| --- Log opened Fri May 25 00:00:21 2018 | ||
| -!- iglesiasg [~iglesias@f119189.upc-f.chello.nl] has quit [Quit: Leaving] | 00:56 | |
| -!- Elfarouk [695d4ff2@gateway/web/freenode/ip.105.93.79.242] has quit [Ping timeout: 260 seconds] | 02:28 | |
| -shogun-buildbot:#shogun- Build nightly trusty deb #166 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/26/builds/166 | 03:00 | |
| -shogun-buildbot:#shogun- Build clang - undefined behaviour analysis #147 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/18/builds/147 | 03:00 | |
| -shogun-buildbot:#shogun- Build doxygen - nightly #147 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/12/builds/147 | 03:00 | |
| -shogun-buildbot:#shogun- Build clang - static analysis #147 is complete: Exception [exception update shogun (failure) uploading static_analysis (exception)] - http://buildbot.shogun-toolbox.org:8080/#builders/29/builds/147 | 03:00 | |
| -shogun-buildbot:#shogun- Build nightly_bsd_license #161 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/4/builds/161 | 03:00 | |
| -shogun-buildbot:#shogun- Build cookbook - nightly #149 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/14/builds/149 | 03:00 | |
| -shogun-buildbot:#shogun- Build nightly jessie deb #154 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/20/builds/154 | 03:00 | |
| -shogun-buildbot:#shogun- Build nightly_default #149 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/17/builds/149 | 03:00 | |
| -shogun-buildbot:#shogun- Build nightly stretch deb #117 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/38/builds/117 | 03:00 | |
| @sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4291 merged by karlnapf | 08:07 |
|---|---|---|
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/024e4658d3cbc46e9168a1b283b8ee32d7048126 by karlnapf | 08:07 |
| -!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has joined #shogun | 08:09 | |
| shubham808 | Hi, today i will look into using typed_tests to automate premature stopping | 08:10 |
| shubham808 | and also refactoring the meta examples | 08:10 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4293 merged by vigsterkr | 09:31 |
| @sukey | [https://github.com/shogun-toolbox/shogun] vigsterkr pushed 6 commits: | 09:31 |
| @sukey | https://github.com/shogun-toolbox/shogun/commit/da950d59f724ebb93deaf827f2258be63b24beaa | 09:31 |
| @sukey | https://github.com/shogun-toolbox/shogun/commit/8d720dd00276e8efe68e2c8a7b2701e1e671b688 | 09:31 |
| @sukey | https://github.com/shogun-toolbox/shogun/commit/7cab56e736b01918aa03c60c3f95e29c1ba4de42 | 09:31 |
| @sukey | https://github.com/shogun-toolbox/shogun/commit/8b44285611e4240d5898bee847c97b3e6842f4ae | 09:31 |
| @sukey | https://github.com/shogun-toolbox/shogun/commit/b85355c43b33da0f1cba0a79ba857c3e98b7a6bc | 09:31 |
| @sukey | https://github.com/shogun-toolbox/shogun/commit/040692b88ffa69fe2488ef17fe91bb133b3a6348 | 09:31 |
| -shogun-buildbot:#shogun- Build deb4 - python3 #419 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/28/builds/419 | 09:31 | |
| -shogun-buildbot:#shogun- Build trusty - libshogun #385 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/33/builds/385 | 09:31 | |
| -shogun-buildbot:#shogun- Build deb3 - interfaces #341 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/37/builds/341 | 09:32 | |
| -shogun-buildbot:#shogun- Build FC23 - libshogun #485 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/27/builds/485 | 09:33 | |
| -shogun-buildbot:#shogun- Build yakkety - libshogun #477 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/1/builds/477 | 09:33 | |
| -shogun-buildbot:#shogun- Build xenial - libshogun #474 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/16/builds/474 | 09:33 | |
| -shogun-buildbot:#shogun- Build zesty - libshogun #473 is complete: Failure [failed update shogun (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/13/builds/473 | 09:34 | |
| @wiking | huhuuuuu | 09:42 |
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/3776b02c873ecd2b06ac3a5d7dbb96be3149ac58 by vigsterkr | 09:48 |
| @wiking | this should do it... | 09:48 |
| -shogun-buildbot:#shogun- Build trusty - libshogun - viennacl #483 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/6/builds/483 | 09:54 | |
| @wiking | wuwei, are you here? | 09:58 |
| -shogun-buildbot:#shogun- Build deb1 - libshogun #455 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/10/builds/455 | 09:58 | |
