--- Log opened Wed May 21 00:00:57 2014 | ||
thoralf|home | abinash_panda: Ping? | 00:14 |
---|---|---|
-!- pickle27f [~pickle27@216.191.231.218] has quit [Quit: This computer has gone to sleep] | 00:20 | |
-!- iglesiasg [~iglesiasg@194.151.204.16] has joined #shogun | 00:49 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 00:49 | |
@iglesiasg | Hey guys | 00:51 |
kislay | yo iglesiasg :) | 01:05 |
-!- khalednasr [~k.nasr92@41.69.238.139] has quit [Ping timeout: 252 seconds] | 01:19 | |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 01:24 | |
shogun-notifier- | shogun: Parijat Mazumdar :develop * f7b47a9 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/f7b47a9b44aa46e4d50cfaabe5b8e50e13ad1e47 | 01:24 |
shogun-notifier- | shogun: CART added | 01:24 |
shogun-notifier- | shogun: Fernando Iglesias :develop * 3dcc48e / / (3 files): https://github.com/shogun-toolbox/shogun/commit/3dcc48e63784f907adad4c7c5504d5d4d196e39b | 01:24 |
shogun-notifier- | shogun: Merge pull request #2239 from mazumdarparijat/CART | 01:24 |
shogun-notifier- | shogun: | 01:24 |
shogun-notifier- | shogun: CART added | 01:24 |
shogun-buildbot | build #2962 of deb1 - libshogun is complete: Success [build successful] Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/2962 | 01:29 |
-!- iglesiasg [~iglesiasg@194.151.204.16] has quit [Quit: Leaving] | 01:32 | |
shogun-buildbot | build #2299 of bsd1 - libshogun is complete: Success [build successful] Build details are at http://buildbot.shogun-toolbox.org/builders/bsd1%20-%20libshogun/builds/2299 | 01:35 |
-!- thoralf|home [~thoralf@91-65-137-195-dynip.superkabel.de] has quit [Quit: Konversation terminated!] | 01:42 | |
shogun-buildbot | build #632 of precise - libshogun is complete: Success [build successful] Build details are at http://buildbot.shogun-toolbox.org/builders/precise%20-%20libshogun/builds/632 | 01:49 |
-!- kislay [~Abhijeet@103.240.205.174] has quit [Ping timeout: 255 seconds] | 01:53 | |
-!- travis-ci [~travis-ci@ec2-75-101-216-118.compute-1.amazonaws.com] has joined #shogun | 02:20 | |
travis-ci | [travis-ci] it's Fernando Iglesias's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/25649883 | 02:20 |
-!- travis-ci [~travis-ci@ec2-75-101-216-118.compute-1.amazonaws.com] has left #shogun [] | 02:20 | |
-!- abinash_panda [0e8b2998@gateway/web/freenode/ip.14.139.41.152] has quit [Ping timeout: 240 seconds] | 02:55 | |
-!- yorkerlin [81614fb3@gateway/web/freenode/ip.129.97.79.179] has quit [Quit: Page closed] | 02:56 | |
shogun-buildbot | build #10 of nightly_fedora is complete: Failure [failed git] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_fedora/builds/10 blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>, Wu Lin <yorker.lin@gmail.com>, Kevin Hughes <kevinhughes27@gmail.com>, Parijat Mazumdar <mazumdarparijat@gmail.com> | 03:00 |
-!- pickle27 [~pickle27@192-0-136-118.cpe.teksavvy.com] has joined #shogun | 03:55 | |
shogun-buildbot | build #311 of debian wheezy - memcheck is complete: Failure [failed memory check] Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/311 blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>, Parijat Mazumdar <mazumdarparijat@gmail.com> | 03:58 |
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 04:24 | |
-!- Saurabh7 [~Saurabh7@117.222.8.201] has joined #shogun | 05:16 | |
-!- Saurabh7 [~Saurabh7@117.222.8.201] has quit [Ping timeout: 255 seconds] | 06:00 | |
-!- Saurabh7 [~Saurabh7@117.222.8.201] has joined #shogun | 06:02 | |
shogun-buildbot | build #808 of nightly_default is complete: Failure [failed notebooks] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/808 blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>, Wu Lin <yorker.lin@gmail.com>, Kevin Hughes <kevinhughes27@gmail.com>, Parijat Mazumdar <mazumdarparijat@gmail.com> | 06:15 |
-!- abinash_panda [dce36163@gateway/web/freenode/ip.220.227.97.99] has joined #shogun | 06:16 | |
abinash_panda | Hey thoralf ! | 06:16 |
-!- abinash_panda [dce36163@gateway/web/freenode/ip.220.227.97.99] has quit [] | 06:25 | |
-!- Saurabh7 [~Saurabh7@117.222.8.201] has quit [Ping timeout: 255 seconds] | 06:39 | |
-!- Saurabh7 [~Saurabh7@59.88.35.186] has joined #shogun | 06:52 | |
-!- pickle27 [~pickle27@192-0-136-118.cpe.teksavvy.com] has quit [Remote host closed the connection] | 06:58 | |
-!- kislay [~Abhijeet@103.240.206.194] has joined #shogun | 07:05 | |
-!- abinash_panda [d2d43db9@gateway/web/freenode/ip.210.212.61.185] has joined #shogun | 08:04 | |
-!- abinash_panda [d2d43db9@gateway/web/freenode/ip.210.212.61.185] has quit [Ping timeout: 240 seconds] | 08:33 | |
-!- sonne|work [~sonnenbu@sams-office-nat.tomtomgroup.com] has quit [Quit: Leaving.] | 08:38 | |
-!- sonne|work [~sonnenbu@sams-office-nat.tomtomgroup.com] has joined #shogun | 08:39 | |
