Open in new window / Try shogun cloud
--- Log opened Wed May 28 00:00:07 2014
-!- khalednasr [~k.nasr92@] has quit [Read error: Connection reset by peer]00:15
-!- jiaolong [9e6d09e4@gateway/web/freenode/ip.] has quit [Ping timeout: 240 seconds]00:15
-!- khalednasr [~k.nasr92@] has joined #shogun00:17
-!- khalednasr [~k.nasr92@] has quit [Read error: Connection reset by peer]00:26
-!- khalednasr [~k.nasr92@] has joined #shogun00:28
-!- thor|home [] has quit [Ping timeout: 240 seconds]00:32
-!- uzunov [80723d82@gateway/web/freenode/ip.] has joined #shogun01:29
uzunovhi there - I am getting a strange error when running the shogun toolkit through the R modular interface. The toolkit and the interface seemed to compile without issue, but then when I run one of the test examples(mkl_multiclass_modular.R), I get the following error on the very last line of the script (out <- lab$get_labels())01:33
-!- PirosB3 [] has joined #shogun01:33
uzunovError in validObject(.Object) :   invalid class "_p_shogun__SGVectorT_double_t" object: invalid object for slot "ref" in class "_p_shogun__SGVectorT_double_t": got class "numeric", should\   be or extend class "externalptr01:34
uzunovAnyone have an idea of what might be wrong? Maybe I did not install something properly? The rest of the script runs fine though. Any help is greatly appreaciated.01:35
-!- khalednasr [~k.nasr92@] has quit [Quit: Leaving]01:58
-!- PirosB3 [] has quit [Quit: PirosB3]02:21
shogun-buildbotbuild #14 of nightly_fedora is complete: Failure [failed git]  Build details are at  blamelist: Fernando Iglesias <>, Parijat Mazumdar <>03:00
-!- uzunov [80723d82@gateway/web/freenode/ip.] has quit [Quit: Page closed]03:02
-!- kislay [~Abhijeet@] has joined #shogun03:27
-!- kislay [~Abhijeet@] has quit [Ping timeout: 252 seconds]04:12
shogun-buildbotbuild #812 of nightly_default is complete: Failure [failed notebooks]  Build details are at  blamelist: Fernando Iglesias <>, Parijat Mazumdar <>04:36
-!- kislay [~Abhijeet@] has joined #shogun04:55
-!- kislay [~Abhijeet@] has quit [Read error: No route to host]05:16
-!- abinash_panda [75c8e0d7@gateway/web/freenode/ip.] has quit [Ping timeout: 240 seconds]05:26
-!- Saurabh7 [~Saurabh7@] has joined #shogun06:19
-!- Saurabh7 [~Saurabh7@] has quit [Ping timeout: 252 seconds]06:28
-!- Saurabh7 [~Saurabh7@] has joined #shogun06:45
-!- pickle27 [] has joined #shogun07:20
-!- kislay [~Abhijeet@] has joined #shogun07:27
-!- iglesiasg [] has joined #shogun07:29
-!- mode/#shogun [+o iglesiasg] by ChanServ07:29
-!- pickle27 [] has quit [Remote host closed the connection]07:30
Saurabh7hey iglesiasg07:34
@iglesiasghey Saurabh707:34
Saurabh7iglesiasg, i am having issues with the mailing list07:35
@iglesiasgSaurabh7, tell me, what issues?07:35
Saurabh7not recieving mails even though i am subscribed07:35
Saurabh7iglesiasg, i even tried setting the options again07:35
@iglesiasgSaurabh7, mmm, that's weird07:35
@iglesiasgSaurabh7, so, you didn't receive any of the GSoC updates?07:36
Saurabh7iglesiasg, no07:36
@iglesiasgSaurabh7, let me check for your user in the list of subscribers07:36
Saurabh7iglesiasg, i ended double posting too on new thread07:36
@iglesiasgSaurabh7, yeah, I saw that07:36
Saurabh7iglesiasg, thats because  didnt recieve the first thread07:37
Saurabh7iglesiasg, and even the archives didn't show that on gmane07:37
