| --- Log opened Sun Jul 06 00:00:02 2014 | ||
| -!- rajul_ [~rajul@182.68.180.70] has quit [Ping timeout: 260 seconds] | 00:22 | |
| -!- rajul [~rajul@180.151.18.31] has joined #shogun | 00:30 | |
| -!- rajul [~rajul@180.151.18.31] has quit [Quit: Leaving] | 00:40 | |
| -!- rajul [~rajul@180.151.18.31] has joined #shogun | 00:46 | |
| -!- rajul [~rajul@180.151.18.31] has quit [Client Quit] | 00:48 | |
| -!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 01:03 | |
| shogun-buildbot | build #745 of nightly_none is complete: Success [build successful] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_none/builds/745 | 03:09 |
|---|---|---|
| shogun-buildbot | build #740 of nightly_all is complete: Failure [failed compile test] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_all/builds/740 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, lambday <heavensdevil6909@gmail.com>, Sergey Lisitsyn <lisitsyn.s.o@gmail.com>, khalednasr <k.nasr92@gmail.com> | 03:13 |
| -!- khalednasr [~k.nasr92@41.69.213.175] has joined #shogun | 03:30 | |
| shogun-buildbot | build #844 of nightly_default is complete: Success [build successful] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/844 | 05:30 |
| -!- khalednasr [~k.nasr92@41.69.213.175] has quit [Ping timeout: 252 seconds] | 05:30 | |
| -!- rajul [~rajul@182.68.240.221] has joined #shogun | 07:42 | |
| -!- rajul_ [~rajul@180.151.18.31] has joined #shogun | 07:55 | |
| -!- rajul [~rajul@182.68.240.221] has quit [Ping timeout: 272 seconds] | 07:59 | |
| -!- rajul_ [~rajul@180.151.18.31] has quit [Quit: Leaving] | 09:05 | |
| -!- rajul [~rajul@180.151.18.31] has joined #shogun | 09:05 | |
| -!- hushell [~hushell@c-71-237-222-49.hsd1.or.comcast.net] has quit [Ping timeout: 240 seconds] | 09:09 | |
| -!- kislay [~Abhijeet@103.240.205.244] has joined #shogun | 09:30 | |
| -!- kislay [~Abhijeet@103.240.205.244] has quit [Ping timeout: 252 seconds] | 09:36 | |
| -!- khalednasr [~k.nasr92@41.69.164.234] has joined #shogun | 09:38 | |
| -!- kislay [~Abhijeet@27.116.52.192] has joined #shogun | 09:39 | |
| -!- khalednasr [~k.nasr92@41.69.164.234] has quit [Quit: Leaving] | 09:43 | |
| shogun-buildbot | build #169 of osx2 - python is complete: Failure [failed test python modular] Build details are at http://buildbot.shogun-toolbox.org/builders/osx2%20-%20python/builds/169 blamelist: Parijat Mazumdar <mazumdarparijat@gmail.com> | 09:53 |
| -!- kislay [~Abhijeet@27.116.52.192] has quit [Ping timeout: 240 seconds] | 09:54 | |
| shogun-buildbot | build #170 of osx2 - python is complete: Failure [failed examples and unit tests] Build details are at http://buildbot.shogun-toolbox.org/builders/osx2%20-%20python/builds/170 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, Sergey Lisitsyn <lisitsyn.s.o@gmail.com>, Fernando Iglesias <fernando.iglesiasg@gmail.com>, khalednasr <k.nasr92@gmail.com> | 10:06 |
| -!- kislay [~Abhijeet@103.240.205.174] has joined #shogun | 10:06 | |
| -!- rajul [~rajul@180.151.18.31] has quit [Ping timeout: 272 seconds] | 10:08 | |
| -!- witness___ [uid10044@gateway/web/irccloud.com/x-gdaigpteywxptahz] has joined #shogun | 10:14 | |
| -!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 10:48 | |
| shogun-notifier- | shogun: khalednasr :feature/linalg * 42829d4 / / (7 files): https://github.com/shogun-toolbox/shogun/commit/42829d43c2e625fcf2da82d665268b3b39f1664d | 10:48 |
| shogun-notifier- | shogun: refactored linalg's dot product | 10:48 |
| shogun-notifier- | shogun: Soumyajit De :feature/linalg * c14ff65 / / (7 files): https://github.com/shogun-toolbox/shogun/commit/c14ff651e56f9ab2d7a0aa0c1d2aabd51d07cf2f | 10:48 |
| shogun-notifier- | shogun: Merge pull request #2367 from khalednasr/feature/linalg | 10:48 |
| shogun-notifier- | shogun: | 10:48 |
