Open in new window / Try shogun cloud
--- Log opened Thu Jun 27 00:00:06 2013
shogun-buildbotbuild #1271 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at
-!- travis-ci [] has joined #shogun00:27
travis-ci[travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun:
-!- travis-ci [] has left #shogun []00:27
shogun-buildbotbuild #1272 of deb3 - modular_interfaces is complete: Failure [failed test python_modular]  Build details are at  blamelist: van51 <>00:38
-!- iglesiasg [] has quit [Quit: leaving]01:01
-!- van51 [] has quit [Quit: Leaving.]01:51
-!- FSCV [~FSCV@] has quit [Quit: Leaving]02:14
-!- shogun-notifier- [] has quit [Quit: transmission timeout]02:23
-!- nube [~rho@] has quit [Quit: Leaving.]04:51
-!- foulwall [~user@2001:da8:215:6100:bc02:b666:46de:6cba] has joined #shogun05:11
-!- foulwall [~user@2001:da8:215:6100:bc02:b666:46de:6cba] has quit [Ping timeout: 264 seconds]05:24
-!- nube [~rho@] has joined #shogun05:41
-!- foulwall [~user@2001:da8:215:6100:b8bb:f41d:16ab:399b] has joined #shogun06:17
-!- nube1 [~rho@] has joined #shogun06:17
-!- nube [~rho@] has quit [Read error: Connection reset by peer]06:17
-!- foulwall [~user@2001:da8:215:6100:b8bb:f41d:16ab:399b] has quit [Ping timeout: 264 seconds]06:40
-!- zxtx [] has joined #shogun07:14
-!- vgorbati [c3ee5cb1@gateway/web/freenode/ip.] has joined #shogun09:05
-!- Netsplit *.net <-> *.split quits: sanyam, nube109:35
-!- Netsplit over, joins: nube1, sanyam09:37
-!- foulwall [~user@2001:da8:215:6100:6c9c:8f7a:eb03:b755] has joined #shogun09:53
foulwallping sonney2k09:57
-!- iglesiasg [] has joined #shogun10:08
-!- mode/#shogun [+o iglesiasg] by ChanServ10:08
@iglesiasgsonne|work: is the admin interface looking normal in your browser?10:10
@iglesiasgthe login and first page look without the common style of django admin10:11
-!- nube1 [~rho@] has quit [Ping timeout: 246 seconds]10:16
sonne|workfoulwall: hey there sec10:25
sonne|workiglesiasg: yes all good10:25
@iglesiasgprogram updated
@wikingdoh :)10:27
@wikingsonne|work: i did not get a slot!10:27
foulwallhey sonne|work, I rewrote a lot of code, I almost finished the js refractor, and have made a new toy data generator, and changed a bit of demo style... so I spent a bit more time on debugging...10:28
-!- nube [~rho@] has joined #shogun10:29
@iglesiasgwiking: maybe we can squeeze a couple of slots and get one more with half an hour10:30
@wikingi'd be great10:32
@iglesiasgI think half an hour for my talk could be ok too10:33
@iglesiasglet's ask sergey too about it10:34
-!- nube [~rho@] has quit [Client Quit]10:34
-!- hushell [] has joined #shogun10:44
@iglesiasghey hushell10:44
@iglesiasghow are you?10:44
hushellhey iglesiasg10:44
hushellI am good10:44
hushellHow are you10:44
hushellDo you have any idea: ImportError: cannot import name SerializableJsonFile10:45
@iglesiasgI am fine too10:45
@iglesiasgI don't know why it happens10:45
hushellFailed at that example: error in serialization_complex_example.py10:45
@iglesiasglisitsyn mentioned that sonne|work had modified something about json10:45
@iglesiasgmaybe it is something related to that10:46
hushellC interfaces are okay10:46
@iglesiasghushell: it is not your PR's fault I think10:46
hushellokay. I always leave python check for Travis :)10:46
hushellC# modular got problem too...10:48
-!- lisitsyn [] has joined #shogun11:12
@iglesiasghushell: what is gram matrix? I have quite limited knowledge regarding kernels :)11:12
@iglesiasghushell: is it the same askernel matrix?11:12
@iglesiasghi lisitsyn, so what do you think about making our presentations in the workshop shorter and squeezing one more for wiking in the schedule?11:14
@iglesiasgright now we have 2 of 45 min, let's make 3 of 30min?11:14
-!- lambday [67157c4c@gateway/web/freenode/ip.] has joined #shogun11:15
hushelliglesiasg: just the kernel matrix :)11:15
lambdaysonne|work: morning :)11:15
