--- Log opened Tue Feb 12 00:00:45 2013 | ||
-!- cavalier- [~cavalier-@70.118.110.73] has quit [Ping timeout: 260 seconds] | 01:42 | |
-!- cavalier- [~cavalier-@70.118.110.73] has joined #shogun | 01:45 | |
-!- blackburn [~blackburn@188.168.14.94] has quit [Ping timeout: 244 seconds] | 03:07 | |
-!- FSCV [~FSCV@204.45.110.163] has quit [Quit: Leaving] | 03:08 | |
-!- blackburn [~blackburn@188.168.3.25] has joined #shogun | 03:24 | |
-!- FSCV [~FSCV@189.139.150.187] has joined #shogun | 04:49 | |
-!- FSCV [~FSCV@189.139.150.187] has quit [Ping timeout: 240 seconds] | 04:56 | |
-!- FSCV [~FSCV@173.254.212.46] has joined #shogun | 04:56 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 05:54 | |
n4nd0 | good news gsoc announcement | 05:56 |
---|---|---|
n4nd0 | sonney2k, sonne|work, blackburn : so I read it is important to have a release before gsoc, how can I help with that? | 05:56 |
blackburn | n4nd0: hey | 05:57 |
n4nd0 | good morning :) | 05:57 |
blackburn | n4nd0: well we need to close all known bugs | 05:58 |
n4nd0 | blackburn: in github issues? | 05:59 |
blackburn | n4nd0: yes and in general | 05:59 |
-!- blackburn [~blackburn@188.168.3.25] has quit [Quit: Leaving.] | 06:10 | |
-!- FSCV [~FSCV@173.254.212.46] has quit [Quit: Leaving] | 06:13 | |
-!- blackburn [~lisitsin@mxs.kg.ru] has joined #shogun | 06:45 | |
n4nd0 | blackburn: is tests/integration something still in use or deprecated? | 07:14 |
blackburn | n4nd0: we still can use it | 07:22 |
blackburn | but it could fail a lot I expect | 07:22 |
n4nd0 | blackburn: ok | 07:23 |
blackburn | n4nd0: could you please run make check-valgrind and paste the result somewhere? | 07:24 |
blackburn | in libshogun | 07:24 |
n4nd0 | blackburn: ok, but I need a bit of time, I was trying to make all the examples in libshogun | 07:32 |
n4nd0 | but it seems that my Eigen is not detected by configure | 07:32 |
n4nd0 | so converter examples are not properly built | 07:32 |
blackburn | hmm true | 07:34 |
blackburn | it is not properly handled I am afraid | 07:35 |
n4nd0 | blackburn: in configure? | 07:36 |
blackburn | n4nd0: I mean examples are not guarded | 07:37 |
blackburn | n4nd0: but why it doesn't detect your eigen? any ideas? | 07:37 |
n4nd0 | blackburn: my Eigen is in /usr/local/include | 07:38 |
n4nd0 | and it seems that the configure script doesn't look in there | 07:38 |
n4nd0 | I have used the flag --incdir=/usr/loccal/include/eigen3 | 07:39 |
n4nd0 | but I am afraid the incdir variable is not used by check eigen3 in configure | 07:39 |
blackburn | it uses pkg-config as I can see | 07:42 |
n4nd0 | I don't have that program installed here :D | 07:43 |
n4nd0 | blackburn: the thing I am wondering is | 07:43 |
n4nd0 | I can just re-install Eigen and I guess it should be fine | 07:43 |
n4nd0 | but maybe configure should find it if it is in /usr/local/include | 07:44 |
n4nd0 | nvm, pkg-config is installed | 07:47 |
blackburn | n4nd0: yes it should but it currently uses *only* pkg-config | 07:47 |
n4nd0 | yep | 07:49 |
n4nd0 | that's probably fine | 07:49 |
n4nd0 | it's a local problem of mine, I need to make my pkg-config detect eigen3 | 07:49 |
n4nd0 | Eigen3 support .. yes :) | 07:58 |
blackburn | nice | 08:09 |
blackburn | n4nd0: so finally works? | 08:09 |
n4nd0 | yes | 08:10 |
n4nd0 | making check-valgrind on the way | 08:11 |
blackburn | thanks! | 08:15 |
blackburn | I have to go, be back in a few hours | 08:16 |
n4nd0 | ok | 08:17 |
n4nd0 | i will paste the results and take a look | 08:18 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 245 seconds] | 08:38 | |
-!- sumit_ [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has joined #shogun | 08:58 | |
-!- n4nd0 [~nando@n158-p227.kthopen.kth.se] has joined #shogun | 09:06 | |
-!- blackburn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.] | 09:38 | |
n4nd0 | blackburn: valgrind results | 10:18 |