| wuwei | wiking: hi | 09:59 |
| @wiking | https://github.com/shogun-toolbox/shogun/blob/develop/doc/ipython-notebooks/multiclass/naive_bayes.ipynb?short_path=d818fbc#L164 | 09:59 |
| @wiking | is there any specific reason why its RealFeatures here instead of features ? | 10:00 |
| wuwei | i guess no :) but let me check | 10:01 |
| @wiking | kk | 10:01 |
| @wiking | lemme know | 10:01 |
| @wiking | if not then i'll fix it | 10:01 |
| wuwei | actually no, i missed that | 10:03 |
| -!- travis-ci [~travis-ci@ec2-54-161-193-191.compute-1.amazonaws.com] has joined #shogun | 10:03 | |
| travis-ci | it's Giovanni De Toni's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/383570716 | 10:04 |
| -!- travis-ci [~travis-ci@ec2-54-161-193-191.compute-1.amazonaws.com] has left #shogun [] | 10:04 | |
| @wiking | kk | 10:04 |
| @wiking | will fix it | 10:04 |
| @sukey | [https://github.com/shogun-toolbox/shogun] vigsterkr pushed 2 commits: | 10:05 |
| @sukey | https://github.com/shogun-toolbox/shogun/commit/5e4c2e8234b0235d45d19834e7a919456053a261 | 10:05 |
| @sukey | https://github.com/shogun-toolbox/shogun/commit/6b852ce4f152fcbe43c877034c0d087a26b2cf1f | 10:05 |
| -shogun-buildbot:#shogun- Build trusty - libshogun #387 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/33/builds/387 | 10:07 | |
| -shogun-buildbot:#shogun- Build FC23 - libshogun #487 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/27/builds/487 | 10:12 | |
| -shogun-buildbot:#shogun- Build yakkety - libshogun #479 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/1/builds/479 | 10:12 | |
| -shogun-buildbot:#shogun- Build zesty - libshogun #475 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/13/builds/475 | 10:13 | |
| -shogun-buildbot:#shogun- Build deb4 - python3 #421 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/28/builds/421 | 10:14 | |
| -!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has quit [Ping timeout: 276 seconds] | 10:16 | |
| -shogun-buildbot:#shogun- Build xenial - libshogun #476 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/16/builds/476 | 10:20 | |
| -shogun-buildbot:#shogun- Build deb3 - interfaces #342 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/37/builds/342 | 10:48 | |
| -!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has joined #shogun | 10:49 | |
| shubham808 | HeikoS here? | 10:50 |
| @wiking | he isn't | 10:55 |
| @wiking | can i help? | 10:56 |
| shubham808 | so about the preamture testing | 11:02 |
| shubham808 | orignally the idea was to | 11:02 |
| shubham808 | use TYPED_TESTS | 11:02 |
| shubham808 | but since not all classes can be used in the same way (features/labels requirements) | 11:03 |
| shubham808 | i wanted to talk about trying something like we did in serialization unit test | 11:04 |
| @wiking | you mean you cannot use a template because they need different way of training? | 11:05 |
| @wiking | i mean you know that you can use typed_tests | 11:05 |
| shubham808 | yes | 11:05 |
| @wiking | but with different fixtures | 11:05 |
| @wiking | and you name the fixture variables the same | 11:06 |
| @wiking | or can you give me a line | 11:06 |
| @wiking | how you imagine having a problem with different way of doing things with machines? | 11:06 |
| shubham808 | we need all the machine to reach the train step | 11:08 |
| shubham808 | before we can stop it | 11:08 |
| shubham808 | will it not cause problems when i try to train all machines with same kind of features | 11:09 |
| shubham808 | since the automation dont know beforehand what it will need | 11:10 |
| shubham808 | but also | 11:10 |
| shubham808 | since the tests we write will be based on serializing and deserializing models | 11:11 |
| shubham808 | it makes sense to use the code for serialization tests | 11:13 |
| -!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: ZNC 1.6.5 - http://znc.in] | 11:17 | |
| -!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun | 11:19 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 11:19 | |
| shubham808 | am i missing something here? | 11:27 |
| -!- travis-ci [~travis-ci@ec2-54-161-193-191.compute-1.amazonaws.com] has joined #shogun | 11:30 | |
| travis-ci | it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/383580906 | 11:30 |
| -!- travis-ci [~travis-ci@ec2-54-161-193-191.compute-1.amazonaws.com] has left #shogun [] | 11:30 | |