@iglesiasgSaurabh7, well, they do show now. Maybe there is just a delay in the archives?07:38
Saurabh7iglesiasg, yes but shouldn't i be recieving the mails?07:38
Saurabh7iglesiasg, i didn't get the meet up one too07:39
Saurabh7iglesiasg, i don't want to miss something else07:39
@iglesiasgSaurabh7, what address did you use to subscribe?07:40
kislayiglesiasg, me too. I never received a single mail till now. I keep checking the archives everyday..07:40
Saurabh7iglesiasg, my email : saurabh.mahindre@gmail.com07:40
@iglesiasgI see you are subscribed with this one07:41
@iglesiasgthe spelling looks correct07:41
Saurabh7iglesiasg, there seems to be some issue then07:41
@iglesiasgyou have a checkbox called ack with a tick, you seem to be the only one who has it07:41
Saurabh7err, what might that be ?!07:42
@iglesiasgSaurabh7, but I guess that is that you get an automatic response once you publish?07:42
Saurabh7iglesiasg, nope nothing, weird enough i don't get anything07:43
Saurabh7kislay, is also not getting it seems07:43
@iglesiasgSaurabh7, just in case -- I guess you checked already your spam folder07:44
Saurabh7iglesiasg, yup checked, no spam, no filters07:45
@iglesiasgSaurabh7, you have the digest mode on as well07:47
@iglesiasgSaurabh7, that would explain that you don't get every single mail once it is sent07:47
@iglesiasgSaurabh7, but you should get a collection of them at least daily07:47
Saurabh7iglesiasg, nothing from the list so far , not even 1 mail07:48
kislaysame here.07:48
Saurabh7iglesiasg, hope i am not irritating you :)07:48
@iglesiasgSaurabh7, it's ok07:48
Saurabh7iglesiasg, just wanted to inform you about it, maybe we can find a fix07:50
@iglesiasgkislay, what's the mail you subscribed with?07:51
kislayiglesiasg, yup07:52
kislayiglesiasg, exactly!07:52
@iglesiasgkislay, this is the one you are subscribed with07:52
@iglesiasgkislay, the spelling looks all right to me, but please double check07:52
kislayits correct07:53
@iglesiasgkislay, ok you also had the digest mode on07:53
@iglesiasgkislay, Saurabh7, so it might be related to this07:53
kislayiglesiasg, but no mails so for07:53
@iglesiasgkislay, Saurabh7, I just turned off the digest mode for both of you07:53
@iglesiasglet me send a test mail to the mailing list and hope you guys will get07:53
Saurabh7ok , !07:54
@iglesiasgSaurabh7, kislay let me know please07:55
Saurabh7iglesiasg, got it!07:56
kislayiglesiasg, awesome. got it07:57
Saurabh7iglesiasg, thanks for fixing it, and sorry for double posting :)07:58
@iglesiasghappy to hear it is fixed now :)07:59
-!- iglesiasg [] has quit [Quit: Leaving]08:14
-!- HeikoS [~heiko@] has joined #shogun08:32
-!- mode/#shogun [+o HeikoS] by ChanServ08:32
-!- parijat [75dc28f9@gateway/web/freenode/ip.] has joined #shogun08:40
-!- HeikoS [~heiko@] has quit [Quit: Leaving.]08:42
-!- Saurabh7 [~Saurabh7@] has quit [Remote host closed the connection]08:55
-!- zxtx_ [] has joined #shogun10:56
-!- zxtx [] has quit [Ping timeout: 258 seconds]10:57
-!- aboutGod [] has joined #shogun11:14
-!- HeikoS [~heiko@] has joined #shogun11:17
-!- mode/#shogun [+o HeikoS] by ChanServ11:17
-!- aboutGod [] has left #shogun []11:19
-!- HeikoS [~heiko@] has quit [Quit: Leaving.]11:26
-!- HeikoS [~heiko@] has joined #shogun11:29
-!- mode/#shogun [+o HeikoS] by ChanServ11:29
-!- HeikoS [~heiko@] has quit [Client Quit]11:31
