--- Log opened Wed Apr 09 00:00:14 2014 | ||
-!- HeikoS [~heiko@31.85.167.119] has joined #shogun | 00:01 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 00:01 | |
-!- kislay [~kislay@14.139.41.150] has joined #shogun | 00:09 | |
-!- HeikoS1 [~heiko@31.75.97.193] has joined #shogun | 00:11 | |
-!- HeikoS [~heiko@31.85.167.119] has quit [Ping timeout: 240 seconds] | 00:11 | |
-!- HeikoS1 [~heiko@31.75.97.193] has quit [Ping timeout: 240 seconds] | 00:17 | |
-!- HeikoS [~heiko@ra-vpn230.ra-vpn.ucl.ac.uk] has joined #shogun | 00:27 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 00:27 | |
-!- HeikoS [~heiko@ra-vpn230.ra-vpn.ucl.ac.uk] has quit [Client Quit] | 00:31 | |
-!- rho_ [~quassel@217.55.153.97] has quit [Read error: Connection reset by peer] | 00:47 | |
lisitsyn | thoralf: hey | 00:47 |
---|---|---|
thoralf | Hey lisitsyn | 00:47 |
lisitsyn | thoralf: I see some singleton issue :) | 00:47 |
thoralf | Yeah. :) | 00:48 |
thoralf | I opened the battle. | 00:48 |
thoralf | Wanna join? ;) | 00:48 |
lisitsyn | thoralf: actually we can go ahead and just make shogun itself instantiable | 00:49 |
lisitsyn | run multiple shoguns :D | 00:49 |
thoralf | lol | 00:49 |
thoralf | Yeah, that's true. | 00:49 |
lisitsyn | well that's natural extension | 00:49 |
thoralf | We would loose that. | 00:49 |
lisitsyn | everything that's single will get multiple | 00:49 |
lisitsyn | somebody wise should have said that before I bet | 00:49 |
lisitsyn | like karl marx haha | 00:50 |
thoralf | lol | 00:50 |
lisitsyn | OpenSSL is written by monkeys https://www.peereboom.us/assl/assl/html/openssl.html | 00:51 |
lisitsyn | can't resist to share | 00:51 |
thoralf | Hahah. | 00:51 |
lisitsyn | ah and if anybody is interested I dumped my c++ssy brain again | 00:53 |
lisitsyn | http://sergey.lisitsyn.me/2014/04/08/actions-with-raii/ | 00:53 |
thoralf | lisitsyn: The monkey-story is great. | 01:00 |
thoralf | tl;sr; | 01:00 |
lisitsyn | thoralf: yeah they sucked a big time | 01:02 |
lisitsyn | our bugs lead to wrong nothing | 01:03 |
lisitsyn | but their bugs are like important :) | 01:03 |
thoralf | Yeah. :) | 01:03 |
thoralf | But really, I still don't know why every object needs to have all this crap. I never used serialization *or* model-selection *or* parameter framework to compare objects. | 01:04 |
lisitsyn | thoralf: they should have one thing | 01:04 |
lisitsyn | shogun instance ;) | 01:04 |
lisitsyn | which has vresion parallel etc | 01:04 |
thoralf | I was about to fork shogun and rip it of. | 01:05 |
lisitsyn | thoralf: it is a known syndrome | 01:05 |
thoralf | Or making it de-configurable by -DDISABLE_PARAMETER and -DDISABLE_MODEL_SELECTION. | 01:05 |
thoralf | Yeah. | 01:05 |
thoralf | I decided against forking, because I want better code... not worse. ;) | 01:05 |
lisitsyn | it is called shogun forking disorder | 01:06 |
lisitsyn | :D | 01:06 |
lisitsyn | sometimes referred as shogunophrenia | 01:06 |
thoralf | ophrenia? | 01:07 |
thoralf | You're suffering this as well? | 01:07 |
lisitsyn | well I am mentally disabled since I got ill | 01:07 |
lisitsyn | :D | 01:07 |
lisitsyn | sometimes I walk into walls | 01:08 |
lisitsyn | ok actually yes everybody in shogun wanted to rewrite this f'ing s't | 01:08 |
thoralf | Hmm. So why are we (everybody except me) putting more sh't in instead of removing it? | 01:09 |
thoralf | Well, wiking is good in consolidating I guess. | 01:10 |
lisitsyn | have you seen my commits lately? ;) | 01:10 |
thoralf | And cleaning up. | 01:10 |
thoralf | lisitsyn: You're still committing? | 01:10 |
thoralf | I thought you're just hanging up here because you're used to. :D | 01:10 |
lisitsyn | not really :D | 01:10 |
lisitsyn | not really to both | 01:11 |
thoralf | Are you using shogun somewhere? | 01:11 |
lisitsyn | no :) | 01:11 |
lisitsyn | that's a major point | 01:11 |
thoralf | lol | 01:11 |
lisitsyn | thoralf: I have nothing to use it for | 01:12 |
lisitsyn | thoralf: but I am not just hanging up I am just sloooooow | 01:12 |
thoralf | That's a bad starting point for discussing improvements. ;) | 01:12 |
lisitsyn | I am sloowly getting to some solution for modular shogun | 01:13 |
lisitsyn | that's what I devote my life for hahah | 01:13 |
lisitsyn | thoralf: I use it pretty rarely but nothing advanced | 01:13 |
lisitsyn | well though I have some vague understanding what our users need | 01:14 |
lisitsyn | so you are mostly right ;) | 01:14 |
thoralf | :D | 01:15 |
thoralf | So does it mean you're joining the battle to improve something that has no effect on you? ;) | 01:15 |
lisitsyn | thoralf: I am pretty far away from all the battles | 01:15 |