n4nd0 | http://pastebin.com/TYjzrCzJ | 10:18 |
n4nd0 | everything looks fine except from libary_serialization | 10:18 |
n4nd0 | I am guessing that we are OK with suppressed errors | 10:19 |
-!- wiking_ is now known as wiking | 10:32 | |
-!- wiking [~wiking@info2k1.hu] has quit [Changing host] | 10:33 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 10:33 | |
-!- blackburn [~lisitsin@mxs.kg.ru] has joined #shogun | 10:36 | |
n4nd0 | blackburn: hello hello | 10:36 |
blackburn | hey | 10:36 |
n4nd0 | the results of make valgrind-check | 10:36 |
n4nd0 | http://pastebin.com/TYjzrCzJ | 10:37 |
blackburn | n4nd0: thanks | 10:37 |
blackburn | whoops | 10:37 |
n4nd0 | suppressed errors are fine I guess? | 10:37 |
blackburn | yes | 10:37 |
blackburn | nice to know my code is a little b0rken | 10:38 |
n4nd0 | oh fuck | 10:38 |
n4nd0 | converter_kernellocallylinearembedding | 10:38 |
n4nd0 | and some others | 10:39 |
blackburn | hah my bad | 10:39 |
n4nd0 | do you know why can it be? | 10:39 |
blackburn | no - I'd have to investigate | 10:39 |
blackburn | n4nd0: or you can if you want :) | 10:39 |
blackburn | actually it didn't produce so many errors before - quite strange for me | 10:40 |
n4nd0 | blackburn: maybe something is wrong in my system | 10:40 |
n4nd0 | when I run check.sh I get error everywhere | 10:41 |
blackburn | no as it should work on any system :) | 10:41 |
n4nd0 | that cannot be good | 10:41 |
blackburn | yes check.sh doesn't work in libshogun examples | 10:41 |
n4nd0 | why not? | 10:41 |
blackburn | n4nd0: could you try to run converter_kernellocallylinearembedding? | 10:41 |
blackburn | and valgrind it too | 10:41 |
n4nd0 | blackburn: yep | 10:42 |
n4nd0 | what is wrong with check.sh then? | 10:42 |
blackburn | n4nd0: I don't know - I use make check-examples | 10:43 |
n4nd0 | blackburn: http://dl.dropbox.com/u/11020840/shogun/valgrind.log | 10:46 |
n4nd0 | kernellocallylinearembedding | 10:46 |
blackburn | thanks | 10:47 |
n4nd0 | blackburn: I can investigate a bit now | 10:47 |
n4nd0 | if you are not already on it? | 10:47 |
blackburn | n4nd0: no I am now under windows at job :( | 10:48 |
blackburn | n4nd0: looks surprising to me | 10:48 |
n4nd0 | blackburn: I am not sure if I ever valgrind-ed tapkee | 10:49 |
n4nd0 | did you? | 10:49 |
blackburn | n4nd0: that matrix is not properly initialized strange | 10:49 |
blackburn | yes I did | 10:49 |
blackburn | there were no errors :) | 10:49 |
blackburn | well the fact is LLE works ok but KLLE works bad | 10:49 |
blackburn | as they are the same in tapkee I find that quite strange | 10:49 |
blackburn | n4nd0: I expect I forgot to initialize something or so | 10:51 |
n4nd0 | blackburn: do you happen to know if resize in Eigen initializes the matrix or vector elements? | 11:43 |
n4nd0 | from the doc it seems to me that it doesn't | 11:43 |
blackburn | n4nd0: no I don't know | 11:57 |
n4nd0 | blackburn: take a look | 12:14 |
n4nd0 | http://eigen.tuxfamily.org/dox/SimplicialCholesky_8h_source.html | 12:14 |
n4nd0 | valgrind complains about line 819, so d must be unintiliazed | 12:15 |
n4nd0 | and I think this is caused because m_diag is not initialized | 12:15 |
n4nd0 | it is resized in 764 | 12:16 |
n4nd0 | but the values are not set | 12:16 |
n4nd0 | and they are used in 803 in an intermediate computation of d | 12:16 |
blackburn | n4nd0: scary! | 12:21 |
n4nd0 | yeah... | 12:21 |
n4nd0 | but I think it is not due to that probably | 12:22 |
n4nd0 | it must be another thing | 12:22 |
n4nd0 | blackburn: a think that confuses me | 12:22 |
n4nd0 | DefaultSparseSolver in tapkee might be SimplicialCholesky or SimplicialLDLT | 12:23 |
n4nd0 | SimplicialCholesky is used at the moment | 12:23 |
n4nd0 | but in the method where valgrind complains, DoLDLT is set to true | 12:23 |
n4nd0 | it seems a bit contradictory, doesn't it? | 12:23 |
blackburn | yeah | 12:24 |