hushelliglesiasg: some people use that name, so I learned from them11:16
@iglesiasghushell: good to know!11:17
sonne|workwiking: what do you want to talk about?11:17
hushelliglesiasg: Are you going to present something in the workshop?11:17
sonne|worklambday: moin moin!11:18
@iglesiasghushell: yeah! I will talk about structured learning in Shogun11:18
sonne|workfoulwall: I am not too happy about doing the toy data generation in the .js11:18
-!- nube [~rho@] has joined #shogun11:18
lisitsyniglesiasg: hmm I don't mind11:18
sonne|workfoulwall: I know it is a lot faster to run on the client but we have some datagenerator class that we might extend to have the data generation in one place11:19
hushellsonne|work: Are you going to record the talks?11:19
@iglesiasghushell: let me know if you have any suggestion ;)11:19
@iglesiasgthat would be very nice indeed, to record them11:19
hushellYeah, I'd like to watch you guys online :)11:19
sonne|workfoulwall: and then can use it also for examples11:20
sonne|workhushell: I wish we had but we don't have equipment11:21
hushellsonne|work: any webcam works :)11:22
hushellmaybe we can have some unofficial videos11:22
lisitsynsonne|work: what equipment do we need?11:24
sonne|worksome camera to do video recording of the whole workshop11:25
sonne|workas in talks /11:25
sonne|workfoulwall: do you understand why it is better to have the data generators in python not .js ?11:26
sonne|worklisitsyn: wiking also wants to give a talk11:26
foulwallsonne|work: ok, I'll change that back to server, the generator is now modular, so it's easy to do that11:26
sonne|worklisitsyn: luckily only us want to talk :D11:26
sonne|workfoulwall: very good!11:27
foulwallsonne|work: why we need do that in python11:27
sonne|workfoulwall: for now we do it in python but later we do the data generation via shogun's DataGenerator class11:28
sonne|workfoulwall: the reason is that we can have stand-a-lone examples to run on the same data11:28
lisitsynsonne|work: well I will be there with some camera for sure but it won't last that long I guess11:28
foulwallsonne|work: gotcha11:29
@iglesiasghushell: so what do you think about the features issue?11:34
@iglesiasghushell: apart from that, from my part the pull request is ready11:35
hushelliglesiasg: I think we have had a m_models in machine, so now we can access m_features by m_models11:36
@iglesiasghushell: yes11:36
hushellbut you are right, we finally will move m_features to LinearStructMachine11:37
hushellDo you think we need keep m_features in LinearSOMachine right now?11:37
@iglesiasgwell not really need11:37
hushellBecause currently it will not be used anywhere11:38
@iglesiasgbut my point is, why to remove it now if we know we are going to get it back?11:38
@iglesiasgok, I understand that too11:38
hushellI saw it was in LinearSOMachine but didn't registered11:38
@iglesiasgso it is up to you, remove it if you prefer that for the moment :) I am fine with that too11:38
@iglesiasghushell: I probably forgot to do register it :(11:39
hushellIn shogun, for a NULL pointer, do we need to register it?11:39
hushellI mean always NULL11:39
-!- foulwall [~user@2001:da8:215:6100:6c9c:8f7a:eb03:b755] has quit [Ping timeout: 264 seconds]11:39
@iglesiasghushell: when you register a variable, you register the variable, not its value11:40
hushelliglesiasg: That's ok. We have so many things here :)11:40
hushelliglesiasg: okay. I was thinking this registration is kind of memory management11:40
@iglesiasghushell: I am not actually aware of everything the registration entails11:41
@iglesiasgI know part of it is related to model selection and cross validation11:42
@iglesiasgbut most of the init methods (or register parameters) happen before the variables have even been given a value11:42
hushelliglesiasg: maybe not keep this pointer in LinearSOMachine right now, we will refine this after the GSOC or in the end :)11:43
@iglesiasghushell: all right, as you prefer11:43