-!- thoralf|home [~thoralf@91-65-137-195-dynip.superkabel.de] has joined #shogun | 09:13 | |
@besser82 | thoralf|home, g'Morning! :D | 09:14 |
thoralf|home | Moin. | 09:14 |
thoralf|home | Yo besser | 09:14 |
thoralf|home | besser82: What's up? | 09:15 |
@besser82 | thoralf|home, see query ;) | 09:16 |
thoralf|home | SELECT count(*) FROM irc | 09:17 |
thoralf|home | 123456 | 09:17 |
-!- jiaolong [9e6d09e4@gateway/web/freenode/ip.158.109.9.228] has joined #shogun | 09:21 | |
-!- thoralf|home [~thoralf@91-65-137-195-dynip.superkabel.de] has quit [Quit: Konversation terminated!] | 10:01 | |
-!- kislay [~Abhijeet@103.240.206.194] has quit [Ping timeout: 240 seconds] | 10:22 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun | 12:01 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 12:01 | |
Saurabh7 | hey HeikoS ! | 12:06 |
@HeikoS | Saurabh7: hey! how are things | 12:06 |
@HeikoS | nice to see you here | 12:06 |
Saurabh7 | i am usually around | 12:06 |
Saurabh7 | but our time dos\esnt match i think | 12:06 |
@HeikoS | Saurabh7: I haven't been recently since very busy, but I plan to hang around most of the time | 12:06 |
Saurabh7 | :) | 12:06 |
@HeikoS | Saurabh7: yeah indeed :) | 12:06 |
@HeikoS | what is yours? | 12:06 |
Saurabh7 | HeikoS, right now its 3 30 noon | 12:07 |
Saurabh7 | gmt + 530 i think | 12:07 |
@HeikoS | ah wow here we have 1108 | 12:08 |
@HeikoS | am | 12:08 |
@HeikoS | Saurabh7: I am a bit busy now, but shall we try to talk a bit later today? | 12:08 |
@HeikoS | I would like to discuss a few things with you | 12:08 |
@HeikoS | will you be around for a while? | 12:08 |
Saurabh7 | HeikoS, yup | 12:09 |
@HeikoS | Saurabh7: cool! | 12:09 |
@HeikoS | see you in a few hours then! :) | 12:09 |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Quit: Leaving.] | 12:09 | |
Saurabh7 | HeikoS, okay will be here | 12:09 |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun | 12:13 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 12:13 | |
-!- Saurabh7 [~Saurabh7@59.88.35.186] has quit [Ping timeout: 264 seconds] | 12:15 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Quit: Leaving.] | 12:24 | |
-!- Saurabh7 [~Saurabh7@117.217.103.228] has joined #shogun | 12:38 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun | 12:55 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 12:55 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Client Quit] | 12:58 | |
-!- kislay [~Abhijeet@103.228.76.4] has joined #shogun | 13:02 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun | 13:34 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 13:34 | |
kislay | hey HeikoS . | 13:37 |
-!- khalednasr [~k.nasr92@41.69.214.145] has joined #shogun | 13:45 | |
lisitsyn1 | khalednasr: hey there | 14:08 |
khalednasr | lisitsyn1, hey | 14:08 |
lisitsyn1 | khalednasr: how is it going? | 14:09 |
khalednasr | lisitsyn1, done with the graph stuff, waiting for travis to clear before I send the PR :) | 14:10 |
lisitsyn1 | khalednasr: oh cool | 14:10 |
-!- [1]Saurabh7 [~Saurabh7@117.217.103.228] has joined #shogun | 14:26 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Quit: Leaving.] | 14:28 | |
-!- Saurabh7 [~Saurabh7@117.217.103.228] has quit [Ping timeout: 252 seconds] | 14:28 | |
-!- [1]Saurabh7 is now known as Saurabh7 | 14:28 | |
khalednasr | lisitsyn1, any suggestions as to how I should approach that file-loading thing? | 14:34 |
lisitsyn1 | khalednasr: yeah that's interesting question | 14:36 |
lisitsyn1 | khalednasr: actually we had issue like that with some model file | 14:36 |
lisitsyn1 | just this gsoc iirc | 14:36 |
khalednasr | which model? | 14:37 |
lisitsyn1 | khalednasr: https://github.com/shogun-toolbox/shogun/blob/develop/src/shogun/io/UAIFile.cpp | 14:37 |
lisitsyn1 | this is the thing | 14:38 |
lisitsyn1 | I don't know whether it is very good designed already | 14:38 |
lisitsyn1 | but afaik it is something similar to what you'd need | 14:38 |
khalednasr | cool! I'll take a look at it | 14:39 |
-!- travis-ci [~travis-ci@ec2-54-237-169-250.compute-1.amazonaws.com] has joined #shogun | 14:48 | |
travis-ci | [travis-ci] it's khalednasr's turn to pay the next round of drinks for the massacre he caused in khalednasr/shogun: http://travis-ci.org/khalednasr/shogun/builds/25687467 | 14:48 |
-!- travis-ci [~travis-ci@ec2-54-237-169-250.compute-1.amazonaws.com] has left #shogun [] | 14:48 | |
khalednasr | lisitsyn1, I took a look at the UAI file stuff, It would work if we only wanted to specify the connectivity of the network, but we also need to specify the type of layers and the hyper parameters | 14:59 |
khalednasr | lisitsyn1, maybe I could use JSON? | 14:59 |
