IRC logs of #shogun for Tuesday, 2011-07-26

« Monday, 2011-07-25 Index Wednesday, 2011-07-27 »
--- Log opened Tue Jul 26 00:00:21 2011
@sonney2kheiko1, still awake?00:32
heiko1yes00:32
heiko1but, i really  want to go home now :)00:32
heiko1whats up?00:32
heiko1just sent a pull request00:33
heiko1sonney2k, so lets speack tomorrow :)00:35
@sonney2kheiko1, it is quasi schon morgen :)00:35
@sonney2kheiko1, ok cu!00:35
heiko1sonney2k, you should also go to bed :)00:35
@bettyboohoho heiko100:36
heiko1sonney2k, byebye00:36
-!- heiko1 [~heiko@134.91.52.56] has quit [Ping timeout: 258 seconds]00:40
-!- f-x [~user@117.192.204.152] has joined #shogun02:31
-!- f-x [~user@117.192.204.152] has quit [Ping timeout: 260 seconds]04:25
-!- f-x [~user@117.192.204.152] has joined #shogun04:37
-!- gsomix [~gsomix@109.169.131.11] has joined #shogun05:46
-!- gsomix [~gsomix@109.169.131.11] has quit [Client Quit]05:46
serialhexfor some reason when i compile just the ruby_modular interface, it locks here: https://gist.github.com/110595506:12
serialhexsorry i've been MIA, lots of work & not enough sleep06:13
-!- f-x [~user@117.192.204.152] has quit [Ping timeout: 260 seconds]06:43
-!- f-x [~user@117.192.215.140] has joined #shogun08:37
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun09:26
-!- blackburn [~blackburn@109.226.76.87] has joined #shogun09:49
-!- sploving1 [~sploving@124.16.139.194] has joined #shogun10:04
sploving1sonney2k, in python, swig_typemaps.i, it has "#include <numpy/arrayobject.h>". when i write ruby swig_typemaps.i, #include <narray.h>, why it said: cannot find ../../shogun/narray.h10:05
sploving1i have make installed narray. and narray.h and narray.so is in /usr/local/lib/site_ruby/1.8/i486-linux:.10:06
@sonney2kserialhex, it will just takes a few minutes to compile that's all - be patient10:06
blackburnsonney2k: I wrote small test generator for distances, preprocessors and kernels :D10:06
sploving1I set $PATH="$PATH:/usr/local/lib/site_ruby/1.8/i486-linux:."10:06
sploving1but it still cannot find narray.h, why??10:06
@sonney2ksploving1, please open an editor and edit shogun/.config10:07
sploving1serialhex, are you here?10:07
@sonney2ksploving1, sorry src/.config10:07
@sonney2ksploving1, btw did you test lua string support yet?10:08
sploving1not yet.10:08
@sonney2ksploving1, please do this first - I can then in the meantime do some configure magic to detect narray10:08
@sonney2kfor ruby10:08
sploving1my machine cannot show all the columns of the results10:09
sploving1like yesterday, it just show the first 8 columns10:09
sploving1so I cannot decide whether it is right or not10:09
@sonney2ksploving1, I meant just creating StringCharFeatures etc10:10
@sonney2ksploving1, python_modular/features_string_char_modular.py10:10
@sonney2kjust translate this to lua and test10:11
sploving1I will take a look at that10:12
sploving1sonney2k, can you support narray ?10:12
sploving1I have change the typemap, and would like to see narrya works well or not10:13
@sonney2ksure10:13
sploving1I compiled several times, and cannot find narray.h or "/home/sploving/shogun/src/interfaces/ruby_modular/modshogun_wrap.cxx:425348: undefined reference to `na_to_array'10:14
sploving1"10:14
sploving1/home/sploving/shogun/src/interfaces/ruby_modular/modshogun_wrap.cxx:425612: undefined reference to `cNArray'10:14
@sonney2ksploving1, please do the lua string test first - when that is done we discuss about narray OK?10:14
sploving1okay10:14
sploving1I will add it and you can hele me take a test. as you know, the kernel results in you machines show the same results. but in my machines it does not show the right10:16
@sonney2ksploving1, why don't you pipe it to a file and then open these files in an editor?10:16
sploving1I will have a try10:17
@sonney2kblackburn, you are awake?10:29
@sonney2kalready?10:29
blackburnyes10:29
blackburnis it strange?10:29
@sonney2ktotally missed you message10:29
@sonney2kyes you usually work until 5 am or so and thus sleep a bit longer10:29
blackburnwhich message?10:29
blackburnI was messing with some java examples last night and wrote a simple template generator10:30
@sonney2ktemplate generator?10:30
blackburnwrong word order10:30