hushelliglesiasg: Travis failed again :(11:43
hushelliglesiasg: Good to know more about Shogun!11:44
hushelliglesiasg: I am going to sleep now, let me know if there are other problems about the PR11:47
@iglesiasghushell: in python modular it is the SerializableJsonFile11:47
@iglesiasgand in C# I have no idea11:47
@iglesiasgpython modular is not your PR's fault11:48
hushellso every merge need to make sure no errors in Travis?11:48
@iglesiasghushell: ideally yes11:48
@iglesiasgbut your PR is ready now11:48
@iglesiasgI am going to merge it soon11:48
@iglesiasgI just want to know if the C# fail is due to your PR or not11:48
hushellmaybe wait other PR merges11:49
hushellone thing about git11:49
hushellonce I fetch from upstream, I just need to run git rebase upstream/develop, right?11:50
hushellokay, then I think I did right thing11:50
@iglesiasgthat is preferable to doing pull as we avoid a commit11:50
@iglesiasghushell: so after this PR you are ready to be coding the new stuff right?11:51
hushellas you suggested, I use --amend11:51
hushelliglesiasg: Actually I am coding new stuffs11:51
@iglesiasgthe factor graph I think it was the first thing to do11:51
hushellthey are in parallel11:51
hushellbut I want to know more about the SSVM in Shogun, so I choose to do this first11:52
@iglesiasgI understand11:52
hushellyes, factor graph should be new stuff to Shogun11:53
@iglesiasgI am afraid tha the C# fail might actually be due to your PR11:55
hushellThe OpenGM guys got a CVPR paper about their graphical model lib, I asked them about the factor graph implementation11:56
hushelllet me check11:56
hushellIs it because I didn't update the Swig for C#?11:59
@iglesiasgI have no idea11:59
hushelleverything about C# in src/interfaces/csharp_modular?12:01
@iglesiasgI think that is mostly autogenerated12:01
@iglesiasgthe actual SWIG code is in src/interfaces/modular12:01
@iglesiasgbut you did not create any new class so there should be no need to update SWIG12:02
hushellthat's a bit strange12:03
@iglesiasgthe only thing I come up with is that maybe any of the new includes has interfered12:03
hushelllet me test C# tomorrow locally then we think about the merge again12:04
@iglesiasghushell: why did you change the position of the includes here btw?12:04
@iglesiasghushell: ok, have a rest now :)12:04
hushellthe define __H__ should be before include files, isn't it?12:05
hushellotherwise the macro trick would not make sense12:06
@iglesiasgI don't know, how is it done in the other files? :)12:06
@iglesiasgit makes sense what you are saying though12:06
@iglesiasgyeah ok that should be right12:06
hushell#ifndef _STRUCTURED_MODEL__H__                                                                                         ||12:06
hushell#define _STRUCTURED_MODEL__H__                                                                                         ||-   struct12:06
hushell                                                                                                                       |||     TMultipleCPinfo [shogun12:06
hushell#include <shogun/base/SGObject.h>                                                                                      |||     CResultSet [shogun]12:07
hushell#include <shogun/features/Features.h>                                                                                  ||12:07
hushell#include <shogun/labels/StructuredLabels.h>12:07
hushellsorry, I copy from vi directly12:07
hushellthis is an example from StructuredModel12:07
hushellokay, I'll test the csharp thing tomorrow and see what's wrong over there12:08
@iglesiasgsounds good12:08
hushellI only test cmd and python_modular currently on my computer12:08
@iglesiasgyeah, me too12:08
@iglesiasgI check with travis and the buildbot the others hehe12:09
hushellthis is the best part of travis :D12:09
hushellokay see you later12:09
@iglesiasgbye, and very nice job!12:09
