IRC logs of #shogun for Wednesday, 2018-02-21

« Tuesday, 2018-02-20 Index Thursday, 2018-02-22 »
--- Log opened Wed Feb 21 00:00:11 2018
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has joined #shogun01:48
travis-ciit'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/34384366801:48
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has left #shogun []01:48
-!- travis-ci [~travis-ci@ec2-54-167-169-20.compute-1.amazonaws.com] has joined #shogun01:53
travis-ciit'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/34409934001:53
-!- travis-ci [~travis-ci@ec2-54-167-169-20.compute-1.amazonaws.com] has left #shogun []01:53
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has joined #shogun02:16
travis-ciit'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/34409934002:16
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has left #shogun []02:16
-!- travis-ci [~travis-ci@ec2-54-166-176-157.compute-1.amazonaws.com] has joined #shogun02:22
travis-ciit'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/34409934002:22
-!- travis-ci [~travis-ci@ec2-54-166-176-157.compute-1.amazonaws.com] has left #shogun []02:22
-shogun-buildbot:#shogun- Build trusty - libshogun - viennacl #334 is complete: Failure [failed compile (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/6/builds/33402:26
-!- travis-ci [~travis-ci@ec2-54-166-176-157.compute-1.amazonaws.com] has joined #shogun03:41
travis-ciit'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/34412343003:41
-!- travis-ci [~travis-ci@ec2-54-166-176-157.compute-1.amazonaws.com] has left #shogun []03:41
-shogun-buildbot:#shogun- Build nightly_all #98 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/22/builds/9803:50
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has joined #shogun04:05
travis-ciit'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/34412343004:05
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has left #shogun []04:05
-!- HeikoS [~heiko@host86-129-231-92.range86-129.btcentralplus.com] has joined #shogun10:30
-!- mode/#shogun [+o HeikoS] by ChanServ10:30
@HeikoSwiking: jojo10:30
@HeikoS33: Test command: /home/heiko/git/shogun/build/tests/meta/meta_example_integration_tester "binary" "kernel_support_vector_machine.dat" "java" "generated_results" "reference_results"10:31
@HeikoS33: Test timeout computed to be: 9.99988e+0610:31
@HeikoS33: [WARN] In file /home/heiko/git/shogun/src/shogun/io/SerializableFile.cpp line 117: Could not read `array/value' from `generated_results/java/binary/kernel_support_vector_machine.dat'!10:31
@HeikoS33: [WARN] In file /home/heiko/git/shogun/src/shogun/base/Parameter.cpp line 2210: TParameter::new_sgserial(): Class `CWrappedBasic<float64>' was not listed during compiling Shogun :( ...  Can not construct it for `array'!10:31
@HeikoSthe thing is that the old results are not loaded anymore, as the CWrappedBasic class doesnt exist anymore10:31
@wikingHeikoS, ?10:34
@wikingwhere is this?10:34
@wikingin Parameter.cpp?10:35
@wikingi mean this check itself10:35
@HeikoSthis is if you run integration test10:35
@HeikoSsee command10:35
@HeikoSso the thing is10:35
@wikingyeah i'm just checking10:36
@wikingwhere the return value10:36
@HeikoSthe file kernel_support_vector_machine.dat10:36
@wikingof that function isnt handled10:36
@HeikoScontains CWrappedBasic10:36
@HeikoSand then TParameter trys to create a new instance of it10:36
@HeikoSwhich fails10:36
@HeikoS(while de-serializing)10:36
@HeikoSso nothing is loaded10:36
@wikingyep that i get10:36
@wikingbut this should fail with a fatal error10:36
@HeikoSwhat I don't get is10:36
@wikinghence i'm looking where the false10:37
@HeikoSwhat do you mean?10:37
@wikingis being ignored10:37
@wikingi mean if you10:37
@HeikoSah yeah10:37