blackburnby template10:30
blackburne.g. most of dim.reduction preprocessors works similar10:31
@sonney2ksimilar is within eps 1e-15 or so?10:31
blackburnno, I mean they are used in similar way10:32
blackburn%CLASS% preprocessor = %CLASS%();10:32
blackburnand so on10:32
blackburnso, just replacing %CLASS% - no need to write tests manually10:32
blackburnI would like to automate all the tests if possible10:33
@sonney2kblackburn, yes that would be great... but it needs some thought how to do that properly10:34
@sonney2kI mean ideally one would be able to use that same system on all languages10:35
@sonney2ks/on/for/10:35
blackburnfor now I'm going to get it work for java, may be later we could write some automagic10:36
blackburnbtw, distances are working10:37
blackburnstring distances to go though10:37
sploving1sonney2k, [ERROR] assertion a failed in file features/Alphabet.cpp line 9810:37
sploving1init();ASSERT(a);set_alphabet(a->get_alphabet());copy_histogram(a);10:37
sploving1what does ASSERT(a) mean?10:38
@sonney2kit is non-zero10:38
blackburnokay sorry gtg10:40
-!- blackburn [~blackburn@109.226.76.87] has quit [Quit: Leaving.]10:40
CIA-87shogun: Baozeng Ding master * rf5b05f4 / examples/undocumented/lua_modular/features_string_char_modular.lua : add string features lua example - https://github.com/shogun-toolbox/shogun/commit/f5b05f473c275c9cc78d2d102ee8101c80fac3cb11:27
CIA-87shogun: Soeren Sonnenburg master * r16fc2a1 / examples/undocumented/lua_modular/features_string_char_modular.lua :11:27
CIA-87shogun: Merge pull request #234 from sploving/master11:27
CIA-87shogun: add lua string features modular - https://github.com/shogun-toolbox/shogun/commit/16fc2a1b93f5e2121524a6d468a6f972b2f2d8c211:27
CIA-87shogun: Soeren Sonnenburg master * r93a63c7 / src/configure : add ruby narray detection in configure - https://github.com/shogun-toolbox/shogun/commit/93a63c749ac5994478b04611d696a6cc73eb807611:28
@sonney2ksploving1, and does the script return the same result?11:28
sploving1no, Wrong arguments for overloaded function 'StringCharFeatures_set_feature_vector'11:56
sploving1 Possible C/C++ prototypes are:11:56
sploving1    set_feature_vector(shogun::CStringFeatures< char > *,shogun::SGVector< char >,int32_t)11:57
sploving1    set_feature_vector(shogun::CStringFeatures< char > *,int32_t,char *,int32_t)11:57
sploving1I am looking why11:57
-!- sploving1 [~sploving@124.16.139.194] has left #shogun []12:01
-!- in3xes [~in3xes@180.149.49.227] has quit [Read error: Connection reset by peer]12:08
CIA-87shogun: Baozeng Ding master * r0654cea / (3 files in 2 dirs): support SGVector<char> - https://github.com/shogun-toolbox/shogun/commit/0654cead46731267087bad0b73baf5f51e5d7ac512:18
CIA-87shogun: Soeren Sonnenburg master * rcb17d3a / (3 files in 2 dirs):12:18
CIA-87shogun: Merge pull request #235 from sploving/master12:18
CIA-87shogun: support lua sgvector<char> - https://github.com/shogun-toolbox/shogun/commit/cb17d3ad839c39d942986d5955d165151e77482d12:18
CIA-87shogun: Baozeng Ding master * rc4fc6c5 / src/interfaces/ruby_modular/swig_typemaps.i : support ruby narray library - https://github.com/shogun-toolbox/shogun/commit/c4fc6c56b4039b949f0a939559a46ebad65c066212:23
CIA-87shogun: Soeren Sonnenburg master * rc9e1279 / src/interfaces/ruby_modular/swig_typemaps.i :12:23
CIA-87shogun: Merge pull request #236 from sploving/master12:23
CIA-87shogun: support ruby narray library - https://github.com/shogun-toolbox/shogun/commit/c9e1279f9fb46924a77d276ca155a1a20acbf91212:23
-!- sploving1 [~sploving@124.16.139.194] has joined #shogun12:31
sploving1sonney2k,preprocessor/LocallyLinearEmbedding.cpp:210: error: ‘arpack_dsaupd’ was not declared in this scope12:31
sploving1when I compile the upstream, it shows the above error.12:31
@sonney2ksploving1, hopefully fixed12:33
CIA-87shogun: Soeren Sonnenburg master * r2387fee / src/shogun/preprocessor/LocallyLinearEmbedding.cpp : fix compile failure w/o arpack - https://github.com/shogun-toolbox/shogun/commit/2387fee9f7071e46aaf9c70e4171dfb870473c4f12:33
sploving1I will try again.12:34
-!- sploving1 [~sploving@124.16.139.194] has left #shogun []12:34
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun13:33
