--- Log opened Wed May 08 00:00:35 2013 | ||
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 00:11 | |
pickle27 | iglesiasg: here is a new pastebin http://pastebin.com/KsR7kLw6 | 00:44 |
---|---|---|
pickle27 | iglesiasg: true, I meant the example runs | 00:45 |
iglesiasg | pickle27: the pastebin is still not loading here for some reason | 00:46 |
pickle27 | iglesiasg: out of curiosity I just ran valgrind on classifier_qda and I got the same unrecognized instruction line | 00:47 |
pickle27 | iglesiasg: strange | 00:47 |
@sonney2k | iglesiasg, it is loading here | 00:48 |
@sonney2k | try a different browser | 00:48 |
@sonney2k | pickle27, are you sure you compile w/o optimization? | 00:49 |
iglesiasg | sonney2k: I should get a proper internet connection rather :S | 00:49 |
@sonney2k | iglesiasg, or that | 00:49 |
pickle27 | yes I've compiled without optimization and with debugging | 00:49 |
iglesiasg | sonney2k: my mom has not got wifi at home so I get connected here with a 3G dongle | 00:49 |
pickle27 | ooooj | 00:50 |
pickle27 | hold up | 00:50 |
@sonney2k | pickle27, and you make ; make installed? | 00:50 |
pickle27 | I just realized that I have not make installed | 00:50 |
pickle27 | wow | 00:50 |
@sonney2k | and then also compiled the example again? | 00:50 |
@sonney2k | haha | 00:50 |
iglesiasg | pickle27: hehe | 00:50 |
@sonney2k | gotcha | 00:50 |
pickle27 | literally realized the second before you typed that | 00:50 |
iglesiasg | pickle27: alias make=make; sudo make install | 00:50 |
iglesiasg | or sth like that, please :P | 00:51 |
@sonney2k | why root? | 00:51 |
@sonney2k | I usually install somewhere where that is not needed | 00:52 |
iglesiasg | sonney2k: just got the habit to do it for python interface | 00:52 |
pickle27 | okay now I have something | 00:52 |
@sonney2k | pickle27, show us | 00:52 |
@sonney2k | well me | 00:52 |
pickle27 | http://pastebin.com/Ek8PW8cZ | 00:52 |
iglesiasg | sonney2k: so is it actually normal that after compiling with optimizations leaks may appear?? | 00:52 |
@sonney2k | spanish can't afford :P | 00:53 |
iglesiasg | :( | 00:53 |
iglesiasg | pickle27: maybe if you run it without -v we still got useful info and my connection can handle that :) | 00:54 |
pickle27 | iglesiasg: http://pastebin.com/EwEKypN0 | 00:54 |
@sonney2k | iglesiasg, nope but valgrind may crash | 00:55 |
iglesiasg | sonney2k: Isee | 00:55 |
@sonney2k | pickle27, is S empty? | 00:55 |
@sonney2k | pickle27, maybe doing | 00:56 |
pickle27 | sonney2k: sorry new to using valgring where are you reading | 00:57 |
@sonney2k | while ( rank < m_dim) && S[rank] > m_tolerance) | 00:57 |
@sonney2k | would fix it? | 00:57 |
pickle27 | oh in code | 00:57 |
@sonney2k | well line 341 | 00:57 |
@sonney2k | in MCLDA.cpp | 00:57 |
@sonney2k | you access S[rank] | 00:57 |
pickle27 | yeah I see | 00:58 |
pickle27 | S shouldn't be empty it should have been filled by wrap_dgesvd | 00:58 |
@sonney2k | btw no idea who merged this but please use { } brackets around for loop's with >1 line | 00:58 |
@sonney2k | and indention seems to be broken | 00:58 |
@sonney2k | you have spaces & tabs mixed | 00:58 |
@sonney2k | we use tabs only | 00:59 |
@sonney2k | size 4 | 00:59 |
@sonney2k | and you can write | 00:59 |
@sonney2k | instead of SGVector< float64_t > S = SGVector< float64_t >(m_dim); | 00:59 |
@sonney2k | SGVector< float64_t > S(m_dim); | 00:59 |
pickle27 | yeah I know my tabs are off, is there a better editor for fixing that? gedit displays it fine | 00:59 |
iglesiasg | pickle27: vim is better than gedit indeed :) | 01:00 |
@sonney2k | pickle27, well open gedit's preferences and go to editor -> set tab width to 4 | 01:00 |
iglesiasg | but gedit should be just fine for the tabs honestly | 01:00 |
pickle27 | yeah I have done that but it makes it appear to be formatted okay when it isn't | 01:00 |
@sonney2k | then maybe enable the plugin | 01:01 |
iglesiasg | pickle27: is there an option to show whitespace characters? | 01:01 |
@sonney2k | 'draw spaces' | 01:01 |
pickle27 | sonney2k: and you prefer braces around all loops? | 01:02 |
@sonney2k | pickle27, when they are > 1 line yes | 01:04 |
@sonney2k | in particular with the current indention massacre it is hard to read | 01:04 |
@sonney2k | that indention == part of the language thing of python is really sth I miss in other $LANGs | 01:05 |
pickle27 | yeah Im working on that sorry | 01:05 |
@sonney2k | pickle27, btw in python spaces are now mandatory or do they allow tabs? | 01:05 |
pickle27 | I think they allow tabs | 01:06 |
@sonney2k | in 2.x they did | 01:06 |
@sonney2k | but we all know how dangerous this is | 01:06 |
-!- Ingvaras [522c194e@gateway/web/freenode/ip.82.44.25.78] has quit [Ping timeout: 245 seconds] | 01:06 | |
iglesiasg | sonney2k: why do you like more tabs than spaces btw? | 01:07 |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 01:08 | |
@sonney2k | iglesiasg, nowadays it is probably no longer reasonable | 01:09 |
@sonney2k | but back in the days (recall I am old ;) | 01:09 |
@sonney2k | tab == 1 byte | 01:09 |
@sonney2k | editors didn't have smart tabs | 01:09 |
@sonney2k | replacing spaces with tabs etc | 01:09 |
@iglesiasg | oh interesting | 01:10 |
@sonney2k | it's like linux kernel | 01:10 |
@sonney2k | they use tabs too | 01:10 |
@sonney2k | though the kernel is a few more decades older than me | 01:11 |
@iglesiasg | heh | 01:11 |
pickle27 | hey I think it might be good now! | 01:12 |
@sonney2k | pickle27, what did you change? | 01:13 |
@iglesiasg | both indentantion and the bug? | 01:13 |
pickle27 | http://pastebin.com/M7HaiUpE | 01:13 |
pickle27 | yes both | 01:13 |
@sonney2k | uahhah! | 01:13 |
pickle27 | thats the entire output of valgrind | 01:13 |
@sonney2k | pickle27, so was it what I suggested? | 01:13 |
@sonney2k | or what? | 01:13 |
pickle27 | got to hand it to you that S was a good catch | 01:13 |
pickle27 | I also figured out my gedit problem I think | 01:14 |
pickle27 | you need to restart it after changing from spaces to tabs | 01:14 |
@sonney2k | I have wasted^H^H^H^Hspent too much time of my life hunting bugs | 01:14 |
@sonney2k | so I know a good part of their species | 01:15 |
@iglesiasg | you have gotten a 6th sense indeed | 01:15 |
pickle27 | sonney2k: ah but the thrill when you catch one! | 01:15 |
@sonney2k | and I am collecting them as trophies | 01:15 |
@sonney2k | but only the very rare hard to catch ones nowadays | 01:15 |
@sonney2k | pickle27, so keep that one for yourself | 01:15 |
@sonney2k | I prefer multithreading bugs these days | 01:16 |
@sonney2k | they just shiiiiinnne | 01:16 |
pickle27 | sonney2k: :) | 01:16 |
@sonney2k | oh dammn I need my medication | 01:16 |
@sonney2k | and sleep | 01:16 |
@sonney2k | good night everyone! | 01:16 |
@sonney2k | and hurray 8 slots! | 01:17 |
@iglesiasg | night | 01:17 |
pickle27 | night | 01:17 |
@iglesiasg | sonney2k: let's see if we get any more with the re-allocation | 01:17 |
@iglesiasg | providing that last year we gave one, we have the karma with us :) | 01:18 |
pickle27 | alright I'm out for the night too now | 01:19 |
pickle27 | later! | 01:19 |
@iglesiasg | pickle27: talking about funny bugs | 01:19 |
@iglesiasg | https://github.com/shogun-toolbox/shogun/commit/2234df4636c1e6061fcc19209fe279850efa7354 | 01:19 |
@iglesiasg | see you! | 01:19 |
-!- iglesiasg [d58f3213@gateway/web/freenode/ip.213.143.50.19] has quit [Quit: Page closed] | 01:27 | |
-!- foulwall [~foulwall@2001:da8:215:503:1943:d77b:b38a:a922] has joined #shogun | 02:08 | |
-!- madhura [~madhura@59.95.237.64] has quit [Quit: Leaving] | 02:10 | |
-!- shogun-toolbox [~shogun@7nn.de] has quit [Ping timeout: 264 seconds] | 02:20 | |
--- Log closed Wed May 08 02:20:30 2013 | ||
--- Log opened Wed May 08 02:20:39 2013 | ||
-!- shogun-t1olbox [~shogun@7nn.de] has joined #shogun | 02:20 | |
-!- Irssi: #shogun: Total of 17 nicks [3 ops, 0 halfops, 0 voices, 14 normal] | 02:20 | |
-!- Irssi: Join to #shogun was synced in 8 secs | 02:20 | |
-!- shogun-toolbox [~shogun@7nn.de] has quit [Ping timeout: 264 seconds] | 02:20 | |
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has quit [Ping timeout: 252 seconds] | 02:46 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 03:32 | |
-!- poojits [75d35d4a@gateway/web/freenode/ip.117.211.93.74] has quit [Ping timeout: 245 seconds] | 03:41 | |
-!- zxtx [~zv@68.125.89.66] has joined #shogun | 03:59 | |
-!- votjakovr [~votjakovr@host-46-241-3-209.bbcustomer.zsttk.net] has joined #shogun | 04:39 | |
-!- nube [~rho@49.244.128.165] has quit [Ping timeout: 245 seconds] | 04:58 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds] | 05:04 | |
-!- hushell [43bd6474@gateway/web/freenode/ip.67.189.100.116] has joined #shogun | 05:24 | |
-!- foulwall [~foulwall@2001:da8:215:503:1943:d77b:b38a:a922] has quit [Remote host closed the connection] | 05:27 | |
-!- lisitsyn [~blackburn@188.168.3.6] has quit [Quit: Leaving.] | 05:47 | |
-!- foulwall [~foulwall@2001:da8:215:503:c16d:6025:8431:c642] has joined #shogun | 05:53 | |
-!- zxtx [~zv@68.125.89.66] has quit [Ping timeout: 248 seconds] | 06:55 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 07:13 | |
-!- hushell [43bd6474@gateway/web/freenode/ip.67.189.100.116] has quit [Ping timeout: 245 seconds] | 07:18 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 07:24 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 07:29 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds] | 07:44 | |
foulwall | Hi sonney2k? | 08:00 |
@sonney2k | foulwall, morning | 08:21 |
foulwall | sonney2k: sorry for my absence, I left for my grandpa's funeral. I'll start my work today. | 08:22 |
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has joined #shogun | 08:23 | |
foulwall | sonney2k: I'm trying to do the kernel matrix visualization work. | 08:24 |
@sonney2k | foulwall, my condolences | 08:26 |
@sonney2k | foulwall, just for the record - it is no problem if you are away but please tell us upfront to avoid misunderstandings in the future | 08:26 |
foulwall | sonney2k: I'll let you know when I am away! sorry for not telling. | 08:28 |
@sonney2k | so I hope you recovered from the funeral.. losing someone is always #%$! | 08:28 |
-!- hushell [43bd6474@gateway/web/freenode/ip.67.189.100.116] has joined #shogun | 08:28 | |
shogun-buildbot | build #926 of bsd1 - libshogun is complete: Failure [failed configure] Build details are at http://www.shogun-toolbox.org/buildbot/builders/bsd1%20-%20libshogun/builds/926 blamelist: Kevin Hughes <kevin@kevin-M3920.(none)> | 08:28 |
foulwall | sonney2k: so anything will be all right. I return to work. | 08:29 |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 08:30 | |
deerishi | hi everyone!! | 08:30 |
deerishi | neone der? | 08:30 |
deerishi | i wanna work on Cross-Product for lists of kernels bug | 08:31 |
deerishi | i have some difficulty understanding the problem statement | 08:31 |
@sonney2k | deerishi, I think van51 worked on that already | 08:32 |
shogun-buildbot | build #927 of bsd1 - libshogun is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/bsd1%20-%20libshogun/builds/927 | 08:42 |
shogun-buildbot | build #761 of cyg1 - libshogun is complete: Failure [failed configure] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/761 blamelist: Soeren Sonnenburg <sonne@debian.org> | 08:44 |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has joined #shogun | 09:14 | |
-!- mode/#shogun [+o lisitsyn] by ChanServ | 09:14 | |
