IRC logs of #shogun for Monday, 2018-02-05

« Sunday, 2018-02-04 Index Tuesday, 2018-02-06 »
--- Log opened Mon Feb 05 00:00:49 2018
-!- RIANRFteamextr [~fxpmeg@176.215.246.142] has joined #shogun00:52
RIANRFteamextr???????????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  vmrac: shogun-buildbot ironstark durovo ??????????00:52
RIANRFteamextr???????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  zzqwwyqbc: durovo sukey tctara ?????????????????00:52
RIANRFteamextr??????????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  wdqjtklih: durovo shogun-buildbot witness ??????????????00:52
RIANRFteamextr???????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  xciomr: sonney2k_ rcurtin sukey ?????????????00:52
RIANRFteamextr???????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  sgoaw: durovo rcurtin sonney2k_ ????????????00:52
RIANRFteamextr????????????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  ehtpfcvw: nikhilweee shogun-buildbot ironstark ????????????00:52
RIANRFteamextr?????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  pywdca: nikhilweee sukey shogun-t1olbox ???????????????????00:52
RIANRFteamextr??????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  aybfhzfja: zxtx tctara shogun-t1olbox ????????????????????00:52
RIANRFteamextr??????????????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  oknrawydam: rcurtin shogitter ironstark ????????????00:52
RIANRFteamextr???????????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  ktkxpiart: nikhilweee shogun-buildbot shogun-t1olbox ?????????????00:52
RIANRFteamextr??????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  bpyxqekt: lisitsyn shogun-buildbot rcurtin ????????????????00:52
RIANRFteamextr?????????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  ftwmuzrwc: zxtx sonney2k_ durovo ?????????????????00:53
RIANRFteamextr??????????????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  zyekkb: shogun-t1olbox rcurtin tctara ???????????00:53
RIANRFteamextr???????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  nnhicgvi: Trixis shogun-buildbot sonney2k_ ?????????????????00:53
RIANRFteamextr???????????????????? IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  xsinvr: shogun-t1olbox wiking durovo ???????????00:53
RIANRFteamextrâ–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„ IRC.SUPERNETS.ORG #SUPERBOWL SUPERBOWL PARTY RIGHT NOW LIVE CHATTING AND PLAY BY PLAY COMMENTARY GOING ON NOW!! ASK CHRONO FOR DETAILS!!  tkeehlb: sukey ironstark Trixis â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â–„â00:53
-!- RIANRFteamextr [~fxpmeg@176.215.246.142] has quit [Client Quit]00:53
-!- HeikoS [~heiko@host86-129-231-92.range86-129.btcentralplus.com] has joined #shogun00:56
-!- mode/#shogun [+o HeikoS] by ChanServ00:56
rcurtinman what is with the spam these days00:58
rcurtinbasically every channel I am in is being hit00:58
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4146 closed by karlnapf00:59
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4146 reopened by karlnapf00:59
-!- HeikoS [~heiko@host86-129-231-92.range86-129.btcentralplus.com] has quit [Ping timeout: 276 seconds]01:04
-shogun-buildbot:#shogun- Build nightly_all #88 is complete: Failure [test (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/22/builds/8803:46
-!- saputro [44c88d72@gateway/web/freenode/ip.68.200.141.114] has joined #shogun05:07
-!- witness [uid10044@gateway/web/irccloud.com/x-tkiutxzylirbower] has quit [Quit: Connection closed for inactivity]05:27
-!- saputro [44c88d72@gateway/web/freenode/ip.68.200.141.114] has quit [Ping timeout: 260 seconds]06:03
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4085 synchronized by vinx1308:17
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/408508:18
-!- HeikoS [~heiko@host86-129-231-92.range86-129.btcentralplus.com] has joined #shogun09:24
-!- mode/#shogun [+o HeikoS] by ChanServ09:24
-!- HeikoS [~heiko@host86-129-231-92.range86-129.btcentralplus.com] has quit [Ping timeout: 240 seconds]09:28
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4085 synchronized by vinx1309:35
