? | 20:17 |
@sonney2k | sqlite does auto_increment per default on primary (integer!) keys | 20:17 |
@sonney2k | but mysql does not | 20:17 |
blackburn | ehm | 20:21 |
blackburn | then enable it | 20:21 |
blackburn | :) | 20:21 |
@sonney2k | that is whay I did | 20:25 |
blackburn | sonney2k: ah so it is just a bad matrix | 20:37 |
blackburn | not a real bug | 20:37 |
blackburn | sonney2k: here you go | 20:40 |
CIA-35 | shogun: Sergey Lisitsyn master * r2e41877 / examples/undocumented/libshogun/converter_localitypreservingprojections.cpp : Better data matrix for LPP example - http://git.io/2FlrvQ | 20:40 |
@sonney2k | blackburn, ? | 20:42 |
@sonney2k | what is a bad matrix? | 20:42 |
blackburn | sonney2k: should work now | 20:42 |
blackburn | sonney2k: 0 1 2 3 4 5 6 ... | 20:42 |
blackburn | is a bad feature matrix | 20:42 |
blackburn | for *any* dim red algorithm | 20:42 |
@sonney2k | so data was too tough for this example to ever work? | 20:42 |
blackburn | yes | 20:42 |
blackburn | unsolvable generalized eigenvalue problem | 20:42 |
blackburn | we could regularize but I am lazy to integrate regularization to LPP | 20:43 |
blackburn | and I am not sure it is not an idiot way | 20:43 |
@sonney2k | yeah ok good enough now | 20:44 |
@sonney2k | I fixed a few issues on the website | 20:44 |
blackburn | sonney2k: I'll try to come up with pics and may be skeleton of post | 20:48 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 20:48 | |
blackburn | mighty fernando | 20:49 |
@sonney2k | n4nd0, hey there | 20:50 |
-!- hoijui [~hoijui@141.23.65.251] has quit [Ping timeout: 260 seconds] | 20:50 | |
n4nd0 | sonney2k: hi | 20:50 |
n4nd0 | hi all :) | 20:50 |
@sonney2k | I guess it is time that you both blackburn, n4nd0 do some edits on the website yourself | 20:50 |
blackburn | sonney2k: give me give me | 20:50 |
n4nd0 | aham ok | 20:51 |
n4nd0 | sonney2k: what do you want to see new? | 20:51 |
blackburn | sonney2k: may be we could afford to put ocr example online? | 20:51 |
blackburn | how to make an octave plot stay? | 20:53 |
shogun-buildbot | build #8 of rpm1 - libshogun is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/rpm1%20-%20libshogun/builds/8 | 20:54 |
blackburn | --persist okay | 20:55 |
blackburn | sonney2k: n4nd0: have you seen that ticket? http://developer.pidgin.im/ticket/15308 | 20:56 |
@sonney2k | blackburn, yay! you did it ;) | 20:57 |
blackburn | sonney2k: I've run hmm graphical exmaple | 20:58 |
blackburn | sonney2k: what is going on here? | 20:58 |
blackburn | should estimated path be equal to true path? | 20:58 |
blackburn | sonney2k: https://dl.dropbox.com/u/10139213/shogun/Screenshot%20from%202012-09-06%2022%3A58%3A51.png | 20:59 |
blackburn | I am a kind of lobotomy victim too so hard to understand | 21:00 |
blackburn | meanwhile in the scikits ML guys discuss large scale data - 1G | 21:01 |
blackburn | lol | 21:01 |
@sonney2k | blackburn, yes please add the flash thing under the applications link | 21:03 |
@sonney2k | (good exercise btw :) | 21:03 |
@sonney2k | blackburn, regarding hmm | 21:03 |
blackburn | sonney2k: no I meant may be | 21:04 |
@sonney2k | perfect result! | 21:04 |
blackburn | put it online really? | 21:04 |
@sonney2k | blackburn, not maybe do it! | 21:04 |
blackburn | *active* | 21:04 |
@sonney2k | active? | 21:04 |
@sonney2k | n4nd0, I want to see fixes ... the site has many bad/broken links and missing info | 21:04 |
blackburn | yes, I think how difficult is to make an java application or so | 21:04 |
blackburn | can we run a java application out there? | 21:05 |
@sonney2k | blackburn, no just the flash back is sufficient! | 21:05 |
@sonney2k | no java crap | 21:05 |
blackburn | what is not a crap? | 21:05 |
@sonney2k | (flash crap is bad enough) | 21:05 |
@sonney2k | exactly | 21:05 |
@sonney2k | blackburn, regarding hmm - it shows the states 1...3 corresponding to when which cube was drawn | 21:06 |
blackburn | sonney2k: so hmm is not broken? | 21:06 |