lisitsyn1 | khalednasr: nono I mean it is something that serves similar purpose | 14:59 |
khalednasr | lisitsyn1, I see. what do you think about using json or xml? | 15:02 |
lisitsyn1 | khalednasr: I am quite ok with either of them | 15:02 |
khalednasr | lisitsyn1, cool | 15:02 |
lisitsyn1 | khalednasr: the question is | 15:02 |
lisitsyn1 | you know we have json | 15:03 |
lisitsyn1 | but it is something that serializes our shogun internal things into json | 15:03 |
lisitsyn1 | and back | 15:03 |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun | 15:04 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 15:04 | |
khalednasr | yeah I saw it | 15:04 |
-!- Saurabh7 [~Saurabh7@117.217.103.228] has quit [Quit: HydraIRC -> http://www.hydrairc.com <- The alternative IRC client] | 15:09 | |
-!- lambday [67157f4f@gateway/web/freenode/ip.103.21.127.79] has joined #shogun | 15:14 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Ping timeout: 252 seconds] | 15:15 | |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 15:15 | |
shogun-notifier- | shogun: lambday :feature/linalg * 115cc99 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/115cc9998f55d8856ac9eb61b778a386924bf101 | 15:15 |
shogun-notifier- | shogun: added rowwise and colwise sum methods in linalg | 15:15 |
shogun-notifier- | shogun: Soumyajit De :feature/linalg * a676eb3 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/a676eb30d42f9562833f82936e743c5749d6ce63 | 15:15 |
shogun-notifier- | shogun: Merge pull request #2235 from lambday/feature/linalg | 15:15 |
shogun-notifier- | shogun: | 15:15 |
shogun-notifier- | shogun: added rowwise and colwise sum methods in linalg | 15:15 |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun | 15:18 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 15:19 | |
-!- khalednasr [~k.nasr92@41.69.214.145] has quit [Ping timeout: 252 seconds] | 15:24 | |
-!- khalednasr [~k.nasr92@41.69.237.19] has joined #shogun | 15:25 | |
@HeikoS | kislay: hey, sorry for not getting back earlier, was in a meeting. Now going out for lunch but back soon! | 15:29 |
kislay | HeikoS, oh. np! | 15:29 |
-!- cuphrody [cuphrody@211.69.205.236] has joined #shogun | 15:35 | |
-!- Saurabh7 [~Saurabh7@59.88.33.86] has joined #shogun | 15:45 | |
-!- Saurabh7 [~Saurabh7@59.88.33.86] has quit [Ping timeout: 276 seconds] | 15:55 | |
-!- Saurabh7 [~Saurabh7@117.222.11.53] has joined #shogun | 16:21 | |
-!- pickle27f [~pickle27@216.191.231.218] has joined #shogun | 16:23 | |
-!- Saurabh7 [~Saurabh7@117.222.11.53] has quit [Ping timeout: 265 seconds] | 16:27 | |
-!- iglesiasg [~iglesiasg@2001:6b0:1:1041:25ea:1e69:30f2:e06c] has joined #shogun | 16:46 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 16:46 | |
@iglesiasg | wow back to .se | 16:46 |
lisitsyn1 | iglesiasg: you back to .se? ;) | 16:47 |
@iglesiasg | lisitsyn1, http://www.sais.se/sais2014/ | 16:47 |
@iglesiasg | lisitsyn1, workshop starting tomorrow at my old uni | 16:47 |
lisitsyn1 | ahh cool | 16:48 |
thoralf | Hey :) | 16:57 |
* thoralf just got a confirmation that we can do the SHOGUN-Hands-On on Sunday, 27th of July at ReasearchGate. | 16:59 | |
thoralf | Berlin-Mitte, close to Hauptbahnhof and other transports. | 16:59 |
thoralf | I'll send a more official announcement for this soon. | 17:01 |
@iglesiasg | thoralf, great! | 17:03 |
lisitsyn1 | ah berlin berlin | 17:05 |
lisitsyn1 | ah guys I have to announce I probably won't be there :( | 17:05 |
@wiking | unfortunately possible me neither | 17:05 |
lisitsyn1 | it seems I am moving to moscow | 17:07 |
lisitsyn1 | just around this date | 17:07 |
lisitsyn1 | will have to do a lot of stuff | 17:07 |
* wiking moving as well | 17:08 | |
lisitsyn1 | wiking: where to? :) | 17:08 |
@wiking | or seems like at least | 17:08 |
@wiking | lisitsyn1: heheh dont want to jinx it | 17:08 |
lisitsyn1 | japan? | 17:08 |
lisitsyn1 | :D | 17:08 |
@wiking | lisitsyn1: close... | 17:08 |
lisitsyn1 | really? | 17:08 |
@wiking | yep, but i really dont want to jinx it | 17:09 |
lisitsyn1 | wiking: sure ;) | 17:09 |
thoralf | wiking going to Baidu? ;) | 17:10 |
-!- lisitsyn1 [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.] | 17:11 | |
@HeikoS | thoralf: nice!!! | 17:17 |
thoralf | HeikoS: I think we should wait a few weeks to announce this. | 17:17 |
thoralf | HeikoS: First announce workshop. | 17:18 |
thoralf | wait 2 weeks | 17:18 |
thoralf | Announce hands-on. | 17:18 |
thoralf | Just spreading it a bit over time. | 17:18 |
-!- Saurabh7 [~Saurabh7@117.200.183.38] has joined #shogun | 17:26 | |
@HeikoS | thoralf: yep! | 17:27 |
@HeikoS | thoralf: a man, time overkill here, would love to push this more | 17:28 |
@HeikoS | thoralf: could you update the google doc? | 17:28 |
thoralf | HeikoS: Don't have my google credentials at work... maybe on the weekend. | 17:29 |