lisitsyn | thoralf: well I have some idea how software should look like - that's what I am based on | 01:16 |
thoralf | A good-code based life form? :) | 01:17 |
thoralf | I know carbon-based life forms... they don't perform well. | 01:17 |
lisitsyn | thoralf: I am not sure what you mean ;) I don't put my opinion on things that affect user experience | 01:19 |
lisitsyn | if someone says X would be better I would believe that | 01:19 |
thoralf | Hmm. Interesting approach. | 01:20 |
thoralf | Saves a lot of hassle. | 01:20 |
lisitsyn | so I believe I am not that harmful ;) | 01:20 |
-!- parijat [671b082a@gateway/web/freenode/ip.103.27.8.42] has quit [Quit: Page closed] | 01:21 | |
thoralf | I should adapt this. | 01:22 |
thoralf | Cool thing. | 01:23 |
thoralf | Will try this out tomorrow at work. :D | 01:23 |
-!- Khalednasr [~khalednas@41.69.201.100] has quit [Quit: Leaving] | 02:07 | |
-!- HeikoS [~androirc@90.201.62.238] has joined #shogun | 02:41 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 02:41 | |
-!- thoralf [~thoralf@91-65-136-14-dynip.superkabel.de] has quit [Ping timeout: 240 seconds] | 02:59 | |
-!- zxtx [~zv@129-79-241-102.dhcp-bl.indiana.edu] has quit [Ping timeout: 240 seconds] | 03:49 | |
-!- mode/#shogun [+o sonney2k] by ChanServ | 03:49 | |
-!- soumyaC [uid15286@gateway/web/irccloud.com/x-dtywjmpzbjyduxgi] has quit [Quit: Connection closed for inactivity] | 04:45 | |
-!- zxtx [~zv@c-98-223-196-32.hsd1.in.comcast.net] has joined #shogun | 04:56 | |
-!- zxtx_ [~zxtx@sunnythinking.org] has joined #shogun | 05:27 | |
-!- zxtx__ [~zv@c-98-223-196-32.hsd1.in.comcast.net] has joined #shogun | 05:35 | |
-!- zxtx__ [~zv@c-98-223-196-32.hsd1.in.comcast.net] has quit [Remote host closed the connection] | 05:36 | |
-!- zxtx__ [~zv@c-98-223-196-32.hsd1.in.comcast.net] has joined #shogun | 05:40 | |
-!- zxtx__ [~zv@c-98-223-196-32.hsd1.in.comcast.net] has quit [Client Quit] | 05:40 | |
-!- zxtx_ [~zxtx@sunnythinking.org] has quit [Quit: ZNC - http://znc.in] | 05:41 | |
-!- zxtx_ [~zxtx@sunnythinking.org] has joined #shogun | 05:42 | |
-!- zxtx [~zv@c-98-223-196-32.hsd1.in.comcast.net] has quit [Quit: Leaving] | 06:14 | |
-!- zxtx [~zv@c-98-223-196-32.hsd1.in.comcast.net] has joined #shogun | 06:15 | |
-!- zxtx [~zv@c-98-223-196-32.hsd1.in.comcast.net] has quit [Client Quit] | 06:16 | |
-!- zxtx_ [~zxtx@sunnythinking.org] has quit [Quit: Strong teeth for a strong nation] | 06:22 | |
-!- zxtx [~zxtx@sunnythinking.org] has joined #shogun | 06:23 | |
-!- kislay [~kislay@14.139.41.150] has quit [Remote host closed the connection] | 06:28 | |
-!- kislay [~kislay@220.227.97.99] has joined #shogun | 06:29 | |
-!- kprah [67157d55@gateway/web/freenode/ip.103.21.125.85] has joined #shogun | 06:48 | |
-!- kprah [67157d55@gateway/web/freenode/ip.103.21.125.85] has quit [Ping timeout: 240 seconds] | 06:53 | |
-!- kislay [~kislay@220.227.97.99] has quit [Remote host closed the connection] | 07:02 | |
-!- pl8787 [9fe22b60@gateway/web/freenode/ip.159.226.43.96] has joined #shogun | 08:39 | |
lisitsyn | HeikoS: hey | 09:19 |
-!- kislay [~kislay@14.139.41.150] has joined #shogun | 09:19 | |
lisitsyn | HeikoS: it seems we have some weight, they always give us what we want :D | 09:20 |
-!- kislay [~kislay@14.139.41.150] has quit [Ping timeout: 240 seconds] | 09:45 | |
-!- thoralf [~thoralf@91-65-136-14-dynip.superkabel.de] has joined #shogun | 09:48 | |
thoralf | Hey. | 09:56 |
-!- thoralf [~thoralf@91-65-136-14-dynip.superkabel.de] has quit [Ping timeout: 240 seconds] | 10:00 | |
-!- soumyaC [uid15286@gateway/web/irccloud.com/x-lddjzlfvibszeghc] has joined #shogun | 10:04 | |
sonne|work | lisitsyn: yeah we are a fat cow that cannot be stopped | 10:34 |
lisitsyn | sonne|work: a fat cow on speed | 10:34 |
sonne|work | yeah magic potion overdose | 10:35 |
-!- lisitsyn [~lisitsyn@80.252.20.67] has quit [Ping timeout: 246 seconds] | 10:53 | |
-!- lambday [67157f54@gateway/web/freenode/ip.103.21.127.84] has joined #shogun | 11:38 | |
-!- HeikoS1 [~heiko@90.201.62.238] has joined #shogun | 11:39 | |
lambday | HeikoS1: hi | 11:39 |
lambday | I was just going to mail you and Dino.. | 11:39 |
-!- HeikoS1 [~heiko@90.201.62.238] has quit [Ping timeout: 250 seconds] | 11:43 | |
-!- HeikoS1 [~heiko@ra-vpn-94.ra-vpn.ucl.ac.uk] has joined #shogun | 11:59 | |
lambday | HeikoS1: hi.. there? | 12:03 |
-!- HeikoS2 [~heiko@90.201.62.238] has joined #shogun | 12:04 | |
-!- HeikoS1 [~heiko@ra-vpn-94.ra-vpn.ucl.ac.uk] has quit [Ping timeout: 246 seconds] | 12:04 | |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has joined #shogun | 12:05 | |
lisitsyn | HeikoS HeikoS2 ! | 12:05 |