@sonney2k | blackburn, so it is *perfect* result | 21:06 |
@sonney2k | why should it be broken? | 21:06 |
@sonney2k | only parallel hmm did not work | 21:06 |
blackburn | no, tests are failing still | 21:07 |
@sonney2k | ahh and note that it cannot know whether dice 2 is dice 1 or vice versa | 21:07 |
@sonney2k | so things might be swapped (just have different name) | 21:07 |
@sonney2k | and distributions should also match pretty well | 21:07 |
blackburn | I feel braindamaged | 21:08 |
blackburn | :D | 21:08 |
blackburn | time to do something stupid | 21:08 |
blackburn | ahh I wanted to make edrt use eigen3 | 21:08 |
@sonney2k | blackburn, but don't commit this in master branch... | 21:10 |
@sonney2k | bugfix mode for now | 21:10 |
blackburn | sonney2k: I will once it is ready | 21:10 |
blackburn | in month or two | 21:10 |
blackburn | :D | 21:10 |
blackburn | recall it is my prolonged gsoc project? | 21:10 |
@sonney2k | we should probably really create a dev branch an do things there | 21:11 |
@sonney2k | blackburn, but more important we need a polished web site | 21:11 |
blackburn | yeah I received a strange passphrase | 21:11 |
blackburn | for schtazi | 21:11 |
blackburn | knock-knock who's there | 21:12 |
@sonney2k | n4nd0, too of course | 21:12 |
@sonney2k | heiko already tried to do some edits :D | 21:13 |
n4nd0 | it will be my first contact with django! | 21:14 |
@sonney2k | n4nd0, there is no django contact here ... just editing pages | 21:14 |
@sonney2k | *argh* I overwrote the doc page with the unfixed one | 21:14 |
@sonney2k | and there is no versioning nor undo :/ | 21:15 |
blackburn | sonney2k: what is username for me? :D | 21:15 |
@sonney2k | sergey | 21:16 |
@sonney2k | blackburn, you didn't read your emails *tsss* | 21:17 |
blackburn | sonney2k: can you find a mail where you mentioned I am a sergey? | 21:17 |
blackburn | :D | 21:17 |
blackburn | I received only schtazi passphrase | 21:18 |
n4nd0 | blackburn: where do you log in? | 21:18 |
blackburn | admin/ | 21:18 |
blackburn | n4nd0: ^ | 21:18 |
blackburn | admin!!!/!!! | 21:18 |
blackburn | :D | 21:18 |
@sonney2k | heh | 21:20 |
@sonney2k | n4nd0, for example we need links to the *latest* documentation, the current one and basically all older versions... | 21:21 |
@sonney2k | 0.10.0/ 1.0.0/ 1.1.0/ 2.0.0/ 2.0.1/ current@ latest@ | 21:21 |
n4nd0 | ok | 21:21 |
n4nd0 | sonney2k: why is it there a current and latest version? | 21:40 |
n4nd0 | is latest cutting edge code in the repo and current the latest release? | 21:41 |
blackburn | exacltly | 21:41 |
n4nd0 | blackburn: thanks ;) | 21:42 |
blackburn | sonney2k: do you know what is a 'position paper'? | 21:42 |
blackburn | for conference | 21:42 |
blackburn | ah found in description | 21:43 |
blackburn | A position paper presents an arguable opinion about an issue. The goal of a position paper is to convince the audience that your opinion is valid and worth listening to, without the need to present completed research work and/or validated results. It is, nevertheless, important to support your argument with evidence to ensure the validity of your claims. | 21:43 |
n4nd0 | sonney2k: here we go with links to the different doc versions | 21:48 |
n4nd0 | http://shogun-toolbox.org/page/documentation/information | 21:48 |
n4nd0 | time to bed | 21:53 |
n4nd0 | good night! | 21:53 |
blackburn | n4nd0: may you have no segfaults in your dreams | 21:54 |
n4nd0 | haha | 21:54 |
n4nd0 | thanks | 21:55 |
blackburn | :D | 21:55 |
n4nd0 | I won't leak either :P | 21:55 |
n4nd0 | exit | 21:55 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 21:55 | |
blackburn | men do not leak | 21:55 |
blackburn | we've got to reduce some code I think | 21:58 |
-!- hoijui [~hoijui@dslb-088-074-107-101.pools.arcor-ip.net] has joined #shogun | 22:26 | |
-!- blackburn [~blackburn@109.226.119.190] has quit [Quit: Leaving.] | 22:57 | |
--- Log closed Fri Sep 07 00:00:10 2012 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!