thoralf | Sorry. | 17:29 |
@HeikoS | thoralf: no worries, just in general we should push this a bit in the near future | 17:30 |
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 18:15 | |
-!- pickle27f [~pickle27@216.191.231.218] has quit [Quit: This computer has gone to sleep] | 18:24 | |
@HeikoS | lambday: hey | 18:27 |
@HeikoS | just reading your patch | 18:27 |
lambday | HeikoS: hi! | 18:27 |
lambday | HeikoS: yeah just read your comments | 18:27 |
lambday | HeikoS: so I added sum_rowwise, sum_colwise as you suggested | 18:27 |
@HeikoS | lambday: cool | 18:28 |
lambday | got this feeling that you'd like it this way :) | 18:28 |
@HeikoS | lambday: haha :) | 18:28 |
@HeikoS | lambday: so lets step a bit back and get the big picture of the current state now .... | 18:28 |
lambday | HeikoS: regarding square, though, I'm not sure! | 18:28 |
@HeikoS | lambday: ok, why? | 18:28 |
@HeikoS | lambday: this square thing is nice to have, but it will create a new matrix right? | 18:28 |
lambday | HeikoS: yeah that's true | 18:29 |
@HeikoS | lambday: ok | 18:29 |
lambday | for a very brief period thogh | 18:29 |
@HeikoS | lambday: so, what is missing for the MMD stuff to work? | 18:29 |
@HeikoS | lambday: but allocating takes time | 18:29 |
@HeikoS | lambday: whereas summing up squares does not | 18:29 |
lambday | HeikoS: alright but that | 18:30 |
lambday | 's a way too specific task to go to linalg, no? | 18:30 |
lambday | umm.. although squared_sum can be done | 18:31 |
-!- cuphrody [cuphrody@211.69.205.236] has quit [Ping timeout: 240 seconds] | 18:31 | |
lambday | so - I'll change that! | 18:31 |
@HeikoS | lambday: yeah sum up squares is sueful | 18:31 |
@HeikoS | useful | 18:31 |
lambday | HeikoS: as of now, nothing is missing for mmd to work - in fact I just modified quadratic time mmd to compute statistic and variance on the fly | 18:32 |
@HeikoS | lambday: ok cool | 18:32 |
@HeikoS | lambday: what about all those different MMD null approximations? | 18:32 |
@HeikoS | lambday: thats where we left the thing a while ago | 18:32 |
lambday | HeikoS: yeah! - so as we discussed, I added these enums | 18:32 |
@HeikoS | lambday: cool | 18:33 |
lambday | so, there are 2 enums for MMD_SPECTRUM now | 18:33 |
@HeikoS | lambday: nice | 18:33 |
@HeikoS | lambday: do you have an estimate for how long these things will take to wrap up? then dino and I can plan a bit | 18:33 |
@HeikoS | lambday: we will probably start writing some experiments with all of this later in June | 18:33 |
lambday | HeikoS: well, quadratic time mmd I'll finish by tonight or tomorrow | 18:34 |
lambday | then BTest and linear time mmd - by next week! | 18:34 |
lambday | actually I'll make a doc to keep track of things | 18:34 |
@HeikoS | lambday: ok, no rush, I am just asking | 18:34 |
@HeikoS | lambday: yeah a doc would be great | 18:34 |
@HeikoS | lambday: since GSoC also started officially now, we should soon transition to the feature selection stuff :) | 18:35 |
lambday | HeikoS: since I deviated from this for a while I lost track :/ | 18:35 |
@HeikoS | lambday: haha :) | 18:35 |
@HeikoS | lambday: yeah no worries, you are doing great work here | 18:35 |
@HeikoS | lambday: I already might have a customer for feature selection stuff, so that is nice | 18:35 |
lambday | HeikoS: wow that's cool! :d | 18:35 |
@HeikoS | lambday: do you need any input on something? | 18:35 |
lambday | HeikoS: umm.... can't think of any right now - if I remmebe something, will mail you and dino? | 18:36 |
@HeikoS | lambday: yeah sure | 18:37 |
lambday | HeikoS: so, we had a whole bunch of tasks planned - I'll reschedule things a bit after finishing with MMD... as per earlier estimation this would have been done long ago while I was doing btest :( | 18:37 |
@HeikoS | lambday: yeah I know | 18:37 |
@HeikoS | I am actually thinking of postponing a few things thats why I am asking you | 18:37 |
@HeikoS | lambday: but also great to have, so its fine if you spend another week on those things | 18:38 |
@HeikoS | then later in the summer we will make use of those things and maybe request some updates/find problems ,etc | 18:38 |
lambday | HeikoS: yeah that sounds good | 18:38 |
lambday | well I was kind of worried to catch up on the schedule that I planned | 18:38 |
@HeikoS | which of the feature selection things would you like to start with? | 18:39 |
@HeikoS | lambday: dont worry, rather take time to do things properly, thats worth more :) | 18:39 |
lambday | HeikoS: we can start with smola's paper, right? feature selection via dependence maximization | 18:40 |