-!- HeikoS2 [~heiko@90.201.62.238] has quit [Ping timeout: 240 seconds] | 12:08 | |
-!- HeikoS1 [~heiko@ra-vpn-100.ra-vpn.ucl.ac.uk] has joined #shogun | 12:22 | |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 12:27 | |
shogun-notifier- | shogun: Wu Lin :develop * 39c3e90 / / (6 files): https://github.com/shogun-toolbox/shogun/commit/39c3e90918e666f90f2f5197c1f21dbd4e5cc6bf | 12:27 |
shogun-notifier- | shogun: add the helper function for unit test | 12:27 |
shogun-notifier- | shogun: Wu Lin :develop * 6181673 / / (2 files): https://github.com/shogun-toolbox/shogun/commit/6181673e3560631daebe3e3e62c4b69ea0739727 | 12:27 |
shogun-notifier- | shogun: fix | 12:27 |
shogun-notifier- | shogun: Wu Lin :develop * 8160ed1 / src/shogun/mathematics/Math.cpp: https://github.com/shogun-toolbox/shogun/commit/8160ed18a93d7d7adf96c8fbe3f53e01211a549b | 12:27 |
shogun-notifier- | shogun: minor fix | 12:27 |
shogun-notifier- | shogun: Heiko Strathmann :develop * 59e9f13 / / (7 files): https://github.com/shogun-toolbox/shogun/commit/59e9f13bfe051f3da303c6b8ce6dec12addfebd8 | 12:27 |
shogun-notifier- | shogun: Merge pull request #2115 from yorkerlin/develop | 12:27 |
shogun-notifier- | shogun: | 12:27 |
shogun-notifier- | shogun: a helper function for unit test | 12:27 |
-!- newlido [~quassel@193.227.20.2] has joined #shogun | 12:46 | |
-!- PirosB3 [~pirosb3@89-97-67-159.ip16.fastwebnet.it] has joined #shogun | 12:48 | |
lisitsyn | we need kaggle team | 12:54 |
lisitsyn | :D | 12:54 |
lisitsyn | HeikoS: wiking: besser82: lambday: sonne|work: we need kaggle team! | 12:54 |
lambday | lisitsyn: who's competing? :-o | 12:55 |
wiking | lisitsyn: for which task? | 12:56 |
lisitsyn | wiking: any | 12:57 |
wiking | https://www.kaggle.com/c/risky-business | 12:57 |
wiking | ? | 12:57 |
wiking | :) | 12:57 |
wiking | we cannot do that ;P | 12:58 |
lisitsyn | well small steps | 12:58 |
lisitsyn | :D | 12:58 |
lisitsyn | we can just start from these simple test tasks | 12:58 |
wiking | https://www.kaggle.com/c/allstate-purchase-prediction-challenge | 12:58 |
lisitsyn | that's the way we can improve our API | 12:58 |
HeikoS1 | lisitsyn: I agree :) | 13:00 |
wiking | lisitsyn: let's start a python notebook ;) | 13:00 |
wiking | *ipython | 13:00 |
wiking | for the task | 13:01 |
wiking | ) | 13:01 |
wiking | that would be the best i think | 13:01 |
HeikoS1 | wiking: I agree :) | 13:01 |
wiking | lisitsyn: new repo? | 13:04 |
wiking | shogun-kaggle ? | 13:04 |
lisitsyn | yes | 13:04 |
lambday | sounds great! | 13:05 |
lisitsyn | let me create that | 13:05 |
lisitsyn | https://github.com/shogun-toolbox/shogun-kaggle | 13:06 |
lisitsyn | HeikoS: wiking: besser82: lambday: sonne|work: ok and another thing - we have to get distributed | 13:06 |
wiking | lisitsyn: hah u were faster :) | 13:06 |
wiking | lisitsyn: what do u mean? :) | 13:06 |
wiking | lisitsyn: btw we would need a team as well | 13:07 |
wiking | i mean on kaggle | 13:07 |
lisitsyn | yes | 13:07 |
wiking | i have a kaggle account | 13:07 |
wiking | anybody else? | 13:07 |
lisitsyn | I have | 13:07 |
lambday | I do | 13:07 |
lisitsyn | hmm guys | 13:07 |
lisitsyn | just an idea | 13:07 |
lisitsyn | can we create some roles in our team? | 13:07 |
wiking | like? | 13:08 |
lisitsyn | hmm e.g. HeikoS knows about X | 13:08 |
lisitsyn | and you wiking know about Y | 13:08 |
lisitsyn | so we delegate responsibilitie | 13:08 |
lisitsyn | I have vague idea but want to share | 13:08 |
lisitsyn | :D | 13:08 |
wiking | k | 13:09 |
lisitsyn | wiking: ok actually may be we should denote maintainers for modules | 13:09 |
lisitsyn | what do you think about that? | 13:09 |
wiking | sounds like a plan | 13:09 |
wiking | btw anybody has a fucking dashboard on kaggle? ) | 13:10 |
lisitsyn | I mean I should be responsible for multiclass things | 13:10 |
lisitsyn | and everybody should kick me to care about that | 13:10 |
wiking | yeah that'd be good | 13:10 |
lisitsyn | :D | 13:10 |
lisitsyn | okay may be we can go on later | 13:10 |
wiking | so that we dont have scary merges | 13:10 |
lisitsyn | I have no clue what's the dashboard on kaggle | 13:10 |
wiking | me neither | 13:11 |
wiking | ok fuck | 13:12 |
wiking | i've found it :D | 13:12 |
wiking | anybody care to give me his kaggle account? | 13:13 |
wiking | or rather the email addr u use for kaggle? | 13:14 |
lambday | wiking: mine is heavensdevil6909@gmail.com | 13:14 |
lambday | rahul-de is the username | 13:14 |
wiking | lisitsyn: ping? | 13:14 |
lisitsyn | my login is no-idea | 13:14 |
lisitsyn | just a sec | 13:14 |
wiking | lisitsyn: email u used for kaggle | 13:14 |