-!- iglesiasg [c1934d18@gateway/web/freenode/ip.193.147.77.24] has joined #shogun | 09:23 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 09:26 | |
@lisitsyn | foulwall: condolences | 09:43 |
-!- foulwall [~foulwall@2001:da8:215:503:c16d:6025:8431:c642] has quit [Remote host closed the connection] | 09:45 | |
@sonney2k | lisitsyn, I am considering to switch the mailing list manager for shogun toolbox.org | 09:55 |
@lisitsyn | sonney2k: why? | 09:55 |
@sonney2k | I am currently setting up a new US based server | 09:55 |
@sonney2k | the mailing list manager is very slim & neat admittedly | 09:56 |
@sonney2k | but it only works well with qmail | 09:56 |
@sonney2k | and qmail is not really maintained | 09:56 |
@sonney2k | so I am going for the default -> exim | 09:56 |
@sonney2k | I hate mailman for doing .mbox files | 09:57 |
@sonney2k | I mean one big mbox file growing and growing | 09:57 |
@sonney2k | not one file / mail | 09:57 |
@sonney2k | that is so stupid | 09:57 |
@lisitsyn | I realize I have absolutely no clue what is you are talking about :D | 09:58 |
@sonney2k | now when you do backups you backup the gigabyte big file | 09:58 |
@sonney2k | if someone just sent a *single* new mail to the list | 09:58 |
-!- foulwall [~foulwall@2001:da8:215:503:c47e:67cf:6bd7:93a9] has joined #shogun | 10:14 | |
@sonney2k | lisitsyn, I just did the switch | 10:28 |
@sonney2k | trouble ahead... | 10:29 |
@lisitsyn | sonney2k: what kind of? | 10:29 |
@sonney2k | mailinglist berserk / mail not working | 10:29 |
@sonney2k | foulwall, the good news is that we will have some separate virtual server to play with for the demos :D | 10:30 |
-!- Yanglittle [deb20afa@gateway/web/freenode/ip.222.178.10.250] has joined #shogun | 10:30 | |
foulwall | sonney2k: That's great1 | 10:30 |
-!- nube [~rho@116.90.239.3] has joined #shogun | 10:34 | |
-!- HeikoS [~heiko@nat-179-224.internal.eduroam.ucl.ac.uk] has joined #shogun | 10:47 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 10:47 | |
Yanglittle | excuse me i can't load multiclasslabels from shogun.Features under ubuntu 13.04 | 10:48 |
Yanglittle | but others can load such as combinedfeatures | 10:49 |
@lisitsyn | Yanglittle: hey. what do you mean by load? | 10:51 |
@lisitsyn | HeikoS: hey | 10:51 |
@HeikoS | hi!° | 10:51 |
@HeikoS | finally back :) | 10:51 |
@lisitsyn | we haven't seen you for a while ;) | 10:51 |
Yanglittle | i mean import | 10:51 |
@HeikoS | thanks for doing all the replies | 10:51 |
@lisitsyn | HeikoS: have you seen we've got 8 slots? | 10:51 |
foulwall | maybe eigen3 dependency? | 10:51 |
@HeikoS | lisitsyn: yes, told you I was in Germany and had only minor internet access | 10:51 |
@HeikoS | lisitsyn: I have seen it | 10:52 |
@lisitsyn | I remember | 10:52 |
@HeikoS | this is pretty much what we wanted right? | 10:52 |
@HeikoS | Yanglittle: might be eigen or another dependency indeed | 10:52 |
@lisitsyn | HeikoS: we should have some decision on two unlucky we left out | 10:52 |
@lisitsyn | I don't think it is eigen dependency | 10:52 |
@lisitsyn | but let me check | 10:52 |
@lisitsyn | no, no dependency here | 10:53 |
@HeikoS | lisitsyn: could you explain? | 10:54 |
@HeikoS | lisitsyn: why did you already give feedback btw? isnt that supposed to happen after its fixed? | 10:54 |
Yanglittle | eigen support is yes. | 10:54 |
@lisitsyn | HeikoS: just to push students to be more active | 10:54 |
@lisitsyn | that's ok | 10:54 |
@sonney2k | HeikoS, what? | 10:55 |
@sonney2k | hey there | 10:55 |
@sonney2k | HeikoS, we give feedback to figure out whom to accept | 10:56 |
@sonney2k | if everyone just sits there and wait then we will never find out who fits best | 10:56 |
@HeikoS | sonney2k: hi! | 10:57 |
@HeikoS | sonney2k: sergey told me, I agree with that, just wondered when I first saw it | 10:58 |
@sonney2k | well last night :D | 11:00 |
@HeikoS | sonney2k: I just read stuff on my phone, no point in writing on it :D | 11:00 |
@sonney2k | HeikoS, I was thinking a bit about our current tests. | 11:02 |
@HeikoS | sonney2k: yes? | 11:02 |
@sonney2k | HeikoS, what we do is serialize and unserialize | 11:02 |
@sonney2k | then compare if results are exactly the same | 11:03 |
@HeikoS | yes | 11:03 |
@sonney2k | now we have the issue w/ gaussian_process_regression | 11:03 |
@sonney2k | that 3 out of 100 labels deviate by 1e-15 | 11:03 |
@HeikoS | yep | 11:03 |
@HeikoS | that should be fine though | 11:04 |
@HeikoS | just update the test | 11:04 |
@HeikoS | annyoing | 11:04 |
@sonney2k | no | 11:04 |
@HeikoS | but as you said, its a one to one comparison | 11:04 |
@sonney2k | it works fine everywhere except on travis | 11:04 |
@HeikoS | I like this regarding the states of the instances | 11:04 |
@HeikoS | ah | 11:04 |
@HeikoS | what about a comparison class then? | 11:04 |
@HeikoS | that compares all parameters to be close to each other? | 11:04 |
@sonney2k | so lets assume that this result is OK (which is not clear to me btw...) | 11:04 |
@sonney2k | yes exactly | 11:05 |
@sonney2k | that is what I wanted to ask | 11:05 |
@sonney2k | we should have a class / method that utilizes the parameter framework | 11:05 |
@lisitsyn | sonney2k: we could check if their virtual machine image is available somewhere | 11:05 |
@sonney2k | and then compares parameters with a certain precision | 11:05 |
-!- gsomix [~gsomix@188.168.3.6] has quit [Ping timeout: 252 seconds] | 11:05 | |
@sonney2k | lisitsyn, I suspect that it is due to numpy 1.6.1 | 11:06 |
@lisitsyn | sonney2k: random? | 11:06 |
@sonney2k | well everything is the same (6 or seven results returned) | 11:06 |
@sonney2k | but labels in 3 things?! | 11:06 |
@sonney2k | or it could be multithreading | 11:06 |
@lisitsyn | hah yes totally crazy | 11:06 |
@sonney2k | I mean the mean of the GP and everythign | 11:07 |
@HeikoS | sonney2k: yes that would be nice to have | 11:07 |
@HeikoS | works fine for all numerical stuff | 11:07 |
@HeikoS | and for SGOnbjects, we do recursion | 11:07 |
@sonney2k | HeikoS, we know the type right? | 11:07 |
@sonney2k | yeah TParameter | 11:07 |
@HeikoS | yes | 11:07 |
@HeikoS | some work though | 11:07 |
@HeikoS | but I can do that | 11:08 |
@HeikoS | at some near point | 11:08 |
@sonney2k | so we could just have a SGObject equals() method | 11:08 |
Yanglittle | excuse me, are there any methods to solve the problem? | 11:08 |
@sonney2k | that takes an SGObject | 11:08 |
@lisitsyn | Yanglittle: oh sorry | 11:08 |
@HeikoS | sonney2k: yes | 11:08 |
@lisitsyn | Yanglittle: so this fails: | 11:08 |
@sonney2k | then works the same way as list_parameters | 11:08 |
@lisitsyn | from modshogun import MulticlassLabels | 11:08 |
@lisitsyn | right? | 11:08 |
@HeikoS | sonney2k I dont see any big obstacles with that | 11:08 |
@sonney2k | HeikoS, if we had this - we could even extend the testsuite to other languages | 11:09 |
@sonney2k | so we no longer need pickle | 11:09 |
@sonney2k | but can directly compare shogun objects created in python_modular with java_modular | 11:09 |
Yanglittle | from modshogun import MulticlassLabels Traceback (innermost last): File "<stdin>", line 1, in <module> ImportError: cannot import name MulticlassLabels | 11:09 |
@sonney2k | etc | 11:09 |
@HeikoS | sonney2k, yes, that sounds good, although we pay with a more complicated thing that might be buggy | 11:09 |
@HeikoS | but still I like that | 11:09 |
@HeikoS | I would do it after fixing the migration stuff however | 11:09 |
@HeikoS | because currently we cannot alter SGObjects without breaking the file format | 11:10 |
@sonney2k | well a unit test would ensure that comparing stuff works | 11:10 |
@HeikoS | and the tests | 11:10 |
@HeikoS | sonney2k, yes | 11:10 |
@sonney2k | both is critical | 11:10 |
@sonney2k | we can hardly afford the buildbot failing etc | 11:10 |
@HeikoS | yes | 11:11 |
@HeikoS | so the equals method should be easy and nice to write | 11:11 |
@HeikoS | the migration stuff is a nightmare to solve | 11:11 |
@HeikoS | no idea where the problem is etc | 11:11 |
-!- nube [~rho@116.90.239.3] has quit [Ping timeout: 264 seconds] | 11:11 | |
@lisitsyn | Yanglittle: is that latest version? I see no reason to fail there actually | 11:12 |
Yanglittle | yes the latest version, | 11:13 |
@sonney2k | HeikoS, shouldn't a small unit test for the migration stuff expose that already? | 11:15 |
@sonney2k | I mean if it is small - one can debug it nicely | 11:15 |
-!- nube [~rho@116.90.239.3] has joined #shogun | 11:16 | |
sijin | lisitsyn: if Travis build fails, how to figure out where is the problem? it only tells that one file conflicts | 11:18 |
@HeikoS | sonney2k: yes maybe | 11:18 |
@HeikoS | its just that the code is so complicated that I have trouble understanding it | 11:19 |
@HeikoS | but in principle you are right | 11:19 |
@lisitsyn | sijin: conflicts? like? | 11:19 |
@HeikoS | sonney2k: but you are right, this is the way to go | 11:19 |
@HeikoS | start with the very fundamental parts and work up | 11:20 |
sijin | I got this https://travis-ci.org/shogun-toolbox/shogun/builds/6634337 several days ago? | 11:20 |
@HeikoS | some parts cannot be checked via unit tests though | 11:20 |
@HeikoS | since you need classes that alter etc | 11:20 |
@HeikoS | better to it in an example | 11:20 |
@lisitsyn | sijin: oh that's not caused by your changes I think | 11:21 |
-!- nube1 [~rho@116.90.239.3] has joined #shogun | 11:21 | |
-!- nube [~rho@116.90.239.3] has quit [Ping timeout: 276 seconds] | 11:22 | |
sijin | lisitsyn: so what I can do for it? I didn't find test for Kpca | 11:22 |
@sonney2k | sijin, write one! | 11:24 |
@lisitsyn | sijin: yeah we have no test | 11:24 |
@lisitsyn | sijin: I just glanced over your PR - could you please fix tabs/spaces? it is wavy sometimes | 11:24 |
sijin | sonney2k: ok. What need to be tested? I mean there is no ground truth for kpca, compare the result with another package? | 11:25 |
sijin | lisitsyn: Fine, I will do that. | 11:26 |
-!- Yanglittle [deb20afa@gateway/web/freenode/ip.222.178.10.250] has quit [Quit: Page closed] | 11:46 | |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 11:46 | |
shogun-notifier- | shogun: hushell :develop * d39547b / / (11 files): https://github.com/shogun-toolbox/shogun/commit/d39547b50c0e97c690d228c9f5f30c8c408c71ca | 11:46 |
shogun-notifier- | shogun: multiclass heuristics for probabilistic outputs estimations | 11:46 |
shogun-notifier- | shogun: hushell :develop * 2bb00e1 / / (6 files): https://github.com/shogun-toolbox/shogun/commit/2bb00e1745c7192c8c6449bdd784b574ff970ae9 | 11:46 |
shogun-notifier- | shogun: added unit-test | 11:46 |
shogun-notifier- | shogun: hushell :develop * a3faafe / src/shogun/multiclass/ (5 files): https://github.com/shogun-toolbox/shogun/commit/a3faafeb2fc5110288fa68c39c775b5218be3484 | 11:46 |
shogun-notifier- | shogun: dropped SGVector reference | 11:46 |
shogun-notifier- | shogun: Heiko Strathmann :develop * e726ca6 / / (12 files): https://github.com/shogun-toolbox/shogun/commit/e726ca641900bf838974aafe90488bd0d0706fc1 | 11:46 |
shogun-notifier- | shogun: Merge pull request #1017 from hushell/issues | 11:46 |
shogun-notifier- | shogun: | 11:46 |
shogun-notifier- | shogun: [update] Probabilistic Outputs for SVM | 11:46 |
-!- nube1 [~rho@116.90.239.3] has quit [Ping timeout: 256 seconds] | 11:48 | |
shogun-buildbot | build #762 of cyg1 - libshogun is complete: Failure [failed configure] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/762 blamelist: hushell <hushell@hushell-U510.(none)> | 11:50 |
@iglesiasg | ooops | 11:50 |