| shogun-notifier- | shogun: refactored linalg's dot product | 10:48 |
| -!- kislay [~Abhijeet@103.240.205.174] has quit [Ping timeout: 255 seconds] | 11:23 | |
| -!- PirosB3 [~pirosb3@91-64-37-165-dynip.superkabel.de] has joined #shogun | 11:34 | |
| -!- kislay [~Abhijeet@103.240.205.186] has joined #shogun | 11:42 | |
| -!- PirosB3 [~pirosb3@91-64-37-165-dynip.superkabel.de] has quit [Quit: PirosB3] | 12:17 | |
| -!- rajul [~rajul@182.68.102.194] has joined #shogun | 12:36 | |
| -!- PirosB3 [~pirosb3@91-64-37-165-dynip.superkabel.de] has joined #shogun | 12:37 | |
| shogun-notifier- | shogun: Abhijeet :develop * fe6e8c6 / src/shogun/preprocessor/FisherLDA.cpp,src/shogun/preprocessor/FisherLDA.h: https://github.com/shogun-toolbox/shogun/commit/fe6e8c67e4dcd642cef82c74104399f4441f620d | 12:40 |
| shogun-notifier- | shogun: FisherLDA coverty defects fixed | 12:40 |
| shogun-notifier- | shogun: Fernando Iglesias :develop * 352b606 / src/shogun/preprocessor/FisherLDA.cpp,src/shogun/preprocessor/FisherLDA.h: https://github.com/shogun-toolbox/shogun/commit/352b60635d07c8468e526e7c684ba5a52ec5ccca | 12:40 |
| shogun-notifier- | shogun: Merge pull request #2369 from kislayabhi/fisher_dafects | 12:40 |
| shogun-notifier- | shogun: | 12:40 |
| shogun-notifier- | shogun: FisherLDA coverty defects fixed | 12:40 |
| shogun-buildbot | build #171 of osx2 - python is complete: Failure [failed test python modular] Build details are at http://buildbot.shogun-toolbox.org/builders/osx2%20-%20python/builds/171 blamelist: Abhijeet <abhijeetkislay@gmail.com> | 13:08 |
| shogun-buildbot | build #172 of osx2 - python is complete: Failure [failed test python modular] Build details are at http://buildbot.shogun-toolbox.org/builders/osx2%20-%20python/builds/172 blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com> | 13:29 |
| -!- travis-ci [~travis-ci@ec2-54-81-195-230.compute-1.amazonaws.com] has joined #shogun | 13:30 | |
| 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/29241552 | 13:30 |
| -!- travis-ci [~travis-ci@ec2-54-81-195-230.compute-1.amazonaws.com] has left #shogun [] | 13:30 | |
| -!- PirosB3 [~pirosb3@91-64-37-165-dynip.superkabel.de] has quit [Read error: No route to host] | 13:31 | |
| -!- PirosB3 [~pirosb3@91-64-37-165-dynip.superkabel.de] has joined #shogun | 13:31 | |
| -!- PirosB3 [~pirosb3@91-64-37-165-dynip.superkabel.de] has quit [Quit: PirosB3] | 14:14 | |
| -!- lisitsyn [~qdrgsm@80.252.20.67] has joined #shogun | 14:45 | |
| -!- HeikoS [~heiko@chy77-1-88-183-41-61.fbx.proxad.net] has joined #shogun | 14:45 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 14:45 | |
| shogun-notifier- | shogun-data: Abhijeet :master * ca0adc5 / testsuite/python2-tests/ (2 files): https://github.com/shogun-toolbox/shogun-data/commit/ca0adc53a5d25cb4a47f94ced95abf0b9caaf92e | 14:54 |
| shogun-notifier- | shogun-data: added data files for LDA.cpp integration | 14:54 |
| shogun-notifier- | shogun-data: Heiko Strathmann :master * 527e601 / testsuite/python2-tests/ (2 files): https://github.com/shogun-toolbox/shogun-data/commit/527e6018d99652cacfd07cc05e0ad93ef08943dc | 14:54 |
| shogun-notifier- | shogun-data: Merge pull request #61 from kislayabhi/master | 14:54 |
| shogun-notifier- | shogun-data: | 14:54 |
| shogun-notifier- | shogun-data: added data files for LDA.cpp integration | 14:55 |
| shogun-notifier- | shogun-demo: Saurabh :master * 456a8ae / / (4 files): https://github.com/shogun-toolbox/shogun-demo/commit/456a8aeac4c203af2272ee6c5264effe53588480 | 15:08 |
| shogun-notifier- | shogun-demo: add predictive plot | 15:08 |
| shogun-notifier- | shogun-demo: Heiko Strathmann :master * a4795ee / / (4 files): https://github.com/shogun-toolbox/shogun-demo/commit/a4795eeac63d75d121accd60c9e61ed84a33cc7d | 15:08 |