lisitsyn | I feel idiot when clicking forgot username | 13:15 |
lisitsyn | :D | 13:15 |
wiking | lambday: u should receive an email | 13:15 |
lambday | got it | 13:15 |
lisitsyn | wiking: lisitsyn.s.o@gmail.com | 13:15 |
wiking | lisitsyn: sent the invite | 13:15 |
wiking | seems we cannot have permanent teams | 13:15 |
wiking | only per competition teams | 13:15 |
lisitsyn | oh | 13:16 |
lisitsyn | really? | 13:16 |
wiking | lisitsyn: yeah seems like | 13:16 |
lambday | holy crap that's a long list of rules | 13:16 |
lisitsyn | sucks | 13:16 |
wiking | heheh accept it | 13:16 |
wiking | :) | 13:16 |
lambday | yeah did :D | 13:16 |
wiking | ok so now we have a team | 13:16 |
wiking | and we can do submissions | 13:16 |
lambday | only the outputs are to be submitted iirc | 13:18 |
wiking | yep | 13:18 |
wiking | so now the task is pretty easy | 13:18 |
lisitsyn | yeah win it | 13:18 |
lisitsyn | that's the easy part | 13:18 |
lambday | $50k | 13:18 |
wiking | i mean read in file | 13:18 |
wiking | and the output file creator | 13:18 |
wiking | that's like step -1 | 13:18 |
lisitsyn | ... | 13:18 |
lisitsyn | profit | 13:18 |
lisitsyn | :D | 13:18 |
wiking | :D | 13:18 |
wiking | and then we'll see the rest | 13:18 |
wiking | btw any ideas of the structure for the repo ? | 13:19 |
wiking | like each competition a separate directory? | 13:19 |
lisitsyn | competition-name/... | 13:19 |
lisitsyn | yeah | 13:19 |
wiking | and the data should be always like | 13:19 |
wiking | competiton-name/data directory | 13:19 |
wiking | since we are not allowed to distribute the data | 13:19 |
lisitsyn | uhmm github will kill us | 13:20 |
lisitsyn | if we put data | 13:20 |
wiking | so everybody has to download that for ourselves | 13:20 |
lisitsyn | yeah | 13:20 |
wiking | lisitsyn: nana we are not allowed i think | 13:20 |
lambday | btw why notebook? I mean the data is usually *H-U-G-E*.. | 13:20 |
lisitsyn | wiking: yeah we are not allowed as well | 13:20 |
wiking | lambday: well the notebook was just for having a great way to present our solution already | 13:20 |
wiking | so it's not just messy code | 13:20 |
lambday | alrighty | 13:20 |
wiking | but somebody can right away like check the whole solution | 13:20 |
wiking | of course it'll make things much more complicated | 13:21 |
wiking | so maybe it's not the good idea | 13:21 |
wiking | as honestly coding and doing experiments with a notebook | 13:21 |
wiking | i think is a bit pain... | 13:21 |
lambday | we need to submit multiple solutions with different approaches.. | 13:22 |
lambday | and the best one is our score | 13:22 |
wiking | lambday: well that's easy | 13:22 |
wiking | random() | 13:22 |
wiking | :D | 13:22 |
wiking | that'll put us somewhere on the leaderboard already :PPP | 13:22 |
lambday | lol... | 13:23 |
lisitsyn | ah guys | 13:23 |
lisitsyn | language? | 13:23 |
lambday | doesn't matter | 13:23 |
wiking | yeah that's what i was about to ask | 13:23 |
lisitsyn | hmm matters | 13:23 |
wiking | go hardcore and do everything in c++ | 13:23 |
lisitsyn | hahah | 13:23 |
lisitsyn | no | 13:23 |
wiking | or be more subtle and python | 13:23 |
wiking | ? | 13:23 |
lisitsyn | python | 13:23 |
wiking | yeah i'm for python as well | 13:23 |
lambday | will take ages to process GBs of data | 13:23 |
wiking | lambday: where do u see the gigs of data? | 13:24 |
lisitsyn | yeah | 13:24 |
lisitsyn | 10 mb | 13:24 |
lisitsyn | :D | 13:24 |
lisitsyn | 5.63mb train | 13:24 |
lisitsyn | 2.02 test | 13:24 |
lambday | last competition I participated in had 500MB compressed data | 13:24 |
lisitsyn | that's not big data | 13:24 |
lisitsyn | :D | 13:24 |
lambday | this one is easy then :D | 13:24 |
lambday | python +1 | 13:24 |
wiking | l-rw-r--r-- 1 wiking staff 46M Feb 18 09:37 train.csv | 13:25 |
lisitsyn | oops | 13:25 |
lisitsyn | I told you zip | 13:25 |
wiking | so the first thing we need to do is fix the readline :))))) | 13:25 |
lisitsyn | I can start with IO actually | 13:25 |
HeikoS1 | just talked to dino, who is also interested | 13:25 |
lisitsyn | tonight | 13:25 |
lisitsyn | HeikoS1: are you in? | 13:26 |
-!- sejDino [~rudo@cpc1-dals14-2-0-cust31.hari.cable.virginm.net] has joined #shogun | 13:26 | |
lisitsyn | hey sejDino | 13:26 |
wiking | ok please register on kaggle and give me email addr that u used for registering on kaggle.com | 13:26 |
wiking | so i can invite u guys to the team | 13:26 |
lambday | sejDino: hi :) | 13:26 |
HeikoS1 | lisitsyn: yes I would like to join this stuff | 13:26 |
sejDino | hey! | 13:26 |
HeikoS1 | good way to make shogun more usable for some more applied things :) | 13:26 |