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun10:36
-!- mode/#shogun [+o HeikoS] by ChanServ10:36
@HeikoSlisitsyn: you there?10:51
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4148 merged by karlnapf10:51
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/f3ad96b0d2a36e7f4a53a61dbc911d590fe41eb3 by karlnapf10:51
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4149 opened by vinx1311:28
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4150 opened by karlnapf11:28
-!- travis-ci [~travis-ci@ec2-54-234-36-74.compute-1.amazonaws.com] has joined #shogun11:41
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/33747190111:41
-!- travis-ci [~travis-ci@ec2-54-234-36-74.compute-1.amazonaws.com] has left #shogun []11:41
-!- travis-ci [~travis-ci@ec2-54-226-10-225.compute-1.amazonaws.com] has joined #shogun12:04
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/33747190112:04
-!- travis-ci [~travis-ci@ec2-54-226-10-225.compute-1.amazonaws.com] has left #shogun []12:04
@HeikoSwiking: jojo12:08
@HeikoSwiking: you said you had removed the clone_equals tests locally? You can push that into develop now, I put a replacement test up that is much faster to compile12:08
@HeikoSI can do the same with serialization soon12:08
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4151 opened by shubham80812:17
@wikingHeikoS, ho12:33
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4152 opened by karlnapf12:36
@HeikoSwiking: btw serialization ascii is already covered so that can be removed as well12:38
@HeikoSlisitsyn:13:19
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4150 merged by karlnapf13:23
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/acb321d3e5eab746055b11d6e111fb5ade239443 by karlnapf13:23
lisitsynHeikoS: ya13:26
@HeikoSlisitsyn: I need two things from you :)13:27
@HeikoSthey are small, promise13:27
lisitsynyeah yeah sure13:27
@HeikoShttps://github.com/shogun-toolbox/shogun/pull/414813:27
@HeikoSclone for arrays13:27
lisitsynaha13:27
@HeikoSthe tests are there and the new clone as well, they just fail13:28
lisitsynthat's on my todo list13:28
@HeikoSjust one case, I can fill in gaps once that is there13:28
lisitsynwhat else?13:28
@HeikoSsecond thing:13:28
@HeikoSSG_ERROR("Type error when setting parameter %s::%s: expected %s but got %s.\n",13:28
@HeikoSget_name(), _tag.name().c_str(),13:28
@HeikoS"X", "Y");13:28
lisitsynah ok ok13:28
@HeikoSin SGObject.h13:28
lisitsyntype13:28
@HeikoSlisitsyn: type is faster I guess?13:29
@HeikoSso maybe then that would be good first13:29
lisitsynHeikoS: yeah let me dive into that once I wait for something to compile or compute13:29
lisitsyn:)13:29
@HeikoSI can do stuff on meta API while you do the clone13:29
@HeikoSkk13:29
@HeikoSthx13:29
@HeikoSIll have lunch now, will check back later this afternoon13:29
@HeikoSvoid put(const Tag<T>& _tag, const T& value) throw(ShogunException)13:30
@HeikoSthis is the method that should print type, both expected and what it got13:30
@HeikoSlisitsyn: uuuuh13:36
@HeikoSlisitsyn: problems with the type system ..13:36
@HeikoSwill tell you later13:36
lisitsynuhm?13:37
lisitsynHeikoS: what's wrong?13:37
Trixiswiking: ill submit the requested cleanup later today. its ready needs formatting fixc13:46
@wikingTrixis, great! :)13:52
@HeikoSlisitsyn: the tag system is not aware of subclasses14:05
@HeikoSsoooo14:05
@HeikoSsvm.put("kernel", k)14:05
@HeikoSwhat type is k?14:05
lisitsynHeikoS: putKernel14:06
@HeikoSah14:06
lisitsynthen it is Kernel14:06
@HeikoSuh14:06
lisitsynyes14:06
@HeikoSso goodbye kwargs14:06
lisitsynHeikoS: not really14:06
lisitsynstill possible14:06
@HeikoSbut then need to keep track of the type14:06
lisitsynHeikoS: we can dispatch to most specific type14:07
lisitsynit is like that:14:07
lisitsyndef put(name, obj):14:07
lisitsyn   if ... then putKernel()14:07
lisitsynlala14:07
lisitsynso some hackery just for python14:07
@HeikoSyeah14:07
@HeikoSmmmh14:07
@HeikoSbut the meta kwargs thing then depends on that14:07
@HeikoSpitty14:07
@HeikoScurrently, it just translates to "put"14:08
lisitsynoh well14:08
lisitsynHeikoS: it depends on the target language then14:08
