| --- Log opened Mon Apr 30 00:00:46 2018 | ||
| -!- witness [uid10044@gateway/web/irccloud.com/x-rydifwwwdgcrtboa] has joined #shogun | 00:36 | |
| -!- witness [uid10044@gateway/web/irccloud.com/x-rydifwwwdgcrtboa] has quit [Quit: Connection closed for inactivity] | 03:46 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 10:12 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 10:12 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 260 seconds] | 11:46 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 12:02 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 12:02 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 255 seconds] | 12:10 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 12:20 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 12:20 | |
| @HeikoS | lisitsyn: jojo | 12:25 |
|---|---|---|
| @HeikoS | lisitsyn: you there? | 12:25 |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 240 seconds] | 12:32 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 12:36 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 12:36 | |
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/4692a7473fbfc66ed715d98079d659e5ddbebc6f by karlnapf | 12:45 |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 264 seconds] | 12:54 | |
| -!- witness [uid10044@gateway/web/irccloud.com/x-kcvicylabzhbudpt] has joined #shogun | 12:54 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 12:58 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 12:58 | |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4265 opened by fyoda | 13:03 |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 260 seconds] | 13:09 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 13:13 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 13:14 | |
| @HeikoS | lisitsyn: let me know when youre back | 13:21 |
| lisitsyn | HeikoS: hey | 13:24 |
| @HeikoS | lisitsyn: ah cool! | 13:24 |
| @HeikoS | how are things? | 13:24 |
| lisitsyn | all fine | 13:24 |
| lisitsyn | what about you? | 13:24 |
| @HeikoS | yeah pretty good | 13:25 |
| @HeikoS | lisitsyn: so question | 13:25 |
| lisitsyn | yeap | 13:25 |
| @HeikoS | combined kernels | 13:25 |
| @HeikoS | they represent some problems that need solve | 13:25 |
| @HeikoS | this is first point | 13:25 |
| @HeikoS | second is: lazy evaluated functions via 'get' | 13:25 |
| lisitsyn | ok first thing first? :) | 13:26 |
| @HeikoS | let me restart my inet | 13:26 |
| @HeikoS | yep | 13:26 |
| @HeikoS | 1 sec | 13:26 |
| lisitsyn | ok | 13:26 |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Read error: Connection reset by peer] | 13:28 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 13:29 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 13:29 | |
| @HeikoS | lisitsyn: jo | 13:29 |
| @HeikoS | can hear me? | 13:29 |
| lisitsyn | yes | 13:29 |
| @HeikoS | ok so combined kernel | 13:29 |
| @HeikoS | has the API "append_kernel" | 13:29 |
| @HeikoS | so the combined kernel doesnt really act like a kernel | 13:29 |
| @HeikoS | i mean it does | 13:30 |
| @HeikoS | it has weights for every subkernel | 13:30 |
| @HeikoS | but how to do this with put get | 13:30 |
| @HeikoS | lisitsyn: am I still online? | 13:31 |
| lisitsyn | HeikoS: yes yes :) | 13:31 |
| lisitsyn | HeikoS: it is an interesting thing actually | 13:32 |
| lisitsyn | I see two ways - make combined kernel a base class or implement get/put/add | 13:32 |
| @HeikoS | about base class | 13:32 |
| @HeikoS | I thought so as well, but then there will be problems with settting it | 13:32 |
| @HeikoS | because it is not a kernel anymore | 13:32 |
| @HeikoS | so then svm.put(combined_kernel) doesnt work | 13:33 |
| @HeikoS | not a CKernel anymore I meant | 13:33 |
| @HeikoS | i.e. CombinedKernel cannot be used as a drop in replacement for CKernel | 13:33 |
| lisitsyn | ah true | 13:33 |
| @HeikoS | we can either use CKernel or CCombinedKernel and need to fix that in the algos | 13:33 |
| lisitsyn | hacky hack: make CKernel a CCombinedKernel with one kernel | 13:34 |
| lisitsyn | :) | 13:34 |
| @HeikoS | yeah | 13:34 |
| @HeikoS | actually just checking | 13:34 |
| @HeikoS | the algorithms that support combinedkernel all do static casts anyways | 13:34 |
| @HeikoS | so it could be a base class | 13:35 |
| @HeikoS | though that is not really nice | 13:35 |
| @HeikoS | lisitsyn: but I mean the problem is there in general | 13:35 |
| @HeikoS | with any form of list | 13:35 |
| @HeikoS | where you want to add things | 13:35 |
| lisitsyn | HeikoS: we can put add(T) to SGObject that tries to get vector, add an element and put | 13:36 |
| @HeikoS | so I feel it would be good to solve it | 13:36 |
| @HeikoS | yeah so put would do the appending automagically | 13:36 |
| lisitsyn | oh | 13:37 |
| lisitsyn | automatic thing I don't really like | 13:37 |
| lisitsyn | I mean it would be not really obvious | 13:38 |
| @HeikoS | so what do you mean then, I dont understand | 13:38 |
| lisitsyn | I'd make it explicit | 13:38 |
| lisitsyn | 'add' | 13:38 |
| @HeikoS | ah sorry | 13:38 |
| @HeikoS | you want to actually do 'add' != put | 13:38 |
| lisitsyn | yes | 13:38 |
| @HeikoS | how does it know that something can be added? | 13:38 |
| lisitsyn | we can either go down to any and hack it there | 13:38 |