| shogun-notifier- | shogun-demo: Merge pull request #51 from Saurabh7/predictive | 15:08 |
| shogun-notifier- | shogun-demo: | 15:09 |
| shogun-notifier- | shogun-demo: add predictive plot | 15:09 |
| -!- HeikoS [~heiko@chy77-1-88-183-41-61.fbx.proxad.net] has quit [Ping timeout: 260 seconds] | 15:44 | |
| -!- hushell [~hushell@c-71-237-222-49.hsd1.or.comcast.net] has joined #shogun | 16:28 | |
| -!- pickle27 [~pickle27@192-0-136-118.cpe.teksavvy.com] has joined #shogun | 17:53 | |
| -!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 18:09 | |
| -!- witness___ [uid10044@gateway/web/irccloud.com/x-gdaigpteywxptahz] has quit [Quit: Connection closed for inactivity] | 18:14 | |
| -!- khalednasr [~k.nasr92@41.69.241.215] has joined #shogun | 18:27 | |
| -!- lambday [67157e4c@gateway/web/freenode/ip.103.21.126.76] has joined #shogun | 18:37 | |
| lambday | khalednasr: hi | 18:40 |
| khalednasr | lambday, hi | 18:40 |
| -!- rajul_ [~rajul@182.68.113.105] has joined #shogun | 18:40 | |
| lambday | khalednasr: could you please explain the issue with returning a new matrix for gpu? | 18:40 |
| lambday | khalednasr: sorry but I just want to understand :D' | 18:41 |
| -!- rajul [~rajul@182.68.102.194] has quit [Ping timeout: 248 seconds] | 18:41 | |
| khalednasr | lambday, If you have the square method always returning an SGMatrix, it'll do the computation on the gpu, create a new matrix on gpu memory and then create another matrix on cpu memory and copy the data | 18:42 |
| khalednasr | lambday, which would take a while :) | 18:42 |
| lambday | khalednasr: ah I see! | 18:42 |
| lambday | khalednasr: great catch! | 18:42 |
| lambday | khalednasr: so you're suggesting that if we use gpu backend to compute something | 18:43 |
| lambday | then we should better stick with gpu memory instead and continue the rest of the work with that | 18:43 |
| khalednasr | yup | 18:43 |
| lambday | khalednasr: what I am wondering is that, how often would we actually use a gpu backend for non gpu matrices... | 18:44 |
| -!- witness___ [uid10044@gateway/web/irccloud.com/x-dggkmmlzocjcxfox] has joined #shogun | 18:44 | |
| khalednasr | I'm not sure, the overhead of copying data back and forth between gpu and cpu is quite large | 18:44 |
| lambday | khalednasr: I am just thinking that creating the result matrix in my code and then passing it to the method is too much work :( | 18:45 |
| lambday | khalednasr: plus, I gotta do it this way... square(SGMatrix.... a, GPUMatrix.... b) | 18:45 |
| lambday | right? | 18:45 |
| lambday | result goes into b | 18:45 |
| khalednasr | yeah, If we add another template parameter for the return type | 18:46 |
| khalednasr | for the result* | 18:46 |
| lambday | yeah | 18:46 |
| lambday | but I guess more often than not, when someone is *knowingly* working with gpu backend, he'd be using gpu matrices | 18:47 |
| -!- rajul_ is now known as rajul | 18:47 | |
| khalednasr | lambday, yeah | 18:47 |
| khalednasr | so another template parameter for the result is probably not necessary | 18:48 |
| lambday | khalednasr: I am just wondering if we need to do it this way for backend independent cases.. | 18:48 |
| lambday | I mean, yes, technically its possible to set global backend to be viennacl | 18:48 |
| lambday | in that case any method using global backend is doomed | 18:48 |
| lambday | but if in my code I want to use with *only* gpu matrices, then maybe in the code I should use explicit backend and use gpu matrices | 18:49 |
| lambday | which is safe no matter what global backend is set | 18:49 |
| khalednasr | lambday, yup, make making it possible to select backends for invidiual algorithms would be better | 18:49 |
| khalednasr | maybe making it possible* | 18:50 |
| lambday | khalednasr: that's always possible.. for each method | 18:50 |