| -!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] | 11:45 | |
| -!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 11:52 | |
| -!- mode/#shogun [+o wiking] by ChanServ | 11:52 | |
| @wiking | HeikoS, have you missed shubam by any chance? | 11:52 |
| @wiking | i had lost connection | 11:52 |
| -!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 240 seconds] | 12:23 | |
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/51b403c87d0f82c35dd9b4fa1b9afb414867bcde by besser82 | 12:46 |
| -!- wiking [~wiking@huwico/staff/wiking] has quit [Read error: Connection reset by peer] | 13:05 | |
| -!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 13:09 | |
| -!- mode/#shogun [+o wiking] by ChanServ | 13:09 | |
| @sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4285 synchronized by vinx13 | 15:27 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4306 opened by vigsterkr | 15:35 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4306 | 15:35 |
| -!- Elfarouk [695d4ff2@gateway/web/freenode/ip.105.93.79.242] has joined #shogun | 16:12 | |
| Elfarouk | wiking, are you here? | 16:14 |
| @wiking | helo | 16:14 |
| @wiking | yes | 16:14 |
| Elfarouk | Hey! So I have sth weird in the new API that I wanna ask about | 16:14 |
| Elfarouk | so basically in the example (.sg file), I have the following line: | 16:15 |
| Elfarouk | StratifiedCrossValidationSplitting splitting_strategy(labels_train, 2) | 16:15 |
| Elfarouk | now when the labels_train is in the old API (BinaryLabels) it works okay | 16:16 |
| Elfarouk | but if I change it to Labels (new API) then the .cpp file crashes | 16:16 |
| Elfarouk | now I traced it back | 16:16 |
| Elfarouk | in the .cpp file to this line of code: | 16:16 |
| @wiking | mmm is this the xval notebook pr? | 16:16 |
| Elfarouk | if (labels->get_num_labels()<num_subsets) { SG_ERROR("Only %d labels for %d subsets!\n", labels->get_num_labels(), num_subsets); } | 16:17 |
| Elfarouk | yea it is | 16:17 |
| @wiking | lemme see | 16:17 |
| Elfarouk | for some reason if I use the old API, labels->get_num_labels() is correct, but if I use the new API, labels->get_num_labels() is equal to 0 | 16:17 |
| Elfarouk | This line of code is in SplittingStrategy.cpp btw (The base class of StratifiedCrossValidationSplitting ) | 16:18 |
| @wiking | ok so you changed this line | 16:19 |
| @wiking | BinaryLabels labels_train(f_labels_train) | 16:19 |
| @wiking | to | 16:19 |
| @wiking | Labels labels_train = labels(f_labels_train) | 16:19 |
| @wiking | ? | 16:19 |
| Elfarouk | yes exactly | 16:19 |
| @wiking | or yeah i see label_train_new | 16:19 |
| Elfarouk | yea thats the only way i managed to make it run fine | 16:20 |
| @wiking | kk | 16:21 |
| @wiking | ok but wait | 16:21 |
| @wiking | i'm just wondering | 16:21 |
| @wiking | that this used to work? | 16:21 |
| @wiking | so if you use label_train | 16:21 |
| @wiking | then everything is good? | 16:21 |
| Elfarouk | Yes, if you use labels_train, everything is good | 16:22 |
| @wiking | mmm | 16:22 |
| Elfarouk | switch to labels_train_new and you get that error | 16:22 |
| @wiking | mmmm | 16:22 |
| @wiking | this is weird | 16:22 |
| @wiking | as there's no assertation on anything | 16:22 |
| @wiking | about the label type | 16:23 |
| @wiking | but lemme see | 16:23 |
| Elfarouk | Alright then, thanks a lot! | 16:23 |
| @wiking | Elfarouk, it's the cpp-evaluation-cross_validation right? | 16:39 |
| Elfarouk | wiking, yes | 16:39 |
| Elfarouk | ctest -R generated_cpp-evaluation-cross_validation -V | 16:40 |
| @wiking | ok i see :) | 16:40 |
| @wiking | managed to get it reproduced locally | 16:40 |
| Elfarouk | So I am guessing this is a bug no? | 16:42 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4307 opened by micmn | 16:43 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4308 opened by vigsterkr | 16:49 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4308 vigsterkr added label: "meta examples" | 16:49 |
| @wiking | Elfarouk, ok | 16:52 |
| @wiking | no | 16:52 |
| @wiking | so the thing is that you do this | 16:52 |
| @wiking | auto f_labels_train = wrap(csv_file("../../data/classifier_binary_2d_linear_labels_train.dat")); | 16:53 |
| @wiking | and then use the same object twice | 16:53 |
| @wiking | BinaryLabels labels_train(f_labels_train) | 16:53 |
| @wiking | auto label_train_new = wrap(labels(f_labels_train)); | 16:53 |