@HeikoSbut if we keep track of the type and then have a list of "putKernel", "putVla", thats not nice14:08
lisitsynsome languages already know how to dispatch that overloading14:08
lisitsynwe have to have putKernel, putBla putLa14:08
@HeikoSyeah14:09
@HeikoSso that is not part of the c++ code14:09
lisitsynat least for langs that do not support overloading properly14:09
@HeikoSbut interface generated code14:09
lisitsynyes14:09
@HeikoSkk14:09
@HeikoSlisitsyn: ok let's see14:09
@HeikoSfor the basic types, the kwargs thing should still work14:10
@HeikoSlisitsyn: where is putKernel defined?14:10
lisitsynHeikoS: macro in swig14:11
lisitsynSGBase.i14:11
lisitsyniirc14:11
@HeikoSkk14:11
@HeikoSkernel.set("width", "???? ????");14:12
@HeikoSshogun wiki :D14:12
lisitsyn:D14:12
lisitsynHeikoS: hahah lol14:12
lisitsynyou better now know wwhat it means14:12
lisitsynnot*14:12
@HeikoSLOL14:12
@HeikoSlisitsyn: so in that doc14:14
@HeikoSwe said that every base class will get its own tag14:14
@HeikoSin SWIG we would have IntTag, StringTag, KernelTag (stands for Tag<int>, Tag<string>, Tag<CKernel>)14:14
lisitsynHeikoS: yes14:15
@HeikoSso there seems to be some stuff already14:16
@HeikoShttps://github.com/shogun-toolbox/shogun/wiki/README_tags14:17
@HeikoShttps://github.com/shogun-toolbox/shogun/wiki/README_tags#python-api14:17
@HeikoSw = gkernel.getsFloat("log_width")14:17
@HeikoSwhere is that implemented... ? in the .i files?14:17
lisitsynHeikoS: yes, get<float> renamed to getFloat14:18
lisitsynin SGBase.i14:18
lisitsyndid you found?14:18
@HeikoSSUPPORT_TAG(Object, object, CSGObject*)14:19
lisitsynyes exactly14:20
@HeikoSok Ill reverse engineer this a bit14:20
@HeikoSdidnt know about it yet :)14:20
lisitsynHeikoS: well it is just rename for swig14:20
lisitsynin C++ there are templates14:20
lisitsynso we rename for a limited set of types14:20
@HeikoSso if I add14:20
@HeikoSSUPPORT_TAG(Kernel, kernel, CKernel*)14:20
lisitsynyes14:20
lisitsynputKernel getKernel14:20
@HeikoStrying14:21
@HeikoSlisitsyn: okok14:22
@HeikoSlisitsyn: but C++14:22
@HeikoS?14:22
@HeikoSthis is only swig14:22
lisitsynget<CKernel*>14:22
@HeikoSmeta lang14:22
@HeikoSI cannot put putKernel(...) then14:23
@HeikoSsince it is translated to c++ as well14:23
@HeikoSso then meta lang needs another extension I guess14:23
@HeikoSput<CKernel>("kernel", k)14:23
@HeikoSwhich is translated to putKernel in the other interfaces14:23
@HeikoShallelujah14:23
@HeikoSlisitsyn: or you have other ideas?14:24
@HeikoSlisitsyn: actually14:25
lisitsynHeikoS: don't get it14:25
@HeikoSthe meta language14:25
lisitsynah you mean we don't have putKernel in C++14:25
@HeikoSneeds to express this API somehow14:25
@HeikoSand this is translated to c++ and swig itnerfaces14:26
@HeikoSyeah14:26
@HeikoSlisitsyn: why can't we have a put(CSGObject*, ...)14:26
@HeikoSthat tries to convert the argument to the type of the parameter?14:26
@HeikoSso I can actually do svm.put("kernel", k)14:26
lisitsynwe can have one14:26
@HeikoSlike in C++14:26
lisitsynyeah we can14:26
@HeikoSjust thinking how that would look like14:27
@HeikoSlisitsyn: can you add that on top of your list? :D14:27
lisitsynyeah but that I have to think about14:27
@HeikoSlisitsyn: I mean14:28
@HeikoSwe can also modify the meta lang14:28
@HeikoSto accept14:28
@HeikoSsvm.put<CKernel>("kernel", k)14:29
@HeikoSbut it is not really nice14:29
@HeikoSmakes it even more complex14:29
@HeikoSI think we should keep that down14:29
@HeikoSrather have the svm.put(...)14:29
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4146 synchronized by grg12114:32
@HeikoSlisitsyn: let me know if you have an idea how to solve that14:32
lisitsynHeikoS: it's a bit tricky as there are things like float and double14:32
lisitsynfor python they are just the same14:33
@HeikoSso both are accepted14:34
@HeikoS?14:34
-!- travis-ci [~travis-ci@ec2-54-226-10-225.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/33752385314:40
-!- travis-ci [~travis-ci@ec2-54-226-10-225.compute-1.amazonaws.com] has left #shogun []14:40
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4149 merged by karlnapf14:42