| lambday | linalg::foo<Backend::VIENNACL>(a,b); | 18:50 |
| khalednasr | yeah, I mean on the user side | 18:50 |
| lambday | the developer just has to be concerned about the gpu-ness of a and b :D | 18:50 |
| khalednasr | like, the user should be able select which backend neural nets run on, an so on | 18:51 |
| lambday | khalednasr: you mean module wise backend setup? | 18:51 |
| lambday | I mean module-wise global backend setup? | 18:51 |
| khalednasr | lambday, yeah | 18:52 |
| lambday | khalednasr: my main concern is this - say, we're using eigen3 matrix for example... and I have a SGMatrix... so I want the result to be SG, not Eigen::Matrix | 18:53 |
| lambday | I mean the non gpu cases shouldn't suffer | 18:53 |
| khalednasr | lambday, ah I see | 18:54 |
| lambday | khalednasr: if I understood your concern correctly, then you want to avoid data copy back and forth from cpu <--> gpu.. which is possible if we use in our code CGPUMatrix and explicit backend for the methods we call on them | 18:56 |
| lambday | but the module-wise global backend setup for those methods won't work!.. we can't use CUDA for that, for example | 18:56 |
| lambday | hmm | 18:57 |
| khalednasr | lambday, yeah, the main problem is with the return type | 18:59 |
| lambday | khalednasr: if we do things explictly then there's no problem though, right? | 19:00 |
| lambday | I pass a gpu matrix, use gpu backend to compute things which constructs a new matrix in gpu,... and it returns a gpu matrix only | 19:00 |
| khalednasr | yeah, but we have to pick a return type for the wrapper method | 19:00 |
| lambday | khalednasr: if we use the same matrix type then what's the problem? when I am passing CGPUMatrix then return type is also CGPUMatrix in which case we're steering clear of the data copy thing | 19:02 |
| khalednasr | lambday, ahh I see, sorry I was confused about something | 19:03 |
| khalednasr | lambday, yeah there won't be a problem | 19:03 |
| lambday | maybe we need to rethink this a bit... using another namespace to separete things as wiking suggested... | 19:04 |
| lambday | haha happens all the time :D | 19:04 |
| khalednasr | lambday, but we should probably have another wrapper methods that uses a pre-allocated result matrix | 19:04 |
| lambday | so I guess it should be mentioned in a SG_INFO that data copy is a disaster | 19:05 |
| lambday | khalednasr: yeah that's fine | 19:05 |
| lambday | khalednasr: in those operators/copy constructors where you copy cpu <-----> gpu... could you please put a SG_INFO in there? | 19:05 |
| khalednasr | lambday, yeah sure | 19:06 |
| lambday | SG_INFO("Caution: You're laptop's gonna curse you in 3...2...1... BOOM!"); | 19:06 |
| lambday | :D | 19:06 |
| khalednasr | haha :D | 19:06 |
| khalednasr | lambday, another thing: in the Block struct, the matrix is copied into a struct member | 19:08 |
| lambday | khalednasr: I think no.. :/ | 19:08 |
| lambday | block just carries indices | 19:08 |
| lambday | let me check | 19:08 |
| lambday | if I did that then I'm stupid | 19:08 |
| lambday | khalednasr: block just wraps around the matrix | 19:10 |
| lambday | memory is allocated via pointer to that ptr is copied not the data | 19:10 |
| khalednasr | yeah that's the case for SGMatrix and GPUMatrix | 19:10 |
| khalednasr | but eigen3 does deep copies | 19:10 |
| lambday | khalednasr: err.. we should use const& | 19:11 |
| khalednasr | lambday, yeah | 19:12 |
| lambday | khalednasr: you're a lifesavor | 19:12 |
| khalednasr | lambday, haha :) | 19:13 |
| lambday | khalednasr: if you add a new module, please make appropriate changes in the cmake also :) | 19:15 |
| khalednasr | lambday, sure | 19:15 |
| lambday | I'm psyched about getting feature/linalg merged with develop... | 19:15 |