lisitsyn | okay employ more guys and we will win | 13:27 |
lisitsyn | :D | 13:27 |
wiking | :D | 13:27 |
HeikoS1 | I guess we will come up with a list of things that would be useful to have and discover loads of bugs ;) | 13:27 |
lisitsyn | we need a tracker | 13:29 |
lisitsyn | to distribute tasks and so on | 13:29 |
lisitsyn | who's up to use github? | 13:29 |
wiking | lisitsyn: github/issues and create a new label called kaggle? | 13:30 |
wiking | or since it's a different repo | 13:30 |
wiking | we can do that there | 13:30 |
wiking | https://github.com/shogun-toolbox/shogun-kaggle/issues | 13:30 |
lisitsyn | created milestone | 13:30 |
lisitsyn | and label | 13:30 |
wiking | and if we need to fix a bug within shogun because we found it because of kaggle task we can still label it kaggle | 13:31 |
wiking | so that we know what might be a priority bug to fix | 13:31 |
wiking | and for later statistics | 13:31 |
wiking | to see how many bugs we actually found/fixed because of kaggling | 13:31 |
wiking | lisitsyn: sounds like a plan | 13:31 |
sejDino | so is the plan to go for allstate challenge or risky business or both? :) | 13:32 |
wiking | sejDino: we cannod to risky business | 13:32 |
wiking | as it's only for master members of kaggle | 13:32 |
wiking | and none of us has that XP points on kaggle :P | 13:33 |
wiking | sejDino: but we can do other 'educational' task as well | 13:33 |
sejDino | I see - oh, well - now we'll start building those XP pts quickly :) | 13:33 |
wiking | just to promote shogun | 13:33 |
lisitsyn | https://github.com/shogun-toolbox/shogun-kaggle/issues | 13:33 |
wiking | lisitsyn: is on fire | 13:33 |
sejDino | that sounds like a good idea as well | 13:33 |
lisitsyn | https://github.com/shogun-toolbox/shogun-kaggle/issues/2 | 13:34 |
lisitsyn | dump your mind here | 13:34 |
lisitsyn | who's up describing the input /output data? | 13:34 |
lisitsyn | in the #2 | 13:34 |
lisitsyn | sooo whoo | 13:35 |
lisitsyn | do we need a cluster? | 13:35 |
wiking | lisitsyn: no ... worst case we use fatbot :P | 13:36 |
wiking | teh amount of data is really not so much | 13:36 |
lisitsyn | wiking: want to describe dataset? | 13:36 |
wiking | 665250 data/train.csv | 13:36 |
wiking | -1 because it has header | 13:36 |
wiking | lisitsyn: yeah sure i'm up for that | 13:36 |
lisitsyn | wiking: https://github.com/shogun-toolbox/shogun-kaggle/issues/3 | 13:37 |
lisitsyn | sejDino: HeikoS: HeikoS1: you are scientific guys | 13:38 |
lisitsyn | so waiting your input https://github.com/shogun-toolbox/shogun-kaggle/issues/2 | 13:38 |
lisitsyn | ;) | 13:38 |
HeikoS1 | lisitsyn: happy to do :) Ill talk to sejDino a bit later today and then well give some output | 13:39 |
lisitsyn | HeikoS: assigned you responsible for this issue ;) | 13:39 |
sejDino | lisitsyn: yep, will work on it :) | 13:39 |
lisitsyn | ok that's going much funny now | 13:40 |
lisitsyn | :D | 13:40 |
sejDino | it seems like an interesting problem actually - each customer samples a bunch of these "options" and then they pick one of those (or a new one) | 13:41 |
lisitsyn | HeikoS1: okay lets organize it this way | 13:41 |
sejDino | so it's not really a standard classification/regression task | 13:41 |
lisitsyn | we discuss it in #2 | 13:41 |
lisitsyn | then everything that is worth trying | 13:41 |
lisitsyn | is forked to a new task | 13:41 |
lisitsyn | agree? | 13:42 |
sejDino | lisitsyn: sounds good | 13:42 |
lisitsyn | sejDino: cool thanks | 13:42 |
lisitsyn | lambday: mind to start notebook? | 13:43 |
lisitsyn | or other words | 13:44 |
lisitsyn | want to be responsible for notebook? | 13:44 |
lisitsyn | ;) | 13:44 |
lambday | lisitsyn: what should we start with? | 13:44 |
lambday | yeah I can do that | 13:44 |
lisitsyn | lambday: hmm okay actually | 13:44 |
lisitsyn | you could create some draft | 13:44 |
lisitsyn | and make sure everybody can work on notebook in one click | 13:45 |
wiking | lambday: use the template and put it in shogun-kaggle/allstate | 13:45 |
lisitsyn | so makefile or sth like that | 13:45 |
wiking | lambday: and u can have the assumption that the data is in shogun-kaggle/allstate/data | 13:45 |
lisitsyn | I want to organize it that way | 13:45 |
lisitsyn | so we have some 'library' | 13:45 |
lisitsyn | that is used in notebook | 13:46 |
wiking | lisitsyn: yeah | 13:46 |
lisitsyn | so we can hide some routine things | 13:46 |
lisitsyn | load_dataset() | 13:46 |
lisitsyn | that's it | 13:46 |
wiking | lisitsyn: yeah i think all the .py + ipynb should be in shogun-kaggle/allstate | 13:46 |
lisitsyn | yes | 13:46 |
wiking | and then u can from . import load_dataset | 13:46 |