n4nd0 | maybe that method should be called with DoLDLT=false? | 12:25 |
n4nd0 | just guessing no real idea :D | 12:25 |
blackburn | I don't know actually | 12:28 |
-!- sumit_ [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has quit [Ping timeout: 245 seconds] | 12:28 | |
blackburn | that all sounds strange | 12:31 |
blackburn | again, LLE and KLLE should both fail | 12:31 |
blackburn | or not | 12:31 |
blackburn | anyway I'll take a look tonight | 12:32 |
n4nd0 | that's true | 12:32 |
n4nd0 | I am going to check if I discover why it fails in one but not the other | 12:35 |
blackburn | makes sense, thanks | 12:45 |
n4nd0 | blackburn: look at this, it confuses me but maybe it tells you something :) | 12:48 |
n4nd0 | to simply to std::cout << matrix << std::endl | 12:48 |
n4nd0 | of the matrix we are giving to the solver | 12:48 |
n4nd0 | rises the conditional jump error in klle and not in lle | 12:49 |
n4nd0 | it must be something related to how this matrix is created | 12:49 |
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Ping timeout: 240 seconds] | 12:50 | |
-!- shogun-buildbot_ [~shogun-bu@7nn.de] has joined #shogun | 12:50 | |
blackburn | yes | 12:50 |
blackburn | n4nd0: https://github.com/shogun-toolbox/shogun/blob/master/src/shogun/converter/LocallyLinearEmbedding.cpp#L87 | 12:52 |
blackburn | absent in https://github.com/shogun-toolbox/shogun/blob/master/src/shogun/converter/KernelLocallyLinearEmbedding.cpp#L65 | 12:52 |
n4nd0 | blackburn: you are a professional :) | 12:53 |
n4nd0 | let me try it | 12:53 |
blackburn | n4nd0: probably that's the reason - this should fire some warning though | 12:53 |
blackburn | n4nd0: added the issue to not forget about setting that | 12:56 |
n4nd0 | blackburn: that solved valgrind's complains indeed | 12:57 |
blackburn | I expect others are caused by something similar.. | 12:57 |
n4nd0 | probably | 12:57 |
n4nd0 | blackburn: about the warning thing | 12:58 |
n4nd0 | you mean that we should enable that in tapkee right? | 12:58 |
n4nd0 | if some parameters are not set, say it | 12:59 |
blackburn | yes sure n4nd0 | 13:02 |
n4nd0 | I apply the change if so | 13:03 |
-!- n4nd0 [~nando@n158-p227.kthopen.kth.se] has quit [Ping timeout: 256 seconds] | 13:14 | |
-!- Hausas [~admin@185.6.154.4] has joined #shogun | 13:19 | |
-!- Hausas [~admin@185.6.154.4] has quit [Changing host] | 13:19 | |
-!- Hausas [~admin@unaffiliated/hausas] has joined #shogun | 13:19 | |
-!- Hausas [~admin@unaffiliated/hausas] has left #shogun [] | 13:30 | |
-!- n4nd0 [~nando@n158-p227.kthopen.kth.se] has joined #shogun | 13:30 | |
wiking | mmm | 14:01 |
wiking | blackburn: one gsoc idea: recommendation system | 14:01 |
blackburn | wiking: mentor? | 14:05 |
wiking | blackburn: happy to be | 14:17 |
wiking | moreover i can get one person i think | 14:17 |
wiking | from a company that is working on recommendation systems | 14:18 |
wiking | i'll have a chat with him today | 14:19 |
wiking | i can ask him what he would have in his mind | 14:19 |
wiking | for sure he would be ok to comentor with me a project like that | 14:19 |
wiking | (co-mentor as he is rather busy) | 14:19 |
-!- FSCV [~FSCV@204.45.110.163] has joined #shogun | 14:32 | |
-!- Guest84830 [73fe7104@gateway/web/freenode/ip.115.254.113.4] has joined #shogun | 15:52 | |
Guest84830 | Hi everyone, I am Parthasarathy G, a computer science under-grad student from IIT Madras, I am interested in contributing for Shogun, can anyone please guide me regarding this? | 15:53 |
n4nd0 | Guest84830: hey | 15:59 |
n4nd0 | the roadmap normally starts forking the repo, fetching a local copy and building it | 16:00 |
n4nd0 | then you could work on something you are interested in or we could suggest you some kind of contribution | 16:01 |
Guest84830 | Is there an existing git repo i could pull from? | 16:01 |
n4nd0 | Guest84830: yeah sure | 16:01 |
n4nd0 | the repository is in github | 16:01 |