@iglesiasgI like how the StructuredModel is getting12:09
@iglesiasgrather clean12:09
hushellvery limited contribution indeed :( but I am happy to join you guys! Have a nice day12:10
-!- foulwall [~user@2001:da8:215:6100:d93a:3d1:5210:fbf] has joined #shogun12:17
-!- lambday [67157c4c@gateway/web/freenode/ip.] has quit [Ping timeout: 250 seconds]12:23
-!- HeikoS [] has joined #shogun12:35
-!- mode/#shogun [+o HeikoS] by ChanServ12:35
-!- lambday [67157f4c@gateway/web/freenode/ip.] has joined #shogun12:38
lambdayHeikoS: hi12:38
@HeikoSlambday:  hi!12:39
@HeikoSsaw your mail, sounds reasonable12:39
lambdayHeikoS: yes, I guess makes things more easy12:39
@HeikoSyep totally12:39
lambdaywe can use lanczos later then12:39
lambdaysending the PR12:39
lisitsynI heard someone is solving eigenproblems here12:40
lambdaylisitsyn: we we! :D12:42
@HeikoSlambday: always ask the master on those :)12:43
@HeikoSaka lisitsyn :)12:43
lisitsynI don't know anything but I worked on that a bit12:43
lisitsynHeikoS: I am thinking how can we re-use the things I have in tapkee12:44
@HeikoSlisitsyn: we will need to solve gigantic eigenvalue problems at some point12:44
@HeikoSbut only largest/smallest12:44
@HeikoSsparse matrices12:44
@HeikoSwith horrible horrible condition numbers12:44
@HeikoSso we need something stable (priority over speed)12:45
lisitsynHeikoS: not arpack then12:45
lisitsynbut you still need something with reverse interface12:45
lisitsynor specialized on sparse12:45
lisitsynHeikoS: this one could be more numericall stable12:46
lisitsynHeikoS: I am thinking if I can extract eigen things from tapkee to separate lib..12:47
lisitsynthen we can work together on that12:47
@HeikoSlisitsyn: ok would be good12:47
@HeikoSI can provide a matrix to test it12:47
lisitsynHeikoS: I mean I already did some things so this should not be lost12:48
@HeikoSlisitsyn, lambday its very important that all the eigensolvers are based on linear operators that are applied to vectors rather than storing matrices btw12:48
lisitsynbut using dimension reduction library from some other code is wrong12:48
lambdayHeikoS: yes12:48
lisitsynHeikoS: well that's arpack and feast12:48
lisitsynI am not sure I know other solvers with reverse interface12:49
lisitsynHeikoS: MKL uses FEAST btw12:50
lisitsynstarting from this year february12:50
@HeikoSno idea about these things :)12:51
@HeikoSmatlab fails on the matrices I have here btw12:51
@HeikoSI want things up to machine precision >:-|12:51
lisitsynHeikoS: but you know what MKL is, right?12:51
@HeikoSlisitsyn:  not really, some sparsity stuff right? :D12:52
lisitsynHeikoS: no, huge intel librar12:52
lisitsynthat does everything12:52
@HeikoSah this one12:52
@HeikoSno I don tknow that12:52
lambdayI should checkout FEAST then.. :-/12:53
lambdaylisitsyn: what do you use in tapkee?12:53
lambdayHeikoS: I sent the PR, please have a look12:53
lisitsynlambday: arpack, randomized and dense method from eigen312:54
lisitsynrandomized is out :D12:55
lisitsyndense is too12:55
lambdayyeah :(12:55
lambdaywe want sparse :(12:55
lambdayFEAST has support for that?12:55
lisitsynlambday: arpack supports sparse12:56
lisitsynfeast too12:56
@HeikoSlambday: why do you add all those debug messages to the constructors? do you have memory leaks often=?13:00
@HeikoSlambday: you can compile with --enable-trace-mallocs then shogun will print all allocated objects at the end of a program so its easy to track which ones are still in mem13:01
lambdayumm.. initially I had when I was trying to get SG_REF/UNREF.. now I just copy paste :D13:01
lambdayI thought of removing all of them when the whole code is ready13:01
@HeikoSthey dont hurt us just wondering :)13:01
@HeikoSnono leave them its fine13:01
@HeikoSjust wonderiung13:01
lambdayalright, I'll try enable-trace-mallocs13:02
@HeikoSlambday: hey I just read about this indian mathematician that went to england before WW2, do you know about that guy?13:02