| khalednasr | lambday, me too, can't wait to crash my computer's gpu with a huge neural net :D | 19:16 |
| lambday | khalednasr: haha... my poor laptop can't now run viennacl.. I am using pocl | 19:17 |
| lambday | it shows pocl error : unimplemented opencl method with your new stuffs :( | 19:17 |
| khalednasr | lambday, oh, what should we call the new module? I don't think Util is very suitable | 19:19 |
| lambday | I am the last person to ask about names :( | 19:19 |
| lambday | you suggest | 19:19 |
| lambday | Ideally we want general stuffs to be there... matrix matrix multiplications, etc | 19:19 |
| lambday | matrixfunctions? | 19:19 |
| lambday | nah | 19:20 |
| -!- rajul [~rajul@182.68.113.105] has quit [Ping timeout: 248 seconds] | 19:20 | |
| lambday | argh | 19:20 |
| khalednasr | I hate naming stuff too | 19:20 |
| lambday | I always turn to heiko about names :( | 19:21 |
| lambday | lisitsyn: SOS | 19:21 |
| khalednasr | maybe Core? or Basic? | 19:21 |
| lambday | basic yeha | 19:21 |
| lambday | core sounds good too | 19:21 |
| lambday | let us steal names from eigen3 | 19:21 |
| khalednasr | lambday, good idea :) | 19:22 |
| khalednasr | brb | 19:23 |
| lambday | oki | 19:24 |
| -!- lisitsyn [~qdrgsm@80.252.20.67] has quit [Ping timeout: 252 seconds] | 19:25 | |
| -!- soumyaC [uid15286@gateway/web/irccloud.com/x-aamotnusdfjjathv] has joined #shogun | 19:28 | |
| khalednasr | lambday, back | 19:36 |
| lambday | khalednasr: wb | 19:36 |
| khalednasr | lambday, seems like Eigen3 has that stuff in their Core module | 19:37 |
| lambday | khalednasr: do you know if we can use ccache with clang? | 19:37 |
| lambday | khalednasr: yeha... and many more things | 19:37 |
| lambday | core is literally the core module there :/ | 19:37 |
| khalednasr | lambday, nope, no idea | 19:37 |
| lambday | basic is cool | 19:40 |
| lambday | no advanced stuffs... just basic things :D | 19:40 |
| khalednasr | lambday, cool :) | 19:40 |
| lambday | khalednasr: later when you send the PR then maybe we can ask peers to suggest names | 19:41 |
| -!- pickle27 [~pickle27@192-0-136-118.cpe.teksavvy.com] has quit [Remote host closed the connection] | 19:42 | |
| khalednasr | lambday, yeah | 19:42 |
| -!- Jiaolong [9e6d09e4@gateway/web/freenode/ip.158.109.9.228] has quit [] | 19:53 | |
| -!- lambday [67157e4c@gateway/web/freenode/ip.103.21.126.76] has quit [Quit: Page closed] | 20:15 | |
| -!- lambday [67157f4f@gateway/web/freenode/ip.103.21.127.79] has joined #shogun | 20:15 | |
| -!- kislay [~Abhijeet@103.240.205.186] has quit [Quit: Leaving] | 20:49 | |
| -!- rajul [~rajul@182.68.160.237] has joined #shogun | 21:49 | |
| -!- jiaolong [9e6d1f01@gateway/web/freenode/ip.158.109.31.1] has joined #shogun | 22:24 | |
| -!- lisitsyn [~qdrgsm@80.252.20.67] has joined #shogun | 22:28 | |
| lisitsyn | oops | 22:28 |
| lisitsyn | lambday: khalednasr: I am back :D | 22:28 |
| lisitsyn | anything I can help with? | 22:28 |
| khalednasr | lisitsyn, hey :) | 22:38 |
| khalednasr | lisitsyn, we were talking about what we should name a module in linalg that contains matrix mulitplications, additions and elementwise operations, etc | 22:38 |
| lisitsyn | khalednasr: uhmm and what were suggestions? | 22:54 |
| khalednasr | lisitsyn, core, basic, util, got any others? | 22:55 |
| lisitsyn | doesn't sound util | 22:56 |
| lisitsyn | core seems to be ok | 22:56 |
| khalednasr | yeah, I agree | 22:57 |
| -!- lambday [67157f4f@gateway/web/freenode/ip.103.21.127.79] has quit [Ping timeout: 246 seconds] | 23:01 | |
| -!- rajul [~rajul@182.68.160.237] has quit [Quit: Leaving] | 23:13 | |
| -!- thoralf_ [~thoralf@91-65-140-81-dynip.superkabel.de] has joined #shogun | 23:15 | |
| thoralf_ | Heho. | 23:15 |
| --- Log closed Mon Jul 07 00:00:04 2014 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!