| --- Log opened Sat May 02 00:00:07 2015 | ||
| -!- abinashm [uid66179@gateway/web/irccloud.com/x-fzdkorchplaibrbf] has quit [Quit: Connection closed for inactivity] | 04:03 | |
| -!- pickle27 [~pickle27@192-0-134-235.cpe.teksavvy.com] has joined #shogun | 05:44 | |
| -!- besser82 [~besser82@fedora/besser82] has joined #shogun | 06:41 | |
| -!- mode/#shogun [+o besser82] by ChanServ | 06:42 | |
| -!- dataVinci [~Akash@115.248.64.169] has joined #shogun | 06:58 | |
| -!- dataVinci [~Akash@115.248.64.169] has quit [Read error: Connection reset by peer] | 06:58 | |
| -!- dataVinci [~Akash@14.139.121.55] has joined #shogun | 06:58 | |
| -!- Akash_ [~Akash@123.63.47.173] has joined #shogun | 06:58 | |
| -!- Akash_ [~Akash@123.63.47.173] has quit [Client Quit] | 07:01 | |
| -!- dataVinci [~Akash@14.139.121.55] has quit [Client Quit] | 07:01 | |
| -!- dataVinci [~Akash@117.239.210.100] has joined #shogun | 07:02 | |
| -!- dataVinci [~Akash@117.239.210.100] has quit [Read error: Connection reset by peer] | 07:02 | |
| -!- dataVinci [~Akash@123.63.47.173] has joined #shogun | 07:02 | |
| @wiking | besser82: ping | 07:09 |
|---|---|---|
| @wiking | sorry yesterday i fell asleep | 07:09 |
| @wiking | i have an almost done patch | 07:09 |
| @wiking | question | 07:09 |
| @wiking | where do we define the swig releated macros? | 07:10 |
| @wiking | see for example SWIG_TYPE_TABLE=shogun SHOGUN | 07:10 |
| @wiking | or USE_BOOL | 07:10 |
| @wiking | etc | 07:10 |
| @wiking | if we can solve this then i can push my patch | 07:11 |
| @besser82 | USE_BOOL can be taken from config.h | 07:14 |
| @wiking | it's not there yet | 07:15 |
| @besser82 | The other's can done in modshogun.i afaik | 07:15 |
| @wiking | :) | 07:15 |
| @wiking | but USE_BOOL is only used in swig related stuff | 07:15 |
| @wiking | this is only used in .i | 07:15 |
| @wiking | LIST(APPEND DEFINES USE_CHAR) | 07:15 |
| @wiking | LIST(APPEND DEFINES USE_BOOL) | 07:15 |
| @wiking | LIST(APPEND DEFINES USE_UINT8) | 07:15 |
| @wiking | LIST(APPEND DEFINES USE_UINT16) | 07:15 |
| @wiking | #LIST(APPEND DEFINES USE_UINT32) | 07:15 |
| @wiking | LIST(APPEND DEFINES USE_UINT64) | 07:16 |
| @wiking | #LIST(APPEND DEFINES USE_INT8) | 07:16 |
| @wiking | #LIST(APPEND DEFINES USE_INT16) | 07:16 |
| @wiking | LIST(APPEND DEFINES USE_INT32) | 07:16 |
| @wiking | LIST(APPEND DEFINES USE_INT64) | 07:16 |
| @wiking | LIST(APPEND DEFINES USE_FLOAT32) | 07:16 |
| @wiking | LIST(APPEND DEFINES USE_FLOAT64) | 07:16 |
| @wiking | LIST(APPEND DEFINES USE_COMPLEX128) | 07:16 |
| @wiking | #LIST(APPEND DEFINES USE_FLOATMAX) | 07:16 |
| @besser82 | wiking, that shouldn't matter ;) | 07:16 |
| @besser82 | wiking, let's stuff all in config.h | 07:17 |
| @besser82 | wiking, and "%include <shogun/lib/config.h>" that in modshogun.i :D | 07:17 |
| @wiking | okeeey | 07:18 |
| @wiking | will do it then now | 07:18 |
| @wiking | and push | 07:18 |
| @wiking | gimme 5 mins | 07:18 |
| @besser82 | okie... will check locally then... =) | 07:19 |
| @wiking | besser82: this is a problematic thing to put in config.h: | 07:31 |
| @wiking | #cmakedefine HAVE_PYTHON 1 | 07:31 |
| @wiking | #cmakedefine HAVE_OCTAVE 1 | 07:31 |
| @wiking | as they are only relevant if we compile modular stuff | 07:31 |
| @wiking | and if we want to decouple libshogun from modular interfaces | 07:32 |
| @wiking | it's going to be a problem | 07:32 |