lisitsyn | okay will be back a bit later | 13:47 |
wiking | in the notebook | 13:47 |
lisitsyn | yes | 13:47 |
wiking | lisitsyn: thing is though | 13:47 |
wiking | lisitsyn: this dataset is more or less a timeline | 13:47 |
wiking | lisitsyn: so actually a pandas dataframe would be the best option to store the data ;) | 13:47 |
wiking | not CFeatures of any kind | 13:48 |
wiking | although this is just the first thought i had now looking at the dataset | 13:48 |
wiking | bbl | 13:48 |
-!- HeikoS1 [~heiko@ra-vpn-100.ra-vpn.ucl.ac.uk] has quit [Ping timeout: 246 seconds] | 13:51 | |
pl8787 | hi, see issue https://github.com/shogun-toolbox/shogun/issues/2119. Dose any one has any suggestion? | 13:51 |
-!- newlido [~quassel@193.227.20.2] has quit [Ping timeout: 246 seconds] | 13:53 | |
pl8787 | I think add a subclass of CFeatures may solve this problem. Such as CIndexFeatures | 13:53 |
pl8787 | If CIndexFeatures pass in CCustom::init, we just change the subset of the kernel matrix, and then get_kernel_matrix will return the right value. | 13:55 |
-!- HeikoS1 [~heiko@90.201.62.238] has joined #shogun | 14:18 | |
-!- sejDino [~rudo@cpc1-dals14-2-0-cust31.hari.cable.virginm.net] has left #shogun [] | 14:25 | |
-!- PirosB3 [~pirosb3@89-97-67-159.ip16.fastwebnet.it] has quit [Quit: PirosB3] | 14:36 | |
-!- HeikoS1 [~heiko@90.201.62.238] has quit [Ping timeout: 246 seconds] | 14:43 | |
-!- HeikoS1 [~heiko@ra-vpn-81.ra-vpn.ucl.ac.uk] has joined #shogun | 14:58 | |
-!- PirosB3 [~pirosb3@89-97-67-159.ip16.fastwebnet.it] has joined #shogun | 15:23 | |
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 15:27 | |
-!- kislay [~kislay@117.239.94.112] has joined #shogun | 15:30 | |
-!- HeikoS1 [~heiko@ra-vpn-81.ra-vpn.ucl.ac.uk] has quit [Read error: Connection reset by peer] | 15:30 | |
-!- newlido [~quassel@193.227.20.2] has joined #shogun | 15:47 | |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has quit [Read error: Connection reset by peer] | 16:00 | |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has joined #shogun | 16:01 | |
-!- pl8787 [9fe22b60@gateway/web/freenode/ip.159.226.43.96] has quit [Quit: Page closed] | 16:07 | |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.] | 16:25 | |
-!- prams [~pratheekm@122.167.124.235] has joined #shogun | 17:13 | |
-!- newlido [~quassel@193.227.20.2] has quit [Remote host closed the connection] | 17:27 | |
-!- prams [~pratheekm@122.167.124.235] has quit [Ping timeout: 246 seconds] | 17:42 | |
-!- dhruv13j [dhruv13j@gateway/shell/bnc4free/x-zivjvwoyvehjepue] has joined #shogun | 18:06 | |
-!- lisitsyn [~lisitsyn@80.252.20.67] has joined #shogun | 18:07 | |
-!- PirosB3 [~pirosb3@89-97-67-159.ip16.fastwebnet.it] has quit [Quit: PirosB3] | 18:25 | |
-!- PirosB3 [~pirosb3@89-97-67-159.ip16.fastwebnet.it] has joined #shogun | 18:26 | |
-!- PirosB3 [~pirosb3@89-97-67-159.ip16.fastwebnet.it] has quit [Client Quit] | 18:29 | |
-!- kislay [~kislay@117.239.94.112] has quit [Quit: Leaving] | 18:36 | |
@besser82 | wiking, ping? | 18:43 |
-!- kislay|exams [~kislay@210.212.61.185] has joined #shogun | 18:48 | |
wiking | pong | 18:50 |
-!- khalednasr [~k.nasr92@41.69.241.51] has joined #shogun | 18:56 | |
-!- kislay|exams [~kislay@210.212.61.185] has quit [Read error: Connection reset by peer] | 19:23 | |
-!- pl8787 [9fe22b60@gateway/web/freenode/ip.159.226.43.96] has joined #shogun | 19:27 | |
-!- lambday [67157f54@gateway/web/freenode/ip.103.21.127.84] has quit [Ping timeout: 240 seconds] | 19:38 | |
-!- zxtx [~zxtx@sunnythinking.org] has quit [Quit: Strong teeth for a strong nation] | 19:57 | |
-!- zxtx [~zxtx@sunnythinking.org] has joined #shogun | 19:58 | |
wiking | khalednasr: thnx for joinng the pr review process | 20:05 |
khalednasr | wiking, no problem, I'm happy to participate :) | 20:06 |
wiking | cool | 20:06 |
wiking | khalednasr: u can join as well the kaggle team if u fancy and have time :) | 20:07 |
khalednasr | wiking, there's a kaggle team? :) | 20:07 |
@besser82 | wiking, WTF is kaggle?!? | 20:08 |
wiking | khalednasr: we've just started today | 20:08 |
wiking | besser82: kaggle.com | 20:08 |
@besser82 | wiking, thx! Will have a look ;) | 20:09 |
khalednasr | wiking, do you guys have a specific competition in mind? | 20:12 |
wiking | khalednasr: we've started with allstate | 20:14 |
khalednasr | wiking, thanks for inviting me! I'll take a look at it | 20:17 |
wiking | nw | 20:18 |
-!- khalednasr [~k.nasr92@41.69.241.51] has quit [Quit: Leaving] | 20:32 | |
-!- Khalednasr [~khalednas@41.69.241.51] has joined #shogun | 20:36 | |
-!- Khalednasr [~khalednas@41.69.241.51] has quit [Ping timeout: 240 seconds] | 20:41 | |