@iglesiasg | aah it seems nothing :) | 11:50 |
@lisitsyn | iglesiasg: chris is back in town (tuebingen) | 11:53 |
@iglesiasg | lisitsyn: how busy is he? | 11:55 |
@iglesiasg | maybe it is a good moment to ping him, no idea | 11:55 |
@lisitsyn | iglesiasg: he will take a look at the paper today | 11:56 |
@iglesiasg | lisitsyn: this is good news then! | 11:56 |
@lisitsyn | yeah | 11:56 |
shogun-notifier- | shogun: Roman Votyakov :develop * f926f91 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/f926f91b6ae22fd37e8755b4f7301f5a8e178ca4 | 11:57 |
shogun-notifier- | shogun: Replaced dynamic_cast with obtain_from_generic in FITCInferenceMethod class | 11:57 |
shogun-notifier- | shogun: Heiko Strathmann :develop * 7c302bc / / (3 files): https://github.com/shogun-toolbox/shogun/commit/7c302bc4ae97fcce6bb1f928ddb36093666a6e15 | 11:57 |
shogun-notifier- | shogun: Merge pull request #1062 from votjakovr/develop | 11:57 |
shogun-notifier- | shogun: | 11:57 |
shogun-notifier- | shogun: Replaced dynamic_cast with obtain_from_generic in FITCInferenceMethod class | 11:57 |
@iglesiasg | HeikoS has set merge mode on :) | 11:58 |
@HeikoS | iglesiasg: yes :) | 11:58 |
@HeikoS | its crazy | 11:58 |
@HeikoS | 3 days away and so many patches :) | 11:59 |
@lisitsyn | votjakovr: hey, I am just curious | 11:59 |
@lisitsyn | are there kinda professors doing some *modern* NN research | 11:59 |
@lisitsyn | or do you do it by your own? | 11:59 |
shogun-notifier- | shogun: van51 :develop * 0ec800c / / (3 files): https://github.com/shogun-toolbox/shogun/commit/0ec800ce08ba59659db85a450c53a3f4c1bca24d | 12:04 |
shogun-notifier- | shogun: Issue #1042 - Cross-Product for lists of kernels | 12:04 |
shogun-notifier- | shogun: Heiko Strathmann :develop * a641e16 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/a641e16a836b3843fae913e5416b0dd6550b050a | 12:04 |
shogun-notifier- | shogun: Merge pull request #1059 from van51/develop | 12:04 |
shogun-notifier- | shogun: | 12:04 |
shogun-notifier- | shogun: Issue #1042 - Cross-Product for lists of kernels | 12:04 |
-!- travis-ci [~travis-ci@ec2-107-21-134-219.compute-1.amazonaws.com] has joined #shogun | 12:05 | |
travis-ci | [travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/6984285 | 12:05 |
-!- travis-ci [~travis-ci@ec2-107-21-134-219.compute-1.amazonaws.com] has left #shogun [] | 12:05 | |
shogun-buildbot | build #763 of cyg1 - libshogun is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/763 | 12:06 |
-!- travis-ci [~travis-ci@ec2-23-20-196-170.compute-1.amazonaws.com] has joined #shogun | 12:18 | |
travis-ci | [travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/6984521 | 12:18 |
-!- travis-ci [~travis-ci@ec2-23-20-196-170.compute-1.amazonaws.com] has left #shogun [] | 12:18 | |
shogun-buildbot | build #1062 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1062 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, hushell <hushell@hushell-U510.(none)> | 12:21 |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Quit: Page closed] | 12:26 | |
shogun-buildbot | build #1063 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1063 blamelist: hushell <hushell@hushell-U510.(none)> | 12:27 |
votjakovr | lisitsyn: Hey :) My university's research is connected with Hierachical Temporal Memory (originally developed by Jeff Hawkins). It's a kind of biologically inspired cognitive architecture. | 12:29 |
-!- foulwall [~foulwall@2001:da8:215:503:c47e:67cf:6bd7:93a9] has quit [Remote host closed the connection] | 12:30 | |
shogun-buildbot | build #765 of cyg1 - libshogun is complete: Failure [failed configure] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/765 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, van51 <vangelis_51@hotmail.com>, Roman Votyakov <votjakovr@gmail.com> | 12:31 |
-!- travis-ci [~travis-ci@ec2-54-234-229-115.compute-1.amazonaws.com] has joined #shogun | 12:39 | |
travis-ci | [travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/6984680 | 12:39 |
-!- travis-ci [~travis-ci@ec2-54-234-229-115.compute-1.amazonaws.com] has left #shogun [] | 12:39 | |
votjakovr | HeikoS: Hi, There is something unclear in squared exponential covariance function with isotropic distance measure (covSEiso) in GPML matlab toolbox. There is no bug with partial derivative by ell parameter? | 12:44 |
-!- foulwall [~foulwall@2001:da8:215:202:d49d:918b:1418:6e43] has joined #shogun | 12:44 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 12:46 | |
@sonney2k | ok I killed shogun's mailinglists | 12:49 |
@wiking | boooooyaaaaa | 12:51 |
@wiking | dontkiiiiliiit | 12:51 |
@sonney2k | too late | 12:51 |
foulwall | (⊙o⊙) | 12:52 |
@sonney2k | I am thinking about having shogun-toolbox.org running its own mail server | 12:52 |
@wiking | sonney2k: why where was it running? | 12:52 |
@sonney2k | wiking, thoughts? | 12:53 |
@sonney2k | my private one | 12:53 |
@wiking | sonney2k: mmm need one? | 12:53 |
@wiking | sonney2k: i have an up and running mailman on my server | 12:53 |
@sonney2k | need one what? | 12:53 |
@wiking | sonney2k: it's like 1 command to get a new mailinglist running on it | 12:53 |
@sonney2k | wiking, I don't like mailman - hard to backup | 12:53 |
@sonney2k | I mean it has a single .mbox file per mailinglist | 12:54 |
@sonney2k | which grows and grows and grows | 12:54 |
@wiking | sonney2k: well yeah i backup the whole fs :P | 12:54 |
@sonney2k | so each and every day you run a backup | 12:54 |
@sonney2k | you backup *everything* | 12:54 |
@wiking | sonney2k: yeah but still i can do incremental backups thanks to zfs :P | 12:54 |
@sonney2k | wiking, incremental if the file is different? | 12:54 |
@wiking | sonney2k: yes | 12:55 |
@sonney2k | how so? | 12:55 |
@wiking | sonney2k: as it's incremental on the fs level | 12:55 |
@wiking | sonney2k: dont ask me, ask zfs developers | 12:55 |
@sonney2k | crazy | 12:55 |
@sonney2k | my experience with zfs was rather crash and burn | 12:55 |
@wiking | i just know that i can create with one comment an incremental backup of the fs | 12:55 |
@wiking | sonney2k: heheh i love it | 12:55 |
shogun-buildbot | build #1064 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1064 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com>, van51 <vangelis_51@hotmail.com>, Roman Votyakov <votjakovr@gmail.com> | 12:55 |
@wiking | sonney2k: but u need a lot of ram | 12:55 |
@wiking | sonney2k: and it's rather unstable in linux | 12:56 |
@wiking | sonney2k: hence the fbsd ;p | 12:56 |
@sonney2k | ahh screw it I put that thing on my server | 12:56 |
@wiking | sonney2k: anyhow it's time to do something about the PRNG | 12:56 |
@wiking | :D | 12:56 |
@wiking | sonney2k: so shall we go with take a random element of the vector and check whether it's the same? :P | 12:56 |
@sonney2k | wiking, you mean now that heiko broke all the tests you can do more ? | 12:56 |
@wiking | hehehehehehe | 12:57 |
@wiking | no sorry i just still populate my commints in a different branch | 12:57 |
@wiking | ;) | 12:57 |
@sonney2k | wiking, as I said I think generating 10k numbers and checking the 10k'th to be the same with the same seed is good enough for me | 12:57 |
@lisitsyn | magic number 10k ;) | 12:58 |
* sonney2k so now the question is how I get from ezmlm & qmail to mailman | 12:58 | |
@sonney2k | wiking, ohh mailman has a daily patch :D | 12:58 |
@sonney2k | lisitsyn, does you shogun-toolbox.org mail address still work? | 12:59 |
@sonney2k | lisitsyn, ohh btw please send me an email under mine | 12:59 |
@lisitsyn | testing | 12:59 |
@sonney2k | no idea if it still works | 12:59 |
@lisitsyn | sonney2k: probably not | 12:59 |
@lisitsyn | I sent you an email | 12:59 |
@sonney2k | I changed too much... | 13:00 |
@sonney2k | greylisting, spam, virtual domains, mail server blabla | 13:00 |
* sonney2k just realizes this server has ipv6 | 13:00 | |
deerishi | @sonny2k: what is sth?the "data" was not actually a directory when i git cloned shogun's repo.it was just a broken file. | 13:04 |
@wiking | tada | 13:05 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * 4255c47 / src/shogun/mathematics/Math.h: https://github.com/shogun-toolbox/shogun/commit/4255c47b3529c4ebe407975b8841180f2de5906d | 13:05 |
shogun-notifier- | shogun: Use global random as extern to avoid SG_REF/UNREF calls | 13:05 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * 88ce738 / src/shogun/lib/external/SFMT/SFMT.h: https://github.com/shogun-toolbox/shogun/commit/88ce7386923423202cda597aff608a6436e05c36 | 13:05 |
shogun-notifier- | shogun: Ignore SFMT headers in doxygen generation | 13:05 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * 701eb5f / src/shogun/mathematics/Random.cpp,src/shogun/mathematics/Random.h: https://github.com/shogun-toolbox/shogun/commit/701eb5f0e6e55ebaa7003e3ded3221401b9e58cb | 13:05 |
shogun-notifier- | shogun: Random: use two SFMT struct instead of one | 13:05 |
shogun-notifier- | shogun: SFMT random regerator requires reinitalization if one wants | 13:05 |
shogun-notifier- | shogun: to generate a 64-bit random after generating a 32-bit random | 13:05 |
shogun-notifier- | shogun: and vice-versa. In order to avoid reinitalization, rather have | 13:05 |
shogun-notifier- | shogun: 2 different SFMT structs for the 32-bit and the 64-bit RNG. | 13:05 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * c7e5f9b / src/shogun/lib/external/dSFMT/ (14 files): https://github.com/shogun-toolbox/shogun/commit/c7e5f9bf86b3824a89a351330e1f28f87335e525 | 13:05 |
shogun-notifier- | shogun: Initial import of dSFMT | 13:05 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * 0628564 / src/shogun/lib/external/dSFMT/ (3 files): https://github.com/shogun-toolbox/shogun/commit/0628564f867f78b4e5a53cebc63bd63dc65bd20d | 13:05 |
shogun-notifier- | shogun: Change include paths in dSFMT for shogun | 13:05 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * e659378 / src/shogun/lib/external/dSFMT/dSFMT.h: https://github.com/shogun-toolbox/shogun/commit/e659378eada060dcaaa1870c6d4738367106c02d | 13:05 |
shogun-notifier- | shogun: Ignore dSFMT in doxygen generation | 13:05 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * c5f7097 / src/shogun/lib/external/dSFMT/ (3 files): https://github.com/shogun-toolbox/shogun/commit/c5f7097cdad5d1219acc9c6928a71e90cf999f69 | 13:05 |
shogun-notifier- | shogun: Fix redefinition of W128_T in dSFMT | 13:05 |
@wiking | hehe | 13:05 |
@wiking | i think shogun-notifier- went on flood protection | 13:05 |
@wiking | :> | 13:05 |
@sonney2k | deerishi, well why did you git add and remove it then? | 13:06 |
@sonney2k | deerishi, it should just not be in the commit / PR | 13:06 |
deerishi | yes. | 13:06 |
@wiking | let's see what travis does with this | 13:07 |
deerishi | https://github.com/deerishi/shogun/tree/m16 | 13:07 |
deerishi | this is the branch i sent pr from | 13:07 |
@wiking | HeikoS: GMM will need a bit of redesign :P | 13:08 |
@wiking | ok i mean Gaussian | 13:08 |
@HeikoS | wiking: that might be | 13:09 |
@wiking | HeikoS: like this for (int32_t i=0; i<m_mean.vlen; i++) random_vec[i]=CMath::randn_double(); | 13:09 |
@sonney2k | wiking, feel free :P | 13:09 |
@wiking | HeikoS: as the new Random has functions like filling a vector of doubles | 13:10 |
@wiking | from various intervalls | 13:10 |
@lisitsyn | m16 :D | 13:10 |
@wiking | *intervals | 13:10 |
@lisitsyn | like machine gun? | 13:10 |
@wiking | but then again u need there std distrib | 13:10 |
@wiking | ok i'll see what i can do about that | 13:10 |