-!- pickle27f [~pickle27@216.191.231.218] has joined #shogun | 18:40 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Quit: Leaving.] | 18:41 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun | 18:41 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 18:41 | |
@HeikoS | lambday: BAHSIC you mean? | 18:42 |
lambday | HeikoS: yeah | 18:42 |
@HeikoS | lambday: about the Block stuff | 18:42 |
@HeikoS | lambday: (sorry to jump back here=) | 18:42 |
lambday | no worries | 18:43 |
@HeikoS | lambday: if one computes the sum of a matrix block | 18:43 |
@HeikoS | does this copy the block around? | 18:43 |
lambday | HeikoS: nah | 18:43 |
@HeikoS | or how does this work? | 18:43 |
@HeikoS | lambday: sorry for asking (I could also read code, but since you are here thats faster) | 18:43 |
lambday | well, as of now I just used eigen3, which refers to blocks in the orig matrix via those block() calls | 18:44 |
lambday | HeikoS: so, no copying | 18:44 |
@HeikoS | lambday: and SGMatrix? | 18:44 |
-!- Saurabh7 [~Saurabh7@117.200.183.38] has quit [Ping timeout: 252 seconds] | 18:44 | |
@HeikoS | ah sorry | 18:44 |
@HeikoS | so that creates a map first | 18:44 |
@HeikoS | lambday: so the naive implementation could just map indices then? | 18:45 |
@HeikoS | like a subset? | 18:45 |
lambday | HeikoS: yeah that's the best way I guess | 18:45 |
lambday | HeikoS: I haven't added naive implementation though | 18:45 |
@HeikoS | lambday: cool! | 18:45 |
@HeikoS | lambday: no need for now | 18:45 |
@HeikoS | (entrance task :) ) | 18:45 |
lambday | hahaha :D | 18:45 |
@HeikoS | maybe make a list of future entrance tasks somewhere | 18:45 |
@HeikoS | seriously, I always forget them, its important to grow those on the fly | 18:46 |
lambday | umm yeah might be good - we always forget stuffs | 18:46 |
@HeikoS | and its a good entrance task | 18:46 |
@HeikoS | one learns a lot | 18:46 |
@HeikoS | lambday: ok then | 18:46 |
@HeikoS | BAHSIC is the first algo for feature selection | 18:46 |
@HeikoS | lambday: some design questions naturally appear | 18:46 |
@HeikoS | lambday: Preprocessor | 18:47 |
@HeikoS | variable-selection-preprocessor | 18:47 |
@HeikoS | we dont have any of those | 18:47 |
lambday | HeikoS: yeah - I have to read the paper in more details to get how do they do it.. currently not having much idea how do add these in preprocessors... | 18:49 |
@HeikoS | lambday: its just backward elimination of features using HSIC as a measure of dependence | 18:49 |
@HeikoS | lambday: super easy | 18:49 |
@HeikoS | lambday: and design is independent of paper | 18:49 |
@HeikoS | lambday: maybe dino and I can share some thoughts, we discussed this a while ago | 18:50 |
@HeikoS | will try to squeeze this in | 18:50 |
lambday | alright - so basically among different feature selection algo, the measure of dependence is what differs, right? | 18:50 |
@HeikoS | lambday: yes | 18:53 |
@HeikoS | the algorithm always removes the feature that is least informative | 18:54 |
lambday | HeikoS: so the different independence measures that we planned to have, will be used by this feature selection framework | 18:54 |
lambday | sorry for sounding super obvious - just trying to get clear thoughts :) | 18:56 |
-!- Saurabh7 [~Saurabh7@117.200.183.38] has joined #shogun | 18:57 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Read error: No route to host] | 19:00 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun | 19:00 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 19:00 | |
@HeikoS | lambday: yes exactly | 19:00 |
@HeikoS | lambday: sorry got disconnected | 19:00 |
@HeikoS | lambday: yes these measure can then be combined with different ways to select fetaures | 19:00 |
lambday | HeikoS: alright... so, this would all be done via some subclass(es) of preprocessor | 19:01 |
-!- iglesiasg [~iglesiasg@2001:6b0:1:1041:25ea:1e69:30f2:e06c] has quit [Quit: Leaving] | 19:02 | |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 19:02 | |
shogun-notifier- | shogun: Heiko Strathmann :develop * 9d77bda / src/shogun/kernel/ANOVAKernel.cpp,src/shogun/kernel/ANOVAKernel.h: https://github.com/shogun-toolbox/shogun/commit/9d77bda196015fd125c9010c5ed5e5de78e996b9 | 19:02 |
shogun-notifier- | shogun: added obtain_from_generic method | 19:02 |
shogun-notifier- | shogun: Heiko Strathmann :develop * c23f2ec / src/shogun/kernel/ANOVAKernel.cpp,src/shogun/kernel/ANOVAKernel.h: https://github.com/shogun-toolbox/shogun/commit/c23f2ecf1e16850da3b685be37364162807def1f | 19:02 |
shogun-notifier- | shogun: Merge pull request #2242 from karlnapf/develop | 19:02 |
shogun-notifier- | shogun: | 19:02 |
shogun-notifier- | shogun: added CANOVAKernel::obtain_from_generic | 19:02 |
@HeikoS | lambday: yes | 19:04 |
@HeikoS | lambday: so we need one that selects a number of variables from the input features | 19:04 |