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/82c499bea7e49a92ce0287be94ea4b470a1f1aab by karlnapf14:42
@HeikoSlisitsyn: any eta or thoughts? then I can plan how much I get involved into this stuff now?14:53
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4085 synchronized by vinx1314:59
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4153 opened by dhelekal15:07
Trixiswont pass style checker :P15:07
Trixis(because i cant get the 3.8 version, and the latest produces different results)15:08
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/408515:17
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/408515:17
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4153 synchronized by dhelekal15:19
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4153 synchronized by dhelekal15:19
-!- travis-ci [~travis-ci@ec2-54-80-202-110.compute-1.amazonaws.com] has joined #shogun15:42
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/33752385315:42
-!- travis-ci [~travis-ci@ec2-54-80-202-110.compute-1.amazonaws.com] has left #shogun []15:42
@HeikoSlisitsyn: sorry to pester you .... ;D15:56
durovoHeikoS: I am trying to fix fit_sigmoid.. would it be alright to remove scores_to_probabilties from Cbinarylabels? I have already written CalibrationMethod and SigmoidCalibrationMethod, wouldn't it be better to use them to calibrate the predictions? I think that this approach would be more consistent in the long run if new calibration methods are added later.15:57
@HeikoSabsolutely15:58
@HeikoSI also suggest different names: Calibration, SigmoidCalibration ...15:58
@HeikoSwithout the "Method"15:58
durovoOk15:58
-!- travis-ci [~travis-ci@ec2-54-80-202-110.compute-1.amazonaws.com] has joined #shogun16:26
travis-ciit's Wuwei Lin'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/33755290716:26
-!- travis-ci [~travis-ci@ec2-54-80-202-110.compute-1.amazonaws.com] has left #shogun []16:26
-!- travis-ci [~travis-ci@ec2-54-226-10-225.compute-1.amazonaws.com] has joined #shogun16:57
travis-ciit's Wuwei Lin'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/33755290716:57
-!- travis-ci [~travis-ci@ec2-54-226-10-225.compute-1.amazonaws.com] has left #shogun []16:57
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 248 seconds]17:08
-!- HeikoS [~heiko@82-132-230-62.dab.02.net] has joined #shogun17:11
-!- mode/#shogun [+o HeikoS] by ChanServ17:11
-!- HeikoS [~heiko@82-132-230-62.dab.02.net] has quit [Remote host closed the connection]17:16
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4152 merged by karlnapf17:31
@sukey[https://github.com/shogun-toolbox/shogun] karlnapf pushed 2 commits:17:31
@sukeyhttps://github.com/shogun-toolbox/shogun/commit/902ee810d5c0eae0361045e87bbe33899b3b8fc517:31
@sukeyhttps://github.com/shogun-toolbox/shogun/commit/91adea73f9b0b9355e9ce5e240cf8ba2d03ee14c17:31
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4146 merged by karlnapf17:34
@sukey[https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/daa95fa1cfe41e97a229dc241925183c7cb6dbed by karlnapf17:34
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun17:35
-!- mode/#shogun [+o HeikoS] by ChanServ17:35
@HeikoSlisitsyn: jo!17:35
lisitsynHeikoS: I have nothing yet17:36
lisitsyn:)17:36
@HeikoSlisitsyn: I have a q17:36
lisitsynshoot17:36
@HeikoSwhat if we just added setKernel to the c++17:36
@HeikoSthat would solve the meta lang shiatz17:36
@HeikoSrather than doing that in swig?17:36
@HeikoSlisitsyn: the alternative I see is just giving up on typed setters, even from c++17:37
lisitsynHeikoS: yeah adding setKernel to C++ is also fine17:37
lisitsynand might be even better17:37
lisitsynat least we have uniform interface this way17:38
@HeikoSyep17:38
@HeikoSthough we can then also do some tricks inside the "put" method17:38
lisitsynyess17:38
lisitsynlike17:38
@HeikoSlike getting type info at runtime and then just cast in there17:39
lisitsyndispatching things into something17:39
lisitsynyeah17:39
lisitsynif it is kernel then putKernel17:39
@HeikoSyeah17:39
@HeikoSnot super clean :D17:39
@HeikoSI just wonder17:39
@HeikoSdo we want uniform API from c++ and swig17:39