CIA-87shogun: Soeren Sonnenburg master * r19b3616 / src/interfaces/python_modular/swig_typemaps.i : take care of dealloc of python objects in python typemaps and remove unused functions - https://github.com/shogun-toolbox/shogun/commit/19b3616126b6507227003aff3753a7c075d632e913:43
CIA-87shogun: Soeren Sonnenburg master * r5ee1b7e / src/configure : fix ruby linkflags - https://github.com/shogun-toolbox/shogun/commit/5ee1b7e12b32b6da3ddfc81c24197fb95ea77f7513:43
CIA-87shogun: Soeren Sonnenburg master * r595aee7 / src/shogun/lib/DataType.h :13:44
CIA-87shogun: Introduce destroy_vector/matrix. This function should be called only be13:44
CIA-87shogun: the 'owner' of the vector/matrix object. - https://github.com/shogun-toolbox/shogun/commit/595aee74bcf5dd8bfec404d6a36b87b4269cbc0313:44
-!- sploving1 [~sploving@210.77.14.219] has joined #shogun14:06
sploving1how to print f.get_features in python.14:07
sploving1here is :strings <modshogun.CharStringList; proxy of <Swig Object of type 'shogun::SGStringList< char > *' at 0xb6524b18> > if we use print method14:07
sploving1sonney2k, any other print method in features_string_char_modular.py14:08
@sonney2ksploving1, ohh14:12
@sonney2kI didn't notice that get_features for strings is nto workign in python14:13
@sonney2kyou could do ''.join(f.get_feature_vector(num))14:13
@sonney2kfor now14:13
sploving1num?14:14
sploving1so strange14:14
@sonney2knum = 0, ..., number_of_strings14:15
sploving1what is the result expected?14:15
@sonney2ka string14:16
sploving1i mean f:get_features() should get which string?14:16
sploving1strings = {'hey','guys','i','am','a','string'}?14:16
@sonney2kno14:16
@sonney2ka number14:16
@sonney2ke.g. 014:17
@sonney2kthen the output should be 'hey'14:17
sploving1f.set_feature_vector(array(['t','e','s','t']), 0) does not effect the result?14:17
@sonney2kwhen you have that in then the output should be 'test'14:18
sploving1vector means columns? then I know14:19
@sonney2kvector == one string14:19
sploving1so after that it should be strings = {'test','guys','i','am','a','string'}14:19
@sonney2kahh stupid bug in python typemaps14:20
@sonney2kyes14:20
@sonney2ksploving1, btw14:24
@sonney2kthe delete[] is wrong in the string typemap14:25
@sonney2kalso in python14:25
@sonney2ksploving1, so just remove the delete[] str[i].string and delete[] str;14:25
sploving1sonney2k, okay14:25
@sonney2ksploving1, what does SWIG_arg++ do?14:26
sploving1that is why when I test, it crashes14:26
@sonney2ksploving1, yes14:26
@sonney2k(I just got the python typemap to work again)14:26
sploving1SWIG_arg++, maybe related to the number of  lua stack14:27
sploving1I am not quite sure about it14:27
@sonney2ksploving1, but in typemaps will be translated for every argument right?14:27
sploving1yeap. then?14:28
@sonney2kI mean all 3 arguments of      foo(SGVector, SGVector, SGMatrix) would be translated or?14:28
sploving1yeap. they can use one typemap14:28
sploving1SGVector use vector typemap and SGMatrix use matrix typemap14:29
sploving1is similar to foo(int, int, int),  just write an int typemap14:30
CIA-87shogun: Soeren Sonnenburg master * r6ffbbae / src/interfaces/python_modular/swig_typemaps.i : fix python out typemap for strings - https://github.com/shogun-toolbox/shogun/commit/6ffbbae774955b7de9e581b1718b8479823b9eb214:36
@sonney2ksploving1, ok14:36
sploving1sonney2k, c++: Internal error: killed (program cc1plus)14:39
sploving1Please submit a full bug report.14:39
@sonney2kseems to require too much memory...14:39
sploving1oh14:40
sploving1I will shut down other app(include irc) to re compile it14:40
-!- sploving1 [~sploving@210.77.14.219] has left #shogun []14:40
-!- sploving1 [~sploving@210.77.14.219] has joined #shogun15:23
sploving1hmm, compiled,but still not work well15:23
CIA-87shogun: Baozeng Ding master * re79b597 / src/interfaces/lua_modular/swig_typemaps.i : splite SGVector<char> from other types, since there is no char type in lua, we need use one-len string in lua to map a char in C - https://github.com/shogun-toolbox/shogun/commit/e79b5978ee002a6c210cb9312ad913503b47deee15:24
CIA-87shogun: Soeren Sonnenburg master * r6536a37 / src/interfaces/lua_modular/swig_typemaps.i :15:24