lambday | HeikoS: okay... so, something inside apply_to_feature_matrix (under DensePreprocessor) would return a feature matrix with selected features | 19:07 |
@HeikoS | lambday: I guess yes | 19:08 |
@HeikoS | lambday: in fact we can think about avoiding the copy | 19:08 |
lambday | HeikoS: use subsets instead | 19:08 |
@HeikoS | lambday: we already have subsets for the features | 19:08 |
@HeikoS | lambday: but that might be tricky | 19:08 |
@HeikoS | lambday: not sure whether its worth it or too messy | 19:08 |
@HeikoS | lambday: a copy is definitely the first choiu | 19:08 |
@HeikoS | choice | 19:08 |
lambday | alright - yeah we gotta think a bit about how this would work | 19:09 |
@HeikoS | lambday: ok cool | 19:09 |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Quit: Leaving.] | 19:11 | |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun | 19:11 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 19:11 | |
lambday | HeikoS: okay - so, we need one hierarchy of classes FeatureSelection - some base class - subclasses use different dependence measures - provides means for applying the subsetting | 19:14 |
@HeikoS | lambday: yes! | 19:14 |
lambday | HeikoS: and then one preprocessor having an instance of this feature selection inside it | 19:14 |
@HeikoS | lambday: I think it would be nice if it was so modular that we can just plug in different measures into different algorithms | 19:14 |
lambday | HeikoS: umm... yeah that's better - so may be we provide some method in those dependence measures only :/ | 19:15 |
lambday | HeikoS: then feature selection works with those methods | 19:15 |
lambday | not sure though - just thinking | 19:15 |
@HeikoS | lambday: no wait, the preprocessor does the feature selection | 19:16 |
@HeikoS | and it gets an instance of the dependence measure | 19:16 |
lambday | HeikoS: yeah - so, it all works with C<dependence-measure-class>::sample_null() ?? | 19:18 |
@HeikoS | lambday: mmh goo dpoint | 19:19 |
@HeikoS | lambday: I am not sure wheather we need that | 19:19 |
lambday | I mean are the null samples the only things that we need to do feature selection? | 19:19 |
lambday | HeikoS: I'm not sure - may be I'll read the paper today and try to get an idea :/ | 19:19 |
@HeikoS | for BAHSIC, one just computes the statistic doesnt one? (not sure currently, hehe) | 19:19 |
@HeikoS | lambday: see also arthurs lecture slides | 19:20 |
@HeikoS | http://www.gatsby.ucl.ac.uk/~gretton/coursefiles/rkhscourse.html | 19:20 |
@HeikoS | lceture 4,5,6 slides | 19:20 |
lambday | HeikoS: checking... | 19:20 |
lambday | (I took printed copies of those :D) | 19:20 |
@HeikoS | slide 177 | 19:21 |
@HeikoS | hehe | 19:21 |
lambday | HeikoS: yeah - just the arg max of statistic | 19:23 |
lambday | then remove that one | 19:23 |
lambday | ah got it | 19:23 |
lambday | so, other algorithms would work this way only - then we can have it this way - FeatureSelection ---> DensePreprocessor, have an instance of IndependenceTest, overrides apply_to_feature_matrix, inside this it uses IndependenceTest::compute_statistic and remove the irrelevent ones iteratively | 19:26 |
lambday | so this loop would go inside that | 19:26 |
@HeikoS | yeah, so that should work with any dependence measure right? :) backwards elemination | 19:26 |
@HeikoS | exactly | 19:26 |
lambday | HeikoS: yeah - although I am stuck at (Why backwords?) :D | 19:27 |
lambday | I mean, if backword elimination works for all then this simple structure would suffice | 19:28 |
@HeikoS | lambday: if you go forward and start adding features, you will miss pairwise dependencies | 19:28 |
@HeikoS | lambday: or what did you mean? | 19:28 |
lambday | aaahhh | 19:28 |
lambday | yeah | 19:28 |
lambday | got your pt | 19:28 |
lambday | HeikoS: yeah I meant that only | 19:28 |
lambday | cool! | 19:30 |
lambday | HeikoS: alright! feeling a bit relaxed.. got the main points - I'll try to finish quadratic time mmd (almost done - except spectrum).. | 19:34 |
@HeikoS | lambday: cool! yeah relax, this will be fun, no need to freak out, rather enjoy, its cool stuff! :) | 19:34 |
lambday | HeikoS: I'll read the soon and start with this issue first - so we'll have the framework ready - later we can add different dependence measures | 19:35 |
lambday | HeikoS: oh another thing - will it work similarly for conditional dependence measures? | 19:36 |
-!- jiaolong [9e6d09e4@gateway/web/freenode/ip.158.109.9.228] has quit [] | 19:38 | |
@HeikoS | lambday: good point! | 19:40 |
@HeikoS | lambday: when you set up the doc, put those things in | 19:40 |
@HeikoS | we can then give feedback and discuss | 19:40 |
lambday | HeikoS: alright that sounds good! | 19:40 |
@HeikoS | lambday: I agree that we should draft base classes first, without implementations, | 19:41 |
