IRC logs of #shogun for Tuesday, 2018-05-29

« Monday, 2018-05-28 Index Wednesday, 2018-05-30 »
--- Log opened Tue May 29 00:00:26 2018
-!- wuwei [wuweilinma@gateway/shell/matrix.org/x-pfqiuxsyziajvgrl] has quit [Ping timeout: 277 seconds]04:42
-!- Trixis [~Trixis@unaffiliated/trixis] has quit [Ping timeout: 260 seconds]04:45
-!- Netsplit *.net <-> *.split quits: @sukey, tctara_, lisitsyn, shogun-buildbot, durovo1, @ChanServ04:47
-!- Netsplit *.net <-> *.split quits: @wiking04:47
-!- Netsplit *.net <-> *.split quits: zxtx, rcurtin, sonney2k_, micmn, shogitter04:47
-!- Netsplit over, joins: lisitsyn, shogun-buildbot, durovo1, @wiking, @ChanServ, @sukey, micmn, tctara_, rcurtin, zxtx (+2 more)04:48
-!- Trixis [~Trixis@unaffiliated/trixis] has joined #shogun04:48
-!- Netsplit *.net <-> *.split quits: @sukey, tctara_, durovo104:50
-!- Netsplit over, joins: durovo1, tctara_04:51
-!- Netsplit over, joins: @sukey04:52
-!- sukey2 [~nodebot@ks312251.kimsufi.com] has joined #shogun04:52
-!- sukey [~nodebot@ks312251.kimsufi.com] has quit [Remote host closed the connection]04:52
-!- durovo1 [~durovo@31.8f.559e.ip4.static.sl-reverse.com] has quit [Write error: Broken pipe]04:52
-!- durovo [~durovo@31.8f.559e.ip4.static.sl-reverse.com] has joined #shogun04:52
-!- wuwei1 [wuweilinma@gateway/shell/matrix.org/x-lwvcxvbgbdmrbwiy] has joined #shogun05:12
-!- durovo2 [~durovo@31.8f.559e.ip4.static.sl-reverse.com] has joined #shogun08:27
-!- durovo [~durovo@31.8f.559e.ip4.static.sl-reverse.com] has quit [Remote host closed the connection]08:27
@wikingwuwei1, hey09:49
wuwei1wiking: hey10:21
sukey2[https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4311 vigsterkr added label: "BUG"10:37
sukey2[https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4311 vigsterkr added label: "performance improvement"10:37
sukey2[https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4311 opened by vigsterkr10:37
sukey2[https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4311 vigsterkr added label: "development tasks"10:37
@wikingwuwei1, still here?11:11
@wikingmicmn, ping?11:13
@wiking:)11:13
wuwei1wiking: i'm here11:27
@wikingwuwei1, sent an email instead :)11:28
-!- Farouk [9cc5aeaf@gateway/web/freenode/ip.156.197.174.175] has joined #shogun13:06
sukey2[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4285 synchronized by vinx1313:11
-!- Farouk [9cc5aeaf@gateway/web/freenode/ip.156.197.174.175] has quit [Ping timeout: 260 seconds]13:15
-!- Farouk [9cc5aeaf@gateway/web/freenode/ip.156.197.174.175] has joined #shogun13:19
wuwei1wiking: could u explain a bit composite in pipeline here https://github.com/shogun-toolbox/shogun/wiki/Hackathon-2017-base-api#some-cool-stuff13:21
sukey2[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4285 synchronized by vinx1313:28
Faroukwiking, are you here? Have a quick question on the new API13:36
sukey2[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4312 opened by abhishek-199114:23
@wikinghello hello14:29
@wikingFarouk, sorry i had some offline things to do14:29
@wikingbut now i'm here14:29
@wikingwuwei1, oh yeah14:29
FaroukHey wiking. No worries. I think I found a similar problem with RegressionLabels while changing some examples to the new API14:34
FaroukIf I use the old API with RegressionLabels it works fine14:34
Faroukbut if I switch to the new API I get:14:34
@wikingwuwei1, i guess your question is what Composite stands for?14:34
Farouk[ERROR] In file /tmp/src/shogun/src/shogun/labels/Labels.h line 143: Sorry, not yet implemented .14:35
shogitter(vinx13) wiking: yeah14:35
@wikingFarouk, do you have part of this pushed into a pr?14:35
FaroukNo its the examples I've been porting since yesterday. Wait let me push it and I will let you know14:36
@wikingwuwei1, ok so that is about somehow be able to compose multiple machines14:36
@wikingFarouk, that'd be good thanks14:36
shogitter(vinx13) wiking: i'm not very sure about how things work, if not linear but DAG14:36
@wikingwuwei1, because say we have Bagging you could do a Bagging(LibSVM, NN)14:36
@wikingand then you have an 'averaging' in Bagging14:37