@HeikoSbecause what you have come up with is nice actually, as it is typesafe in c++17:40
@HeikoSbut how much do we care that the examples do have the same api?17:40
@HeikoSI think this is a design question17:40
@HeikoSas if we want uniform, then we need to accept some ugly things17:40
@HeikoSif we don't like that, we have to accept that the c++ meta examples are maybe about to die17:40
-!- travis-ci [~travis-ci@ec2-54-226-10-225.compute-1.amazonaws.com] has joined #shogun18:20
travis-ciit's Wuwei Lin'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/33755290718:20
-!- travis-ci [~travis-ci@ec2-54-226-10-225.compute-1.amazonaws.com] has left #shogun []18:20
-!- travis-ci [~travis-ci@ec2-54-226-10-225.compute-1.amazonaws.com] has joined #shogun18:23
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/33762833218:23
-!- travis-ci [~travis-ci@ec2-54-226-10-225.compute-1.amazonaws.com] has left #shogun []18:23
@HeikoSlisitsyn: python and java work with just "put", c++ and octave don't18:26
lisitsynHeikoS: yeah that's what I expected18:27
lisitsynoverloading is a bit different in different langs18:27
@HeikoSlisitsyn: yeah18:27
@HeikoSlisitsyn: so this speaks against this SUPPORT_TAG(Object, object, CSGObject*)18:28
@HeikoS thing18:28
lisitsynHeikoS: I think `base` should be putKernel putObject18:28
lisitsynyes18:28
lisitsynbut if language supports it18:28
lisitsynput() should also work18:28
@HeikoSok then I will leave that in18:28
@HeikoSand I will just add a CSGObject::putKernel18:28
@HeikoSon the other hand, we could modify put to do the casting for us by trial and error :D18:28
lisitsynHeikoS: yeah if you're at it drop SUPPORT_TAG and add all the put/gets18:28
@HeikoSlisitsyn: not sure what you mean by the last thing18:29
lisitsynHeikoS: I mean you can drop SUPPORT_TAG from swig18:29
lisitsynand add putKernel etc18:29
lisitsynto SGObject18:29
lisitsyngetKernel putKernel18:29
lisitsynsome different macro18:29
lisitsyn:)18:29
@HeikoSok18:30
@HeikoSwill do18:30
lisitsynHeikoS: this also helps resolving some issue I had18:33
lisitsynwith templates18:33
lisitsynit didn't work for me with SGVectors and stuff18:33
@HeikoSok18:33
lisitsynthis way it will work18:33
@HeikoSlet me send a PR18:33
@HeikoSand then we discuss from there18:34
lisitsynokie18:34
@HeikoSah18:34
@HeikoSah annoying18:35
@HeikoSif I dont want to include all the headers in the SGObject.h, i need to forwrad declare18:36
@HeikoSbut then I need another macro for the .cpp18:36
@HeikoSlisitsyn: where would you do the dispatching?18:40
@HeikoSlisitsyn: argh, actually I don't see a need for putKernel18:41
@HeikoSI think the put should just try to do a cast of the passed object to the type of the tag instead18:41
@HeikoSfor SGObject that is fine18:41
@HeikoSsince all passed objects will be castable to SGObject, no?18:42
@HeikoSlisitsyn: this is messed up :D18:43
@HeikoSlet me know when you have some time to discuss, I think there is need for discussion before something is done18:44
-!- travis-ci [~travis-ci@ec2-54-80-202-110.compute-1.amazonaws.com] has joined #shogun19:04
travis-ciit's Francisco Navarro'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/33762979219:04
-!- travis-ci [~travis-ci@ec2-54-80-202-110.compute-1.amazonaws.com] has left #shogun []19:04
@sukey[https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4151 synchronized by shubham80819:13
-!- travis-ci [~travis-ci@ec2-54-234-36-74.compute-1.amazonaws.com] has joined #shogun19:15
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/33762833219:15
-!- travis-ci [~travis-ci@ec2-54-234-36-74.compute-1.amazonaws.com] has left #shogun []19:15
-!- travis-ci [~travis-ci@ec2-54-80-202-110.compute-1.amazonaws.com] has joined #shogun19:15
travis-ciit's Wuwei Lin'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/33755290719:15
-!- travis-ci [~travis-ci@ec2-54-80-202-110.compute-1.amazonaws.com] has left #shogun []19:15
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 248 seconds]19:36
--- Log closed Tue Feb 06 00:00:50 2018
« Sunday, 2018-02-04 Index Tuesday, 2018-02-06 »

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