lambday | yeah! | 19:41 |
-!- lisitsyn [~qdrgsm@80.252.20.67] has joined #shogun | 19:48 | |
-!- Saurabh7 [~Saurabh7@117.200.183.38] has quit [Ping timeout: 252 seconds] | 19:58 | |
-!- Saurabh7 [~Saurabh7@117.200.183.180] has joined #shogun | 20:01 | |
@HeikoS | Saurabh7: hey! | 20:01 |
Saurabh7 | HeikoS, hi! | 20:01 |
Saurabh7 | HeikoS, did you get my msg? | 20:01 |
@HeikoS | Saurabh7: do you have a minute to discuss? | 20:01 |
@HeikoS | not yet, currently working throgh a batch of emails :) | 20:01 |
Saurabh7 | HeikoS, no i pinged you here | 20:02 |
Saurabh7 | i think got dc first | 20:02 |
@HeikoS | Saurabh7: ah no | 20:02 |
Saurabh7 | yeah have some time | 20:02 |
@HeikoS | Saurabh7: so lets discuss the first week | 20:02 |
@HeikoS | and update the doc | 20:02 |
Saurabh7 | I have prepared a draft of first task | 20:02 |
@HeikoS | Saurabh7: nice! | 20:02 |
Saurabh7 | HeikoS, yes i will update the detail part tomorrow first thing | 20:03 |
@HeikoS | Saurabh7: could you give me a link? | 20:03 |
Saurabh7 | HeikoS, should i send in a pr about the first one? | 20:03 |
@HeikoS | Saurabh7: no need for a PR, just a link for me to see | 20:03 |
Saurabh7 | nbviewer? | 20:03 |
Saurabh7 | ok | 20:03 |
@HeikoS | Saurabh7: but maybe even put the current state into a PR then I can start reviewing | 20:03 |
Saurabh7 | sec | 20:03 |
@HeikoS | Saurabh7: but wait | 20:03 |
@HeikoS | Saurabh7: lets also discuss a bit what to go in there, what is your plan for this notebookss | 20:04 |
@HeikoS | ? | 20:04 |
@HeikoS | but maybe its good to go from notebook, so lets see the link! :) | 20:04 |
Saurabh7 | HeikoS, ah | 20:05 |
Saurabh7 | here it is http://nbviewer.ipython.org/gist/Saurabh7/5557715ec5ad323ffafe# | 20:05 |
@HeikoS | Saurabh7: this already looks very nice | 20:07 |
Saurabh7 | HeikoS, oh ! good.. | 20:08 |
Saurabh7 | any suggestions? | 20:08 |
@HeikoS | Saurabh7: I like this direction a lot | 20:08 |
@HeikoS | Saurabh7: providing some very simple example of things and then pointing to the other notebooks | 20:08 |
Saurabh7 | HeikoS, yes thats what i planned | 20:09 |
@HeikoS | Saurabh7: so this is like basic supervised classification | 20:09 |
Saurabh7 | HeikoS, we can group similar nbs together further | 20:09 |
@HeikoS | Saurabh7: I will try to come up with some feedback soon, but for now this looks good so you can keep on working in this direction | 20:10 |
Saurabh7 | HeikoS, yes its the simplest i could get hold of | 20:10 |
@HeikoS | Saurabh7: yes I agree | 20:10 |
@HeikoS | Saurabh7: great work! :) | 20:10 |
Saurabh7 | so instead of say 'supervised' pointing to wiki , we can point it to the bunch | 20:10 |
@HeikoS | like the reference to the web demo | 20:10 |
@HeikoS | Saurabh7: yeah maybe, thats also good for the website | 20:11 |
@HeikoS | like it scikit | 20:11 |
@HeikoS | Saurabh7: BTW could you condense a list of Shogun's features on the fly | 20:11 |
@HeikoS | maybe put them in the document | 20:11 |
@HeikoS | Saurabh7: since we desperately need this up to date :) | 20:11 |
@HeikoS | Saurabh7: ok this looks good, | 20:12 |
Saurabh7 | HeikoS, yes because the nbs are pretty complicated to browse through | 20:12 |
Saurabh7 | HeikoS, yup i am trying to set up a diagram | 20:12 |
@HeikoS | Saurabh7: I guess you could elaborate a bit on other cases: other features, regression ... etc, but that might come later | 20:12 |
@HeikoS | Saurabh7: yes, great idea! really like it | 20:12 |
@HeikoS | Saurabh7: You should add some section on the interfaces of features | 20:13 |
@HeikoS | Saurabh7: like number of vectors, dimension, etc | 20:13 |
@HeikoS | And maybe also some things on other feature types | 20:13 |
Saurabh7 | HeikoS, ok nice, will try to look into that | 20:13 |
@HeikoS | this notebook can grow a bit, there could also be an example for regression and maybe even a preprocessor | 20:14 |
@HeikoS | since those are very fundamental | 20:14 |
Saurabh7 | yes | 20:14 |
Saurabh7 | i was afraid to make it too complicated | 20:14 |
@HeikoS | like one simple example each? | 20:14 |
@HeikoS | yes I know what you mean | 20:14 |
@HeikoS | but if the thing just grows a bit thats fine | 20:15 |
Saurabh7 | ok i will try to come up with something about regression | 20:15 |
@HeikoS | nice | 20:15 |
@HeikoS | I mean the focus is just on the setting of the method, not so much about details, so you can just use a particular algorithm under the CMachine interface and then point to another notebook for method details, | 20:16 |
@HeikoS | like you already did | 20:16 |
@HeikoS | mean remove, unit variance preprocessor might be good | 20:16 |
@HeikoS | or whitening | 20:16 |
@HeikoS | since those are std things | 20:16 |
@HeikoS | Saurabh7: ok, let me know how it goes | 20:17 |