lambdayumm... no! :(13:02
lambdayI mean13:02
@HeikoSawesome guy!13:02
lambdayyeah !!13:02
lambdayhe's a genius!13:02
lambdayabsolute genius! I heard stories that when he was on his deathbed, someone came by a car to meet him... he looked up the window.. when that guy entered, he told him that your car plate number is a multiple of two huge prime numbers13:03
lambdayand then he told what those prime numbers are13:04
lambday(not sure how far these stories are true anyway)13:04
@iglesiasgit is a pity he passed away so young :(13:05
lambdayyes :(13:05
@HeikoSlambday: yeah13:05
@HeikoSlambday: currently reading a book on prime numbers where he had a chapter13:06
@HeikoSlambday: I think storing the eigenvalues in the class is a bit strange if you provide a matrix in the compute method13:07
@HeikoSlambday: either make the class store the matrix (and the eigenvalues) or return them and dont store them13:07
@HeikoSlambday: you can do this via references13:07
@HeikoSlisitsyn: do we have typemaps for std::pair?13:07
lambdayyes I thought of that13:07
lambdayokay then I'll change it13:08
@HeikoSlambday: ok13:08
@HeikoSand please do some unit tests for the non full rank systems13:08
@HeikoSnote that LLT can solve over and underdetermined systems but not those with negative eigenvalues13:08
lambdayI was thinking of adding a check in the linear solver I added (for eigenvalues), but then thought it will be too slow :-/13:09
@HeikoSlambday: man if you continue in that pace the project will be done in 2 weeks :)13:09
@HeikoSlambday: LLT class should do this in fact13:09
@HeikoStry it with a non-psd matrix13:10
@HeikoSyou can probably ask it whether the matrix is fine13:10
lambdayHeikoS: haha no don't worry my thesis adviser has threatened me to kick me out (because I haven't been doing any of his work since last 2 weeks), so I'll be a bit slow for sometime :(13:10
@HeikoSlambday: careful then! uni is more important than GSoC in the long run13:11
@HeikoSlambday: also you work very fast, so dont worry to slow down a bit13:11
lambdayHeikoS: LLT class :-/13:11
lambdayHeikoS: haha yes..13:11
@HeikoSalthough I really like that things are going so well!13:11
@HeikoSlambday: looks like it doesnt catch that13:12
lambdayHeikoS: you're the mastermind.. I just execute.. I too am really happy with how this is going too :) so, I didn't get much depressed with what thesis adviser told me (I was supposed to be more depressed)13:12
@HeikoSlambday: then its fine, user sets the solver and if user provides wrong matrix its his problem13:13
lambdayHeikoS: yes that's what I thought13:13
@HeikoSthe std solver should be able to solve everything though, so probably QR/SVD13:13
lambdayuser shouldn't call LLT for non psd matrices13:13
@HeikoSlambday: yep LLT should be a conscious decision13:13
@HeikoSlambday: there is ComputationInfo info() though13:14
@HeikoSyou could check this in any case13:14
lambdayin Eigen3?13:14
@HeikoSlambday: its a method of LLT13:14
lambdayokay.. yes I'll check it13:15
@HeikoSonly check for Success13:15
@HeikoSor maybe just print error type in case of no success13:15
lambdayah, that would help me print an error msg if user mistakenly calls LLT on a nonpsd matrices13:15
@HeikoSthere are 3 cases13:15
lambdayyes yes!13:15
@HeikoSlambday: ok gotta do some work now, will be back later in the day13:16
lambdayHeikoS: alright.. I'll check it13:16
lambdaysee you :)13:16
@HeikoSlambday: Ill keep an eye on PRs though13:16
@HeikoSbut I have to switch off irc, otherwise I wont work :D13:16
lambdayokay :)13:16
@HeikoSso see you13:16
-!- HeikoS [] has quit [Quit: Leaving.]13:16
-!- nube [~rho@] has quit [Quit: Leaving.]13:17
-!- nube [~rho@] has joined #shogun13:17
-!- gsomix [~Miranda@] has joined #shogun13:45
gsomix*ba dum tss*13:45
gsomixpain is gone13:47
gsomixexams are passed :)13:47
@iglesiasgcongrats gsomix13:47
sonne|workgsomix: hurray!13:51
-!- nube [~rho@] has quit [Quit: Leaving.]13:56