@wikingof the result14:37
shogitter(vinx13) btw getting some problem with irc client so i'm using gitter now14:37
shogitter(vinx13) yeah, that's equivalent to: bagging = new Bagging(); bagging.add(..) right?14:39
@wikingvinx13 sure no problem at all14:41
@wikingvinx13 exactly14:41
@wikingi mean first we can go with the builder pattern14:41
@wikingfor linear pipelines14:41
shogitter(vinx13) yeah14:42
@wikingand then we can add later the composite thing14:42
@wikingbut yeah as you can see14:42
@wikingi'm not so sure if we need to support this kind of api14:43
@wikingwith(trans, IS_FITTED).14:43
@wikingi.e. the boolean for fit or not fit14:43
@wikingone should assume that it's not fit14:43
shogitter(vinx13) we should ignore those IS_FITTED when calling train_machine?14:44
@wikingnono14:46
@wikingi dont think that api actually is necessary14:46
@wikingwe just have14:46
@wikingwith(trans)14:46
@wikingand that's all14:46
@wikinglets see later if we wanna support that14:46
@wikingbut i'm really not sure if we need an api like that14:47
shogitter(vinx13) sure14:47
shogitter(vinx13) btw the composite should only work with a few machines like Bagging ?14:47
shogitter(vinx13) ah.. have checked BagingMachine, it doesn't support multiple kinds of machines, maybe we should use combination rule for composite ?14:50
sukey2[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4309 synchronized by FaroukY15:02
Faroukwiking, check the synchronized pr in the file rain_forest_regression.sg15:03
FaroukRegressionLabels ==> works ok, Labels, get the error15:04
sukey2[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4309 synchronized by FaroukY15:19
sukey2[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4310 synchronized by shubham80815:33
sukey2[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4285 synchronized by vinx1315:42
@wikingFarouk, ok here sorry15:49
@wikinghad some meetings15:49
@wikingvinx13 yeah that's why i sad for the time being... :)15:49
@wikingvinx13 we'll refactor a bit the Bagging Machine15:49
@wikingbut yeah for the time being just think about pipeline as a linear thing15:49
@wikingFarouk, ok i'll check now the pr15:49
@wikingFarouk, it's this right: https://github.com/shogun-toolbox/shogun/pull/430915:50
Faroukwiking, no worries, im here all day~15:50
@wikingFarouk, https://github.com/shogun-toolbox/shogun/pull/4309/files#diff-a89b037ce72a965cb4fdf35083d4897fR915:50
@wikingsorry15:50
@wikingi dont know why this doesn't wanna take a line15:50
@wikingok so Labels labels_train = labels(f_labels_train)15:51
@wiking-RegressionLabels labels_test(f_labels_test)15:51
@wiking+Labels labels_test = labels(f_labels_test)15:51
@wikingis the thing that gives you the erro?15:51
@wikingkernel_ridge_regression_nystrom.sg15:51
@wiking?15:51
sukey2[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4312 merged by vigsterkr15:53
sukey2[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/7e9699e5a1c7df75271e8dd26fc71e615e8138b4 by vigsterkr15:53
@wikingwuwei1, ok lets see what the CIs are saying for your transformer thingy and then i'll merge15:54
FaroukI have no idea what just happened !15:54
FaroukI just tested it again and somehow it works15:54
Faroukwhen it was the new API it just told me not implemented15:54
shogitter(vinx13) wiking: sure15:55
@wikingFarouk, ok lemme know when you know where it actually happens15:55
@wikingand which example can i test15:55
@wikingvinx13: https://github.com/shogun-toolbox/shogun/pull/4285/files#r19143267715:55
@wiking:P15:55
Faroukno it just worked again! I've been running the test since morning, and its been failing, then i just pushed to github, then recompiled and retested and the tests passed!15:56
Faroukwait make meta_examples generates the tests right?15:56
@wikingshould15:57
@wikinglets see what the CIs react15:57
@wiking :)15:57
shogitter(vinx13) wiking: i will rename apply -> transform now :)15:57
@wikingvinx13 :) thnx15:57
@wikingvinx13 have you had a chance to think about the ngraph story? :)15:58
shogitter(vinx13) had a quick look, the apis look good15:59