n4nd0 | https://github.com/shogun-toolbox/shogun | 16:01 |
Guest84830 | I am currently cloning the repo, thanks for the link. | 16:06 |
n4nd0 | you are welcome | 16:06 |
Guest84830 | n4nd0: I have finished building the code, but not exactly found where the binaries are | 16:23 |
n4nd0 | Guest84830: libshogun.so should be in src/shogun | 16:25 |
n4nd0 | it is a good idea to make install too so you can use it from any part on your system | 16:26 |
-!- blackburn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.] | 16:27 | |
Guest84830 | ah yes, i have finished that. | 16:29 |
n4nd0 | I will be back in afew minutes | 16:30 |
-!- Guest84830 [73fe7104@gateway/web/freenode/ip.115.254.113.4] has quit [Ping timeout: 245 seconds] | 16:33 | |
n4nd0 | re | 16:40 |
-!- blackburn [~blackburn@83.234.54.3] has joined #shogun | 17:36 | |
-!- FSCV_ [~FSCV@173.254.212.46] has joined #shogun | 17:38 | |
-!- FSCV [~FSCV@204.45.110.163] has quit [Ping timeout: 252 seconds] | 17:39 | |
-!- n4nd0 [~nando@n158-p227.kthopen.kth.se] has quit [Quit: leaving] | 17:49 | |
-!- sumit [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has joined #shogun | 17:51 | |
-!- naruto [73fe7104@gateway/web/freenode/ip.115.254.113.4] has joined #shogun | 18:11 | |
-!- naruto is now known as Guest61045 | 18:11 | |
-!- sumit [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has quit [Quit: Page closed] | 18:11 | |
Guest61045 | Hi sonney2k , I was working on building from source some time ago, i have finished the building part, but what about launching the gui? | 18:16 |
-!- blackburn [~blackburn@83.234.54.3] has quit [Quit: Leaving.] | 18:39 | |
-!- Guest61045 [73fe7104@gateway/web/freenode/ip.115.254.113.4] has quit [Quit: Page closed] | 18:48 | |
-!- SCV__ [~FSCV@173.254.212.46] has joined #shogun | 20:19 | |
-!- Netsplit *.net <-> *.split quits: FSCV_ | 20:25 | |
-!- n4nd0 [53b32c87@gateway/web/freenode/ip.83.179.44.135] has joined #shogun | 21:31 | |
-!- KMcQuisten [d8338942@gateway/web/freenode/ip.216.51.137.66] has joined #shogun | 23:01 | |
KMcQuisten | Hello all. Hope things are going well | 23:02 |
KMcQuisten | I have a quick question that I hope I can get answered | 23:03 |
wiking | shoot | 23:05 |
KMcQuisten | I use LibSVR quite a bit, and have generally used the epsilon-SVR flavor. The parameter optimization framework only allows one to optimize nu, the maximum fraction of data points that are support vectors. I know that there's a relationship between nu and the tube epsilon. | 23:05 |
KMcQuisten | I've done the optimization on nu, but when I try to apply those parameters manually, changin nu doesn't seem to be changing the results. | 23:06 |
KMcQuisten | Are the default values for tube_epsilon stepping on my manual setting of nu before training? | 23:07 |
KMcQuisten | And if so how do I fix it? | 23:08 |
KMcQuisten | How does LibSVR determine if it's doing eps-SVR or nu-SVR? | 23:08 |
wiking | mmm | 23:09 |
wiking | <pass> | 23:09 |
wiking | i don't know who would be an expert in libsvr here | 23:09 |
-!- n4nd0 [53b32c87@gateway/web/freenode/ip.83.179.44.135] has quit [Ping timeout: 245 seconds] | 23:30 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 23:39 | |
-!- KMcQuisten [d8338942@gateway/web/freenode/ip.216.51.137.66] has quit [Ping timeout: 245 seconds] | 23:41 | |
-!- KMcQuisten [d8338942@gateway/web/freenode/ip.216.51.137.66] has joined #shogun | 23:41 | |
KMcQuisten | Well damn. | 23:42 |
KMcQuisten | Well, does anyone know why nu is available for optimization in LibSVR, but the tube epsilon isn't? It would certainly be very handy to be able to use the built-in model selection fromework to find the best tube epsilon | 23:43 |
-!- KMcQuisten [d8338942@gateway/web/freenode/ip.216.51.137.66] has quit [Quit: Page closed] | 23:50 | |
--- Log closed Wed Feb 13 00:00:45 2013 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!