Open in new window / Try shogun cloud
--- Log opened Tue May 20 00:00:56 2014
-!- pickle27 [~pickle27@192-0-136-118.cpe.teksavvy.com] has joined #shogun01:06
-!- khalednasr [~k.nasr92@41.69.168.241] has quit [Quit: Leaving]02:14
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun02:16
-!- mode/#shogun [+o HeikoS] by ChanServ02:16
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Quit: Leaving.]02:43
-!- lambday [67157f4f@gateway/web/freenode/ip.103.21.127.79] has quit [Ping timeout: 240 seconds]02:55
shogun-buildbotbuild #9 of nightly_fedora is complete: Failure [failed git]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_fedora/builds/9  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>, Parijat Mazumdar <mazumdarparijat@gmail.com>03:00
kislaypickle27, yo!03:31
pickle27kislay: whats up?03:32
kislaypickle27, http://pastebin.com/Uz2CJi3X03:32
kislaypickle27, A small eg. of DenseSGMatrixFactory.03:32
pickle27kislay: looks good03:33
pickle27can you have a default on the dense factory too plz03:34
kislayyes sure03:34
kislayBut we need CDenseFeatures<float64_t> a lot.03:34
kislaypickle27, So, I think we must convert all of them to <float64_t>03:35
pickle27kislay: actually this is a good point03:36
pickle27this is something we need to think about03:36
pickle27do we want to incorporate switching the type with the conversions03:37
pickle27or do we always do 1 to 1 types and then you can change the shogun type after if you need03:37
kislayI think we should maintain the way the Matrix was originally presented to us.03:38
pickle27okay lets go with that then03:38
pickle27you'll need to extend the current implementation to include getting the type from the input and choosing the appropriate type for the output03:38
kislayBut again, I don't know how to do it. Templates right?03:39
pickle27I don't think you'll need templates03:39
pickle27you can ask a cvMat what its type is03:39
pickle27and then lookup what the corresponding shogun type is from a constant03:39
pickle27then pass that result into the SGMatrix constructor that you are going to return03:40
pickle27make sense?03:40
kislayyeah.03:40
kislayvery appropriate it seems to me. Btw the other method you suggested is good too.03:41
kislaywhat's your views?03:41
pickle27I like simple so I agree we should return a matrix of the same type as was given to us03:41
pickle27is that what you meant>03:42
kislaythat is right. The other method where we do 1 to 1 types..i.e always generate uint8_t but at the last moment convert it into the required type.03:43
pickle27lets just go 1 to 103:44
kislayerr.. the one which you explained to me. that one, right?03:44
pickle27yeah!03:45
kislayyup.03:45
pickle27getting the type of the input and then setting the output type to match03:45
pickle27you'll need a constant that maps cv types to shogun types03:46
kislayyeah. It looks elegant.03:46
pickle27check out std::map03:46
pickle27should do the trick03:46
kislaygreat! thanks03:46
-!- kislay [~Abhijeet@103.240.206.219] has quit [Ping timeout: 255 seconds]04:05
shogun-buildbotbuild #807 of nightly_default is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/80705:27
-!- iglesiasg [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has joined #shogun06:19
-!- mode/#shogun [+o iglesiasg] by ChanServ06:20
@iglesiasgGood morning, people06:21
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun06:27