@HeikoSsure10:37
@wikingwant to deserialize an object10:37
@wikingthat is in a different 'version'10:37
@wikingbecause lets say the object has changed10:37
@wikingsince then10:37
@wikingthen this should be not a nice warning10:37
@HeikoSsomewhere downstream it is ignored in de-serialization I guess10:37
@wikingbut a fatal error10:37
@HeikoSagreed10:37
@HeikoSwhat I dont get is why the array->equals didnt fail10:38
@wikingso i'm trying to see where the bool is ignored10:38
@wikingas there's a bool that is false10:38
@wikingfor 2 depth10:38
@wikingwhen there's this error10:38
@HeikoSbecause it should if the data that is compared against has the new format10:38
@wikingwell equals will naturally fail :)10:38
@wikingah no10:39
@wikingactually no10:39
@HeikoSthe thing is empty10:39
@wikingthe object that you are actually running10:39
@HeikoSbut why is the "generated results" empty?10:39
@wikingshould have all the elements10:39
@wikingyeah indeed10:39
@wikingthat equal should fail10:39
@wikingeven if the deserialization error is silenced10:39
@HeikoSy10:39
@wikingas one of the array should have elements10:39
@wikingthe other shoulnd't10:39
@wikingthat is even better wtf10:40
@wiking?\_(?)_/?10:40
@wiking:>10:40
@HeikoSyep10:40
@wikingHeikoS, i'll check the stuff10:40
@wikingfor where this error10:40
@wikingis ignored in parameter.cpp10:40
@wikingwhen deserializing10:40
@wikingk?10:40
@HeikoSsure10:41
@HeikoSI am checking why equals doesnt fail10:41
@wikingcool10:41
@HeikoSwiking: in any case10:41
@HeikoSwe will need to re-generate all integration testing data from the commit just before the change with the DynamicObjectArray10:41
@wikingyep10:41
@wikingi mean10:41
@HeikoSand then see which tests fail with HEAD10:41
@wikingwe dont have to10:41
@wikingi can fake it10:41
@wikingif you want10:41
@wiking:)10:41
@wikingbut first lets have these tests all fail10:42
@HeikoSyep10:42
@wikingboth your case10:42
@wikingand the case where deserialization fails10:42
@wikingas these supposed to be two separate bugs10:42
@HeikoSso the equals doesnt fail10:42
@HeikoSbecause both arrays are empty10:42
@wikingbut why are those empty?10:42
@wikingi mean as you see from that unit test10:42
@HeikoSdont know, checking that next10:42
@wikingit actually getting filled10:42
@HeikoSthe file contains the stuff also, just checked10:44
@wikingyeah i've chacked that in the past as well10:44
@wikingbut the unit tests does that as well :P10:44
@HeikoS31: [WARN] In file /home/heiko/git/shogun/src/shogun/io/SerializableFile.cpp line 117: Could not read `array/value' from `generated_results/cpp/binary/kernel_support_vector_machine.dat'!10:46
@HeikoSthis is the warning happening when the generated (new format) is loaded10:46
@HeikoSthats why the new version array is empty10:46
@wikingok so10:47
@wikingi've got where the thing is10:47
@wikingignored10:47
@wikingbool CSGObject::load_serializable(CSerializableFile* file,10:47
@wikingconst char* prefix)'10:47
@wikingnow the question is that do we wanna throw an exception10:47
@HeikoSyes I think10:48
@HeikoSah wait10:48
@HeikoSor return false10:48
@HeikoSbut I don't like this returning false stuff actually10:48
@HeikoStends to be forgotten10:48
@HeikoSin tester.cpp it is not used e.g.10:48
@wikingyeah i mean it shoulld be10:48
@wikingas return value check is something you should never ignore10:49
@wikingbut still10:49
@wikingimo this should be an exception10:49
@wikingthat you can catch if you like10:49
@wikingon the other hand10:49
@HeikoSyes agreed10:49
@wikingas you can see10:49
@wikingthe whole load_serializable is catching the exceptions10:49
@wikingi would change it10:49
@wikingthat if there's a load error -> SGERROR10:49
@wikingi.e. exception is thrown10:50
@wikingand either you handle it10:50
