| --- Log opened Mon Jul 16 00:00:10 2018 | ||
| -!- HeikoS [~heiko@2a00:23c5:e10a:5c00:e59f:8ac6:d583:5195] has quit [Ping timeout: 256 seconds] | 00:31 | |
| -shogun-buildbot:#shogun- Build nightly stretch deb #134 is complete: Failure [failed dput (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/38/builds/134 | 03:04 | |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4374 opened by csitaula | 05:55 |
|---|---|---|
| -!- supportvectorsrc [4492768b@gateway/web/freenode/ip.68.146.118.139] has joined #shogun | 08:07 | |
| -!- supportvectorsrc [4492768b@gateway/web/freenode/ip.68.146.118.139] has quit [Ping timeout: 252 seconds] | 09:00 | |
| @wiking | wuwei, ping? | 13:24 |
| wuwei | wiking: hi | 13:26 |
| @wiking | ok so we have a slight problem:) | 13:27 |
| @wiking | with pipelines | 13:27 |
| @wiking | basically it's related the story of eigen solvers | 13:27 |
| @wiking | and their solutions | 13:27 |
| @wiking | "> | 13:27 |
| @wiking | meaning that now in pipeline example you use PCA | 13:27 |
| @wiking | and PCA depends on an eigen solver | 13:27 |
| @wiking | and the eigen vector solutions are dependent on the actual solver | 13:28 |
| @wiking | they give the 'same' solution in mathetmatical sense because ev = -1 * ev | 13:28 |
| @wiking | but of course on the end | 13:28 |
| @wiking | you'll get different transformation | 13:28 |
| @wiking | and thus you'll get different result :( | 13:28 |
| @wiking | and because of this now the pipeline meta example fails when using viennacl :) | 13:29 |
| wuwei | I see | 13:30 |
| wuwei | this happens everywhere that uses eigensolver right? | 13:31 |
| @wiking | yeah i mean this is already a bug | 13:31 |
| @wiking | with GMM | 13:31 |
| @wiking | for example GMM fails when you have eigen but without LAPACK backend | 13:31 |
| @wiking | and it's because of the same reason | 13:31 |
| @wiking | it'll yield a different EV | 13:31 |
| @wiking | ev' = -1 ev_lapack | 13:31 |
| wuwei | yeah | 13:32 |
| wuwei | any idea for solving this? | 13:33 |
| @sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/770f748f4fd3b6db66ac08e699924f7152a8557d by vigsterkr | 16:12 |
| @sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4375 opened by vinx13 | 16:34 |
| -!- travis-ci [~travis-ci@ec2-54-224-183-23.compute-1.amazonaws.com] has joined #shogun | 17:03 | |
| 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/404462727 | 17:03 |
| -!- travis-ci [~travis-ci@ec2-54-224-183-23.compute-1.amazonaws.com] has left #shogun [] | 17:03 | |
| -!- travis-ci [~travis-ci@ec2-54-80-174-49.compute-1.amazonaws.com] has joined #shogun | 18:04 | |
| 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/404462727 | 18:04 |
| -!- travis-ci [~travis-ci@ec2-54-80-174-49.compute-1.amazonaws.com] has left #shogun [] | 18:04 | |
| -!- HeikoS [~heiko@2a00:23c5:e10a:5c00:ec0e:5179:7bd0:9fb0] has joined #shogun | 22:49 | |
| -!- mode/#shogun [+o HeikoS] by ChanServ | 22:49 | |
| @sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4374 closed by karlnapf | 22:58 |
| --- Log closed Tue Jul 17 00:00:12 2018 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!