--- Log opened Mon Jun 27 00:00:57 2016 | ||
-!- arianepaola [~ariane@unaffiliated/arianepaola] has quit [Read error: Connection reset by peer] | 03:52 | |
-!- arianepaola [~ariane@unaffiliated/arianepaola] has joined #shogun | 03:56 | |
shogun-buildbot | build #29 of memleak - valgrind is complete: Failure [failed memory check generate html] Build details are at http://buildbot.shogun-toolbox.org/builders/memleak%20-%20valgrind/builds/29 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, OXPHOS <engelzora@gmail.com> | 05:38 |
---|---|---|
shogun-buildbot | build #1030 of nightly_none is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_none/builds/1030 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, OXPHOS <engelzora@gmail.com> | 05:43 |
shogun-buildbot | build #1160 of nightly_default is complete: Failure [failed test test_1 notebooks] Build details are at http://buildbot.shogun-toolbox.org/builders/nightly_default/builds/1160 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, OXPHOS <engelzora@gmail.com> | 06:27 |
arianepaola | ping wiking | 08:16 |
@wiking | pon | 08:33 |
@wiking | g | 08:33 |
BZDMG | bazdmeg morning | 08:33 |
@wiking | morning :) | 08:33 |
arianepaola | hi | 08:35 |
-!- sanuj [~sanuj@117.203.23.79] has joined #shogun | 09:40 | |
sanuj | BZDMG, there? | 09:40 |
sanuj | I solved the bug | 09:41 |
sanuj | very silly bug | 09:41 |
arianepaola | ping wiking | 10:26 |
arianepaola | wiking https://gist.github.com/arianepaola/d9ea1c39d632128942554bbc3e862de6 | 10:30 |
BZDMG | sanuj: | 10:55 |
BZDMG | so what's the bug | 10:55 |
sanuj | one word | 10:55 |
sanuj | https://github.com/shogun-toolbox/shogun/pull/3221/commits/380f576e3615aea29b56f0afb721efe5485510ae#diff-9c3599c0d2090e493be261b079e9b63eR37 | 10:55 |
sanuj | should be void set(...) | 10:56 |
sanuj | i actually did all the changes step by step on a fresh branch to spot this | 10:57 |
sanuj | hehe | 10:57 |
-!- OXPHOS [c68f0c0c@gateway/web/freenode/ip.198.143.12.12] has joined #shogun | 11:10 | |
-!- arianepaola [~ariane@unaffiliated/arianepaola] has quit [Read error: Connection reset by peer] | 11:14 | |
-!- HeikoS [~heiko@LFbn-1-4373-242.w92-170.abo.wanadoo.fr] has joined #shogun | 11:26 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 11:26 | |
@wiking | HeikoS: yo | 11:27 |
@wiking | OXPHOS: china! | 11:27 |
@HeikoS | jojo | 11:27 |
OXPHOS | wiking: hello! | 11:27 |
@wiking | BAZDMEG | 11:28 |
@wiking | BAAAAAAAAAAAAAAAZDMEG | 11:28 |
@wiking | OXPHOS sanuj Saurabh7__ | 11:28 |
@wiking | where is the weekly email? :) | 11:28 |
@wiking | OXPHOS sanuj Saurabh7__ PING PING PING PING PING | 11:29 |
@wiking | please let's try to have this in your calendars | 11:29 |
@wiking | so i dont have to do this every week | 11:29 |
@wiking | thnx | 11:29 |
OXPHOS | wiking: working on it. trying to locate the revised schedule i wrote yesterday.. | 11:29 |
@wiking | okok | 11:29 |
@wiking | it's 5:28pm | 11:29 |
@wiking | :P | 11:29 |
@wiking | i know where u r | 11:30 |
@wiking | :D | 11:30 |
@HeikoS | in fact, as I said in the last email, pls also send a 1-2 sentence short status every day you work on shogun | 11:30 |
sanuj | okay, sending | 11:30 |
sanuj | i was thinking of reporting what i had done today | 11:30 |
sanuj | since the day is not over, i didn't send | 11:30 |
@HeikoS | the daily thing can me minor, totally fine | 11:30 |
@HeikoS | just for us to keep track | 11:30 |
sanuj | sending to HeikoS BZDMG wiking and lambday | 11:31 |
@wiking | no | 11:31 |
@wiking | mailing list | 11:31 |
sanuj | shogun list | 11:32 |
sanuj | okay | 11:32 |
@wiking | is a perfect place | 11:32 |
@wiking | :) | 11:32 |
@HeikoS | sanuj: so what sup with da bug? | 11:32 |
sanuj | HeikoS, solved it | 11:32 |
sanuj | was silly | 11:32 |
@HeikoS | sanuj: great!! | 11:32 |
@HeikoS | sanuj: what was it? | 11:32 |