CIA-87shogun: Merge pull request #237 from sploving/master15:24
CIA-87shogun: splite SGVector<char> from other types, since there is no char type in lua, we need use one-len string in lua to map a char in C - https://github.com/shogun-toolbox/shogun/commit/6536a37bdeed738599eb1a3d1206878886f470aa15:24
sploving1hey��15:25
sploving1guys���15:25
-!- blackburn [~blackburn@109.226.76.87] has joined #shogun15:25
sploving1after set, t����15:25
sploving1guys���15:25
sploving1every time, I changed typemap and recompile, there usually take a long time and some times the machine crashes!15:31
blackburnsonney2k: why it was failing?15:32
blackburnLLE+arpack15:32
sploving1is anyway to reduce compile time, other than ./configure --interfaces=lua_modular --enable-debug --disable-doxygen --disable-optimization15:32
blackburnah I see15:33
blackburnsploving1: I can't mind any other way15:33
sploving1Checking for Ruby NArray Developer Files ... no why?15:35
sploving1blackburn, oh15:35
sploving1I have installed NArray.15:35
@sonney2ksploving1, we will try to reduce the wrapper code size15:48
@sonney2ksploving1, we just have too many templates15:48
@sonney2kand of course swig typemaps should be as small as possible for all the types15:48
@sonney2klike 1-2 lines per type if possible and calling other fucntions15:49
@sonney2kthis way the generated *.cxx file becomes much smaller...15:49
@sonney2ksploving1, bad examples are stringfeatures currently15:49
sploving1sonney2k, okay.15:49
@sonney2kstring in / out typemap I mean15:49
sploving1Checking for Ruby NArray Developer Files ... no15:49
@sonney2ksploving1, yes15:50
@sonney2kplease install narray to /usr15:50
@sonney2knot /usr/local15:50
@sonney2kthen it should work15:50
sploving1--prefix = /usr/local?15:50
sploving1--prefix = /usr/15:50
sploving1?15:50
@sonney2kyes15:50
@sonney2knot for shogun but for narray15:51
sploving1I will have a try15:51
@sonney2kblackburn, I #ifdef'd your arpack code15:52
@sonney2kblackburn, so it should now work...15:52
sploving1sonney2k, how do you compiled narray. there is no configure file15:52
sploving1so there cannot use prefix15:52
@sonney2kblackburn, btw - trying to fix all memory leaks/borderline issues I realized that it is actually pretty annoying that we use delete[] and new[] ...15:52
sploving1just a Makefile15:52
sploving1 ruby extconf.rb  make   make site-install15:53
sploving1do you mean ruby extconf.rb --prefix=/usr15:54
@sonney2ksploving1, where are your ruby extensions normally?15:54
@sonney2kfor example mine are in /usr/lib/ruby/1.8/x86_64-linux15:55
sploving1 /usr/local/lib/site_ruby/1.8/i486-linux15:55
sploving1sonney2k, this is mine15:55
sploving1you mean I should compile ruby??15:56
@sonney2kblackburn, it might be even better to use just SG_MALLOC / SG_CALLOC / SG_FREE - this way we can always SG_RESIZE etc15:56
@sonney2ksploving1, no15:56
@sonney2ksploving1, but that should then work too15:56
sploving1another simple way is??15:56
@sonney2ksploving1, what does shogun in .config say15:56
@sonney2kgrep -i INCLUDES_RUBY_MOD .config15:57
sploving1INCLUDES_RUBY= -I/usr/lib/ruby/1.8/i486-linux15:57
sploving1INCLUDES_RUBY_MODULAR= -I/usr/lib/ruby/1.8/i486-linux15:57
@sonney2khmmhh that is weird then - why doesn't it detect it to be in /usr/local ...?15:58
@sonney2kdoes the directory /usr/lib/ruby/1.8/i486-linux exist?15:58
blackburnsonney2k: what do you mean?15:58
blackburnwhere do you suggest to use it?15:58
sploving1sonney2k, yeap15:59
@sonney2kblackburn, I think we should replace all new[] / delete[] with SG_MALLOC / SG_FREE15:59
sploving1it exist15:59
@sonney2ksploving1, ok then that is the path where the extensions should go15:59
blackburnsonney2k: what is the +s of this way?15:59
@sonney2kblackburn, no more delete[] / free() mismatches between interfaces, we could then use realloc to resize arrays, calloc to allocate a zeroed array (no need to do all this memset(xxx,xxx,0))16:00
@sonney2kblackburn, and even better we can trace memory allocations since we are using our own macros :D16:01
blackburnhmm okay16:01
blackburnI'll make a tranistion for my preprocessors16:01
@sonney2kblackburn, stop16:01