@wikingor the code is done10:50
@wikingi mean it fails badly10:50
@HeikoSyep10:50
@HeikoSthe method returns false10:50
@HeikoSso asserting load_success in tester makes it fail10:50
@wikingyeah10:51
@wikingi mean10:51
@wikingon the end of the day10:51
@wikingits the same shit :)10:51
@wikingtry-catch10:51
@wikingor check the return value10:51
@wiking:)10:51
@wikingonly thing is return value is easier to ignore10:51
@HeikoSthats why try-catch exists right? :D10:51
@HeikoSok so far so good10:52
@HeikoSnow need to find out why it fails to load the new format10:52
@wikingi mean yeah but you see what i mean right?10:52
@HeikoSI do10:52
@wikingits just more inconvenient10:52
@wikingto ignore an exception10:52
@wikingthan a return value10:52
@wikingi'm fine to convert this to an exception10:52
@wikingnote then i would do symetric10:53
@wikingi.e. CSGObject::save_serializable10:53
@HeikoSyep10:53
@wikingshould do the same10:53
@HeikoS+110:53
@wikingk i'll do this change then now10:53
@HeikoSin develop?10:53
@HeikoSI would wait until we fix the loading10:53
@HeikoSi.e. first fix loading (of new format) and then put exception in10:53
@wikingmmm i can just do it in a branch10:53
@HeikoSkk10:53
@wikingand see what travis does10:54
@wikingthe only thing that this is a moo point10:54
@wikingi hope we realise this10:54
@wiking:D10:54
@wikingas the whole thing should get out ASAP10:54
@wikingwhen we switch to tag based serialization10:54
@wiking:P10:54
@HeikoSwell the format will change again10:55
@HeikoSbut it is easy to re-generate these files10:55
@wikingi meant more about10:55
@wikingwhether we throw10:55
@wikingor return the error10:55
@wiking:)10:55
@HeikoSah10:55
@HeikoSyeah10:55
@HeikoSI hope we did not screw too many algorithms10:56
@HeikoSthere was quite some refactoring that relied on the integration tests10:56
@HeikoScatching changes ...10:56
@HeikoSLMNN is an example10:56
@HeikoSI had a discussion with him where I asked "why did the integration test not fail?"10:56
@HeikoSand he was like "did terminate due to maximum number of iterations, not convergence criterion that I refactored"10:56
-!- sukey [~nodebot@ks312251.kimsufi.com] has joined #shogun11:06
-!- mode/#shogun [+o sukey] by ChanServ11:06
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/8a119a3aa0835dba4432cb1f330df16643b3cc53 by vigsterkr11:06
@sukey[https://github.com/shogun-toolbox/shogun] New branch serialexception created11:06
@wikingthis should now all fail then11:06
@wikingmmm11:07
@wikingok i have more errors11:07
@wikingstopping travis11:07
@HeikoSwiking: the problem is loading array Vector<SGSerializable*>11:09
@wikinghahahah11:09
@wikingjust a sec11:09
@wikingi've just realised some amazing stuff11:09
@wikingin our ShogunException class11:10
@wikingbtw does anybody has any idea why our own exception is not inherited from std::exception?11:10
@HeikoSprobably because that didnt exist back then :D11:12
@wikingit sure did11:12
@wikingit is existing since c++ :)11:13
@HeikoSI dont get it why the loading fails11:13
@HeikoSthe formats are exactly the same11:13
@HeikoSapart from WrappedSGVector -> Serializable11:14
@HeikoSit might be the templating11:15
@HeikoSin CSerializable11:15
@HeikoSset_generic<typename extract_value_type<T>::value_type>();11:16
@HeikoSwiking: i think it might be this11:17
@HeikoSsee in the old file we have11:17
@HeikoSWrappedSGVector float6411:17
@HeikoSor more WrappedSGVector float64 [11:17
@HeikoS  7 value SGVector<float64> 2011:17
@HeikoSi.e. the type is WrappedSGVector with primitive type float6411:17
@HeikoSbut in the new file there is11:18
@HeikoSSerializable float64 [value SGVector<float64> 2011:18
@HeikoSi.e. Serializable where the primitive type is set to float64_t11:18
@HeikoSas well11:18