| @wiking | as actually one needs to override config.h macros... | 07:32 |
| @wiking | so i would actually create a swig_config.h | 07:32 |
| @wiking | that we generate for modular interfaces | 07:33 |
| @besser82 | wiking, if it's just this macro(s), we can pass them on cli | 07:34 |
| @wiking | mmm | 07:34 |
| @besser82 | as when building / generating the swiged cxx | 07:34 |
| @wiking | well all those USE_<types> are the same | 07:34 |
| @wiking | as well as | 07:34 |
| @wiking | HAVE_JBLAS | 07:34 |
| @besser82 | those can go to config.h | 07:35 |
| @wiking | noup | 07:35 |
| @wiking | again | 07:35 |
| @besser82 | so we have one common config.h | 07:35 |
| @wiking | jblas is only relevant | 07:35 |
| @besser82 | why? | 07:35 |
| @wiking | if u create the java_modular interface | 07:35 |
| @besser82 | and otherwise won't do impact, right? | 07:37 |
| @besser82 | since we pass it now to all file being build... | 07:37 |
| @besser82 | so no conflicts, when def'ing this in config.h :P | 07:38 |
| @besser82 | wiking, 1 | 07:38 |
| @besser82 | wiking, ^ | 07:38 |
| -!- nszceta [~nszceta@5ED062E5.cm-7-1b.dynamic.ziggo.nl] has joined #shogun | 07:38 | |
| @besser82 | same is for HAVE_{PYTHON,OCTAVE,...} | 07:39 |
| @besser82 | they don't collide with each other | 07:39 |
| @wiking | yeah but it's useless to define in config.h | 07:39 |
| @besser82 | just, if one wouldn't have the deps installed when building libshogun... :( | 07:39 |
| @besser82 | not really useless... | 07:40 |
| @wiking | well look | 07:40 |
| @wiking | if we build libshogun | 07:40 |
| @wiking | we'll get a config.h | 07:40 |
| @besser82 | yes | 07:40 |
| @wiking | that won't have HAVE_PYTHON + HAVE_JBLAS etc defined | 07:40 |
| @wiking | then you want to separately build the python_modular and java_modular interface | 07:41 |
| @wiking | then you check the dependent libshogun's config.h | 07:41 |
| @wiking | and you will have a problem... | 07:41 |
| @wiking | or you start overriding those variables | 07:41 |
| @wiking | but then i dont see why would you define them there in the first plac | 07:41 |
| -!- pickle27 [~pickle27@192-0-134-235.cpe.teksavvy.com] has quit [Remote host closed the connection] | 07:43 | |
| @besser82 | wiking, I see... It's about build-time config... | 07:44 |
| @besser82 | then you are fully right... | 07:44 |
| @besser82 | and we should generate that into swig_config.h sth... | 07:44 |
| @besser82 | but the USE_* defines can go to config.h ? | 07:45 |
| @wiking | those are the same | 07:46 |
| @wiking | only used for *_modular (swig) stuf | 07:46 |
| @besser82 | then we can put them there as well... | 07:46 |
| @besser82 | swig_config | 07:46 |
| @wiking | ok | 07:51 |
| @wiking | so shall we have a src/interfaces/modular/swig_config.h.in | 07:59 |
| @wiking | ? | 07:59 |
| @besser82 | yes, that would be nice | 08:07 |
| -!- besser82 [~besser82@fedora/besser82] has quit [Remote host closed the connection] | 08:17 | |
| -!- besser82 [~besser82@fedora/besser82] has joined #shogun | 08:19 | |
| -!- mode/#shogun [+o besser82] by ChanServ | 08:19 | |
| -!- nszceta [~nszceta@5ED062E5.cm-7-1b.dynamic.ziggo.nl] has quit [Quit: My Mac has gone to sleep. ZZZzzz...] | 08:41 | |
| -!- dataVinci [~Akash@123.63.47.173] has quit [Quit: Leaving] | 08:51 | |
| @wiking | besser82: okeeey testing now python modular | 09:42 |
| @wiking | if it works i'll commit+psuh | 09:42 |