blackburnaren't you suggesting me to do that? :D16:01
@sonney2kblackburn, that is a big change so we should be pretty sure that this is the right way16:02
@sonney2kwe should e.g. ask f-x too or heiko...16:02
blackburnI could do that for preprocessors16:02
blackburnand if something goes wrong - we will know it earlier16:02
blackburnsome practice ground for SG_*16:03
@sonney2kblackburn, things will definitely go wrong because we will then have delete[] / SG_FREE / new[] / SG_MALLOC mismatches16:03
@sonney2kblackburn, it has to be done at once16:03
@sonney2kreplacing delete[]16:04
blackburnhmm not only delete/new16:04
@sonney2k is trivial sed operation16:04
blackburne.g. if REALLOC was allowed - I would not create new feature matrix16:04
blackburnjust realloc old one16:04
blackburnand so on16:04
@sonney2kreplacing new[] needs some thought. we might want to change SG_MALLOC to contain the type of the thing to be alloc'd as first argument16:04
@sonney2kblackburn, yes correct - that would be possible then16:05
f-xsonney2k: i'm for this kind of transition.. but as you say replacing new[] may not be so direct16:05
@sonney2kf-x, if we replace new int[17] -> SG_MALLOC(int, 17) it can be done with some regular expression too16:06
@sonney2kbut yes there might be tough exceptions...16:06
@sonney2kbetter do this manually16:06
@sonney2kthen we should overload new[] / delete[] to throw an error that this is not allowed ...16:07
f-xgradually - i could do it in my part of the code.. would be easier if you just give us some kind of example to follow first16:07
f-xright16:07
@sonney2ksploving1, hmmhh would it help if we just include /usr/local in include paths too?16:10
sploving1I do not know. I removed the local dir, and I am recompiling narray,using make install instead of make site-install16:12
@sonney2kblackburn, f-x ok then so it is decided. I will probably enter berserk mode and just do this change16:14
blackburnsonney2k: again berserk mode? :D16:14
@bettyboolol16:14
@sonney2kblackburn, berserk mode is the way to go :D16:15
@sonney2kso I will first change SG_MALLOC to take as first argument the data type16:16
@sonney2ksame for SG_REALLOC and SG_CALLOC16:17
sploving1sonney2k, it works! I use : cp  /usr/local/lib/site_ruby/1.8/i486-linux/* /usr/lib/ruby/1.8/i486-linux/16:18
@sonney2ksploving1, OK - great16:18
sploving1simple but effective way16:19
@sonney2ksploving1, I think at some point we will need to support /usr/local too but for now I dont' know how except hardcoding it16:19
sploving1oh. maybe serialhex can help16:19
@sonney2ksploving1, that is true - he might know how to detect the install path(s)16:20
-!- gsomix [~gsomix@109.169.145.83] has joined #shogun16:21
sploving1gtg, bye sonney2k16:21
@sonney2ksploving1, ok cu16:21
-!- sploving1 [~sploving@210.77.14.219] has left #shogun []16:22
@sonney2kblackburn, do you know what gsomix's opinion on SG_MALLOC vs new[] would be?16:22
blackburnI don't know - he is here16:23
-!- gsomix [~gsomix@109.169.145.83] has quit [Ping timeout: 258 seconds]16:28
-!- gsomix [~gsomix@109.169.145.83] has joined #shogun16:29
gsomixsonney2k, i vote for 'new'.16:31
@sonney2kgsomix, new[] you mean?16:31
gsomixyep16:31
@sonney2kgsomix, did you read our discussion in the logs?16:31
gsomixsonney2k, wait a minute16:32
@sonney2k<sonney2k> blackburn, no more delete[] / free() mismatches between interfaces, we could then use realloc to resize arrays, calloc to allocate a zeroed array (no need to do all this memset(xxx,xxx,0))16:33
@sonney2k<sonney2k> blackburn, and even better we can trace memory allocations since we are using our own macros :D16:33
blackburncan't mind any reason why new is better ;)16:37
gsomixsonney2k, i have no arguments other than aesthetics. :) In this case, the macros better.16:38
gsomixafk16:39
@sonney2kgsomix, well we can use int* x=SG_MALLOC(int, 123) vs int* x=new int[123]16:39
blackburnSG_CALLOC I guess?16:41
@sonney2kblackburn, does new[] zero the memory?16:42
@sonney2kI thought not?16:42
blackburnno16:42
blackburndo you mind to use malloc as calloc?16:42
@sonney2kyes16:42
blackburnwhy?16:42
@sonney2kthen better use SG_CALLOC explicitly16:42
@sonney2knon-standard behaviour16:43