@HeikoSso when creating a new instance from this information, it cannot know that the value is SGVector11:18
@HeikoSiirc this is the reason why I had different types for basic/matrix/vector11:18
@HeikoSbecause otherwise the loading wouldnt work11:19
@HeikoSI think you cannot have a single CSerializable actually11:19
@HeikoSas the de-serialization cannot work the way it is implemented right now11:19
@HeikoSwiking: so removing CWrapped* causes errors when loading the old files, and using CSerializable causes error when loading the new files11:20
@HeikoSnot sure how to fix it without having a Serializable for basic/vector/matrix individually11:21
@wikingmmm11:21
@wikingi'll read this11:21
@wikingbut i just wanna fix that other part11:21
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/9977d634f38adc203d1aea10255469be8f4fed16 by vigsterkr11:27
@wikingk afaik now it should be good11:27
@wikingHeikoS, mm i understand why you had copy pasted 3 classes around11:28
@wikingbut with varg that was the ide11:28
@HeikoSyes :/11:28
@wikingthat you can dig out the generic type11:28
@HeikoSthe thing is that when loading the information is not availaing11:28
@HeikoSif you see this: Serializable float64 [ value SGVector<float64> 2011:29
@HeikoSthen you know the value type is SGVector11:29
@HeikoSbut if you just see Serializable float64 [ ..11:29
@HeikoSthen you can't know that it is SGVector11:29
@HeikoSi.e. in shogun we cannot have classes with the same name that have different parameter types11:30
@HeikoSunless it is the template argument (generic type) ifself11:30
@HeikoSbut you cannot have11:30
@HeikoST and SGVector<T>11:30
@wikinglemme fix this11:30
@wikingi'm not so sure why a name of the class would actually has any play in this game11:42
@wikingit's what you register11:42
@wikingin the param/tags11:42
@wikingbut even if it would be the name11:42
@wikingthat this is not a class11:42
@wikingit's a template11:42
@wikingso you can return whatever 'name' you want11:42
@wikingbased on the template param11:42
@HeikoSit is because of the way shogun loads11:52
@HeikoSit first reads the classname11:52
@HeikoSand the primitive type11:52
@HeikoSand then instantiates an empty instance using that info11:52
@HeikoSso in the CSerializable case, the classname and primitive type is the same11:52
@HeikoSwhether value is float64_t or SGMatrix<float64_t>11:52
@HeikoSand since only the primitive type is stored11:53
@HeikoSshogun instantiates an empty instance of CSerializable with type of "value" being float6411:53
@HeikoSand then tries to load a matrix11:53
@HeikoSand thats where the fail comes from11:53
@HeikoSwiking: imo there need to be 3 classes unfortunately11:53
@HeikoSwiking: maybe there is a better way somehow, or we can change serialization11:54
@wikingagain11:54
@HeikoSbut since this is quite urgent, I would just go back to have 3 classes11:54
@wikingif it's the string type11:54
@wikingquite urgent?11:54
@HeikoSintegration tests not working = urgent11:54
@wikingi'm not getting the 'i cannot wait 2 hours' problem11:54
@wikinghere11:54
@HeikoSah 2 hours of course11:54
@wikingand i'm not understanding11:54
@HeikoSI am talking longer time11:54
@wikingwhy the string cannot be differnet11:54
@wikingin case of a template11:55
@wikingas that could be derived from a template parameter11:55
@wikingright?11:55
@HeikoSah11:55
@wikingas Serializable11:55
@wikingis a template11:55
@HeikoSI think it is not the get_name()11:55
@HeikoSbut it is the class_list thing11:55
@wikingmmm11:55
@wikingi mean11:55
@HeikoSwhere a class is created from its name and primitive type11:55
@wikingthese presumptions11:55
@wikingwere already a bit11:55
@wikingbroken11:55
@wikinglast week11:55
@HeikoSif you do11:56
@wikingso lets rather just debug it11:56