-!- khalednasr [~k.nasr92@41.69.156.29] has joined #shogun | 20:51 | |
-!- lambday [67157f54@gateway/web/freenode/ip.103.21.127.84] has joined #shogun | 20:59 | |
-!- sunil1337 [0e8bd055@gateway/web/freenode/ip.14.139.208.85] has joined #shogun | 21:29 | |
sunil1337 | lambday: hey! | 21:30 |
lambday | sunil1337: hi! | 21:31 |
sunil1337 | lambday: please review https://github.com/sunil1337/shogun/commit/ef30f622df44ab51cb97c4a66c5ea3a673348dfe | 21:31 |
lambday | sunil1337: checking | 21:32 |
sunil1337 | lambday: okay! fails at compiling class_list.cpp because we have 3 defs of eigen3 and naive. need a work around there | 21:32 |
lambday | 3 defs of eigen3 and naive? | 21:33 |
sunil1337 | lambday: partial specialization they py scripts adds them to class_list.cpp | 21:33 |
sunil1337 | lambday:*the | 21:33 |
lambday | sunil1337: you could have just left the sparse thing for now | 21:34 |
lambday | the code won't be same for dense and sparse | 21:34 |
lambday | as in, we can't use map | 21:34 |
sunil1337 | lambday: yeah.. i just called existing sparse_dot for now | 21:34 |
sunil1337 | lambday: i've commented the problem please take a look | 21:35 |
lambday | okie | 21:35 |
lambday | sunil1337: could you please split the definition and declarations? | 21:36 |
lambday | into separate .h and .cpp | 21:36 |
lambday | each class - one .h one .cpp | 21:37 |
sunil1337 | lambday: sure.. | 21:37 |
wiking | btw u could use macros | 21:37 |
sunil1337 | wiking: for new calls ? | 21:38 |
lambday | sunil1337: umm you didn't need to add VectorDotOperator already... first just VectorDotProduct and Eigen3DotProduct which would work independently | 21:39 |
lambday | then we would add CLinearAlgebra factory class | 21:39 |
lambday | after that we'll add VectorDotOperator | 21:39 |
lambday | under LinearOperator interface | 21:39 |
lambday | this iterative approach makes it easy to find bugs | 21:40 |
lambday | Naive - we don't really need.. I just added in the gist for demonstration purpose | 21:40 |
sunil1337 | lambday: okay! :) | 21:40 |
lambday | this looks good :) | 21:41 |
lambday | good work :) | 21:41 |
sunil1337 | lambday: thanks :) naive is already there in existing SGVector class... | 21:41 |
lambday | but if you could please send patches containing those two first | 21:41 |
lambday | it would make it so much easier | 21:41 |
sunil1337 | lambday: will do.. :) | 21:41 |
lambday | yeah - no one needs a sophisticated interface to do naive things ;) | 21:41 |
sunil1337 | lambday: :) | 21:42 |
lambday | sunil1337: we *don't* need to put VectorDotProduct under SGObject | 21:45 |
-!- thoralf [~thoralf@91-65-136-198-dynip.superkabel.de] has joined #shogun | 21:46 | |
lambday | and please avoid using __ in #ifndef __CLASSNAME_H__... make it #ifndef CLASSNAME_H_ | 21:46 |
sunil1337 | lambday: okay :) | 21:46 |
thoralf | Hey. | 21:49 |
lambday | thoralf: hi... | 21:53 |
lambday | thoralf: where are we about the singleton issue? | 21:54 |
thoralf | lambday: We didn't decide anything... currently we're collecting arguments. ;) | 21:57 |
thoralf | lambday: Do you have an optinion about it? | 21:57 |
-!- iglesiasg [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has joined #shogun | 21:58 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 21:58 | |
lambday | thoralf: not particularly.. I agree with HeikoS's suggestion | 21:59 |
lambday | but I am eager to see what sonney2k says about this | 22:00 |
thoralf | Me too :) | 22:08 |
-!- soumyaC [uid15286@gateway/web/irccloud.com/x-lddjzlfvibszeghc] has quit [Quit: Connection closed for inactivity] | 22:25 | |
-!- iglesiasg [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has quit [Ping timeout: 252 seconds] | 22:27 | |
-!- iglesiasg [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has joined #shogun | 22:34 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 22:34 | |
thoralf | lisitsyn: Awake? | 22:57 |
lisitsyn | thoralf: yes | 22:57 |
thoralf | Hello :) | 22:57 |
thoralf | I just realized we're having global variables anyway and are just copying them to each object. | 22:57 |
thoralf | So it does means we cannot have multiple shoguns even right now. | 22:58 |
thoralf | And the difference between global variables and singletons is artificial for this discussion. ;) | 22:58 |
lisitsyn | yeah | 23:00 |
lisitsyn | we have it tied together already | 23:00 |
thoralf | We're doomed anyway. So let's change it while everybody is sleeping. :D | 23:04 |
lisitsyn | thoralf: what exactly? | 23:10 |
lisitsyn | I am completely lost in discussions | 23:10 |
lisitsyn | :D | 23:10 |
thoralf | Removing SGIO, Version, Parallel members from CSGObject. | 23:11 |