shogun-notifier-shogun: Wu Lin :develop * 13905c1 / src/shogun/mathematics/Integration.cpp: https://github.com/shogun-toolbox/shogun/commit/13905c1a3679f898e0c4434d32759cd8253888f606:27
shogun-notifier-shogun: minor fix06:27
shogun-notifier-shogun: Fernando Iglesias :develop * b2a43db / src/shogun/mathematics/Integration.cpp: https://github.com/shogun-toolbox/shogun/commit/b2a43dbaf3723248e4eb8a14ae4c8e77b0df7a8106:27
shogun-notifier-shogun: Merge pull request #2237 from yorkerlin/yorker-integration06:27
shogun-notifier-shogun:06:27
shogun-notifier-shogun: minor fix for Integration.cpp06:27
-!- kislay [~Abhijeet@103.240.205.21] has joined #shogun06:34
-!- pickle27 [~pickle27@192-0-136-118.cpe.teksavvy.com] has quit [Remote host closed the connection]06:36
-!- iglesiasg [~iglesiasg@524AE0A7.cm-4-3d.dynamic.ziggo.nl] has quit [Quit: Leaving]06:52
-!- travis-ci [~travis-ci@ec2-54-87-168-245.compute-1.amazonaws.com] has joined #shogun07:19
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/2557204207:19
-!- travis-ci [~travis-ci@ec2-54-87-168-245.compute-1.amazonaws.com] has left #shogun []07:19
shogun-buildbotbuild #309 of debian wheezy - memcheck is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/309  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>, Wu Lin <yorker.lin@gmail.com>09:03
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]09:27
-!- jiaolong [9e6d09e4@gateway/web/freenode/ip.158.109.9.228] has joined #shogun09:30
-!- HeikoS [~heiko@0542590b.skybroadband.com] has joined #shogun10:05
-!- mode/#shogun [+o HeikoS] by ChanServ10:05
-!- HeikoS [~heiko@0542590b.skybroadband.com] has quit [Ping timeout: 264 seconds]10:11
-!- HeikoS [~heiko@nat-247-97.internal.eduroam.ucl.ac.uk] has joined #shogun11:20
-!- mode/#shogun [+o HeikoS] by ChanServ11:21
-!- kislay [~Abhijeet@103.240.205.21] has quit [Ping timeout: 264 seconds]11:29
-!- kislay [~Abhijeet@103.240.205.174] has joined #shogun13:28
-!- khalednasr [~k.nasr92@41.69.238.139] has joined #shogun14:22
-!- Saurabh7 [~Saurabh7@117.203.118.13] has joined #shogun14:38
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun15:22
shogun-notifier-shogun: Kevin Hughes :develop * c94bb2b / README.md: https://github.com/shogun-toolbox/shogun/commit/c94bb2b0de9087240c8ee310760dbaf9bb319ef915:22
shogun-notifier-shogun: add contributions to readme15:22
shogun-notifier-shogun: Fernando Iglesias :develop * f6ca590 / README.md: https://github.com/shogun-toolbox/shogun/commit/f6ca5907947ed90dff05c3b5e6b4fe8efb0a179c15:22
shogun-notifier-shogun: Merge pull request #2229 from pickle27/develop15:22
shogun-notifier-shogun:15:22
shogun-notifier-shogun: add contributions to readme15:22
shogun-buildbotbuild #2298 of bsd1 - libshogun is complete: Failure [failed compile test]  Build details are at http://buildbot.shogun-toolbox.org/builders/bsd1%20-%20libshogun/builds/2298  blamelist: Kevin Hughes <kevinhughes27@gmail.com>15:26
shogun-buildbotbuild #2961 of deb1 - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/deb1%20-%20libshogun/builds/2961  blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com>15:27
shogun-buildbotbuild #631 of precise - libshogun is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/precise%20-%20libshogun/builds/631  blamelist: Kevin Hughes <kevinhughes27@gmail.com>15:41
-!- yorkerlin [81614fb3@gateway/web/freenode/ip.129.97.79.179] has joined #shogun15:45