Saurabh7 | yup seems good | 20:17 |
@HeikoS | Saurabh7: btw you should include nbviewer links in your weekly updates to the list, people will love them :) | 20:17 |
Saurabh7 | i will try to set up a diagram and example by tomorrow | 20:17 |
@HeikoS | cool | 20:17 |
@HeikoS | Saurabh7: and send a PR of this one already, together with the nbviwer link, I will put some comments, | 20:18 |
@HeikoS | Saurabh7: Nice! Glad how this is going! Keep on the good work! | 20:18 |
Saurabh7 | ok will send it now | 20:18 |
Saurabh7 | but its late here so will have to address them tomorrow | 20:18 |
@HeikoS | Saurabh7: I will leave now then. Having a look later today. | 20:18 |
Saurabh7 | thanks! | 20:18 |
@HeikoS | Saurabh7: yeah no problem | 20:18 |
Saurabh7 | ok np | 20:18 |
-!- sonne|osx [~sonne@89.204.139.93] has joined #shogun | 20:21 | |
kislay | Saurabh7, I too saw the nbviewer link! it's cool :) | 20:21 |
Saurabh7 | kislay, hey , thanks :) | 20:22 |
Saurabh7 | kislay, hows ur project going? | 20:22 |
kislay | Saurabh7, nice. | 20:23 |
Saurabh7 | kislay, great | 20:23 |
Saurabh7 | see you later .. pretty late now :) | 20:24 |
kislay | Saurabh7, not much late. | 20:25 |
kislay | ;) | 20:25 |
kislay | Saurabh7, bye | 20:25 |
Saurabh7 | kislay, i am at home now so .. :p | 20:25 |
kislay | Saurabh7, me too actually. but never mind. | 20:25 |
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Quit: Leaving.] | 20:28 | |
-!- Saurabh7 [~Saurabh7@117.200.183.180] has quit [Quit: Leaving] | 20:33 | |
-!- sonne|osx [~sonne@89.204.139.93] has quit [Quit: sonne|osx] | 20:38 | |
-!- thoralf|home [~thoralf@91-65-137-195-dynip.superkabel.de] has joined #shogun | 20:48 | |
thoralf|home | Hey. | 20:48 |
@besser82 | thoralf|home, ho! :D | 21:19 |
-!- pickle27f [~pickle27@216.191.231.218] has quit [Quit: This computer has gone to sleep] | 21:24 | |
-!- pickle27f [~pickle27@216.191.231.218] has joined #shogun | 21:26 | |
thoralf|home | besser82: Bo! | 21:37 |
shogun-buildbot | build #312 of debian wheezy - memcheck is complete: Failure [failed memory check] Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/312 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 21:46 |
@besser82 | thoralf|home, :D | 21:49 |
@besser82 | thoralf|home, ^^^ Heiko hat's kaputt gemacht :-P | 21:49 |
thoralf|home | Nee. Have a look: https://github.com/shogun-toolbox/shogun/pull/2232 | 21:51 |
thoralf|home | This will "fix" memcheck. | 21:51 |
thoralf|home | Finally. | 21:51 |
thoralf|home | Spent so many hours on debugging and fixing the whole mess-up. | 21:51 |
thoralf|home | s/-up// | 21:52 |
@besser82 | thoralf|home, ic ;) | 21:52 |
@besser82 | thoralf|home, to me the pr looks good ;) | 21:52 |
@besser82 | thoralf|home, shall I merge? | 21:52 |
@besser82 | thoralf|home, so they can blame me for doing nasty stuff? | 21:53 |
thoralf|home | If you like. ;) | 21:54 |
@besser82 | thoralf|home, sure ;) | 21:55 |
@besser82 | thoralf|home, travis lloks good so lets-a-go! | 21:55 |
shogun-notifier- | shogun: Thoralf Klein :develop * d61a691 / configs/valgrind.supp: https://github.com/shogun-toolbox/shogun/commit/d61a6914c92de2c7853833f48e7e80220f4eebf1 | 21:55 |
shogun-notifier- | shogun: Removing last memory warnings on buildbot (could not reproduce them locally). | 21:55 |
shogun-notifier- | shogun: Bj?rn Esser :develop * da1698f / configs/valgrind.supp: https://github.com/shogun-toolbox/shogun/commit/da1698f3fb6da96907fcbde3e41d2de9b9148cef | 21:55 |
shogun-notifier- | shogun: Merge pull request #2232 from tklein23/more_valgrind_suppressions | 21:55 |
shogun-notifier- | shogun: | 21:55 |
shogun-notifier- | shogun: Removing last memory warnings on buildbot (could not reproduce them locally) | 21:55 |
@besser82 | thoralf|home, btw. what is the correct procedure to convert the mallet model and stuff to vectors? | 21:57 |
@besser82 | thoralf|home, so I can train a model from the findings with liblinear? | 22:01 |
thoralf|home | besser82: You can. | 22:01 |
thoralf|home | besser82: Create a feature vector, where the i-th component is the probability of the i-th topic. | 22:02 |
@besser82 | thoralf|home, let's get that german ;) | 22:02 |
-!- travis-ci [~travis-ci@ec2-54-234-84-161.compute-1.amazonaws.com] has joined #shogun | 22:39 | |
travis-ci | [travis-ci] it's Bj?rn Esser's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/25726952 | 22:39 |
-!- travis-ci [~travis-ci@ec2-54-234-84-161.compute-1.amazonaws.com] has left #shogun [] | 22:39 | |
-!- pickle27f [~pickle27@216.191.231.218] has quit [Quit: This computer has gone to sleep] | 23:14 | |
-!- khalednasr [~k.nasr92@41.69.237.19] has quit [Quit: Leaving] | 23:51 | |
--- Log closed Thu May 22 00:00:59 2014 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!