lisitsyn | thoralf: want to do that? | 23:12 |
lisitsyn | ;) | 23:12 |
thoralf | Wrong question. | 23:13 |
thoralf | Want to write a ticket for that? :D | 23:13 |
lisitsyn | ahh students | 23:13 |
lisitsyn | thoralf: that's the same thing happens at my job | 23:13 |
lisitsyn | constantly forgetting I have 3 programmers to delegate job | 23:14 |
lisitsyn | :D | 23:14 |
thoralf | lisitsyn: To me honest, I'm a bit lost how to improve this "thing" and I want to discuss which options we have (without making it worse by adding another 1000 classes). | 23:15 |
thoralf | Changing it would only be the technical part. | 23:16 |
lisitsyn | thoralf: I think making SGContext a singleton first | 23:16 |
lisitsyn | would work | 23:16 |
thoralf | I have no concerns about this. | 23:18 |
thoralf | Putting Version, Parallel, SGIO into it. | 23:18 |
thoralf | Most functionality on these things is based on macros, so easy to change without touching many classes. | 23:19 |
lisitsyn | yeah | 23:21 |
thoralf | And then we would naturally lose the duplication of SG_SINFO/SG_INFO, etc., because everything is depending on the singleton. | 23:21 |
lisitsyn | uh | 23:22 |
thoralf | What? Did I miss something? | 23:22 |
lisitsyn | well if we don't want multiple contexts | 23:22 |
lisitsyn | it should work | 23:22 |
lisitsyn | I overlooked the SG_SINFO/SG_INFO thing | 23:22 |
lisitsyn | thoralf: if we use the singleton in SG_INFO | 23:23 |
lisitsyn | this member of SGObject makes no sense | 23:23 |
lisitsyn | won't be used | 23:23 |
thoralf | Yes. | 23:24 |
thoralf | So we can remove it. Hooray :) | 23:24 |
lisitsyn | this denies multiple contexts | 23:25 |
lisitsyn | I would be ok with multiple contexts | 23:25 |
lisitsyn | so that's disputable | 23:25 |
thoralf | Okay, how would it work with multiple contexts? | 23:25 |
thoralf | Lets say, you have machine1 from context1 and machine2 from context2. How would each machine create a new CLabel of the right context? | 23:26 |
lisitsyn | we should ask for labels in context | 23:28 |
lisitsyn | :D | 23:28 |
lisitsyn | that sounds like a factory | 23:28 |
thoralf | Yeah. | 23:28 |
thoralf | Sounds like JavaEE | 23:28 |
thoralf | CLabels l = context.register(new CLabels); | 23:29 |
thoralf | vs | 23:29 |
thoralf | CLabels l = context.getInstanceOf(class CLabels); | 23:29 |
lisitsyn | thoralf: actually there is a lot of beauty in that :D | 23:32 |
lisitsyn | makes things really extensible | 23:32 |
lisitsyn | thoralf: if we go modular | 23:33 |
lisitsyn | we'd need factories anyway | 23:33 |
thoralf | Correct me if I'm wrong, but this doesn't make things simpler. ;) | 23:33 |
lisitsyn | thoralf: this makes things extensible ;) | 23:33 |
thoralf | Correct me if I'm wrong, but this doesn't make things simpler. ;) | 23:33 |
thoralf | lol, okay. | 23:34 |
thoralf | No free lunch. | 23:34 |
lisitsyn | well the simplest thing is no objects | 23:34 |
lisitsyn | ;) | 23:34 |
lisitsyn | but the issue is that we really need factories to go modular | 23:34 |
lisitsyn | say I wrote SuperSVM | 23:34 |
lisitsyn | how do I spawn it? | 23:35 |
lisitsyn | there is no SuperSVM in headers/interfaces | 23:35 |
lisitsyn | I can only ask for SVM | 23:36 |
lisitsyn | (effectively) | 23:36 |
thoralf | Damn. | 23:39 |
-!- newlido [~quassel@41.218.173.235] has joined #shogun | 23:39 | |
thoralf | You're playing in the opposite team. | 23:39 |
thoralf | I'm on simplification (white circle on white background) | 23:39 |
thoralf | You're on, well, you're on maximum extensible. ;) | 23:40 |
lisitsyn | thoralf: we need plugins | 23:41 |
lisitsyn | to not wait \infty before it compiles | 23:41 |
lisitsyn | and to use only what you want :) | 23:41 |
lisitsyn | plugins are not only about extensibility but to put development on speed | 23:42 |
thoralf | Hmm. | 23:43 |
thoralf | But this involves *many* design decisions that haven't been met. For example: simple "data" classes like vectors, matrices, etc. that need to be fast. No need to ask daddy every time you want to create an instance. | 23:44 |
lisitsyn | but they are not sgobjects, right? | 23:45 |
thoralf | Hmm. | 23:45 |
thoralf | Hmm! | 23:45 |
thoralf | Yes. | 23:45 |
lisitsyn | I am talking only about sgobjects because they are not lightweight | 23:45 |
lisitsyn | and never been :) | 23:45 |
thoralf | I got your point. | 23:47 |
lisitsyn | okay destination bed | 23:50 |
thoralf | Good night. | 23:52 |
-!- newlido [~quassel@41.218.173.235] has quit [Remote host closed the connection] | 23:58 | |
--- Log closed Thu Apr 10 00:00:15 2014 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!