deerishi | @HeikoS: u der? | 13:12 |
@wiking | no he is die :P | 13:12 |
@wiking | hahahahaha this was a very lame joke :D | 13:12 |
deerishi | @HeikoS: cud you look at my latest pr? | 13:14 |
@lisitsyn | wiking: haha nice | 13:15 |
@wiking | lisitsyn: when do u arrive to berlin? | 13:16 |
@lisitsyn | wiking: 9th | 13:16 |
@lisitsyn | 21-20 or so | 13:16 |
@wiking | okkey | 13:16 |
@lisitsyn | wiking: staying at hotel tulip | 13:16 |
shogun-notifier- | shogun: Heiko Strathmann :develop * 72ba739 / src/shogun/kernel/PolyKernel.h: https://github.com/shogun-toolbox/shogun/commit/72ba73925ad80e6f98a219801170f4edcda679e5 | 13:16 |
shogun-notifier- | shogun: added getter for degree | 13:16 |
shogun-notifier- | shogun: Heiko Strathmann :develop * 9e178a5 / examples/undocumented/libshogun/ (2 files): https://github.com/shogun-toolbox/shogun/commit/9e178a577e9eae54d290ca407e458c8add504043 | 13:16 |
shogun-notifier- | shogun: added example how to perform modelselection over sub-parameters of combined kernels | 13:16 |
shogun-notifier- | shogun: Heiko Strathmann :develop * 549256e / src/NEWS: https://github.com/shogun-toolbox/shogun/commit/549256e39861f2f735887fdc89478c90884179a9 | 13:16 |
shogun-notifier- | shogun: mention multiclass probabilities and modelselection for combined kernels sub-parameters | 13:16 |
shogun-notifier- | shogun: Heiko Strathmann :develop * 61767d9 / / (4 files): https://github.com/shogun-toolbox/shogun/commit/61767d9d02ad5717fd24b6438ce98ceb5d0cc285 | 13:16 |
@wiking | i'll check my flight options | 13:16 |
shogun-notifier- | shogun: Merge pull request #1066 from karlnapf/develop | 13:16 |
shogun-notifier- | shogun: | 13:16 |
shogun-notifier- | shogun: example for modelselection for combined kernels | 13:16 |
@lisitsyn | wiking: near tiergarten | 13:16 |
@wiking | cool cool | 13:16 |
@sonney2k | some kind of fight going on here? | 13:16 |
@sonney2k | the one who breaks most stays until x-mas? | 13:16 |
@lisitsyn | haha | 13:17 |
@iglesiasg | :D | 13:18 |
@wiking | lisitsyn: ok found it | 13:19 |
@lisitsyn | wiking: found what? | 13:19 |
shogun-buildbot | build #766 of cyg1 - libshogun is complete: Failure [failed configure] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/766 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 13:20 |
@lisitsyn | der HeikoS | 13:22 |
@lisitsyn | :D | 13:22 |
@HeikoS | thats cyg | 13:23 |
@lisitsyn | das HeikoS! | 13:23 |
@lisitsyn | :D | 13:23 |
@HeikoS | DER! | 13:23 |
@HeikoS | deerishi: take a look at my comments | 13:23 |
@lisitsyn | die HeikoS! | 13:23 |
@lisitsyn | :D | 13:23 |
@HeikoS | we just need three files from you | 13:23 |
@HeikoS | sonney2k: I cannot send mails to the list anymore | 13:23 |
@lisitsyn | HeikoS: interesting that we have gender stuff in russian but I have absolutely no idea how to use it in german | 13:24 |
@lisitsyn | well I know you are der but Kaffee? | 13:24 |
@lisitsyn | das/der/die? no idea :D | 13:24 |
@lisitsyn | HeikoS: mailing list is der broken | 13:25 |
@HeikoS | lisitsyn: der Kaffee | 13:25 |
deerishi | @HeikoS:i resent the pr | 13:25 |
@HeikoS | lisitsyn: broken?? | 13:25 |
deerishi | has sparse invariance covariance been solved? | 13:25 |
@lisitsyn | HeikoS: yeah der sonney2k does secret ops | 13:25 |
@lisitsyn | HeikoS: moving our mailing list to langley's CIA HQ | 13:26 |
@lisitsyn | deerishi: yes | 13:26 |
@HeikoS | deerishi: you have to reset your git (do a backup before) | 13:26 |
@HeikoS | lisitsyn: I see | 13:26 |
@HeikoS | lisitsyn: we now have modelselection for sub-kernels in MKL | 13:26 |
@HeikoS | and we have probs for multiclass svm | 13:26 |
@wiking | HeikoS: when u have some time | 13:27 |
@lisitsyn | HeikoS: I have no idea what he does actually - check the logs for more detailed stuff | 13:27 |
@lisitsyn | HeikoS: yeah I noticed that for sure | 13:27 |
@wiking | HeikoS: i will need your help with one example | 13:27 |
@HeikoS | two things that have been asked for often | 13:27 |
@lisitsyn | the latter | 13:27 |
@HeikoS | nice! | 13:27 |
@HeikoS | wiking: hi! | 13:27 |
deerishi | but why? | 13:27 |
@HeikoS | sorry, I wasnt following | 13:27 |
@wiking | statistics_hsic.cpp | 13:27 |
@wiking | HeikoS: it fails with the new prng | 13:27 |
@HeikoS | deerishi: because the PR contains things we dont want | 13:27 |
@wiking | but no idea why atm | 13:27 |
@HeikoS | wiking: with the new random? | 13:27 |
shogun-notifier- | shogun: Heiko Strathmann :develop * d8ededf / src/NEWS: https://github.com/shogun-toolbox/shogun/commit/d8ededf769eae4216881922ff0ad1383f40235c4 | 13:30 |
shogun-notifier- | shogun: mention the recent streaming bugfix | 13:30 |
shogun-notifier- | shogun: Heiko Strathmann :develop * 3cd0318 / src/NEWS: https://github.com/shogun-toolbox/shogun/commit/3cd0318e95f7877bd365cd115e3730892e4c30f0 | 13:30 |
shogun-notifier- | shogun: Merge pull request #1068 from karlnapf/develop | 13:30 |
shogun-notifier- | shogun: | 13:30 |
shogun-notifier- | shogun: update news once more | 13:30 |
@HeikoS | wiking: ok | 13:31 |
@HeikoS | wiking: remove the asserts | 13:31 |
@HeikoS | I will update the example soon and move ciritcal parts to test | 13:31 |
shogun-buildbot | build #767 of cyg1 - libshogun is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/767 | 13:34 |
@sonney2k | lisitsyn, please rename to mr.kalashnikov | 13:35 |
@lisitsyn | sonney2k: why that? | 13:35 |
@sonney2k | you are a killer | 13:36 |
deerishi | @HeikoS:i resent the pr | 13:36 |
@lisitsyn | sonney2k: people around won't understand why :D | 13:36 |
@HeikoS | deerishi: there are still the data commits in there | 13:37 |
@HeikoS | you have to reset your git version | 13:38 |
@HeikoS | otherwise they wont go away | 13:38 |
@HeikoS | but backup your code before | 13:38 |
@lisitsyn | sonney2k: have you ever disassemble and assemble AK-47 btw? :D | 13:39 |
@lisitsyn | as you lived in not the best part of berlin you may have that too | 13:39 |
@sonney2k | sonney2k, I do that blindfolded lefthanded in 10 seconds! | 13:39 |
@sonney2k | you learn that when you protect the wall(tm) | 13:40 |
@lisitsyn | sonney2k: my mother did that in a minute or so :D they have to do that in school | 13:40 |
@lisitsyn | ooh wait I did that too | 13:40 |
@sonney2k | so well earned | 13:40 |
@sonney2k | deserved | 13:40 |
@lisitsyn | sonney2k: where the wall was? | 13:41 |
@lisitsyn | which street? | 13:41 |
@HeikoS | haha, you should use your skills to guard our workshop sergey :) | 13:41 |
-!- travis-ci [~travis-ci@ec2-107-21-134-219.compute-1.amazonaws.com] has joined #shogun | 13:41 | |
travis-ci | [travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/6985876 | 13:41 |
-!- travis-ci [~travis-ci@ec2-107-21-134-219.compute-1.amazonaws.com] has left #shogun [] | 13:41 | |
@HeikoS | wiking: does removing the asserts help? | 13:41 |
@lisitsyn | HeikoS: I did that but I am terribly bad in it I think :D | 13:41 |
@lisitsyn | HeikoS: is friedrichstrasse the street the wall was at? | 13:43 |
votjakovr | HeikoS: Hi, I'd like to ask: is there any bug in covSEiso (GPML matlab toolbox)? | 13:51 |
shogun-buildbot | build #1065 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1065 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 13:52 |
@sonney2k | lisitsyn, yes that too | 13:53 |
shogun-buildbot | build #1066 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1066 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 13:56 |
-!- van51 [~van51@athedsl-318776.home.otenet.gr] has joined #shogun | 14:01 | |
shogun-buildbot | build #769 of cyg1 - libshogun is complete: Failure [failed configure] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/769 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 14:02 |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Quit: Page closed] | 14:11 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 14:11 | |
deerishi | @HeikoS: i am doing git reset | 14:14 |
deerishi | git reset --soft .but what changes should i see in my repo | 14:14 |
@sonney2k | man if you want google to help you with mailman setup you find that http://www.dailyrecord.co.uk/news/crime/mailman-exposed-himself-motorists-mowed-1716575 | 14:14 |
@iglesiasg | hahaha | 14:15 |
-!- iglesiasg [c1934d18@gateway/web/freenode/ip.193.147.77.24] has quit [Quit: Page closed] | 14:16 | |
@sonney2k | HeikoS, btw did you get some kind of email that you couldn't post to the list or so? | 14:17 |
-!- nube [~rho@49.244.99.69] has joined #shogun | 14:21 | |
shogun-buildbot | build #1067 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1067 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 14:23 |
-!- nube [~rho@49.244.99.69] has quit [Quit: Leaving.] | 14:33 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 14:37 | |
@wiking | HeikoS: well if i remove the assert of course it'll fix it :P | 14:37 |
@HeikoS | sonney2k: yes two | 14:40 |
@HeikoS | wiking: I will update the stuff soon, could you add a github issue and assign it to me? | 14:40 |
-!- travis-ci [~travis-ci@ec2-23-20-196-170.compute-1.amazonaws.com] has joined #shogun | 14:43 | |
travis-ci | [travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/6986461 | 14:43 |
-!- travis-ci [~travis-ci@ec2-23-20-196-170.compute-1.amazonaws.com] has left #shogun [] | 14:43 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 14:46 | |
@wiking | HeikoS: ah but i thought i can fix it :) | 14:47 |
@HeikoS | wiking: well its a stupid test | 14:47 |
@HeikoS | so only way to fix it is to rewrite :) | 14:47 |
@wiking | ah no shit | 14:47 |
@HeikoS | fix data and assert correct numbers | 14:48 |
@wiking | why is it there in the first place!? :) | 14:48 |
@HeikoS | since I was not thinking properly | 14:48 |
@wiking | clearly | 14:48 |
@wiking | :) | 14:48 |
@HeikoS | :) | 14:48 |
@wiking | CLEARLY | 14:48 |
@wiking | ok i have to remove it because i want to see what travis does with the unit tests with the new PRNG | 14:49 |
@wiking | but it's only from feature/PRNG branch | 14:49 |
@sonney2k | HeikoS, please forward to me | 14:52 |
@HeikoS | sonney2k: dont you get a copy anyway? | 14:52 |
deerishi | @HeikoS : i did git reset --hard.then i did git commit -m "kpca".it said nothing to commit.working directory clean | 14:53 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * e270af7 / examples/undocumented/libshogun/Makefile: https://github.com/shogun-toolbox/shogun/commit/e270af7eff0a9a9c6333e4861001eee375fda949 | 14:53 |
shogun-notifier- | shogun: Remove statistics_hsic example run | 14:53 |
@wiking | mmm let's see what travis does now | 14:53 |
@HeikoS | sonney2k: could you help deerishi, no idea whats the problem there | 14:54 |
-!- naywhayare [~ryan@spoon.lugatgt.org] has quit [Ping timeout: 245 seconds] | 14:58 | |
@sonney2k | deerishi, --hard erases | 15:02 |
@sonney2k | you need to do --soft | 15:02 |
deerishi | what changes do i expect to see after reset in the repo | 15:03 |
@sonney2k | deerishi, all of them | 15:03 |
@wiking | ok lol | 15:04 |
@wiking | i have not seen this before | 15:04 |
@wiking | https://travis-ci.org/shogun-toolbox/shogun/jobs/6988387 | 15:04 |
@sonney2k | wiking, there is too much to wonder about | 15:05 |
deerishi | @sonny2k: i did git reset --soft in my main repo.after that i did git commit -m ,then git push origin branchname | 15:05 |
deerishi | it still came,everything is upto date | 15:06 |
@sonney2k | deerishi, use git push --force origin | 15:08 |
deerishi | @sonny2k :the data commits still seem to there | 15:12 |
@sonney2k | deerishi, well if you did commit everything ofcourse | 15:13 |