shogitter(vinx13) i'm thinking about cpu/gpu things15:59
@wikingyeah it's much smoother/clearer15:59
@wikingvinx13 i mean they kind of like solved that :)15:59
@wikingbecause the way they create tensors (matrix sgvector etc)16:00
@wikingthat actually you call the backend16:00
@wikingto do that16:00
@wikingso they have Backend::create_tensor(Type, Shape)16:00
@wikingso that assures you that it's on the right device16:00
@wikingbecause if its a GPUBackend16:00
@wikingyou'll get a tensor on the GPU16:00
@wikingetc16:00
@wikingit's a nice and clean design imo16:00
shogitter(vinx13) yeah, i'm thinking how we deal with those operations like index on matrix16:01
@wikingwell16:01
@wikingyou know16:01
@wikingthat's the trick here16:01
@wikingyou actually dont do direct operations on the matrix :)16:01
@wikingeverything is a Function16:01
shogitter(vinx13) of course16:01
shogitter(vinx13) but unfortunately algos need that, for now16:02
@wikingyeah but we can start with a simple algo16:03
shogitter(vinx13) i agree that operations should implemented as objects16:03
@wikingand try to have a design for that16:03
@wikingbecause i mean i hope you've seen16:03
@wikingthat in ngraph16:03
@wikingactually the operations are nothing else16:03
@wikingbut writing C++ code :D16:03
@wikingliterally16:03
@wikingop = streamstream << "Eigen...." << ..... << std::endll;16:04
@wikingand then these string streams are compiled with an JIT :)16:04
@wikingso it's a bit crazy16:04
-!- HeikoS [~heiko@host81-153-166-11.range81-153.btcentralplus.com] has joined #shogun16:04
-!- mode/#shogun [+o HeikoS] by ChanServ16:04
@wikingbut this way it's highly optimized16:04
@wikingwe dont need this atm16:04
@wikingit'd be better already to get rid of those macro generated16:04
@wikingvirtual methods16:04
@wikingand not to have that interface an ever extending thing16:05
@wikingmeaning even if we have a container that is typeless16:05
@wikingthe linalg api is bound to be ever extending if we dont support thing like Function :)16:05
@wikingas then we'll add things to it ad-hoc16:05
@wiking(various decomposers etc.)16:05
@wikingFarouk, ok the cis are done16:06
@wikinglemme check it16:06
@wikingor at least travis is done16:06
FaroukIs there a problem with Java and C# ?16:07
@wikingFarouk, https://travis-ci.org/shogun-toolbox/shogun/jobs/385195492#L325316:07
@wikinghttps://travis-ci.org/shogun-toolbox/shogun/jobs/385195493#L288816:07
@wikingdiffernet places it fails16:07
@wikingin case of java16:07
@wikingit's about EvaluationResult casting16:07
@wikingin case of c# it's16:08
@wiking): error CS0266: Cannot implicitly convert type `Kernel' to `CombinedKernel'. An explicit conversion exists (are you missing a cast?)16:08
@wikinghttps://github.com/shogun-toolbox/shogun/pull/4309/files#diff-850db6d5082e9cf6bbf0ddb8e5ad2ab3R6016:08
@wikingit should be16:09
@wikingKernel k = kernel(machine_mkl.get("kernel"))16:09
@wikinginstead of CombinedKernel k = kernel(machine_mkl.get("kernel"))16:09
@wiking RealVector w = k.get_subkernel_weights()16:09
@wikingno?16:09
FaroukOhh yea! Sorry about that!16:10
Faroukfor CrossValidationResult result = cross.evaluate(), do I need an obtain from generic?16:14
@wikingjust submitted a mini review16:15
@wikingbut yeah16:15
sukey2[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4309 synchronized by FaroukY16:29
-!- sukey2 [~nodebot@ks312251.kimsufi.com] has quit [Remote host closed the connection]16:35
-!- sukey [~nodebot@ks312251.kimsufi.com] has joined #shogun16:35
-!- mode/#shogun [+o sukey] by ChanServ16:35
@wikingsukey, flip16:35
@sukey(/?? ? ?)/? ~ ???16:35
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has joined #shogun17:16
shubham808HeikoS hey17:16
@HeikoSshubham808: hi!17:16
shubham808so lets start with the premature test stuff17:17
@HeikoSshubham808: so tomorrow 15 utc is good17:17
@HeikoSlets meet here17:17
@HeikoSalternatively I also have some time now17:17
@HeikoSok17:17
shubham808now is good17:17
shubham808for me too17:17
shubham808so17:18
shubham808there is a pr to drop jinja2 from serialization tests17:18