-!- sonne|work [] has quit [Quit: Leaving.]14:11
-!- sonne|work [] has joined #shogun14:11
-!- nube [~rho@] has joined #shogun15:01
-!- lambday [67157f4c@gateway/web/freenode/ip.] has quit []15:06
-!- thoralf [] has joined #shogun15:31
thoralfHello everybody.15:31
sonne|workhello hello15:33
-!- nube [~rho@] has quit [Ping timeout: 264 seconds]15:47
-!- zxtx [] has quit [Ping timeout: 256 seconds]15:51
-!- foulwall [~user@2001:da8:215:6100:d93a:3d1:5210:fbf] has quit [Ping timeout: 264 seconds]16:05
-!- pickle27 [] has joined #shogun16:11
-!- gsomix [~Miranda@] has quit [Quit: Miranda IM! Smaller, Faster, Easier.]16:25
-!- FSCV [~FSCV@] has joined #shogun16:26
-!- nube [~rho@] has joined #shogun16:38
-!- FSCV_ [~FSCV@] has joined #shogun16:39
-!- iglesiasg [] has quit [Quit: leaving]16:39
-!- FSCV [~FSCV@] has quit [Ping timeout: 252 seconds]16:41
-!- vgorbati [c3ee5cb1@gateway/web/freenode/ip.] has quit [Quit: Page closed]16:57
@wikinglisitsyn: ping17:06
@wikingsonne|work: pong17:06
@wikingsonne|work: latent (so) solvers17:10
@wikingsonne|work: and maybe about pending design issues of shogun ;P17:11
-!- nube [~rho@] has quit [Ping timeout: 248 seconds]17:13
-!- lisitsyn [] has quit [Quit: Leaving.]17:13
-!- shogun-notifier- [] has joined #shogun17:17
shogun-notifier-shogun: lambday :develop * a7eccd9 / / (5 files):
shogun-notifier-shogun: EigenSolver added, implementation using Eigen3 added17:17
shogun-notifier-shogun: lambday :develop * 8333c76 / / (6 files):
shogun-notifier-shogun: minor changes in EigenSolver, LinearSolver17:17
shogun-notifier-shogun: Heiko Strathmann :develop * faed169 / / (6 files):
shogun-notifier-shogun: Merge pull request #1192 from lambday/feature/log_determinant17:17
shogun-notifier-shogun: EigenSolver added, implementation using Eigen3 added17:17
-!- lambday [67157d4c@gateway/web/freenode/ip.] has joined #shogun17:22
shogun-buildbotbuild #983 of cyg1 - libshogun is complete: Failure [failed configure]  Build details are at  blamelist: Heiko Strathmann <>, lambday <>17:24
-!- foulwall [~user@2001:da8:215:c252:d10:e269:ac4b:bc2f] has joined #shogun17:30
shogun-buildbotbuild #1151 of bsd1 - libshogun is complete: Failure [failed test_1]  Build details are at  blamelist: Heiko Strathmann <>, lambday <>17:31
shogun-buildbotbuild #1152 of bsd1 - libshogun is complete: Failure [failed test_1]  Build details are at  blamelist: lambday <>17:42
-!- lambday [67157d4c@gateway/web/freenode/ip.] has quit []17:43
-!- travis-ci [] has joined #shogun17:44
travis-ci[travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun:
-!- travis-ci [] has left #shogun []17:44
shogun-buildbotbuild #984 of cyg1 - libshogun is complete: Failure [failed compile]  Build details are at  blamelist: lambday <>17:46
shogun-buildbotbuild #1273 of deb3 - modular_interfaces is complete: Success [build successful]  Build details are at
-!- pickle27 [] has quit [Remote host closed the connection]18:28
-!- pickle27 [] has joined #shogun18:30
shogun-buildbotbuild #1274 of deb3 - modular_interfaces is complete: Failure [failed test python_modular]  Build details are at  blamelist: lambday <>18:33
-!- foulwall [~user@2001:da8:215:c252:d10:e269:ac4b:bc2f] has quit [Ping timeout: 245 seconds]19:01
-!- foulwall [~user@2001:da8:215:c252:9a3:d4b7:6d1e:6ecc] has joined #shogun19:58
-!- foulwall [~user@2001:da8:215:c252:9a3:d4b7:6d1e:6ecc] has quit [Ping timeout: 264 seconds]20:04
-!- hushell [] has quit [Ping timeout: 248 seconds]20:05
-!- lisitsyn [~lisitsyn@] has joined #shogun20:12
lisitsynwiking: pong20:12
-!- shogun-notifier- [] has quit [Quit: transmission timeout]20:17
-!- HeikoS [] has joined #shogun20:18
-!- mode/#shogun [+o HeikoS] by ChanServ20:18
@HeikoSsonney2k: around?20:18
@HeikoSsonne|work: around?20:23
@HeikoSlets do the PO tomorrow then, going home20:29
@HeikoSbye all20:30
-!- HeikoS [] has quit [Quit: Leaving.]20:30
-!- gsomix [~Miranda@] has joined #shogun20:41