@sonney2k | deerishi, you should do git reset --soft <id> | 15:13 |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has joined #shogun | 15:13 | |
@sonney2k | then remove unnecessary files | 15:13 |
@sonney2k | or git commit only the needed files | 15:14 |
@lisitsyn | sonney2k: git consulting company sonnenburg and partners | 15:15 |
@lisitsyn | ;) | 15:15 |
@sonney2k | wiking, I *hate* exim & mailman to the deepest of my heart now | 15:16 |
-!- travis-ci [~travis-ci@ec2-107-21-134-219.compute-1.amazonaws.com] has joined #shogun | 15:16 | |
travis-ci | [travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/6988385 | 15:16 |
-!- travis-ci [~travis-ci@ec2-107-21-134-219.compute-1.amazonaws.com] has left #shogun [] | 15:16 | |
@sonney2k | man is qmail *simple* *easy* | 15:16 |
@sonney2k | *fast* | 15:16 |
@lisitsyn | sonney2k: why did you go away from simple easy fast thing - I didn't get the point | 15:16 |
@sonney2k | too bad the author is an %@^%$#! | 15:16 |
@sonney2k | lisitsyn, last update of qmail is 10 years ago or so | 15:17 |
@lisitsyn | haha | 15:17 |
@sonney2k | 0 maintainance | 15:17 |
@lisitsyn | stable release 1998 | 15:17 |
@lisitsyn | lol! | 15:17 |
@sonney2k | running on a machine with 128MB | 15:17 |
@wiking | sonney2k: they are great ;) | 15:17 |
@wiking | sonney2k: u just need to tweak them :DDDD | 15:17 |
vgorbati | lisitsyn: hi, sorry for being afk that long - holidays + preparation for the bachelor's work took too much time:) | 15:19 |
@wiking | HeikoS: any ideas why this could happen: statistics/QuadraticTimMD_unittest.cc:188: Failure | 15:19 |
@wiking | Expected: (CMath::abs(mean-CStatistics::mean(null_samples))) <= (10E-8), actual: 0.043256 vs 1e-07 | 15:19 |
@wiking | statistics/QuadraticTimMD_unittest.cc:189: Failure | 15:19 |
@wiking | Expected: (CMath::abs(var-CStatistics::variance(null_samples))) <= (10E-8), actual: 0.0285305 vs 1e-07 | 15:19 |
@sonney2k | wiking, HeikoS domain... | 15:20 |
@lisitsyn | vgorbati: I see - just let me know when you are not going to be around | 15:20 |
@sonney2k | I guess rng sucks :D | 15:20 |
@wiking | ahhahahah | 15:20 |
@sonney2k | vgorbati, before you are gone :D | 15:20 |
vgorbati | lisitsyn: I thought that you should have these holidays too:) | 15:21 |
@wiking | ah ok i think i know now | 15:21 |
@wiking | although again a funny feature that it doesn't fail on my machine | 15:21 |
@sonney2k | wiking, ? | 15:21 |
@wiking | sonney2k: he reinits the PRNG | 15:22 |
@lisitsyn | vgorbati: not only me but other guys :) | 15:22 |
@sonney2k | vgorbati, we have no holidays | 15:22 |
@sonney2k | we don't even sleep | 15:22 |
@lisitsyn | hah | 15:22 |
@lisitsyn | but actually I was here on holidays | 15:22 |
@wiking | CMath::init_random(1); | 15:22 |
@wiking | calls twice | 15:22 |
-!- abinayam [3d0c137b@gateway/web/freenode/ip.61.12.19.123] has joined #shogun | 15:22 | |
@wiking | so i'll have to change a little bit the code :P | 15:22 |
vgorbati | sonney2k: robots, or something?) | 15:22 |
@lisitsyn | sonney2k: I am going to have holidays tomorrow due to our old deal you know | 15:23 |
@lisitsyn | :D | 15:23 |
@sonney2k | vgorbati, exactly | 15:24 |
@wiking | yeeey i've managed to break it! | 15:24 |
-!- Yanglittle [deb20afa@gateway/web/freenode/ip.222.178.10.250] has joined #shogun | 15:24 | |
-!- foulwall [~foulwall@2001:da8:215:202:d49d:918b:1418:6e43] has quit [Ping timeout: 245 seconds] | 15:25 | |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.] | 15:31 | |
Yanglittle | excuse me , i installed the shogun, but it says can't find the module , does the version of python must be 2.5 or 2.6? | 15:31 |
@sonney2k | Yanglittle, anything from 2.4 should work | 15:32 |
-!- travis-ci [~travis-ci@ec2-23-20-196-170.compute-1.amazonaws.com] has joined #shogun | 15:32 | |
travis-ci | [travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/6988395 | 15:32 |
-!- travis-ci [~travis-ci@ec2-23-20-196-170.compute-1.amazonaws.com] has left #shogun [] | 15:32 | |
@sonney2k | wiking, congrats! | 15:32 |
@sonney2k | wiking, people with your talent are rare these days :P | 15:33 |
Yanglittle | :) | 15:45 |
@wiking | sonney2k: breakers? :) | 15:49 |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 15:52 | |
erlenda | exit | 15:53 |
-!- erlenda [~erlenda@fw-oslo.intelcom.no] has quit [Quit: Leaving] | 15:53 | |
-!- iglesiasg [c1934d18@gateway/web/freenode/ip.193.147.77.24] has joined #shogun | 15:59 | |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 16:00 | |
@iglesiasg | sonney2k: ping | 16:10 |
Yanglittle | it still can't impot the module, and says no the module | 16:10 |
@iglesiasg | Yanglittle: just with MulticlassLabels or also with others? | 16:12 |
Yanglittle | all , shogun.Features, shogun.Classifier | 16:13 |
Yanglittle | all can't find the module | 16:13 |
@iglesiasg | Yanglittle: so you are not able to execute any of the examples in examples/undocumented/python_modular I guess? | 16:14 |
@iglesiasg | Yanglittle: have you compiled, built and installed shogun with python_modular support? | 16:14 |
@iglesiasg | sth like: ./configure --interfaces=python_modular | 16:14 |
@iglesiasg | make | 16:14 |
@iglesiasg | make install | 16:15 |
Yanglittle | i run ./configure make make install | 16:16 |
@iglesiasg | Yanglittle: you need to specify python as I have written above | 16:17 |
Yanglittle | ok. | 16:18 |
@iglesiasg | Yanglittle: otherwise shogun is not built with python support | 16:18 |
@iglesiasg | Yanglittle: note also that install might require special permissions (i.e. sudo make install) | 16:18 |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 16:25 | |
@HeikoS | wiking | 16:26 |
@HeikoS | the broken travis is again due to stupid unit tests by me | 16:27 |
@sonney2k | HeikoS, could you please send your email to the list again? | 16:29 |
@sonney2k | iglesiasg, ? | 16:29 |
deerishi | @HeikoS: i just sent another PR.hopefully all the data commits are missing | 16:29 |
@iglesiasg | sonney2k: just wanted to let you know that I might be a bit absent until Saturday, when I doing my (hopefully) last exam towards my degree :) | 16:30 |
@sonney2k | deerishi, look it is again 9 commits | 16:30 |
@sonney2k | then good luck! | 16:30 |
@HeikoS | iglesiasg: good luck with that | 16:31 |
@iglesiasg | sonney2k: I am planning to do a contribution soon anyway, something for the hmsvm | 16:31 |
@HeikoS | deerishi: you can see yourself whether there is something wrong before sending the PR | 16:31 |
@iglesiasg | sonney2k, HeikoS : thanks! | 16:31 |
@sonney2k | HeikoS, please send the mail again! | 16:31 |
@iglesiasg | it is about automata theory, somewhat entertaining | 16:31 |
@HeikoS | sonney2k: I just did | 16:32 |
@sonney2k | HeikoS, please now! | 16:32 |
* sonney2k checks | 16:32 | |
@HeikoS | didnt come back | 16:32 |
@sonney2k | very good | 16:32 |
@sonney2k | greylisted | 16:32 |
@sonney2k | so it might arrive in >10mins | 16:32 |
@HeikoS | ok? | 16:33 |
@HeikoS | wiking, I correct myself, something is strange with the broken thing | 16:33 |
@sonney2k | I thHeyes | 16:33 |
@sonney2k | HeikoS, yes | 16:33 |
@HeikoS | wiking | 16:34 |
@HeikoS | there is CMath::init_random(1); in line 161 | 16:34 |
@HeikoS | if you also put that line into 175 it might work | 16:35 |
@sonney2k | HeikoS, did you get my email? | 16:42 |
@HeikoS | sonney2k: yes | 16:42 |
@sonney2k | k | 16:42 |
-!- abinayam [3d0c137b@gateway/web/freenode/ip.61.12.19.123] has quit [Ping timeout: 245 seconds] | 16:45 | |
-!- iglesiasg [c1934d18@gateway/web/freenode/ip.193.147.77.24] has quit [Quit: Page closed] | 16:49 | |
-!- gsomix [~Miranda@188.168.3.6] has joined #shogun | 16:51 | |
gsomix | hi, guys | 16:51 |
-!- van51 [~van51@athedsl-318776.home.otenet.gr] has quit [Quit: Leaving.] | 16:53 | |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has quit [Ping timeout: 245 seconds] | 17:10 | |
Yanglittle | i installed python 2.4 and numpy, but it can't find the numpy when running ./configure... | 17:16 |
-!- Yanglittle [deb20afa@gateway/web/freenode/ip.222.178.10.250] has quit [Quit: Page closed] | 17:28 | |
-!- nube [~rho@49.244.120.94] has joined #shogun | 17:28 | |
@wiking | HeikoS: yeah i saw that | 17:30 |
@wiking | HeikoS: you mean line 182 or? | 17:31 |
@wiking | HeikoS: bootstrap_null uses CMath::random in some way i guess | 17:32 |
@wiking | ? | 17:32 |
-!- foulwall [~foulwall@2001:da8:215:c252:d15d:f218:cac9:393e] has joined #shogun | 17:35 | |
-!- nube [~rho@49.244.120.94] has quit [Quit: Leaving.] | 17:48 | |
-!- deerishi_ [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 17:51 | |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has joined #shogun | 17:52 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 17:52 | |
pickle27 | sonney2k: so the merge is all good then? | 17:53 |
-!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 17:53 | |
deerishi_ | @sonny2k : u told me to send the pr as a single commit | 17:57 |
deerishi_ | was that the only problme in the pr?or were there any data commits too? | 17:58 |
deerishi_ | HeikoS: were there any data commits in the pr? | 18:08 |
deerishi_ | sonny2k told me to send them as a single git commit | 18:08 |
deerishi_ | @HeikoS : u der? | 18:11 |
@wiking | deerishi_: trust me people will answer u when they are here | 18:11 |
@wiking | otherwise they wont | 18:11 |
deerishi_ | k | 18:11 |
@wiking | need to be more patient | 18:11 |
deerishi_ | allright.i was just wanted to get over this pr.the code is working fine,but m having some issues with the merging part | 18:12 |
deerishi_ | i'l b more patient | 18:13 |
deerishi_ | :) | 18:13 |
deerishi_ | @wiking : do you have any idea on how to send as a single commit | 18:18 |
deerishi_ | if i have modified 4 files? | 18:19 |
@wiking | well git commit -a | 18:19 |
@wiking | that'll send all the modified files into the commit | 18:20 |
deerishi_ | then i should simply do git push origin branchname,rite? | 18:27 |
@wiking | y | 18:27 |
-!- deerishi_ [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Quit: Page closed] | 18:44 | |
-!- gsomix [~Miranda@188.168.3.6] has quit [Read error: Connection reset by peer] | 18:45 | |
@sonney2k | HeikoS, hurray! so mailinglist works again \o/ | 18:48 |
-!- gsomix [~gsomix@188.168.3.6] has joined #shogun | 18:49 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has joined #shogun | 18:51 | |
deerishi | @wiking : i sent the pr again,but it still showed 11 commits instead of 1 | 18:52 |
-!- foulwall [~foulwall@2001:da8:215:c252:d15d:f218:cac9:393e] has quit [Remote host closed the connection] | 19:00 | |
-!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 19:03 | |
shogun-notifier- | shogun: van51 :develop * 9e2afda / / (4 files): https://github.com/shogun-toolbox/shogun/commit/9e2afdad9387da2e7f13e8771bbbe4129a906c03 | 19:03 |
shogun-notifier- | shogun: Update on #1042. Fixed order of returned kernels | 19:03 |
shogun-notifier- | shogun: Heiko Strathmann :develop * 2a8b6f1 / / (4 files): https://github.com/shogun-toolbox/shogun/commit/2a8b6f124a51e847832eb02566cfcea35af66177 | 19:03 |
shogun-notifier- | shogun: Merge pull request #1070 from van51/develop | 19:03 |
shogun-notifier- | shogun: | 19:03 |
shogun-notifier- | shogun: Update on #1042. Fixed order of returned kernels | 19:03 |
@HeikoS | deerishi: never do git commit -a | 19:04 |
@HeikoS | commit all files by hand | 19:04 |
@HeikoS | otherwise you will always commit these data things | 19:04 |