shubham808we could add our own test there17:20
shubham808like: https://pastebin.com/wS90K7Gx17:21
shubham808u see line 46 ?17:22
shubham808*4317:22
@HeikoSchecking17:25
@HeikoSyes thats a good idea17:26
-!- travis-ci [~travis-ci@ec2-184-73-117-198.compute-1.amazonaws.com] has joined #shogun17:26
travis-ciit's Abhishek Kumar Srivastava'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/38521123017:26
-!- travis-ci [~travis-ci@ec2-184-73-117-198.compute-1.amazonaws.com] has left #shogun []17:26
@HeikoSshubham808: thats a good thing to work towards to, need to merge that PR soon then I guess?17:26
shubham808yeah17:26
shubham808but there are still a few problems17:27
shubham808first is the continue/break thing in the macro17:28
shubham808next is the ASSERT_TRUE(i)17:28
shubham808this is to make sure we are stopped by the callback17:31
shubham808but we will not trigger this if the algorithm terminates in a single iteration17:34
shubham808any ideas for a work around ?17:34
@HeikoSshubham808: sorry something came up17:53
@HeikoSI think algorithms that terminate within a single iteration should not be tested17:53
@HeikoSas it doesnt make sense17:53
shubham808it depends on the dataset17:54
shubham808i think17:54
shubham808yeah we could just not test those17:55
@HeikoSdo you have an example?17:57
@HeikoSlike an algorithm that terminates immediately?17:57
@HeikoSI have no feel;ing for what happens in that case17:57
@HeikoSshubham808: but I think for now, we could just pass/skip the test17:57
@HeikoSwe can deal with this problem later17:57
shubham808alright17:58
shubham808now about the algorithms we will be testing18:00
@HeikoSyep18:00
@HeikoSshubham808: so?18:01
shubham808let me find the line link18:01
shubham808just a sex18:01
shubham808*sec18:01
shubham808https://github.com/micmn/shogun/blob/11b1a342c96a8c1b51c5c8514fd0cb94c09cac88/tests/unit/base/trained_model_serialization_unittest.cc.py#L13018:03
shubham808i have added another type list next to it18:03
shubham808we will be limited by the number of algorithms the serialization test works on18:04
shubham808as a start i made the new type list identical to Machine list18:05
-!- travis-ci [~travis-ci@ec2-54-81-11-238.compute-1.amazonaws.com] has joined #shogun18:06
travis-ciit's Abhishek Kumar Srivastava'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/38521123018:06
-!- travis-ci [~travis-ci@ec2-54-81-11-238.compute-1.amazonaws.com] has left #shogun []18:06
@HeikoSok sure18:07
@HeikoSseems ok18:07
@HeikoS(for now)18:07
shubham808yeah18:07
shubham808for now18:07
shubham808alright this should be enough for a wip pr18:07
shubham808we could rebase it later when we merge the drop jinja2 pr18:08
@HeikoSyep exactly18:08
@HeikoSIm keen on getting that one in18:08
@HeikoSprobably tomorrow or so18:08
shubham808okay cool18:08
@HeikoS\18:09
@HeikoSshubham808: I will have to run soon, shall we still have our meeting tomorrow?18:10
@HeikoSTo discuss next steps and stuff?18:10
shubham808yes sure18:10
shubham808bye18:11
@HeikoSshubham808: thx!18:11
@HeikoSIll have more time tomorrow!18:11
shubham808great :)18:12
@HeikoSjust catching up with all those emails >(18:12
@HeikoS:)18:12
-!- 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]18:17
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4310 synchronized by shubham80818:22
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4313 opened by karlnapf18:42
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4285 synchronized by vinx1319:41
-!- HeikoS [~heiko@host81-153-166-11.range81-153.btcentralplus.com] has quit [Ping timeout: 240 seconds]20:33
-!- HeikoS [~heiko@host81-153-166-11.range81-153.btcentralplus.com] has joined #shogun20:41
-!- mode/#shogun [+o HeikoS] by ChanServ20:41
-!- HeikoS [~heiko@host81-153-166-11.range81-153.btcentralplus.com] has quit [Ping timeout: 256 seconds]20:49
-!- Farouk [9cc5aeaf@gateway/web/freenode/ip.156.197.174.175] has quit [Ping timeout: 260 seconds]21:40
--- Log closed Wed May 30 00:00:28 2018
« Monday, 2018-05-28 Index Wednesday, 2018-05-30 »

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