-!- Saurabh7 [~Saurabh7@117.203.118.13] has quit [Read error: Connection reset by peer]15:49
-!- pickle27f [~pickle27@216.191.231.218] has joined #shogun15:52
-!- lisitsyn [~qdrgsm@80.252.20.67] has joined #shogun15:52
-!- pickle27f [~pickle27@216.191.231.218] has quit [Quit: Leaving]15:57
-!- travis-ci [~travis-ci@ec2-184-73-98-218.compute-1.amazonaws.com] has joined #shogun16:00
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/2560204116:00
-!- travis-ci [~travis-ci@ec2-184-73-98-218.compute-1.amazonaws.com] has left #shogun []16:00
-!- khalednasr [~k.nasr92@41.69.238.139] has quit [Quit: Leaving]16:02
-!- pickle27f [~pickle27@216.191.231.218] has joined #shogun16:03
-!- Saurabh7 [~Saurabh7@117.200.190.189] has joined #shogun16:05
-!- HeikoS [~heiko@nat-247-97.internal.eduroam.ucl.ac.uk] has quit [Ping timeout: 276 seconds]16:18
-!- HeikoS [~heiko@ra-vpn-151.ra-vpn.ucl.ac.uk] has joined #shogun16:32
-!- mode/#shogun [+o HeikoS] by ChanServ16:32
-!- yorkerlin [81614fb3@gateway/web/freenode/ip.129.97.79.179] has quit [Ping timeout: 240 seconds]16:34
-!- pickle27f [~pickle27@216.191.231.218] has quit [Quit: This computer has gone to sleep]17:03
-!- pickle27f [~pickle27@216.191.231.218] has joined #shogun17:13
-!- pickle27_ [~pickle27@216.191.231.218] has joined #shogun17:16
-!- pickle27f [~pickle27@216.191.231.218] has quit [Read error: Connection reset by peer]17:16
shogun-buildbotbuild #310 of debian wheezy - memcheck is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/310  blamelist: Kevin Hughes <kevinhughes27@gmail.com>17:43
kislayhey lisitsyn17:43
lisitsynkislay: hey17:43
@besser82lisitsyn, ho!17:44
lisitsynbesser82: hey17:44
@besser82lisitsyn, how is it?17:44
@besser82lisitsyn, any news on the data?!?17:44
lisitsynbesser82: sorry it was very intense17:44
@besser82lisitsyn, np  ;)17:44
kislayI need some help in Cpp. could you look this http://pastebin.com/qj3BqLTs.17:44
lisitsynI was ill and now trying to catch up with my thesis17:44
@besser82lisitsyn, allrighty!  ;)17:44
lisitsynhave like a few more days to get it polished17:44
@besser82lisitsyn, like me with results on the data :-P17:45
lisitsynbesser82: it would be next thing ;)17:45
@besser82lisitsyn, don't hurry17:45
lisitsynkislay: alright - what is it?17:45
@besser82lisitsyn, the clustering stuff is nearly ready2go  ;)17:45
lisitsynbesser82: cool17:45
@besser82lisitsyn, next would be deep sentiment stuff  =)17:46
kislaylisitsyn, converting opencv matrix into SGMatrix.17:46
lisitsynkislay: yeah that's pretty clear ;) I mean the question17:46
kislaylisitsyn, I am not getting how to get those mapped values into arguments of SGMatrix< > templates!!17:47
kislaylisitsyn, the String I have shown there is obviously wrong. Just to print the output I did tha17:48
kislaylisitsyn, what should be the  map<int, *type*>17:49
lisitsynwhat is this map intended for?17:49
kislayto invoke the correct SGMatrix constructor... i.e according to the type of the opencv matrix.17:49
lisitsynI guess you need templated getMatrix function17:50
lisitsynlike template <typename T> SGMatrix<T> getMatrix(Mat);17:50
lisitsynI see no other way as Mat has erased its type17:50
kislayah!17:50
kislaylisitsyn, okay. I will hack a bit. thanks.17:52
lisitsynkislay: it would be possible if we could make SGMatrix like 'polymorphic'17:58