@wikingand see what happens11:56
@HeikoScreate("Serializable", float64_t)11:56
@HeikoSwhat is the type of "value"?11:56
@wiking(i'm referring to get_name()11:56
@wikingand that object cnnot have it11:56
@wikingso i would say that i'm happy to fix it11:56
@wikingi dont see the urgency here11:56
@HeikoSnot urgent as in days11:56
@wikingthat this has to be done in this very second11:56
@wikingbut if you believe strongly11:56
@HeikoSbut I would do that before e.g. changing the serialization stuff11:57
@wikingthat that's the case11:57
@wikingit's just a flip in the switch11:57
@HeikoSnot this very second11:57
@wikingi.e. revert that one commit11:57
@HeikoSyeah sure11:57
@wikingbut i firmly beliee11:57
@wikingthat it's not that urgent11:57
@wikingthis is not blocking anybody imo11:57
@HeikoSwell, I would not merge any more PRs now11:57
@HeikoSat least those that are refactoring11:58
@HeikoSas errors slip there11:58
@HeikoSthat's just all I mean11:58
@wikingi can take on11:58
@wikingthose fixing11:58
@HeikoSI know it happened11:58
@HeikoSlmnn11:58
@wikingi can fix11:58
@HeikoSthere might be more11:58
@wikingand review11:58
@HeikoSyeah sure11:58
@wikingthe last 3 weeks11:58
@wikingprs11:58
@HeikoSI didnt mean to stress11:58
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has joined #shogun11:58
travis-ciit'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/34425067611:58
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has left #shogun []11:58
@wikingif that's your concern11:58
@HeikoSwhen was the CSerializable introduced?11:58
@HeikoSyou remember?11:58
@wikinggit log11:59
@wikingit's a july commit11:59
@wikingbut merged between11:59
@wikingDate:   Fri Jan 19 10:04:35 2018 +000011:59
@wikingDate:   Tue Jan 16 09:09:29 2018 +000011:59
@HeikoSah yes jan12:00
@HeikoSah ok12:00
@HeikoScool, well let me know if you need help or something12:00
@wikingbtw i'm not so sure that this change12:02
@wikingfrom bool to exception12:02
@wikingis actually worth to do12:02
@wikingas everywhere12:02
@wikingthis is being done12:02
@wikingi.e. check the return value12:02
@wikingsee all the jinja2 unit test templates12:02
@HeikoSyes12:02
@wikingand since this anyways goes into the trash12:02
@HeikoSleave it12:02
@wikingi would rather just fix one place the problem12:03
@HeikoSwe can just put an assertion into the tester.cpp12:03
@wikingi.e. the meta tester12:03
@wikingfor the return value12:03
@HeikoS++12:03
@wikingas this one now has multiple places12:03
@wikingthat i would need to do more fixe12:03
@wikings12:03
@wikingfor changing to void from bool12:03
@wikingi mean the other way round12:03
@wikingfrom bool to void12:03
@HeikoSyeah it is less work and all this will go anyways, so agreed12:03
@sukey[https://github.com/shogun-toolbox/shogun] vigsterkr pushed 2 commits:12:12
@sukeyhttps://github.com/shogun-toolbox/shogun/commit/77f0d29ee8bab68f743d856e718127cf887327ff12:12
@sukeyhttps://github.com/shogun-toolbox/shogun/commit/ea175886d9deeecc2e10b5dac130369e76e6fe2112:12
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/5158986a4aa12ad68c6343fc4de8f797bcbade0b by karlnapf12:25
@HeikoSlisitsyn: https://github.com/shogun-toolbox/shogun/commit/5158986a4aa12ad68c6343fc4de8f797bcbade0b#diff-095bc6624c9b34fd7828144d24f4133c12:26
@HeikoSwiking: ^12:26
@HeikoSfirst example works with new api12:26
@HeikoSat least partly12:26
@HeikoSneed some factories for features12:26
@wikingfactories for features?12:27
@HeikoSfeature = Features(Loader("filename"))12:28
@HeikoSand features has type CFeatures12:28
@wikingHeikoS, what i dont get why cant we just wrap the watch_param under SG_ADD macro/12:28
@wiking?12:28
@HeikoSwiking: not possible with the current approach12:29
@HeikoSsince m_parameters->add can only accept CSGObject atm12:29