blackburnSG_MALLOC(int, 123) is kinda non-standart16:43
gsomixsonney2k, ok. I will fix own code.16:43
blackburngsomix: you shouldn't16:43
blackburnwe have specially trained berserk for that kind of ops16:44
gsomixwow16:44
@sonney2kblack mode ops16:44
blackburnI heard he have six 24' displays and 12 eyes16:45
blackburnand 5 hands of course16:45
@sonney2kblackburn, and is being sponsored by CIA, FBI, KGB... and google!16:45
blackburnyes, especially by non-existent KGB  :D16:46
* gsomix remembered the joke about Vim and a third arm.16:46
@sonney2kgsomix, you name it + vim powered16:48
@sonney2kblackburn, you would think!16:48
blackburnhmm I use wrong kernel now16:50
blackburn:D16:50
-!- blackburn [~blackburn@109.226.76.87] has quit [Quit: Leaving.]16:50
gsomixI was without electricity today. =___=16:57
gsomixAnd yesterday.16:57
-!- blackburn [~blackburn@109.226.76.87] has joined #shogun16:59
-!- sploving1 [~sploving@210.77.14.219] has joined #shogun17:04
sploving1sonne2y, I tested minimal_test.rb17:05
sploving1sonney2k, but it does not work17:05
sploving1require 'modshogun'17:05
sploving1gk=modshogun::GaussianKernel.new(), is this right??17:05
sploving1serialhex, any suggestion??17:11
-!- in3xes [~in3xes@180.149.49.227] has quit [Quit: Leaving]17:21
-!- [1]warpy [~warpy@bzq-79-181-43-167.red.bezeqint.net] has joined #shogun17:22
[1]warpyhey sonney2k17:23
-!- sploving1 [~sploving@210.77.14.219] has left #shogun []17:23
[1]warpyi have the updated typemap and all the java converted examples. can i issue a pull request even though they are a little broken. i just want to get them on the master branch so that sploving can use them when he starts with the c# thing17:24
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun18:07
CIA-87shogun: Sergey Lisitsyn master * re0df385 / (2 files): Merge branch 'master' of git://github.com/gsomix/shogun - https://github.com/shogun-toolbox/shogun/commit/e0df3851f64846855e56dde8e2092f22e0ef697718:12
CIA-87shogun: Evgeniy Andreev master * r72b9f6c / (2 files): Added get_key function for Fibonacci heap - https://github.com/shogun-toolbox/shogun/commit/72b9f6c797f73f1f83007de84c5f4b5a0e6cb11818:12
-!- f-x [~user@117.192.215.140] has quit [Ping timeout: 260 seconds]18:26
-!- gsomix [~gsomix@109.169.145.83] has quit [Read error: Connection reset by peer]18:43
-!- gsomix [~gsomix@109.169.154.188] has joined #shogun18:50
CIA-87shogun: Evgeniy Andreev master * rf8fba71 / src/shogun/lib/FibonacciHeap.cpp : Fix fatal bug - https://github.com/shogun-toolbox/shogun/commit/f8fba71e0d82189adb15a49df3003013a6d5439618:56
CIA-87shogun: Sergey Lisitsyn master * r0370104 / src/shogun/lib/FibonacciHeap.cpp : Merge branch 'master' of git://github.com/gsomix/shogun - https://github.com/shogun-toolbox/shogun/commit/0370104d67d2da5f5fd734d984459700354ac19018:56
CIA-87shogun: Sergey Lisitsyn master * r08e35cf / (2 files): Moved isomap_distance to .cpp - https://github.com/shogun-toolbox/shogun/commit/08e35cf3e7d080c3469a412458013fae92d8627419:36
[1]warpysonney2k alive ?19:47
CIA-87shogun: Soeren Sonnenburg master * rcc045e9 / (30 files in 11 dirs):19:53
CIA-87shogun: Introduce new SG_MALLOC/REALLOC/CALLOC macros calling sg_malloc etc but19:53
CIA-87shogun: have the type information as first argument. Change all of shogun to use19:53
CIA-87shogun: these new macros. - https://github.com/shogun-toolbox/shogun/commit/cc045e9cfb9e08c439f5fad300c81a6623661c9219:53
@sonney2kblackburn, please have a look at the new SG_MALLOC macros19:55
@sonney2kthey are in shogun/lib/memory.h19:55
@sonney2kf-x_, you too please ^19:55
@sonney2kif these are ok with you f-x_ blackburn I continue the berserk and do the delete[] new[] stuff19:56
blackburnlooks much better to me19:57
@sonney2kblackburn, it even saved quite a bit of code already19:57
@sonney2kutilizing the SG_CALLOC that is19:58
@sonney2k$ grep 'new [^[]*\[' shogun/ -r | wc -l19:58
@sonney2k138219:58
@sonney2khmmhhhh19:58
blackburnshhhhhh19:58
@sonney2kand I thought there were many ALLOC's20:00
@sonney2k$ grep 'SG_.*ALLOC' shogun/ -r | wc -l20:00
@sonney2k22020:00
@sonney2kbut no...20:00
@sonney2k$ grep 'delete\[\]' shogun/ -r | wc -l20:01
@sonney2k154020:01