-!- hushell [~hushell@] has joined #shogun20:46
-!- hushell [~hushell@] has quit [Ping timeout: 276 seconds]21:03
-!- van51 [] has joined #shogun21:25
-!- gsomix [~Miranda@] has quit [Quit: Miranda IM! Smaller, Faster, Easier.]21:53
lisitsynpickle27: hey!22:30
pickle27lisitsyn: hey!22:30
lisitsynpickle27: how is it going?22:30
pickle27lisitsyn: good! I finished my code that tests jadiag vs the R version22:31
pickle27and they matched!22:31
pickle27today I have been working on ffdiag and I think I am getting close to being done22:31
lisitsynpickle27: cool!22:31
lisitsynmore time to get some examples done22:32
pickle27lisitsyn: yeah!22:32
pickle27lisitsyn: I think I am going to try and get a BSS example going after ffdiag and before porting uwedge22:33
lisitsynpickle27: what example?22:33
pickle27lisitsyn: for now I just want it to work on the scikit-learn example data22:33
pickle27I haven't got to it yet but it looks like ADJ is just one of the step for BSS so I think theres more to do22:34
pickle27and I want to get started looking at that part22:35
lisitsynpickle27: I like that mixed signal thing22:38
pickle27lisitsyn: I have the example ported to C++ I just need to finish the methods and run them on it22:39
lisitsynpickle27: ohh you don't like easy ways ;)22:39
pickle27lisitsyn: oh like running the whole thing in python?22:40
lisitsynyeah would be easier for you22:40
pickle27lisitsyn: hah true that would have been easier22:40
pickle27lisitsyn: although it'll be easier to debug now that its all c++22:40
lisitsynpickle27: same thing if you are using gdb22:41
lisitsynpickle27: ah btw try cgdb22:41
pickle27lisitsyn: yeah Im not that great with gdb I tend to just use printing and cin.get for debugginh22:42
lisitsynpickle27: oh that's mostly gdb that is not nice enough22:42
lisitsynI mean it is rather difficult to see a matrix there22:43
pickle27I'll have to check out cgdb22:43
lisitsynpickle27: it is a editor (or text viewer) combined with gdb22:44
lisitsynso you may put breakpoints22:45
@sonney2kvan51, how is it going?22:45
lisitsynusing just spacebar22:45
pickle27cool yeah I've used gdb a few times22:45
van51sonney2k: it's been a slow day today tbh :$22:46
van51sonney2k: now I'm converting the densematrix to a sparse representation like you commente22:46
van51and after that I'm thinking of starting to see the streaming features22:46
van51btw, is there a specific way I should expect a document collection to be represented?22:47
-!- iglesiasg [] has joined #shogun22:56
-!- mode/#shogun [+o iglesiasg] by ChanServ22:56
-!- FSCV_ [~FSCV@] has quit [Quit: This computer has gone to sleep]22:58
lisitsynpickle27: yes23:00
pickle27lisitsyn: does memory from memcpy need to be free'd manually?23:00
lisitsynpickle27: what memory?23:01
pickle27I've got a seg fault that throws after my code runs and this memcpy I've got is the only thing I suspect23:01
pickle27its from one SGNDArray to another23:01
van51does a SGSparseMatrix expect its SGSparseVectors to have the same size?23:01
lisitsynpickle27: hmm still not clear23:02
van51because I see it can take a num_feat as a parameter, but it's not documented on its meaning23:02
pickle27this is the code I have:23:03
van51does it mean the size of the vectors or the size of the dimension they represent?23:03
pickle27SGNDArray<float64_t> C(C0.dims,3);23:04
pickle27memcpy(C.array, C0.array, C0.dims[0]*C0.dims[1]*C0.dims[2]*sizeof(float64_t));23:04
pickle27lisitsyn: ^23:04
pickle27C0 is passed by ref to the function23:04
lisitsynpickle27: I expect C0.dims may be deleted before C23:05
lisitsynmemcpy part looks ok23:05
pickle27ahh so I should do a deep copy of that too23:05
pickle27okay I'll try that23:05
-!- FSCV_ [~FSCV@] has joined #shogun23:06
-!- FSCV_ [~FSCV@] has quit [Client Quit]23:06
pickle27lisitsyn: that fixed it thanks23:07
pickle27lisitsyn: I don't know if I would have caught that23:07