@HeikoSand but watch_param should get the base class12:29
@wikingyes12:29
@HeikoSI can probably hack that12:29
@HeikoSthat SG_ADD casts12:29
@HeikoSyeah good point will do that12:29
@wikingi mean i think that macro could stay12:30
@wikingand just some partial things under it12:30
@HeikoSyes it can stay12:30
@wikingor around it12:30
@wikingas this is an ever changing thing12:30
@HeikoSbut the (CSGObject**) needs to be moved inside the macro12:30
@HeikoSso the line needs change12:30
@HeikoSbut yeah12:30
@HeikoSah12:30
@wikingyeah unfortunately yes12:30
@HeikoSnot sure  actually12:30
@wikingbut that's 1 change12:31
@HeikoSif it works12:31
@wikingright?12:31
@wikingi mean you can regexp that12:31
@HeikoSI  can cast in macro12:31
@wikingbecause now it will be watch12:31
@HeikoSbut then everything is casted12:31
@HeikoSso first need to check type12:31
@wikingbut then tomorrow it's asdf12:31
@HeikoSjaja12:31
@wikingi mean this is why it'd be better to follow interfaces12:32
@wikingor shit like this12:32
@wiking(sg_add)12:32
@HeikoSneed to check the type of the macro parameter12:33
@HeikoSnot sure how that would work12:33
@wikingyou cannot12:33
@wiking:)12:33
@wikingthat's a preprocessor12:33
@HeikoSthe type is different for "watch" and old "add"12:33
@wikinghas no clue about anything just strings12:33
@HeikoSyeah that's why I split it12:34
@wikingbut this is actually12:34
@wikinga change12:34
@HeikoSyes12:34
@wikingin the sg_add12:34
@wikingi mean you get that out12:35
@wikingand under it you do whatever you want to12:35
@wikingyou can even put there a template12:35
@wikingthat then actually can match for a type12:35
@wikingsee all the traits12:35
@HeikoSbut the old SG_ADD just gets a base class pointer12:35
@wikingyeah but that's easy to drop right?12:35
@HeikoSSG_ADD( (CSGObject**) bla ...)12:35
@wikingyes12:35
@HeikoSso you are saying drop that12:35
@wikings/(CSGObject**)//g12:35
@HeikoSand then deal with the rest inside12:35
@wikingyes12:35
@HeikoSyeah sure12:36
@HeikoSbetter12:36
@wikingbecause this is adding more lines12:36
@wikingas well as12:36
@wikingyou'll ahve to do dropping12:36
@wikingmultiple files12:36
@HeikoSthe m_parameter will go soon12:36
@wikingwhen Parameters.cpp12:36
@wikingis out12:36
@wikingright?12:36
@HeikoSyes12:36
@HeikoSall m_parameter calls will go12:36
@wikinginstead youi could jsut drop that12:36
@HeikoSonly watch will stay12:36
@wikingbehind SG_ADD12:36
@HeikoSyes12:36
@wikingand that's 1 change12:36
@HeikoSaye aye12:36
@wikingand if tomorrow it's watchMagicLand12:37
@wikingthen its sstill one line12:37
@HeikoSwatchMagicStringToTypeConverter_TM___12:37
@HeikoSwiking: you know what will be fun12:37
@HeikoSchanging all the notebooks12:37
@HeikoSwith new api12:37
@wikingmmm12:37
@HeikoScan't wait12:37
@wikingset/get?12:37
@HeikoSyes12:38
@HeikoSthere is awesome stuff12:38
@wikingthat is a bit regex-able12:38
@HeikoSlike in GaussianKernel12:38
@wikingsome parts12:38
@HeikoSyou have12:38
@wikingnot all12:38
@HeikoSset_width12:38
@HeikoSwhich does12:38
@HeikoSm_log_width=CMath::log(w/2.0)/2.0;12:38
@HeikoSbut with put/get, we have to directly set log_width12:38
@wikingi mean for the time being12:39
@wikingif you done wanna be that harsh of changing stuff12:39
@wikingyou can just put the put under the setter12:39
@HeikoSyep12:39
@HeikoSthats the way12:39
@HeikoSbut thats all later I guess12:39
@HeikoSfirst one example needs to work and be clean :)12:40
@wikingHeikoS, http://buildbot.shogun-toolbox.org:8080/#/builders/4/builds/110/steps/7/logs/stdio13:26
@wikingthis doesn't look like a problem with some python package13:26
@wikingor?13:26
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has joined #shogun13:29
travis-ciit'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/34426632313:29