@sonney2khuh ?! we delete[] more often than we new[] ...20:01
blackburn:D20:01
blackburn???20:02
@sonney2kI guess cleanups in error handling...20:02
@sonney2kanyways20:02
blackburnфр20:02
* sonney2k goes berserk again20:02
blackburnah20:02
blackburndestructors and so on20:02
-!- gsomix [~gsomix@109.169.154.188] has quit [Quit: Ухожу я от вас (xchat 2.4.5 или старше)]20:06
* sonney2k issues find shogun -type f -exec sed -ie 's/delete\[\][ \t]*\([^ ;]*\)[ \t]*;/SG_FREE(\1);/g' {} \;20:07
@sonney2kdelete[]'s are gone...20:20
@sonney2kblackburn, gotcha - what is this in LLE?20:29
@sonney2kSGMatrix<float64_t>(new float64_t[N*N],N,N,true); ?20:29
@sonney2kjust write SGMatrix<flaot64_t>(N,N,true)20:29
@sonney2kand it will be alloc'd20:29
blackburnseriously?20:29
blackburndidn't know20:29
blackburn:)20:29
@sonney2klook at lib/DataType.h20:29
blackburnfeel free to fix ;)20:30
@sonney2kblackburn, I have to because my sed script wont' work with new xxx[], bla, foo);20:30
blackburnokay20:30
[1]warpysonney2k, did you see my pull request20:38
@sonney2kblackburn, so instead of writing20:38
@sonney2kSGMatrix<float64_t> M_matrix = SGMatrix<float64_t>(new float64_t[N*N],N,N,true);20:38
@sonney2kyou can write20:38
@sonney2kSGMatrix<float64_t> M_matrix(N,N,true);20:38
blackburnhmm20:38
blackburnokay20:39
@sonney2kblackburn, looks much better doesn't it?20:39
blackburnexactly it is20:39
-!- gsomix [~gsomix@109.169.154.188] has joined #shogun20:39
blackburnjust habits of using this way with other objects20:40
@sonney2k[1]warpy, yes but no time right now  - this transition has prio20:41
@sonney2kblackburn, heiko is doing the same thing :(20:42
[1]warpyok20:57
-!- [1]warpy [~warpy@bzq-79-181-43-167.red.bezeqint.net] has quit [Quit: HydraIRC -> http://www.hydrairc.com <- s0 d4Mn l33t |t'z 5c4rY!]20:57
blackburnhooray, managed ltsa to work21:00
blackburnhttp://dl.dropbox.com/u/10139213/shogun/ltsa-k8.png21:02
blackburnsonney2k: how it is going?21:50
@sonney2kblackburn, I had an important phone call so I couldn't do anythin in the last hour :(21:56
* sonney2k back to berserk21:56
blackburnI see21:56
@sonney2kblackburn, I have high hopes that I can solve everything with sed :)21:59
@sonney2kit is just not looking too beautiful I must admit...21:59
blackburnI hope so21:59
* sonney2k runs find shogun -type f -exec sed -i -e 's/new[ \t]*\([^[(]*\)\[\(.*\)\];/SG_MALLOCX(\1, \2);/g' {} \;22:17
* sonney2k reverts the change22:21
blackburnfail?22:22
@sonney2kblackburn, berserk not fail :D22:26
@sonney2kit will be so super cool to ./configure --trace-mallocs22:26
@sonney2kand then to list all malloc'd size's22:26
blackburnany way to automagically detect memory leak?22:28
-!- in3xes [~in3xes@180.149.49.227] has quit [Quit: Leaving]22:30
@sonney2kblackburn, at least we could store a table with sizes of the allocated memory22:34
@sonney2kblackburn, otherwise valgrind...22:34
@sonney2kof course if we modify SG_MALLOC to take as last parameter __LINE__ then we can really trace it...22:35
@sonney2kactually that is easy - we just do this when TRACE_MALLOCS is on22:36
@sonney2kso we see the lines where SG_MALLOC got called including the memory size :D22:37
@bettyboointeresting sonney2k22:37
@sonney2kbettyboo, you are damn right22:37
@bettyboosonney2k: but on the other hand these are damn cool features.22:37
@sonney2kblackburn, it did compile with my changes in...22:38
@sonney2kso the above grep doesn't return anything now :)22:39
@bettyboointeresting22:39
blackburnsonney2k: I have an idea22:39
@sonney2k?22:39
blackburnif we store all the allocated things in some 'table' or so22:40
blackburnwe could force free it with warning22:40
blackburnon exit22:40
@sonney2kblackburn, we do already22:40
blackburnso no crashes22:40
@sonney2kwell ohh no22:40
@sonney2knot a good idea...22:40
blackburnwhy?22:40
@sonney2kbecause we don't know when we exit22:40
@sonney2khmmhh 2 examples segfault now22:41
@sonney2kall the others seem to work just fine :)22:41
blackburntoo bad we can't do auto-cleaning :(22:43
@sonney2kahh examples crashed because I forgot to convert them :D22:49
blackburndid you restrict new somehow?22:50
@sonney2kblackburn, the problem really is that if you want to do efficient memory handling you have to do it manually22:50