deerishi | The PR i had sent ,did it contain any data commits?i think i had removed all the data commits before sending the pr. | 19:05 |
deerishi | all the broken files named data,i had removed | 19:05 |
@HeikoS | deerishi: you can see this yourself | 19:06 |
@HeikoS | just click on diff | 19:06 |
deerishi | also i need your advice ,on the next bug .as in what should do next? | 19:08 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * 43af113 / src/shogun/mathematics/Math.cpp,src/shogun/mathematics/Math.h: https://github.com/shogun-toolbox/shogun/commit/43af113771b39b83fd4409f40ef28f9bf67bab27 | 19:09 |
shogun-notifier- | shogun: Change CMath::init_random to use CRandom's set_seed | 19:09 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * 9ea5c08 / tests/unit/statistics/QuadraticTimeMMD_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/9ea5c08da92a1a9d6f0492676c9e3ad77aa2e27f | 19:09 |
shogun-notifier- | shogun: Fix QuadraticTimeMMD unit test | 19:09 |
shogun-notifier- | shogun: the seed wasn't set in the right way | 19:09 |
deerishi | @HeikoS: also sonny2k told me to submit as a single commit | 19:10 |
@wiking | letssee if that one fixed it | 19:10 |
@HeikoS | deerishi: yes that is all three files, not more, so add them by hand with one single commit, dont do commit -a | 19:10 |
@HeikoS | deerishi: what do you mean by "on the next bug?" | 19:11 |
deerishi | i want to contribute more.should i start working on the proposal or should i solve more issues | 19:11 |
@HeikoS | deerishi: so the first thing you should do is to get this patch submitted | 19:17 |
@HeikoS | since proposal deadline is over, the only thing that is left is to submit more good patches | 19:18 |
shogun-buildbot | build #770 of cyg1 - libshogun is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/770 | 19:20 |
-!- travis-ci [~travis-ci@ec2-107-22-117-126.compute-1.amazonaws.com] has joined #shogun | 19:24 | |
travis-ci | [travis-ci] it's Heiko Strathmann's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/6995347 | 19:24 |
-!- travis-ci [~travis-ci@ec2-107-22-117-126.compute-1.amazonaws.com] has left #shogun [] | 19:24 | |
deerishi | @HeikoS: this time i manually added the files: git add -f [names of the 4 files] ,them did git commit -m "kpca",then git push origin branchname | 19:37 |
shogun-buildbot | build #1068 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1068 blamelist: Heiko Strathmann <heiko.strathmann@gmail.com> | 19:38 |
deerishi | @HeikoS : could you take a look at this https://github.com/deerishi/shogun/compare/mm5 | 19:38 |
-!- lisitsyn [~blackburn@188-122-243-31.clients.tlt.100megabit.ru] has joined #shogun | 19:41 | |
-!- mode/#shogun [+o lisitsyn] by ChanServ | 19:41 | |
-!- travis-ci [~travis-ci@ec2-107-22-117-126.compute-1.amazonaws.com] has joined #shogun | 19:41 | |
travis-ci | [travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/6995546 | 19:41 |
-!- travis-ci [~travis-ci@ec2-107-22-117-126.compute-1.amazonaws.com] has left #shogun [] | 19:41 | |
@wiking | mmmm can i serialize actually mutliclassliblinear? | 19:47 |
@sonney2k | wiking, you will find out when you tried :D | 19:49 |
@wiking | sonney2k: i have like a fucking 300M file | 19:50 |
@wiking | serialized out | 19:50 |
@sonney2k | wiking, ohh yeah | 19:50 |
@sonney2k | there is room for improvement | 19:50 |
@sonney2k | we save everything multiple times | 19:51 |
@sonney2k | if e.g. you have 10 different kernels all using the same features | 19:51 |
@sonney2k | -> 10times the features | 19:51 |
@sonney2k | deerishi, one suggestion: | 19:51 |
@sonney2k | deerishi, copy the files you intended to submit | 19:51 |
@sonney2k | then create a fresh clone | 19:51 |
@sonney2k | copy the files back into the clone | 19:51 |
@sonney2k | do a git commit and push | 19:52 |
@sonney2k | and do a PR | 19:52 |
@lisitsyn | wiking: if you had 2x more classes | 19:52 |
@lisitsyn | you would have 600M | 19:52 |
@lisitsyn | :D | 19:52 |
deerishi | @sonny2k : but wont that also get back the data commits too. | 19:52 |
@sonney2k | HeikoS, just to get some feedback | 19:52 |
deerishi | @sonny2k : could you take a look at this https://github.com/deerishi/shogun/compare/mm5 | 19:52 |
@sonney2k | deerishi, no | 19:52 |
@sonney2k | deerishi, yeah it is horrible just don't touch all these data etc files | 19:53 |
@sonney2k | HeikoS, say we wanted to improve serialization by avoiding duplicates | 19:53 |
@sonney2k | lisitsyn, and that yes | 19:53 |
@sonney2k | HeikoS, so how would we do it - since you are the master of disaster - I would say we introduce some running number of objects | 19:54 |
shogun-buildbot | build #1069 of deb3 - modular_interfaces is complete: Failure [failed compile csharp_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1069 blamelist: van51 <vangelis_51@hotmail.com> | 19:54 |
@sonney2k | whenever objects are the same (meaning same pointer!) they will just be referenced by the running number ptr | 19:55 |
@sonney2k | lisitsyn, have you seen? the mailinglist works again :D | 19:55 |
@wiking | lisitsyn: when i deserialize the machine the w (one of the class) equal original feature size + numclasses | 19:55 |
@sonney2k | ohh and new record | 19:55 |
@sonney2k | 315 subscribers | 19:55 |
@lisitsyn | sonney2k: I received test message | 19:55 |
@sonney2k | yeah and I the answer | 19:56 |
@lisitsyn | wiking: oops? | 19:56 |
@sonney2k | so I think mail works fine again | 19:56 |
@wiking | lisitsyn: yeah i have no clue why is that happening | 19:56 |
@wiking | lisitsyn: but i'm getting an assertation fail | 19:56 |
@lisitsyn | dissertation fail | 19:56 |
@wiking | that w.vlen != feautresize | 19:56 |
@lisitsyn | we need a hero! | 19:57 |
@lisitsyn | :D | 19:57 |
@sonney2k | HeikoS, does that sound reasonable? back then when you wrote the migration framework you did have to load all original data anyways | 19:57 |
@sonney2k | HeikoS, so the same thing would have to be done and then the objects need to be filled | 19:58 |
@sonney2k | wiking, let me guess - sparse data? | 19:58 |
@sonney2k | wiking, btw one way to make the model small is the clear the features away | 19:59 |
@sonney2k | wiking, then you just have the w | 19:59 |
@sonney2k | 's | 19:59 |
@wiking | sonney2k: noup dense | 19:59 |
@wiking | i mean the very interesting thing is that multiclassliblibnear takes like 5+ minutes to train 20k vectors with 170 features | 20:01 |
@lisitsyn | wiking: could be faster.. | 20:05 |
@lisitsyn | wiking: how many classes? | 20:05 |
@HeikoS | sonney2k: I gotta go now, sorry, could you send me an email with this? didnt get it from the chat... | 20:06 |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 20:06 | |
-!- van51 [~van51@77.49.220.109.dsl.dyn.forthnet.gr] has joined #shogun | 20:06 | |
deerishi | @sonny2k: @HeikoS told me to remove all the data commits,if i say what you suggested then the data commits will still reamin | 20:07 |
deerishi | *remain | 20:07 |
@sonney2k | HeikoS, it was just about avoiding duplicates | 20:07 |
@sonney2k | pickle27, did you already refactor that train function into some smaller sub-units? | 20:08 |
@sonney2k | it's been a bit messy | 20:08 |
@sonney2k | deerishi, look current git develop is OK - so if you just git add your 3 files to that and commit & push you have a perfectly legit PR | 20:09 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * 43028bf / examples/undocumented/libshogun/evaluation_cross_validation_multiclass_mkl.cpp: https://github.com/shogun-toolbox/shogun/commit/43028bfaa4d9cd3baf1906080210fcdc4c782c03 | 20:09 |
shogun-notifier- | shogun: Fix evaluation_cross_validation_multiclass_mkl.cpp for new PRNG | 20:09 |
@sonney2k | so just stop wondering and do it and be happy! | 20:09 |
deerishi | alright.i will do that again.although i have done that a couple of times before. | 20:13 |
-!- viksvampire19_ [~chatzilla@115.184.13.194] has joined #shogun | 20:14 | |
-!- viksvampire19_ is now known as viksvampire19 | 20:14 | |
viksvampire19 | hi sonney2k and lisitsyn | 20:17 |
@lisitsyn | hey | 20:19 |
viksvampire19 | As I have submitted my proposal for GSOC , I need ur reviews on it n what should I do before the results come out? | 20:21 |
gsomix | lisitsyn, yo, maaaan | 20:21 |
@lisitsyn | viksvampire19: feel free to contibute to support your proposal with real code | 20:22 |
viksvampire19 | ok..so I should start coding on the same .. I mean I have coded the general csv reader that supports SGMatrix..I will put it up at my github profile..therz only a simple ascii reader.. | 20:25 |
@lisitsyn | please issue a PR so we can comment and discuss it | 20:25 |
viksvampire19 | sure..I gues I have issued PR.. | 20:26 |
-!- van511 [~van51@athedsl-408474.home.otenet.gr] has joined #shogun | 20:27 | |
-!- viksvampire19 [~chatzilla@115.184.13.194] has quit [Quit: ChatZilla 0.9.90 [Firefox 20.0/20130329030832]] | 20:29 | |
-!- travis-ci [~travis-ci@ec2-107-22-117-126.compute-1.amazonaws.com] has joined #shogun | 20:30 | |
travis-ci | [travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/6997381 | 20:30 |
-!- travis-ci [~travis-ci@ec2-107-22-117-126.compute-1.amazonaws.com] has left #shogun [] | 20:30 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds] | 20:30 | |
-!- deerishi [73f88294@gateway/web/freenode/ip.115.248.130.148] has quit [Ping timeout: 245 seconds] | 20:30 | |
-!- van51 [~van51@77.49.220.109.dsl.dyn.forthnet.gr] has quit [Ping timeout: 276 seconds] | 20:31 | |
-!- van511 [~van51@athedsl-408474.home.otenet.gr] has quit [Client Quit] | 20:31 | |
-!- van51 [~van51@athedsl-408474.home.otenet.gr] has joined #shogun | 20:32 | |
-!- hushell [43bd6474@gateway/web/freenode/ip.67.189.100.116] has quit [Ping timeout: 245 seconds] | 20:33 | |
-!- van51 [~van51@athedsl-408474.home.otenet.gr] has quit [Ping timeout: 268 seconds] | 20:45 | |
-!- iglesiasg [d58f32e4@gateway/web/freenode/ip.213.143.50.228] has joined #shogun | 21:06 | |
@wiking | mmmm lisitsyn ^ see that... | 21:06 |
@wiking | lisitsyn: https://travis-ci.org/shogun-toolbox/shogun/jobs/6997383 | 21:06 |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 21:06 | |
@lisitsyn | wiking: can't see what failed | 21:07 |
@wiking | lisitsyn: ff? | 21:07 |
@wiking | lisitsyn: IsomapTest.neighbors_preserving | 21:07 |
@lisitsyn | ohhh | 21:07 |
@wiking | converter/Isomap_unittest.cc:194: Failure | 21:08 |
@wiking | Expected: ((float64_t) similar_elements_count /(float64_t) total_elements_count) >= (min_similarity_level), actual: 0.4 vs 0.8 | 21:08 |
@wiking | #similarElements/#total < minimal similarity level. | 21:08 |
@wiking | i mean it's with the new PRNG | 21:08 |
@wiking | and funny enouhg | 21:08 |
@wiking | it only fails with clang on travis | 21:08 |
@wiking | gcc went w/o problems :D | 21:08 |
@wiking | and of course on my machine it works w/o problems :D | 21:09 |
@lisitsyn | vgorbati: хеллоу | 21:10 |
vgorbati | lisitsyn: hi) | 21:11 |
@lisitsyn | vgorbati: houston has got a problem | 21:11 |
vgorbati | lisitsyn: what problem? | 21:11 |
@lisitsyn | vgorbati: check previous messages | 21:11 |
@lisitsyn | :) | 21:11 |
@lisitsyn | isomap tests fails sometimes now | 21:12 |
@lisitsyn | :D | 21:12 |
@lisitsyn | wiking: okay it is not landmark isomap | 21:12 |
@lisitsyn | so it is not random based | 21:12 |
vgorbati | lisitsyn: it can't be that bad | 21:13 |
vgorbati | lisitsyn: I mean 0.4 | 21:14 |
@lisitsyn | vgorbati: yeah I wonder what happened | 21:14 |
@lisitsyn | wiking: so ONLY clang + travis + prng | 21:14 |
@lisitsyn | what is so special in travis? why it fails big time | 21:15 |