| Elfarouk | Ohh!! | 16:53 |
| Elfarouk | so you can't use it twice | 16:53 |
| @wiking | now the problem with this is that f_labels_train is a file descriptor | 16:53 |
| @wiking | so basically it reads in the data into | 16:54 |
| @wiking | BinaryLabels labels_train(f_labels_train) | 16:54 |
| @wiking | but then there's nothing more to be read for label_train_new | 16:54 |
| Elfarouk | Ohhhhhhh | 16:54 |
| Elfarouk | that makes a ton more sense now! | 16:54 |
| @wiking | buut | 16:54 |
| @wiking | once i'm done | 16:54 |
| @wiking | we are still here | 16:54 |
| @wiking | [ERROR] In file /home/wiking/shogun/src/shogun/evaluation/StratifiedCrossValidationSplitting.cpp line 44: Multiclass or binary labels required for stratified crossvalidation | 16:54 |
| @wiking | terminate called after throwing an instance of 'shogun::ShogunException' | 16:54 |
| @wiking | what(): [ERROR] In file /home/wiking/shogun/src/shogun/evaluation/StratifiedCrossValidationSplitting.cpp line 44: Multiclass or binary labels required for stratified crossvalidation | 16:54 |
| @wiking | :) | 16:54 |
| Elfarouk | :O | 16:55 |
| @wiking | that's because labels(csvfile) generates a DenseLabel | 16:55 |
| @wiking | which has LT_DENSE_GENERIC | 16:56 |
| Elfarouk | can we use binary_labels() to change DenseLabels to BinaryLabels ? | 16:56 |
| Elfarouk | in xvalidation.cpp | 16:56 |
| @wiking | yeah | 16:56 |
| @wiking | but actually this is just a bad design | 16:56 |
| @wiking | :))) | 16:56 |
| Elfarouk | Ohh, do u mean a bad design referring to the labels? | 16:58 |
| @wiking | yes | 16:59 |
| @wiking | the factory | 16:59 |
| @wiking | because still | 16:59 |
| @wiking | i mean which one do you wanna have | 16:59 |
| @wiking | multiclass or binary | 16:59 |
| @wiking | because you can cast a binary imput | 16:59 |
| @wiking | into both a BinaryLabels | 17:00 |
| @wiking | and MulticlassLabels | 17:00 |
| Elfarouk | Ohhh I see what you mean | 17:00 |
| @wiking | ok | 17:05 |
| @wiking | are yo ublock | 17:05 |
| @wiking | with other stuff? | 17:05 |
| @wiking | because this one is now up in the air | 17:05 |
| @wiking | i've sent an email to Heiko | 17:05 |
| @wiking | but i would prefer till then you work on something else | 17:05 |
| @wiking | ? | 17:05 |
| @wiking | i mean the cost function? | 17:05 |
| Elfarouk | Yea for now I'm good, I will fix the line 44 | 17:05 |
| @wiking | i mean line 44 is something | 17:06 |
| @wiking | that needs some design decisions | 17:06 |
| @wiking | so i would not actually touch that | 17:06 |
| Elfarouk | ohh | 17:06 |
| Elfarouk | I thought its a minor fix like change it to binary_labels | 17:06 |
| @wiking | yeah not really | 17:06 |
| @wiking | because you would need to do this | 17:06 |
| @wiking | try | 17:06 |
| @wiking | { | 17:06 |
| @wiking | binary_labels(input) | 17:06 |
| @wiking | } catch () { | 17:06 |
| @wiking | multiclass_labels(input) | 17:06 |
| @wiking | } | 17:06 |
| @wiking | and that's pretty shitty | 17:07 |
| Elfarouk | :) | 17:07 |
| Elfarouk | I can pause this for now and switch to the cost function then | 17:07 |
| @wiking | great | 17:07 |
| @wiking | thnx | 17:07 |
| @wiking | i have to run now | 17:07 |
| @wiking | ttyl | 17:07 |
| Elfarouk | I have a lot of comments there already | 17:07 |
| Elfarouk | ttyl | 17:07 |
| Elfarouk | thanks for the help again! | 17:07 |
| wuwei | hi, here are update for today: 1) updated doc of transformers 2) refactored ica converters with fit+apply | 17:56 |
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/fb1b5d342823e35f9cf8973c30db58c5cffaa737 by vigsterkr | 18:09 |
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/2afac8303953173975f2a21aec53e16af0ed3a80 by besser82 | 18:23 |
| -!- Elfarouk [695d4ff2@gateway/web/freenode/ip.105.93.79.242] has quit [Quit: Page closed] | 18:23 | |
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/af2ea4c54309abbde8b85cdd563e864a5c33ec2b by vigsterkr | 19:15 |
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/d84907709fa872a0209d7c269bddec7468f44815 by vigsterkr | 19:27 |
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/4666249f3224b473c193ba0897f5e42b789b0133 by besser82 | 20:29 |
| -!- Elfarouk [9cc4a6d7@gateway/web/freenode/ip.156.196.166.215] has joined #shogun | 22:32 | |
| Elfarouk | wiking are you here? | 23:57 |
| --- Log closed Sat May 26 00:00:22 2018 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!