-!- jiaolong [9e6d09e4@gateway/web/freenode/ip.] has joined #shogun11:41
-!- PirosB3 [] has joined #shogun11:51
-!- PirosB3 [] has quit [Quit: PirosB3]13:36
-!- HeikoS [~heiko@] has joined #shogun13:39
-!- mode/#shogun [+o HeikoS] by ChanServ13:39
-!- parijat [75dc28f9@gateway/web/freenode/ip.] has quit [Quit: Page closed]14:05
-!- HeikoS [~heiko@] has quit [Ping timeout: 252 seconds]14:14
-!- parijat [75dc28f9@gateway/web/freenode/ip.] has joined #shogun14:32
-!- lambday [67157d4c@gateway/web/freenode/ip.] has joined #shogun14:45
-!- PirosB3 [] has joined #shogun14:48
-!- Saurabh7 [~Saurabh7@] has joined #shogun15:22
-!- khalednasr [~k.nasr92@] has joined #shogun15:23
-!- parijat [75dc28f9@gateway/web/freenode/ip.] has quit [Ping timeout: 240 seconds]15:23
-!- pickle27 [] has joined #shogun15:53
-!- pickle27 [] has quit [Remote host closed the connection]15:58
-!- pickle27f [~pickle27@] has joined #shogun16:17
-!- Saurabh7 [~Saurabh7@] has quit [Read error: Connection reset by peer]16:33
-!- Saurabh7 [~Saurabh7@] has joined #shogun16:45
-!- khalednasr [~k.nasr92@] has quit [Quit: Leaving]16:46
-!- khalednasr [~k.nasr92@] has joined #shogun17:02
khalednasrlambday, hey17:09
lambdaykhalednasr: hey17:09
lambdaykhalednasr: how's it going?17:10
khalednasrlambday, good, you?17:10
lambdaykhalednasr: just started with vector-sum thingi17:11
lambdaykhalednasr: you know if there's a method in viennacl that computes sums of values?17:11
lambdaysum of all elements in vector I mean17:11
khalednasrlambday, nope, they don't have a method for that :(17:12
lambdaykhalednasr: I was afraid so... found till hell without any luck17:12
lambdayI'm using std::accumulate17:12
khalednasrso you copy stuff back to the cpu?17:13
lambdaykhalednasr: umm.. not sure what to do.. may be you have an idea? this won't work when vector loaded on gpu?17:14
khalednasrlambday, it can be done with an custom opencl kernel17:15
lambdaykhalednasr: alright! that's something I don't have any idea -17:16
lambdaykhalednasr: so I'm just adding eigen3 here... may be you can add viennacl version when you have time/when you'll need it?17:16
khalednasrlambday, yeah, I'm gonna need to implement it later on anyway17:16
lambdaykhalednasr: cool!17:16
khalednasrlambday, there's a problem that I ran into when I tried to do what we talked about last time17:17
lambdaykhalednasr: what is it?17:17
khalednasrlambday, some compile error that I can't figure out17:18
lambdaykhalednasr: checking17:18
lambdaykhalednasr: argh - this is because in the specialization we're using a matrix type that cannot be matched with the generic one17:19
lambdayextra 2nd template arg -17:19
lambdaycrap man17:19
khalednasrit compiles as long as you don't try to call the function17:20
lambdaykhalednasr: the only way I can imaigne is to use a proxy type17:21
lambdaymay be because some crazy compiler optimization thing17:21
khalednasrlambday, that's why I thought it worked, when we talked last time17:22
lambdaykhalednasr: ah I see17:22
lambdaykhalednasr: let me play with this a bit17:22
khalednasrlambday, have fun :)17:22
lambdaykhalednasr: but unless we use a fake type I don't see any other way we can handle it17:23
khalednasrlambday, yeah probably17:23
khalednasrlambday, we could write a SGMatrix-style wrapper around it17:23
lambdaykhalednasr: no but wait! I misunderstood something17:24
lambdaythis *should* work17:24