lisitsynpickle27: two years of catching such errors ;)23:07
lisitsynslowed down though23:08
@wikinglisitsyn: yo23:14
@wikinglisitsyn: can i borrow 15 minutes of your time?23:16
lisitsynwiking: yes sure23:16
@wikinglisitsyn: thnx heaps23:16
lisitsynwhat's up?23:16
@wikingi mean your workshop presentation time ;)23:16
lisitsynwiking: ahha :D23:17
lisitsynyes sure23:17
lisitsynwiking: what do you want to present?23:17
@wikingwell mostly the yet to be commited shit i was working between sept-dec23:17
@wikinggive a little demo23:17
@wikingwhat i was doing with it23:17
lisitsynwiking: DVOR images please23:17
lisitsynI can't forget that category23:18
lisitsynremember that?23:18
@wikingi doo23:18
@wikingit's watermarked in my eyes23:18
@sonney2kwiking, so what did you want to talk about?23:18
lisitsynhaha yes same too23:18
lisitsynI just wake up and see that23:18
@wikingsonney2k: so if possible would like to give a little demo of the stuff i was doing with shogun23:19
@sonney2kvan51, I would rather want you to finish up the stuff and get it merged ASAP23:19
@iglesiasgwiking: latent stuff?23:19
@wikingiglesiasg: yep23:20
@iglesiasgwiking: yeah, do a latent presentation!23:20
@sonney2kvan51, I was expecting that this takes you 2 days no more...23:20
@iglesiasgthat sounds mighty23:20
lisitsynwiking: iglesiasg: latent like with some gay allusions ;)23:20
@wikingiglesiasg: yeah .. but unfortunately i cannot pythonize it :(23:21
@wikinglisitsyn: :D23:21
lisitsynwiking: what is it in?23:21
@wikinglisitsyn: parts detection in image23:21
lisitsynwiking: I mean written in23:21
lisitsynyou said you cannot pythonize23:21
@wikingah yeah23:21
@wikingremember the discussion back in october23:21
@wikingabout what can be and what cannot be directorized? :)23:21
@wikingby swig23:22
lisitsynwiking: so C++?23:22
@wikingi could create wrapper classes for it23:22
@wikingbut i doubt that anybody would ever use it again in the same way23:22
@sonney2kvan51, it would be cool to move on23:24
@sonney2kwiking, sure but need to schedule it lisitsyn, iglesiasg any thoughts?23:25
lisitsynsonney2k: I am ok to give my 15 minutes23:25
@wikingsonney2k: only thing is that maybe presenting after iglesiasg would make the most sense, if iglesiasg don't mind, as i'll be referring to SOSVM...23:26
-!- FSCV [~FSCV@] has joined #shogun23:28
@wikingbtw anybody realised that u can edit code on github page directly? :)23:29
@wikingi mean actually edit HEAD and do commit there straight away23:29
@sonney2kwiking, ok so then sergey should be last23:29
@sonney2kand you inbetween23:29
@sonney2kwiking, do you have a real demo?23:30
@sonney2kor slides?23:30
lisitsynwiking: I use it23:30
@wikingsonney2k: yeah i'm using it23:30
@wikingsonney2k: i mean i have a real application23:30
@wikingsonney2k: that is actually being used23:30
@sonney2kwiking, that you can show?23:30
lisitsynsonney2k: hopefully not these images I have seen once23:30
@sonney2kiglesiasg, would it be OK if you do just 30 minutes SO, followed by wiking with an SO application?23:31
@iglesiasgsonney2k: sure23:31
@sonney2kwiking, what is the title of the talk then?23:37
@sonney2kiglesiasg, ok then please adjust the schedule on the website23:37
@iglesiasgwiking: give me a title please :)23:38
@wikingsonney2k: mmm23:42
@sonney2kwiking, ?23:43
@wikingiglesiasg: (latent)SOSVM in practice23:43
van51sonney2k: yeah I know.. I have to pick up some speed23:50
@sonney2kvan51, otherwise we cannot do the fun part23:52
@sonney2kit would be stupid to get stuck in the easiest thing already23:52
@sonney2kso van51 go go go!23:52
* sonney2k dead23:53
@sonney2kvan51, btw regarding sparse vectors just alloc number of ngrams/words many elements23:53
@sonney2kand fill them23:53
van51sonney2k: that would leave some spaced unused, wouldn't it?23:54
@iglesiasgSpain to the final guys, we are there!23:59
@iglesiasgwell not me ... whatever23:59
--- Log closed Fri Jun 28 00:00:08 2013