lisitsynbut it is somewhat called value type17:58
lisitsynyou just have its value and its efficient type is the same17:58
lisitsynso actually SGMatrix<int> and SGMatrix<float> are totally different classes with nothing in common17:58
kislaylisitsyn, so you suggest this as viable too, right?18:00
lisitsynkislay: you'd have to put a type in any case18:00
lisitsynso it would be18:00
lisitsyngetMatrix<float>(Mat)18:00
lisitsynor whatever type it is18:00
lisitsynI don't think it is possible to obtain the type of Mat's element18:01
lisitsynyou can only try to convert it and check whether this conversion is ok18:01
kislaylisitsyn, ok18:02
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout]18:22
-!- pickle27_ [~pickle27@216.191.231.218] has quit [Quit: This computer has gone to sleep]18:26
-!- pickle27f [~pickle27@216.191.231.218] has joined #shogun18:34
-!- khalednasr [~k.nasr92@41.69.238.139] has joined #shogun18:44
-!- Saurabh7 [~Saurabh7@117.200.190.189] has quit [Ping timeout: 252 seconds]18:52
-!- Saurabh7 [~Saurabh7@117.222.9.237] has joined #shogun19:05
-!- abinash_panda [75ef5e70@gateway/web/freenode/ip.117.239.94.112] has joined #shogun19:29
abinash_pandathoralf: Hi !19:30
-!- jiaolong [9e6d09e4@gateway/web/freenode/ip.158.109.9.228] has quit []19:52
-!- abinash_panda [75ef5e70@gateway/web/freenode/ip.117.239.94.112] has quit []19:52
-!- Saurabh7 [~Saurabh7@117.222.9.237] has quit [Ping timeout: 252 seconds]19:57
-!- HeikoS [~heiko@ra-vpn-151.ra-vpn.ucl.ac.uk] has quit [Ping timeout: 240 seconds]20:50
@besser82thoralf, ping?21:00
-!- lisitsyn [~qdrgsm@80.252.20.67] has quit [Read error: Connection reset by peer]21:14
-!- abinash_panda [75ef5e6f@gateway/web/freenode/ip.117.239.94.111] has joined #shogun22:18
-!- abinash_panda [75ef5e6f@gateway/web/freenode/ip.117.239.94.111] has quit [Ping timeout: 240 seconds]22:32
-!- abinash_panda [0e8b2998@gateway/web/freenode/ip.14.139.41.152] has joined #shogun22:33
-!- thoralf|home [~thoralf@91-65-137-195-dynip.superkabel.de] has joined #shogun23:16
thoralf|homeHeyho.23:16
-!- thoralf|home [~thoralf@91-65-137-195-dynip.superkabel.de] has quit [Ping timeout: 240 seconds]23:25
-!- thoralf|home [~thoralf@91-65-137-195-dynip.superkabel.de] has joined #shogun23:32
thoralf|homebesser82: pong23:32
thoralf|homeabinash_panda: Hey.23:32
-!- yorkerlin [81614fb3@gateway/web/freenode/ip.129.97.79.179] has joined #shogun23:33
yorkerlinHi Can anyone tell whether my guess about this construct SGVector(T * v,index_t len,bool ref_counting = true) is correct or not?  My guess is the SGVector will RE-USE the memory space in v and will AUTOMATICALLY clean-up (DELETE) the memory space if we use SGVector instance instead of SGVector pointer.23:38
thoralf|homeyorkerlin: You should check the implementation.23:39
thoralf|homeyorkerlin: Tell me what you think (after reading the code) and I'll tell you if it was true.23:39
thoralf|home:)23:39
yorkerlinic23:39
yorkerlinI check the implementation. It seems my guess is correct.23:47
thoralf|homeyorkerlin: Seems it's just copying the provided pointer... and it gets freed.23:50
thoralf|homeyorkerlin: You know that you don't need to do "new SGVector(v, len, true)" but just "new SGVector(v, len)"?23:51
thoralf|homeref_count=true is passed as default parameter23:51
yorkerlinYes23:52
yorkerlinThanks for your quick reply23:52
--- Log closed Wed May 21 00:00:57 2014