sanuj | HeikoS, i actually did all the changes step by step on a fresh branch to spot this as you had suggested ;) | 11:32 |
sanuj | HeikoS, https://github.com/shogun-toolbox/shogun/pull/3221/commits/380f576e3615aea29b56f0afb721efe5485510ae#diff-9c3599c0d2090e493be261b079e9b63eR37 | 11:33 |
sanuj | should have been void set(...) | 11:33 |
sanuj | forgot to do this ^ | 11:33 |
OXPHOS | wiking: haha you caught me | 11:33 |
@wiking | :> | 11:33 |
@HeikoS | sanuj: this is the private method to add a nonexisting tag? | 11:34 |
@HeikoS | sanuj: can you rename this to something better then _set? | 11:34 |
sanuj | no this is in Self struct in SGObject.cpp | 11:34 |
sanuj | HeikoS, i have renamed _set to add() to set a parameter for first time from init_param() | 11:35 |
@HeikoS | sanuj: cool | 11:35 |
@HeikoS | so then, you can re-add the things you had developed | 11:35 |
@HeikoS | or did you already? | 11:35 |
@HeikoS | sanuj: and then we can finally merge this pr | 11:35 |
sanuj | HeikoS, i'm just adding some unit-tests | 11:35 |
sanuj | on MockObject | 11:35 |
@HeikoS | sanuj: great, pls ping me once done | 11:35 |
sanuj | okay | 11:36 |
@HeikoS | I am online for 2 more hours or so | 11:36 |
sanuj | HeikoS, that's good | 11:36 |
@HeikoS | OXPHOS: so how are things going? | 11:36 |
sanuj | back in UK? | 11:36 |
sanuj | no it seems | 11:36 |
sanuj | HeikoS_web, | 11:36 |
@HeikoS | sanuj: no thats my workstation in the office | 11:36 |
@HeikoS | I am still hiding | 11:36 |
sanuj | yeah....that's how in inferred that you are not in london :P | 11:36 |
OXPHOS | HeikoS: good! revising based on your comments | 11:37 |
@HeikoS | OXPHOS: ok, as said, I am here for a while now | 11:37 |
@HeikoS | so we can discuss also, if needed | 11:37 |
@HeikoS | wiking: when do you finally tell us how to get rid of the templating macros? | 11:37 |
OXPHOS | HeikoS: cool! I have a question - | 11:39 |
OXPHOS | template<class T> void SGVector<T>::copy_data(const SGReferencedData &orig) { gpu_vector=std::shared_ptr<GPUMemoryBase<T>>(((SGVector*)(&orig))->gpu_vector); | 11:39 |
@HeikoS | shoot | 11:39 |
OXPHOS | you said: this crashes if I pass SGMatrix, no? If so, this was a bug. Could need a fix there | 11:39 |
@HeikoS | yeah since you static cast | 11:39 |
@HeikoS | and if I passed something else than SGVector what happens? | 11:40 |
@HeikoS | OXPHOS: ever saw this: http://stackoverflow.com/questions/27504280/metaprogramming-with-stdis-same | 11:40 |
OXPHOS | like some other derived of SGRef? | 11:40 |
@HeikoS | OXPHOS: yes, like SGMatrix | 11:41 |
@HeikoS | no type error | 11:41 |
@HeikoS | but static cast will fail | 11:41 |
OXPHOS | HeikoS: yeah i see. But not sure how to avoid that..? | 11:43 |
@HeikoS | OXPHOS: can you give me the link again? | 11:43 |
OXPHOS | HeikoS: is_same also requires a list of all types right? why is it different from the template<T> stuff? | 11:43 |
OXPHOS | HeikoS: https://github.com/shogun-toolbox/shogun/pull/3310/files#diff-9b962870c247ccd3f897b1ceacae8567 | 11:43 |
@HeikoS | OXPHOS: not sure, just was looking at it, never really checked it before | 11:44 |
@HeikoS | OXPHOS: so copy_data is not a method of SGReferencedData, right? Just of SGVector? | 11:45 |
OXPHOS | HeikoS: it's pure virtual in SGRef | 11:46 |
@HeikoS | OXPHOS: I see | 11:46 |
@HeikoS | ok then | 11:46 |
OXPHOS | But if we also have a gpu_data in SGMatrix maybe it is okay..? | 11:46 |
@HeikoS | I kind of dont wanna put a dynamic cast | 11:46 |
@HeikoS | so just leave it for now | 11:46 |
@HeikoS | Ill ask BZDMG | 11:46 |
OXPHOS | sure HKS! | 11:47 |
BZDMG | what? | 11:48 |
@HeikoS | BZDMG: copy_data method | 11:49 |
@HeikoS | purely virtual in SGReferencedData | 11:49 |
-!- lambday [8028b10a@gateway/web/freenode/ip.128.40.177.10] has joined #shogun | 11:49 | |
-!- mode/#shogun [+o lambday] by ChanServ | 11:49 | |
@HeikoS | implemented in SGVector | 11:49 |