-!- iglesiasg [d58f32e4@gateway/web/freenode/ip.213.143.50.228] has quit [Quit: Page closed] | 21:23 | |
vgorbati | lisitsyn: I do not even what to say:) There is even 0.1 similarity level - it is 9 of 10 neighbors missing:) From your previous discussion I see that it can be because of new random number generator, so maybe switch the test to be fully deterministic? | 21:24 |
@lisitsyn | vgorbati: would be nice if we could reproduce the error | 21:25 |
@lisitsyn | vgorbati: is there anything you are working on (shogun/tapkee wise)? | 21:25 |
vgorbati | lisitsyn: well, I am kind a working on the manifold sculpting, but I guess I can put it on hold | 21:26 |
vgorbati | lisitsyn: but I have no idea how to reproduce this issue) | 21:27 |
@lisitsyn | vgorbati: ah! remember :D | 21:27 |
@lisitsyn | how is it going? | 21:27 |
vgorbati | lisitsyn: Given that there are 5 algorithm steps (more or less comparable), I am approaching the 3rd one | 21:28 |
@lisitsyn | I see | 21:29 |
@wiking | lisitsyn: no idea | 21:41 |
@wiking | :( | 21:41 |
@lisitsyn | wiking: okay I'll take a look again tomorrow | 21:43 |
@lisitsyn | if you can just ignore it for now | 21:44 |
@wiking | lisitsyn: yeah i mean i easily ignore it | 21:46 |
@wiking | just that i dont know why it's happening | 21:46 |
@wiking | but yeah the PRNG branch is growing big time | 21:46 |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 21:46 | |
@wiking | it'll need some fair amount of merging :P | 21:46 |
shogun-notifier- | shogun: Viktor Gal :feature/PRNG * 01c0bea / examples/undocumented/libshogun/Makefile: https://github.com/shogun-toolbox/shogun/commit/01c0bea0b8e5bffd435a8ebec76f2997b8bce7eb | 21:48 |
shogun-notifier- | shogun: Add back temporarily removed statistics_hsic | 21:48 |
@wiking | lisitsyn: lol | 21:55 |
@wiking | https://travis-ci.org/shogun-toolbox/shogun/jobs/7000221 | 21:55 |
@wiking | it worked fine now | 21:55 |
@lisitsyn | wiking: ohh | 21:55 |
@lisitsyn | a chance! | 21:55 |
@lisitsyn | wiking: you won a million! | 21:55 |
@wiking | ok i'll have to run like kazillion times | 21:56 |
@lisitsyn | wiking: hmm may be graph was disconnected | 21:56 |
@wiking | buildbot on this | 21:56 |
@lisitsyn | we can fix that I think | 21:56 |
@wiking | ok there's now the integration tests failing | 21:57 |
@wiking | but that's kind of like | 21:57 |
@wiking | normal | 21:57 |
@wiking | as we saved the data for another PRNG | 21:57 |
@lisitsyn | yeah | 21:57 |
@wiking | gcc failed | 21:57 |
@wiking | halleluya | 21:57 |
@wiking | fuckthisshit | 21:58 |
@sonney2k | wiking, haha | 21:58 |
@wiking | hahaha | 21:58 |
@wiking | i know why | 21:59 |
@wiking | :D | 21:59 |
@sonney2k | enlighten us :) | 21:59 |
@wiking | at least! | 21:59 |
@wiking | ok so the order of unit tests are different | 21:59 |
@sonney2k | ahh ... | 21:59 |
@wiking | and there's for sure some random somewhere | 21:59 |
@wiking | but still i think | 21:59 |
@wiking | it should never fail | 21:59 |
@wiking | regardless of the PRNG state | 22:00 |
@wiking | but yeah i mean IsomapTest.neighbors_preserving uses PRNG | 22:00 |
-!- votjakovr [~votjakovr@host-46-241-3-209.bbcustomer.zsttk.net] has quit [Quit: Leaving] | 22:00 | |
@wiking | and in the case where it doesn't fail | 22:01 |
@wiking | prng is reseeded | 22:01 |
@wiking | at least twice | 22:01 |
@wiking | while in the failed case it's not | 22:02 |
@wiking | ok i think this is some prng error | 22:02 |
@wiking | it has to be | 22:02 |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Quit: Page closed] | 22:02 | |
@wiking | lisitsyn: if the test features in IsomapTest.neighbors_preserving are from std distrib | 22:02 |
@wiking | it should never fail | 22:02 |
@wiking | right? | 22:02 |
@lisitsyn | wiking: I can't guarantee that :D | 22:03 |
@wiking | mmmm | 22:03 |
@wiking | lisitsyn: should i fix the seed then there? | 22:03 |
@lisitsyn | wiking: yeah something has to be done here | 22:03 |
@wiking | fill_matrix_with_test_data | 22:03 |
@lisitsyn | wiking: yeah if that helps | 22:03 |
@wiking | because u use CMath::randn_double | 22:04 |
@lisitsyn | wiking: later I think I have an idea what to do | 22:04 |
@wiking | ok i'll try to fix the seed there | 22:04 |
@wiking | and i'll try different seed | 22:04 |
@wiking | maaaybe i find a by chance a seed that makes it fail | 22:04 |
@lisitsyn | wiking: the thing is that it constructs neighborhood graph | 22:04 |
@lisitsyn | and if it is disconnected | 22:04 |
@lisitsyn | the result won't be much helpful | 22:05 |
@lisitsyn | some nans may be | 22:05 |
@lisitsyn | I either could throw an exception if it is not connected | 22:05 |
@wiking | maaaybe i find a by chance a seed that makes it fail | 22:05 |
@wiking | mmmm | 22:05 |
@wiking | no not here :) | 22:05 |
@sonney2k | lisitsyn, well you should seed with something that works then | 22:05 |
@wiking | i'm just trying different seeds | 22:05 |
@sonney2k | and then it is ok | 22:05 |
@lisitsyn | sonney2k: I think I need a routine that connects components | 22:06 |
@lisitsyn | or selects the largest (in means of size) component | 22:06 |
@wiking | ok i'm trying now the 5th differnt seed | 22:06 |
@wiking | couldnt find a failing one | 22:06 |
@wiking | how can we fix the order of unit tests? | 22:06 |
@wiking | maybe i can then reproduce the thing | 22:06 |
@sonney2k | wiking, IMHO before each unit test is run a seed should be set | 22:07 |
@sonney2k | or in each unit test that has some randomness | 22:07 |
@wiking | i mean it's a good thing to have some randomness imho | 22:08 |
@wiking | as this case | 22:08 |
@wiking | because maaaybe i'm still doing something wrong | 22:08 |
@wiking | in the PRNG | 22:08 |
@wiking | the comment for SFMT and dSFMT is done by japanese | 22:08 |
@wiking | and their english is not so much clear in some cases | 22:09 |
@wiking | :D | 22:09 |
@wiking | sonney2k: btw for these apis | 22:10 |
@wiking | random(float32_t min_value, float32_t max_value) | 22:10 |
@wiking | can we do that it's (min_value, max_value) and not the closed interval | 22:10 |
@wiking | as that we the PRNG is much faster | 22:10 |
-!- travis-ci [~travis-ci@ec2-184-72-78-247.compute-1.amazonaws.com] has joined #shogun | 22:10 | |
travis-ci | [travis-ci] it's Viktor Gal's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/7000219 | 22:10 |
-!- travis-ci [~travis-ci@ec2-184-72-78-247.compute-1.amazonaws.com] has left #shogun [] | 22:10 | |
@wiking | i mean that the actual return value of that function is in (min_value, max_value) and not [min_value, max_value] | 22:11 |
vgorbati | lisitsyn: here? | 22:19 |
@lisitsyn | vgorbati: yes | 22:19 |
vgorbati | lisitsyn: I wanted to ask: in the algorithm Gashler says, that he preprocesses the data using PCA, but without reducing the dimensionality: he rotates the points around first n_target_dim principal components. | 22:22 |
@wiking | sonney2k: 0/1? :) | 22:22 |
vgorbati | lisitsyn: I am not sure what he means by this. As I see it, you can't rotate around 2 vectors in 3-D space, you need 3 vectors | 22:22 |
vgorbati | lisitsyn: And since PCA gives perpendicular eigen-vectors, the only way that I see to rotate while preserving dimensions is to use all eigen-vectors to project the data | 22:23 |
@wiking | who fscked up regression? :) | 22:24 |
@lisitsyn | vgorbati: yeah makes sense to me | 22:24 |
@lisitsyn | vgorbati: I do not actually realize how to use only target_dim components here | 22:26 |
@lisitsyn | it won't be preprocessing as it already reduces everything | 22:26 |
@wiking | where is votjakovr | 22:26 |
@lisitsyn | wiking: barnaul is asleep I think | 22:27 |
@wiking | sonney2k: new measure for deciding who gets that 8 slots: number of times you break travis :) | 22:27 |
@lisitsyn | wiking: just check where it is :D | 22:27 |
@wiking | the top 2 guys gets out from the game | 22:27 |
@lisitsyn | not china but half a way | 22:27 |
@wiking | votjakovr++ | 22:27 |
@wiking | :) | 22:27 |
vgorbati | lisitsyn: huuh, because I actually spend quite a lot of time trying to understand how he does it, and did not get it at all:) Maybe an algorithm that he gives looks familiar for you: it is here http://axon.cs.byu.edu/papers/gashler2011smc.pdf, on the second last page | 22:28 |
@wiking | but of course if you fix somebody else's error you get -1 | 22:28 |
@wiking | ;) | 22:28 |
@lisitsyn | vgorbati: let me check | 22:29 |
@wiking | hence u can correct your earlier mistakes :) | 22:29 |
@wiking | sonney2k: ^ totally good measure :P | 22:29 |
@wiking | lisitsyn: i'll hack your github accout and break the whole shogun :))0 | 22:29 |
@lisitsyn | wiking: uhmm and? | 22:29 |
@lisitsyn | :D | 22:29 |
@wiking | you'll have like +1000 :D | 22:30 |
@lisitsyn | wiking: aaand? | 22:30 |
@wiking | and you are out | 22:30 |
@wiking | :) | 22:30 |
@lisitsyn | wiking: out of? | 22:30 |
@wiking | gsoc | 22:30 |
@lisitsyn | wiking: so I won't be admin anymore? | 22:31 |
* wiking is tired... | 22:31 | |
@lisitsyn | :D | 22:31 |
@wiking | lisitsyn: nada | 22:31 |
@wiking | zilch | 22:31 |
@wiking | we'll have to find another russian slave to do all the work | 22:31 |
@wiking | :S | 22:31 |
gsomix | http://www.youtube.com/watch?v=g3qkbc28R_I | 22:31 |
@wiking | gsomix: wtf is this? :D | 22:32 |
gsomix | it's fine | 22:32 |
@wiking | allzgood | 22:32 |
@wiking | damnshiats | 22:32 |
@wiking | we'll never have trees in shogun :( | 22:32 |
@wiking | i saw this and i got facepalmed | 22:32 |
@wiking | #325 | 22:33 |
@wiking | Do you expect a code translation from java from weka to C++? | 22:33 |
@lisitsyn | not sure what she means | 22:33 |
@lisitsyn | vgorbati: again , page 2? | 22:34 |
@wiking | lisitsyn: well i guess it's like 'manual translation' | 22:34 |
vgorbati | lisitsyn: second from the end | 22:34 |
@wiking | lisitsyn: hope not some codegenerator :)))) | 22:34 |
@lisitsyn | wiking: it is pain in da azz | 22:34 |
@wiking | lisitsyn: hehehehe | 22:34 |
@wiking | lisitsyn: that'd be awesomeness :D | 22:34 |
@wiking | lets start codegen evertyhing into shogun | 22:34 |
@wiking | :DDD | 22:34 |
@lisitsyn | vgorbati: do 20 times? | 22:35 |
@lisitsyn | :D | 22:35 |
@lisitsyn | oh I recognize gashler | 22:35 |
vgorbati | lisitsyn: as I said, he loves magic constants) | 22:35 |
@wiking | lisitsyn: like /me wantz linux kernel in shogun | 22:35 |
gsomix | wiking, codegenerator of codegenerators | 22:35 |
@wiking | gsomix: yeah !!!! | 22:35 |
@wiking | and swig it! | 22:35 |
@wiking | as always | 22:35 |
@lisitsyn | vgorbati: did I told you he computes Ax 100 times to compute largest eigenvector? | 22:35 |
@wiking | :) | 22:35 |
@lisitsyn | 100? what the hell.. | 22:36 |
@lisitsyn | do 20 times | 22:36 |
@lisitsyn | argh | 22:36 |
@lisitsyn | vgorbati: I do not recognize the algorithm | 22:37 |
vgorbati | lisitsyn: :D This algorihtm just makes very little sense to me) | 22:37 |
vgorbati | lisitsyn: if any | 22:37 |
@lisitsyn | it is like gram-schmidt stuff | 22:37 |
@lisitsyn | 20 times | 22:37 |
@lisitsyn | :D | 22:37 |
@lisitsyn | vgorbati: as I understand | 22:38 |
@lisitsyn | it is kind of optional | 22:38 |
@lisitsyn | just to help that thing to converge | 22:38 |
@lisitsyn | right? | 22:38 |
vgorbati | lisitsyn: yes | 22:38 |
@lisitsyn | vgorbati: I see something from GS orthogonalization actually here | 22:39 |
vgorbati | lisitsyn: what does gramm-schmidt process do? | 22:39 |
@lisitsyn | vgorbati: finds an orthogonal basis from a set of vectors | 22:39 |