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has left #shogun []13:29
@wikingHeikoS, btw i dont have contact to wu (only his waterloo email that i reckon is dead)13:36
@wikingcould you drop him a gsoc endorsment line13:36
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has joined #shogun13:50
travis-ciit'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/34426632313:50
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has left #shogun []13:50
@HeikoSwiking: will drop wu a line!14:07
@HeikoSwiking: Ill dig out the package error14:08
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has joined #shogun14:10
travis-ciit's Heiko Strathmann'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/34427108814:10
-!- travis-ci [~travis-ci@ec2-54-221-154-197.compute-1.amazonaws.com] has left #shogun []14:10
-!- travis-ci [~travis-ci@ec2-54-167-169-20.compute-1.amazonaws.com] has joined #shogun14:40
travis-ciit's Heiko Strathmann'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/34427108814:40
-!- travis-ci [~travis-ci@ec2-54-167-169-20.compute-1.amazonaws.com] has left #shogun []14:40
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/b6900ce5383361d1cb8b2397f56012b9d181a8db by karlnapf15:05
-shogun-buildbot:#shogun- Build deb1 - libshogun #315 is complete: Failure [failed configure (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/10/builds/31515:05
@HeikoSshogun-buildbot: force build 'nightly_bsd_license'15:05
-!- HeikoS [~heiko@host86-129-231-92.range86-129.btcentralplus.com] has quit [Quit: Leaving.]15:06
@wikingheiko that one broke gpl15:18
@wikingCMake Error at cmake/FindMetaExamples.cmake:52 (LIST):15:18
@wiking  list sub-command REMOVE_DUPLICATES requires list to be present.15:18
@wikingCall Stack (most recent call first):15:18
@wiking  cmake/FindMetaExamples.cmake:61 (get_excluded_meta_examples)15:18
@wiking  tests/meta/CMakeLists.txt:35 (find_meta_examples)15:18
-shogun-buildbot:#shogun- Build nightly_bsd_license #111 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/4/builds/11115:43
-!- slayerjain [uid117577@gateway/web/irccloud.com/x-aplxahevpzmwpkbz] has joined #shogun16:02
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun17:14
-!- mode/#shogun [+o HeikoS] by ChanServ17:14
@HeikoSwiking: http://buildbot.shogun-toolbox.org:8080/#/builders/4/builds/11117:19
@HeikoSnot sure where the error went, but now the thing is green :)17:19
@HeikoSlisitsyn: see email17:25
@HeikoSlisitsyn: also, clone? :D17:25
@wikingHeikoS, yeah but now you broke the other17:25
@wikingHeikoS, http://buildbot.shogun-toolbox.org:8080/#/builders/10/builds/315/steps/3/logs/stdio17:25
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/297f9a0c5d4771154515fef2fac865fdc1a1af5e by karlnapf17:34
-shogun-buildbot:#shogun- Build deb1 - libshogun #316 is complete: Success [build successful] - http://buildbot.shogun-toolbox.org:8080/#builders/10/builds/31617:38
@HeikoSwiking: tada17:44
-!- slayerjain [uid117577@gateway/web/irccloud.com/x-aplxahevpzmwpkbz] has quit [Quit: Connection closed for inactivity]18:12
@HeikoSwiking: http://buildbot.shogun-toolbox.org:8080/#/builders/14/builds/10018:15
@HeikoSthis was the one with the missing python-ply package18:15
-!- travis-ci [~travis-ci@ec2-54-198-188-114.compute-1.amazonaws.com] has joined #shogun18:38
travis-ciit's Heiko Strathmann'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/34440111818:38
-!- travis-ci [~travis-ci@ec2-54-198-188-114.compute-1.amazonaws.com] has left #shogun []18:38
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 264 seconds]19:45
--- Log closed Thu Feb 22 00:00:12 2018
« Tuesday, 2018-02-20 Index Thursday, 2018-02-22 »

Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!