@HeikoS | BZDMG: doesnt check type | 11:49 |
@HeikoS | but static casts to SGVector | 11:49 |
@lambday | erm.. what are we talking about? | 11:50 |
@HeikoS | lambday: minor stuff | 11:50 |
@HeikoS | lambday: whats needed to merge OXPHOS PR? | 11:50 |
@lambday | HeikoS: let me check | 11:50 |
@lambday | well the build is broken. | 11:51 |
@lambday | secondly, this still is for develop branch | 11:51 |
OXPHOS | Do I suppose to split them into .h and .cpp? | 11:51 |
@lambday | OXPHOS: why didn't you just refactor things in the linalg_refactor branch? there you've already added the global instance | 11:51 |
@lambday | OXPHOS: why do we need to keep adding the same thing in multiple branches? :) | 11:52 |
BZDMG | HeikoS: show me the code I don't get it :) | 11:52 |
OXPHOS | lambday: dunno..because some stuff are already merged? | 11:52 |
OXPHOS | lambday: I can pull it against that branch for this update | 11:53 |
@lambday | OXPHOS: merged to the feature branch.. doesn't mean we can't change things later :) | 11:53 |
OXPHOS | lambday: haha sure. also why it always fails some tests? | 11:53 |
@lambday | OXPHOS: you'll have to check that | 11:54 |
@lambday | OXPHOS: but please make the PR against feature/linalg_refactor branch.. | 11:54 |
@lambday | where you already added those stuffs.. just refactor it | 11:54 |
OXPHOS | lambday: sure | 11:54 |
OXPHOS | BZDMG: i think you're @-ed on Github for that line of codes | 11:55 |
@lambday | OXPHOS: also, I don't think you have turned off *all* the SGVector operations when data is on gpu.. we have to make sure that they throw error message instead of just crashing | 11:55 |
Saurabh7__ | Hi | 11:56 |
Saurabh7__ | HeikoS: yo | 11:56 |
OXPHOS | lambday: not all..there're some static ones | 11:56 |
@lambday | OXPHOS: but in case you did that already, just ignore my comments :D | 11:56 |
OXPHOS | lambday: those static methods..I think they should be moved out finally..and they takes T* not SGVectors actually so I think they're harmless for now | 11:58 |
@HeikoS | Saurabh7__: jojo | 11:58 |
@lambday | OXPHOS: what are those methods? | 11:58 |
OXPHOS | lambday: sum..add..norm.. | 11:59 |
@lambday | OXPHOS: yeah they should be moved out! put to linalg | 11:59 |
@lambday | OXPHOS: but let's not do it all in just one patch | 11:59 |
Saurabh7__ | HeikoS: I think I have updated most comments in the PR, | 11:59 |
@HeikoS | Saurabh7__: you feel it is merge ready? | 12:00 |
@HeikoS | will check soon | 12:00 |
OXPHOS | lambday: sure | 12:00 |
@lambday | OXPHOS: it's easier to miss things - we gotta do it in some later patch | 12:00 |
Saurabh7__ | HeikoS: yes | 12:00 |
@lambday | BZDMG: who are your vowels? | 12:00 |
Saurabh7__ | mm lemme check memleak | 12:00 |
@lambday | ate* | 12:01 |
@HeikoS | lambday: jo | 12:03 |
@HeikoS | lambday: can you give me a little tip here | 12:03 |
@lambday | HeikoS: sure.. what is it about? | 12:04 |
@HeikoS | lambday: see pm | 12:04 |
@HeikoS | lambday: can you see? | 12:05 |
@lambday | yes | 12:05 |
@HeikoS | Saurabh7__: got some timings on which changes helped? | 12:06 |
Saurabh7__ | HeikoS: yes will have to put them in | 12:08 |
@HeikoS | Saurabh7__: https://github.com/shogun-toolbox/shogun/pull/3293/files#diff-8faba8d51b4c212ec2cf40d103697caaR571 | 12:19 |
@HeikoS | what about this? | 12:19 |
Saurabh7__ | HeikoS: ah yes, a column is chosen there to go through | 12:20 |
Saurabh7__ | HeikoS: so do we need num_subset of columns | 12:21 |
Saurabh7__ | * we need | 12:21 |
Saurabh7__ | HeikoS: so you recommend sorting the first num_subset elements ? | 12:22 |
@HeikoS | everything that determines the order in which you access memory | 12:22 |
@HeikoS | should be sorted | 12:22 |
@HeikoS | Saurabh7__: all my comments were minor | 12:22 |
@HeikoS | Ill give me OK to merge | 12:22 |
Saurabh7__ | HeikoS: ok thanks, will get those values out and do memcheck here | 12:23 |