lambdayyou're using VCLMatrix17:24
lambdaywhich has 2 template args17:25
khalednasrlisitsyn, the gpu stuff is gonna take more time than I thought it would..17:30
lisitsynkhalednasr: that's ok deadlines are for n00bs :D17:31
lisitsynkhalednasr: what's in particular causing struggles?17:31
khalednasrlisitsyn, :D17:31
khalednasrlisitsyn, viennacl matrices don't fit with the linalg library17:32
khalednasrlisitsyn, they have different template arguments than SGMatrix and eigen3 matrix17:32
lisitsyndifferent like?17:33
lisitsynkhalednasr: ahh you mean they have different number of template arguments?17:33
khalednasrtemplate <class T, class F, unsigned int>17:33
lisitsynwhat's F for btw?17:34
khalednasrrow major vs column major17:34
lisitsynI see17:34
lambdaylisitsyn: we managed this thing for vector with template <class Info, template<class,Info...> class Vector, class T, Info... info> if you remember17:34
lisitsynlambday: yeah17:34
lambdaylisitsyn: this fails for matrix for their f**king different template args for matrices :(17:35
lambdaylisitsyn: so khalednasr suggested that we should have two versions for row-majors and col-majors and fix the type17:35
lisitsynlike a wrapper17:36
lisitsynthat fixes the value17:36
lisitsyne.g. as row major17:36
khalednasrlisitsyn, yup,
lambdaylisitsyn: pretty much17:36
lisitsynah these template aliases17:37
lisitsynI barely understand consequencies but it looks ok17:38
lisitsyndo you see any drawbacks?17:38
khalednasrexcept that it doesn't  work when you try to call the method :D17:38
khalednasrit compiles as long as you don't try to use it17:38
khalednasrlisitsyn, anyway, I started working on conv. nets, in parallel with the gpu stuff17:40
khalednasrlisitsyn, gonna send a PR with some basic methods later today17:40
lisitsynkhalednasr: yeah sounds good17:40
lisitsynkhalednasr: lambday: I could try to come online a bit later to discuss it a bit more17:41
lisitsynmay be we could come up with some solution17:41
lambdaylisitsyn: yep! since linalg is in its infancy, its easier for now to change things17:41
-!- pickle27f [~pickle27@] has quit [Quit: This computer has gone to sleep]18:16
-!- lambday is now known as lambday|afk18:20
-!- pickle27f [~pickle27@] has joined #shogun18:27
-!- PirosB3 [] has quit [Quit: PirosB3]18:45
-!- PirosB3 [] has joined #shogun18:47
-!- Saurabh7 [~Saurabh7@] has quit [Quit: Leaving]19:04
-!- kislay [~Abhijeet@] has quit [Ping timeout: 264 seconds]19:24
-!- naywhayare [] has quit [Remote host closed the connection]19:43
-!- PirosB3 [] has quit [Quit: PirosB3]20:02
-!- iglesiasg [] has joined #shogun20:25
-!- mode/#shogun [+o iglesiasg] by ChanServ20:25
-!- kislay [~Abhijeet@] has joined #shogun20:33
-!- lambday|afk is now known as lambday20:46
-!- iglesiasg [] has quit [Ping timeout: 240 seconds]21:18
-!- pickle27f [~pickle27@] has quit [Quit: This computer has gone to sleep]21:29
-!- pickle27f [~pickle27@] has joined #shogun21:31
-!- kislay [~Abhijeet@] has quit [Ping timeout: 265 seconds]23:01
-!- PirosB3 [] has joined #shogun23:21
-!- pickle27f [~pickle27@] has quit [Quit: This computer has gone to sleep]23:32
-!- travis-ci [] has joined #shogun23:53
travis-ci[travis-ci] it's khalednasr's turn to pay the next round of drinks for the massacre he caused in khalednasr/shogun:
-!- travis-ci [] has left #shogun []23:53
--- Log closed Thu May 29 00:00:08 2014