@lisitsyn | basically the same as QR decomposition | 22:40 |
vgorbati | lisitsyn: but he uses it on principal components - what is the need, or do I miss something? | 22:40 |
@lisitsyn | vgorbati: no I don't mean it does the same - I just mean it resembles some parts of it | 22:41 |
@lisitsyn | well any iterative projection would | 22:41 |
vgorbati | lisitsyn: you are right, because in the code there's even a comment: "Use the modified Gram-Schmidt process to compute the other plane vector") | 22:42 |
@lisitsyn | hah lucky guess | 22:42 |
@lisitsyn | vgorbati: I have no idea how it was published :D | 22:43 |
@lisitsyn | my experience of JMLR submission tells reviewers would not like that | 22:44 |
@lisitsyn | why 20 times :D | 22:44 |
-!- hushell [43bd6474@gateway/web/freenode/ip.67.189.100.116] has joined #shogun | 22:45 | |
@lisitsyn | how can that be fixed in pseudocode.. | 22:45 |
vgorbati | lisitsyn: :) I guess I will skip this step, or maybe just rotate as we discussed? | 22:45 |
@lisitsyn | vgorbati: just skip it for now | 22:45 |
@lisitsyn | vgorbati: actually that 20 times step reminds me power iteration method | 22:45 |
-!- van51 [~van51@athedsl-320452.home.otenet.gr] has joined #shogun | 22:46 | |
@lisitsyn | vgorbati: okay just my thoughts on that algorithm | 22:47 |
@lisitsyn | we subtract mean from the data to center it (it is standard PCA stuff) | 22:47 |
@lisitsyn | we also store copy of vectors as we are going to rotate them a little | 22:48 |
@lisitsyn | next, we project random vectors to each component | 22:48 |
@lisitsyn | then we modify vectors so that we get them orthogonal to random vector | 22:49 |
@lisitsyn | I don't know what is the meaning but that's the thing I notice to be similar to G-S | 22:49 |
@wiking | oh shit i haven't even realized that we have Marsaglia polar method for std normal distrib | 22:49 |
@lisitsyn | and finally we rotate G | 22:50 |
@lisitsyn | actually what we see here is just unrolled Givens (rotation) matrix multiplication | 22:50 |
@lisitsyn | what we have found that way is a transformation matrix that aligns it | 22:50 |
@lisitsyn | by it I mean P | 22:51 |
@lisitsyn | wiking: yes of course | 22:51 |
@lisitsyn | I even remember who did that | 22:51 |
hushell | Hey guys, what's blame list? | 22:51 |
@wiking | lisitsyn: u? | 22:51 |
@lisitsyn | no | 22:51 |
@wiking | lisitsyn: HeikoS i guess | 22:51 |
@lisitsyn | wiking: no! | 22:51 |
@lisitsyn | justin patera | 22:52 |
@wiking | i should do ziggurat | 22:52 |
@lisitsyn | hushell: just a list of emails who could be blamed for breakin things | 22:52 |
@wiking | where's justin? | 22:52 |
@lisitsyn | wiking: he didn't get to gsoc 2011 | 22:53 |
@lisitsyn | so he is not with shogun :) | 22:53 |
@wiking | :( | 22:53 |
hushell | I saw the merge from my branch failed in travis :( and been blamed | 22:54 |
@lisitsyn | hushell: oh yeah we have everything failing | 22:54 |
@lisitsyn | :D | 22:54 |
hushell | :D what's wrong here? anyway to fix? | 22:54 |
@lisitsyn | we need stalin or so I don't know | 22:55 |
@wiking | WE NEED LENIN! | 22:55 |
@wiking | where's good old comrad lenin | 22:55 |
@lisitsyn | kim jong il | 22:55 |
@wiking | mmm we need a lenin bot! | 22:55 |
-!- iglesiasg [d58f32bd@gateway/web/freenode/ip.213.143.50.189] has joined #shogun | 22:55 | |
@wiking | mmm what's a good bot nowadays? | 22:56 |
@lisitsyn | wiking: lenin? in red house right away from kremlin | 22:56 |
@lisitsyn | lying dead | 22:56 |
@lisitsyn | :D | 22:56 |
@wiking | eggdrop still any good? | 22:56 |
-!- mode/#shogun [+o iglesiasg] by ChanServ | 22:56 | |
-!- naywhayare [~ryan@spoon.lugatgt.org] has joined #shogun | 22:56 | |
@lisitsyn | wiking: no idea! | 22:56 |
@lisitsyn | eggdrop is something from my deep youth | 22:56 |
@wiking | hahahah | 22:56 |
@wiking | and u are like younger than me about 10 years | 22:57 |
@wiking | so imagine that eggdrop is from my youth as well | 22:57 |
@wiking | ;P | 22:57 |
@lisitsyn | I think 8 | 22:57 |
@lisitsyn | :D | 22:57 |
@wiking | Initial release December 1993 | 22:57 |
@wiking | :D | 22:57 |
@wiking | niiiize | 22:57 |
@wiking | of course i need tcl for it :DDD | 22:57 |
@wiking | no worries it's already building | 22:57 |
@wiking | lenin bot is coming! | 22:58 |
@sonney2k | lisitsyn, recall konrads bot here | 22:58 |
@lisitsyn | wiking: right two years after my born | 22:58 |
@sonney2k | that was nice | 22:58 |
@sonney2k | back in the days... | 22:58 |
@lisitsyn | sonney2k: yeah cool bot | 22:58 |
@wiking | should we call him lenin or vlad? | 22:58 |
@sonney2k | what was her name? | 22:58 |
@sonney2k | betty? | 22:58 |
@lisitsyn | sonney2k: bettyboo | 22:58 |
@wiking | sonney2k: sooooooo random on (0,1)? | 22:58 |
@wiking | yes yes? | 22:58 |
@sonney2k | wiking, you mean closed interval right? | 22:59 |
@wiking | sonney2k: i mean open | 22:59 |
@lisitsyn | shit lenin says | 22:59 |
@sonney2k | wiking, btw you can provide another non-closed interval function | 22:59 |
@wiking | i can do [0,1) and (0,1] | 22:59 |
@sonney2k | and then we can gradually switch | 22:59 |
@wiking | but not fast [0,1] | 22:59 |
@sonney2k | don't make this change too complex | 22:59 |
@lisitsyn | wiking: just return 0 sometimes | 23:00 |
@lisitsyn | that's ok! | 23:00 |
@wiking | 'sometimes' | 23:00 |
@wiking | :D | 23:00 |
@lisitsyn | yes! | 23:00 |
@lisitsyn | from time to time come on | 23:00 |
@lisitsyn | that's pure random | 23:00 |
@sonney2k | ahh nice my vpn to the states is good | 23:00 |
vgorbati | lisitsyn: I am just curious if I am right in this logical sequence: we can rotate only around a basis of vectors; if, say, we want to align with 1 principal vector in 3D space - we need 2 more vectors that will be perpendicular to that one -> but that will be 2 other principal components -> there's no way you can align with some subset of principal components without getting the other? | 23:00 |
@wiking | if (static counter > 123) counter =0 return - | 23:00 |
@wiking | return 0 | 23:00 |
@wiking | right? :) | 23:01 |
@sonney2k | wiking, I don't know you :P | 23:01 |
@wiking | sonney2k: and that's actually true :D | 23:01 |
@lisitsyn | wiking: yes that's pure random thing | 23:01 |
@lisitsyn | vgorbati: as I understand it is not *exact* align | 23:01 |
@lisitsyn | vgorbati: one remark here | 23:02 |
@lisitsyn | erich maria remarque | 23:02 |
@lisitsyn | :D | 23:02 |
@lisitsyn | okay | 23:02 |
vgorbati | lisitsyn: :D | 23:02 |
@lisitsyn | I mean | 23:02 |
@lisitsyn | the basis is not made of target_dim vectors | 23:03 |
@lisitsyn | it is made of vectors itself | 23:03 |
@sonney2k | wiking, did you ever try seafile? | 23:03 |
@wiking | sonney2k: what'sthat | 23:04 |
@lisitsyn | vgorbati: these vectors are just project so that we get them non-orthogonal | 23:04 |
@lisitsyn | projected* | 23:04 |
@lisitsyn | sorry, cont'd non-orthogonal to principal components | 23:05 |
@wiking | sonney2k: i dont want to actually create more api for cmath | 23:05 |
@wiking | sonney2k: i mean now random calls are still there but actually they shouldn't be... those should be marked deprecated | 23:05 |
@wiking | and everybody shoudl start calling the global sg_rand directly | 23:06 |
@sonney2k | wiking, just refactor that then | 23:07 |
@sonney2k | I am very much fine with that | 23:07 |
vgorbati | lisitsyn: trying to make it clear for me: the algorithm tries to project every data vector so that it won't be orthogonal with any of target_dim principal components? | 23:07 |
@lisitsyn | vgorbati: yes somewhat | 23:08 |
@lisitsyn | vgorbati: it constructs a basis | 23:08 |
@lisitsyn | that is not orthogonal to principal components (as possible) | 23:08 |
@wiking | sonney2k: ok first i add ziggzag to Random as well | 23:08 |
@lisitsyn | so when you project your vectors on that basis | 23:08 |
@sonney2k | wiking, seafile is an opensource dropbox | 23:08 |
@wiking | i.e. to have a fast gaussian generator | 23:09 |
@lisitsyn | you get them 'rotated' so that they are not orthogonal to principal components too | 23:09 |
@lisitsyn | I think the idea is that | 23:09 |
@lisitsyn | is that I described I mean :) | 23:09 |
@lisitsyn | sonney2k: what is wrong with das dropboox? | 23:10 |
vgorbati | lisitsyn: the main question: why?:) Why not just use pure PCA, but for dimension > target_dim | 23:11 |
@lisitsyn | vgorbati: if I knew | 23:11 |
@lisitsyn | :D | 23:11 |
@lisitsyn | wiking: https://github.com/lisitsyn/tapkee_website/blob/master/src/tapkee/core.clj more crazyness for you I know you like it | 23:18 |
@lisitsyn | :D | 23:18 |
@lisitsyn | that's what is being run here http://tapkee.lisitsyn.me/ | 23:19 |
vgorbati | lisitsyn: ok, leaving now, see you later | 23:19 |
@wiking | njiha | 23:19 |
@wiking | lol | 23:19 |
@lisitsyn | vgorbati: alright, see you | 23:20 |
-!- vgorbati [d4029f22@gateway/web/freenode/ip.212.2.159.34] has quit [Quit: Page closed] | 23:20 | |
-!- hushell [43bd6474@gateway/web/freenode/ip.67.189.100.116] has quit [] | 23:20 | |
@lisitsyn | wiking: clojure in action :D | 23:20 |
@wiking | yeps | 23:20 |
@lisitsyn | schlafen! | 23:31 |
@sonney2k | lisitsyn, man if you would spend half the time on shogun ... | 23:33 |
@sonney2k | beautiful examples - can we steal them? | 23:33 |
@lisitsyn | sonney2k: I do not mind doing them for shougn *once I am done with f&S*$#$ paper* | 23:34 |
@sonney2k | ok so in 10 years | 23:34 |
@lisitsyn | sonney2k: chris says he has just a few issues left | 23:35 |
@lisitsyn | :D | 23:35 |
@lisitsyn | sonney2k: I think I could even submit tomorrow | 23:36 |
@lisitsyn | sonney2k: what slows down is that we have iterations over paper (like one per week due to chris being busy) | 23:36 |
@lisitsyn | sonney2k: as for examples - we could put them live easily I think | 23:37 |
@lisitsyn | just copy images, .js-es and jsons | 23:38 |
@sonney2k | lisitsyn, but they are static right? | 23:39 |
@sonney2k | not live? | 23:39 |
@lisitsyn | sonney2k: lets just ask foulwall to merge it from my sources | 23:39 |
@lisitsyn | sonney2k: what do you mean by static/live? | 23:39 |
@sonney2k | not computed on-the-fly | 23:40 |
@lisitsyn | ahh no | 23:40 |
@sonney2k | the embedding I mean | 23:40 |
@sonney2k | that is sth we need | 23:40 |
@lisitsyn | sonney2k: easy - I just host on heroku which is troublesome for that | 23:40 |
@lisitsyn | sonney2k: just give them foulwall I think :) | 23:41 |
@lisitsyn | sonney2k: mnist embedding is ~1m btw | 23:45 |
@lisitsyn | not reasonable for web example | 23:45 |
@lisitsyn | I don't expect it to be any faster with t-sne | 23:45 |
@sonney2k | lisitsyn, we don't need all 60k examples | 23:48 |
@lisitsyn | sonney2k: ohh 60k is much more | 23:51 |
@lisitsyn | I used some subset, 2000 may be | 23:51 |
@lisitsyn | sonney2k: once it callback'ed more I could add a sparse callback which would be nice for mnist | 23:52 |
@sonney2k | hmmh pretty slow | 23:53 |
@sonney2k | then we have to subsample even more | 23:53 |
@sonney2k | anyway bed time | 23:53 |
van51 | sonney2k: can you wait for a couple of questions? :p | 23:56 |
van51 | anyway, I'll shoot and you can answer in the morning. no rush | 23:57 |
van51 | about the conversion in comb features/kernel, there are some methods like "get_first_kernel (CListElement *¤t)" that are dependant on the CList representation. should they be removed? | 23:58 |
van51 | or should they just do what they are supposed to do and be marked as deprecated or something? | 23:59 |
--- Log closed Thu May 09 00:00:37 2013 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!