@HeikoS | Saurabh7__: cool | 12:27 |
@HeikoS | Saurabh7__: once thats done, then we do get rid of the float64_t? | 12:27 |
Saurabh7__ | HeikoS: yes could be | 12:27 |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 12:28 | |
shogun-notifier- | shogun-data: Saurabh7 :master * e27433b / testsuite/meta/ (2 files): https://github.com/shogun-toolbox/shogun-data/commit/e27433b4349c3c2821c80e90deb4bc0a387e0be9 | 12:28 |
shogun-notifier- | shogun-data: update rf data | 12:28 |
shogun-notifier- | shogun-data: Heiko Strathmann :master * b9b2d90 / testsuite/meta/ (2 files): https://github.com/shogun-toolbox/shogun-data/commit/b9b2d9058c22dae86a4181db7271dfa02485af7d | 12:28 |
shogun-notifier- | shogun-data: Merge pull request #111 from Saurabh7/rfnewdata | 12:28 |
shogun-notifier- | shogun-data: | 12:28 |
shogun-notifier- | shogun-data: update rf data | 12:28 |
Saurabh7__ | right now its expected everything is already numerals | 12:28 |
-!- sanuj [~sanuj@117.203.23.79] has quit [Ping timeout: 244 seconds] | 12:32 | |
@HeikoS | OXPHOS: jo | 12:35 |
@HeikoS | could you clean up the doc? | 12:35 |
@HeikoS | lots of deprecated stuff in there | 12:35 |
@HeikoS | OXPHOS: maybe just create a new doc with the revised timeline | 12:35 |
@HeikoS | not sure where to look | 12:35 |
OXPHOS | HeikoS: which one? | 12:36 |
@HeikoS | the doc you sent in the mail to list | 12:37 |
OXPHOS | HeikoS: oh i see. so I'll remove the outdated ones | 12:37 |
@HeikoS | OXPHOS, Saurabh7__, gotta leave soon | 12:50 |
@HeikoS | might be back later | 12:50 |
@HeikoS | but anything you want to discuss? | 12:50 |
OXPHOS | HeikoS: not yet. thx! | 12:51 |
@HeikoS | cool see you later then | 12:52 |
-!- sanuj [~sanuj@117.203.23.79] has joined #shogun | 12:53 | |
Saurabh7__ | HeikoS: ok | 12:55 |
Saurabh7__ | HeikoS: next I will try multithreaded thing and removing float64_t ? | 12:55 |
@HeikoS | Saurabh7__: start with float64 | 12:56 |
@HeikoS | there might be some mini issues we need to solve with this | 12:56 |
@HeikoS | or multithread | 12:56 |
@HeikoS | you can choose :) | 12:56 |
-!- HeikoS [~heiko@LFbn-1-4373-242.w92-170.abo.wanadoo.fr] has quit [Quit: Leaving.] | 12:57 | |
sanuj | BZDMG, there? | 13:11 |
BZDMG | sanuj: a bit | 13:11 |
sanuj | BZDMG, need to discuss one thing | 13:11 |
BZDMG | yeap | 13:11 |
sanuj | BZDMG, now params are set in init_params() for each class | 13:11 |
sanuj | they are declared but not initialized with a value | 13:11 |
sanuj | BZDMG, if someone does get() then it will result in garbage value | 13:12 |
sanuj | without setting a value to it | 13:12 |
sanuj | BZDMG, ? | 13:33 |
BZDMG | sanuj: get should fail | 13:33 |
BZDMG | if there is no such thing | 13:34 |
BZDMG | no garbage | 13:34 |
sanuj | BZDMG, currently we have this | 13:34 |
sanuj | void init_params() | 13:34 |
sanuj | { | 13:34 |
sanuj | m_vector = SGVector<float64_t>(); | 13:34 |
sanuj | add("vector", m_vector); | 13:34 |
sanuj | add("int", m_int); | 13:34 |
sanuj | add("float", m_float); | 13:34 |
sanuj | } | 13:34 |
sanuj | SGVector<float64_t> m_vector; | 13:34 |
sanuj | int32_t m_int; | 13:34 |
sanuj | float64_t m_float; | 13:34 |
sanuj | BZDMG, how to make get() fail? | 13:34 |
sanuj | BZDMG, and we need to consider member variables with default values also.... | 13:35 |
BZDMG | BZDMG: get = if has(..) return else fail | 13:38 |
BZDMG | sanuj: ^ | 13:38 |
BZDMG | not really get the problem | 13:38 |
sanuj | BZDMG, there are two functions now in SGObject.h | 13:38 |
sanuj | set() and add() | 13:38 |
BZDMG | yes | 13:39 |
sanuj | add is protected and set is public | 13:39 |
sanuj | add("vector", m_vector); will do => map[basetag] = any | 13:39 |
sanuj | but m_vector is empty | 13:39 |
sanuj | same is the case for m_int, m_float | 13:39 |
sanuj | so m_int and m_float have random values since they are not initialized | 13:40 |