@sonney2kblackburn, restrict?22:50
blackburnprohibit/disallow/..22:50
blackburnyou said we should do it or so22:51
@sonney2kI know the word but I don't understand what you are asking22:51
blackburnme too22:53
blackburn:D22:53
blackburnnevermind22:53
@sonney2kblackburn, do you have an idea what22:53
@sonney2k            *(CSGObject***) m_parameter22:53
@sonney2k                = new CSGObject*[new_length]();22:53
@sonney2kthis means?22:53
blackburnoh shit22:53
@sonney2kwhat do the extra () do ?22:53
@sonney2kif I put a SG_MALLOC(CSGObject, new_length) there22:53
@sonney2kit will crash22:53
@sonney2khmmhh CALLOC works22:54
@sonney2kso that was just a CALLOC ...22:55
blackburnyes looks like22:56
blackburnwill new double[](10.0); work?22:56
blackburntriple pointer rocks22:57
CIA-87shogun: Soeren Sonnenburg master * r8431a6d / (194 files in 32 dirs): examples all run trough nicely, so rename SG_MALLOCX -> SG_MALLOC (+6 more commits...) - https://github.com/shogun-toolbox/shogun/commit/8431a6db403e11a7b0b678bb348ba5a35b70e0b622:58
blackburnsonney2k: soon you'll have ~1M lines changed :D22:59
@sonney2kblackburn, nope won't work  error: ISO C++ forbids initialization in array new23:00
blackburnhmm23:00
blackburnwith classes it works I guess23:01
@sonney2kbut that works23:01
@sonney2k    double* l=new double[5]();23:01
@sonney2klooks like it is inited with 023:01
@sonney2kjust like CALLOC23:01
blackburnyes23:01
@sonney2kblackburn, crazy sh...23:01
blackburnyes we use kinda strange language :)23:02
blackburnthat's why I suggest you to write shogun in perl23:02
@sonney2kblackburn, I had one more such problem with the SG_MALLOC thingy23:07
@sonney2kthere was a double templated type23:07
@sonney2k sth like new foo<int, int>[17]23:08
@sonney2kSG_MALLOC(foo<int,int>, 17) didn't work because the macro parser thought that due to the additional ',' the macro gets 3 arguments23:08
@sonney2k^ blackburn hefty corner cases right?23:09
blackburnuh23:09
@sonney2kso I worked around this by using typedef's...23:09
blackburnhmmm there are a robust version of LTSA23:11
blackburnif it is good and I implement it - shogun will have it 'exclusively'23:11
@sonney2kblackburn, I am modifying lib/memory now such that we can use TRACE_MALLOCS :)23:11
blackburnnice23:11
blackburnsonney2k: could you write a command to replace new/delete?23:16
@sonney2kyou mean overload the operator?23:16
blackburnnot, that you used23:16
@sonney2k?!23:16
blackburnyou did sed ... blabla to replace it, right?23:17
@sonney2kyes23:17
blackburncan you paste it here?23:17
@sonney2kblackburn, I closed the shell already .. don't have it in history ...23:18
blackburnah oka23:18
@sonney2kblackburn, ahh here23:18
@sonney2kfind shogun -type f -exec sed -i -e  's/new \([a-zA-Z][^[(]*\)\[\(.*\)\];/SG_MALLOCX(\1, \2);/g' {} \;23:18
@sonney2kbut it will replace to MALLOCX23:18
@sonney2k(I did this for debugging to be able to see that this is not a standard MALLOC)23:19
blackburnuh23:19
blackburnhow to use it for one file?23:19
@sonney2k sed -i -e  's/new \([a-zA-Z][^[(]*\)\[\(.*\)\];/SG_MALLOCX(\1, \2);/g' file.cpp23:24
blackburnthanks23:24
@sonney2keasy to read isn't it :D23:29
@sonney2kI wonder when bettyboo will paste that line :D23:29
blackburnI can't understand a little23:29
@bettyboosonney2k: I had been sick for a while just like Chris, such a wonder - cold is going around Samara and Tuebingen both :D23:29
blackburnshh23:30
blackburncan*23:30
blackburnnot can't23:30
@sonney2kbettyboo, now I am sick though and the others are ok23:30
@bettyboosonney2k: ok, you're not getting the spiffy cleaned-up version today, but you are getting the version that works today... it's sleep time yo! nite all, and morning everyone else!!!23:30
blackburnI can say whose phrase does she say23:30
blackburnevery time23:31
blackburnLTSA for shogun is almost done23:32
@sonney2k:)23:36
@sonney2kand :)23:36
-!- gsomix [~gsomix@109.169.154.188] has quit [Ping timeout: 263 seconds]23:45
blackburnhoooooooooooooooray23:58
--- Log closed Wed Jul 27 00:00:31 2011
« Monday, 2011-07-25 Index Wednesday, 2011-07-27 »

Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!