| lisitsyn | or just try to cast to vector<T> | 13:39 |
| lisitsyn | push_back and put back | 13:39 |
| lisitsyn | one thing I don't really like is that | 13:39 |
| lisitsyn | we'd have add | 13:39 |
| lisitsyn | but no remove | 13:39 |
| lisitsyn | :) | 13:39 |
| lisitsyn | but I guess it is fine | 13:39 |
| @HeikoS | yep | 13:39 |
| lisitsyn | we might also have some reset(string) that resets parameter named string | 13:40 |
| @HeikoS | i dont like adding a method to all objects that is only used by a minority | 13:40 |
| lisitsyn | that's true | 13:40 |
| @HeikoS | but i guess this might be useful in other contexts as well | 13:40 |
| lisitsyn | yeah it might be | 13:40 |
| @HeikoS | mmh | 13:40 |
| @HeikoS | and alternatives? | 13:40 |
| @HeikoS | make CKernel distinct from CCombinedKernel ... | 13:41 |
| lisitsyn | HeikoS: one alternative might be to rely on naming | 13:42 |
| lisitsyn | like put("subkernel_gaussian", GaussianKernel) | 13:42 |
| lisitsyn | and then act | 13:42 |
| lisitsyn | not really obvious | 13:42 |
| @HeikoS | so where else would this problem appear? | 13:42 |
| @HeikoS | in the pipelines? | 13:43 |
| @HeikoS | no but that will be solved differently | 13:43 |
| lisitsyn | combined something | 13:43 |
| lisitsyn | IDK | 13:43 |
| lisitsyn | combined distance | 13:43 |
| lisitsyn | :) | 13:43 |
| @HeikoS | whats your fav soluition? | 13:44 |
| lisitsyn | HeikoS: I am thinking of making CKernel a combined kernel | 13:45 |
| lisitsyn | but don't see pretty way of doing that | 13:45 |
| lisitsyn | but in general having repeated parameters | 13:45 |
| lisitsyn | is something good | 13:46 |
| @HeikoS | which speaks for the "add" | 13:46 |
| lisitsyn | yeah | 13:46 |
| @HeikoS | ok let me cook up something there | 13:46 |
| @HeikoS | lisitsyn: so the other point | 13:46 |
| @HeikoS | you wanna prototype the registering a function with 'get' ? | 13:47 |
| @HeikoS | we need that for lambday testing stuff for example | 13:47 |
| lisitsyn | yeah can try to do that | 13:47 |
| @HeikoS | sweet! | 13:47 |
| @HeikoS | alright | 13:47 |
| lisitsyn | we have some kind of holidays next three days so let me try to get some time to do that | 13:47 |
| @HeikoS | that was it from my side | 13:47 |
| @HeikoS | cool | 13:47 |
| @HeikoS | lets think of more stuff for you to do then :) | 13:47 |
| @HeikoS | like have a working plugijn | 13:48 |
| @HeikoS | which we want/need for the release | 13:48 |
| lisitsyn | well there is a PR | 13:48 |
| lisitsyn | I wanted to clean it up | 13:48 |
| lisitsyn | and merge | 13:48 |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 255 seconds] | 14:38 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 14:55 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 14:55 | |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4265 | 15:00 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4265 | 15:13 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4265 | 15:13 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4266 opened by karlnapf | 15:17 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4267 opened by fyoda | 15:50 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4267 | 15:52 |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 256 seconds] | 16:09 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 16:14 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 16:14 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 265 seconds] | 16:23 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 16:29 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 16:29 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 240 seconds] | 17:31 | |
| @sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4266 merged by karlnapf | 17:37 |
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/9a14c52fc41178c28781b251cc130a45e7a7e965 by karlnapf | 17:37 |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 17:47 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 17:47 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 248 seconds] | 17:58 | |
| -!- HeikoS [~heiko@82-132-229-139.dab.02.net] has joined #shogun | 18:23 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 18:23 | |
| -!- HeikoS [~heiko@82-132-229-139.dab.02.net] has quit [Remote host closed the connection] | 18:56 | |
| -!- travis-ci [~travis-ci@ec2-54-160-217-153.compute-1.amazonaws.com] has joined #shogun | 19:00 | |
| travis-ci | it'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/373090310 | 19:00 |
| -!- travis-ci [~travis-ci@ec2-54-160-217-153.compute-1.amazonaws.com] has left #shogun [] | 19:00 | |
| -!- travis-ci [~travis-ci@ec2-54-160-217-153.compute-1.amazonaws.com] has joined #shogun | 19:17 | |
| travis-ci | it'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/373090310 | 19:17 |
| -!- travis-ci [~travis-ci@ec2-54-160-217-153.compute-1.amazonaws.com] has left #shogun [] | 19:17 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has joined #shogun | 22:45 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 22:45 | |
| -!- HeikoS [~heiko@host86-128-122-53.range86-128.btcentralplus.com] has quit [Ping timeout: 264 seconds] | 23:12 | |
| --- Log closed Tue May 01 00:00:47 2018 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!