sanuj | BZDMG, has() will also return true since they exist in the map | 13:40 |
sanuj | BZDMG, got it? | 13:42 |
BZDMG | sanuj: probably yes | 13:57 |
BZDMG | but it is not garbage | 13:57 |
BZDMG | its default | 13:57 |
BZDMG | I think we should explicitly provide what value to put | 13:57 |
sanuj | BZDMG, most members dont have default value | 13:57 |
sanuj | okay | 13:57 |
BZDMG | they all have | 13:57 |
BZDMG | it is either zero or something like that | 13:58 |
sanuj | BZDMG, no worries then | 13:58 |
-!- sanuj [~sanuj@117.203.23.79] has quit [Ping timeout: 276 seconds] | 14:12 | |
-!- sanuj [~sanuj@117.203.23.79] has joined #shogun | 14:38 | |
sanuj | BZDMG, one more question | 14:40 |
-!- OXPHOS [c68f0c0c@gateway/web/freenode/ip.198.143.12.12] has quit [Quit: Page closed] | 15:27 | |
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 15:28 | |
-!- sanuj [~sanuj@117.203.23.79] has quit [Ping timeout: 272 seconds] | 16:11 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has joined #shogun | 17:02 | |
-!- lambday [8028b10a@gateway/web/freenode/ip.128.40.177.10] has quit [Quit: Page closed] | 17:25 | |
-!- sanuj [~sanuj@117.203.23.79] has joined #shogun | 17:26 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has quit [Quit: sonne|osx] | 17:30 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has joined #shogun | 17:32 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has quit [Quit: sonne|osx] | 17:41 | |
-!- lambday [8028b10a@gateway/web/freenode/ip.128.40.177.10] has joined #shogun | 17:43 | |
-!- mode/#shogun [+o lambday] by ChanServ | 17:43 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has joined #shogun | 17:56 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has quit [Client Quit] | 17:57 | |
-!- lambday [8028b10a@gateway/web/freenode/ip.128.40.177.10] has quit [Ping timeout: 250 seconds] | 18:46 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has joined #shogun | 19:06 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has quit [Client Quit] | 19:07 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has joined #shogun | 19:08 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has quit [Quit: sonne|osx] | 19:21 | |
-!- sanuj [~sanuj@117.203.23.79] has quit [Remote host closed the connection] | 20:22 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has joined #shogun | 21:12 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has quit [Quit: sonne|osx] | 21:18 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has joined #shogun | 21:28 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has quit [Quit: sonne|osx] | 21:40 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has joined #shogun | 22:11 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has quit [Quit: sonne|osx] | 22:25 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has joined #shogun | 22:27 | |
-!- HeikoS [~heiko@82.113.183.181] has joined #shogun | 22:29 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 22:29 | |
-!- sonne|osx [~sonne@x4e37463a.dyn.telefonica.de] has quit [Client Quit] | 22:29 | |
-!- HeikoS [~heiko@82.113.183.181] has quit [Quit: Leaving.] | 22:46 | |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 22:46 | |
shogun-notifier- | shogun: Heiko Strathmann :feature/kernel_exp_family * b027178 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/b02717855307fb01ff52fa2041721e2014bfaf10 | 22:46 |
shogun-notifier- | shogun: add nystrom version of xi_norm | 22:46 |
-!- travis-ci [~travis-ci@ec2-54-146-45-46.compute-1.amazonaws.com] has joined #shogun | 23:22 | |
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/140642685 | 23:22 |
-!- travis-ci [~travis-ci@ec2-54-146-45-46.compute-1.amazonaws.com] has left #shogun [] | 23:22 | |
--- Log closed Tue Jun 28 00:00:58 2016 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!