| @besser82 | wiking, allrighty, will then build all interfaces in a row ;) | 10:01 |
| @wiking | hehe | 10:01 |
| @besser82 | wiking, and run testsuite... rpmbuild is tha thing ;) | 10:01 |
| @wiking | ok | 10:03 |
| @wiking | only one error :)) | 10:03 |
| @besser82 | which error? | 10:03 |
| @wiking | /Users/wiking/shogun/build/src/interfaces/python_modular/modshogunPYTHON_wrap.cxx:384556:48: error: allocating an object of abstract class type 'shogun::CLinearARDKernel' | 10:04 |
| @wiking | result = (shogun::CLinearARDKernel *)new shogun::CLinearARDKernel(); | 10:04 |
| @wiking | ^ | 10:04 |
| @wiking | /Users/wiking/shogun/src/shogun/kernel/LinearARDKernel.h:160:30: note: unimplemented pure virtual method 'get_parameter_gradient' in 'CLinearARDKernel' | 10:04 |
| @wiking | virtual SGMatrix<float64_t> get_parameter_gradient(const TParameter* param, | 10:04 |
| @wiking | ^ | 10:04 |
| @wiking | /Users/wiking/shogun/src/shogun/kernel/LinearARDKernel.h:197:20: note: unimplemented pure virtual method 'compute' in 'CLinearARDKernel' | 10:04 |
| @wiking | virtual float64_t compute(int32_t idx_a, int32_t idx_b)=0; | 10:04 |
| @wiking | dunno why | 10:04 |
| @besser82 | push and I'll figure out =) | 10:07 |
| @besser82 | as soon as we're finished, we can rebase the branch && PR... ;) | 10:09 |
| @besser82 | wiking, ^ | 10:10 |
| @wiking | mmm ok pushing | 10:11 |
| @wiking | couldn't compile the python modular though | 10:11 |
| @wiking | in | 10:12 |
| -!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 10:12 | |
| shogun-notifier- | shogun: Viktor Gal :feature/config.h * 74ffb67 / / (29 files): https://github.com/shogun-toolbox/shogun/commit/74ffb670099a07db4e5d4827dc6c17e744d7ec3e | 10:12 |
| shogun-notifier- | shogun: Remove compiler macros defined via compiler flags | 10:12 |
| shogun-notifier- | shogun: Use <shogun/lib/config.h> and swig_config.h for configure paramters | 10:12 |
| @wiking | 74ffb67 | 10:12 |
| -!- travis-ci [~travis-ci@ec2-54-92-244-69.compute-1.amazonaws.com] has joined #shogun | 10:17 | |
| 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/60931544 | 10:17 |
| -!- travis-ci [~travis-ci@ec2-54-92-244-69.compute-1.amazonaws.com] has left #shogun [] | 10:17 | |
| shogun-notifier- | shogun: Viktor Gal :feature/config.h * d8cea97 / src/shogun/lib/RefCount.h: https://github.com/shogun-toolbox/shogun/commit/d8cea97cae978125e1234f3614ac3e8c5ba1aca6 | 10:20 |
| shogun-notifier- | shogun: Fix the order of including config.h in RefCount.h | 10:20 |
| @besser82 | Building CXX object src/shogun/CMakeFiles/libshogun.dir/lib/RefCount.cpp.o | 10:22 |
| @besser82 | cd /home/besser82/shared/git/shogun/build/src/shogun && /usr/bin/ccache /usr/lib64/ccache/c++ -std=c++11 -Wall -Wno-unused-parameter -Wformat -Wformat-security -Wparentheses -Wshadow -Wno-unknown-pragmas -Wno-deprecated -fopenmp -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -m64 -mtune=generic -O3 -fexpensive-optimizations -frerun-cse-after | 10:22 |
| @besser82 | -loop -fcse-follow-jumps -finline-functions -fschedule-insns2 -fthread-jumps -fforce-addr -fstrength-reduce -funroll-loops -mfpmath=sse -fPIC -I/home/besser82/shared/git/shogun/src -I/home/besser82/shared/git/shogun/src/shogun -I/home/besser82/shared/git/shogun/build/src -I/home/besser82/shared/git/shogun/build/src/shogun -isystem /usr/include/eigen3 -I/usr/lib64/../include -I/usr/include/json-c -I/usr/include/libxml2 -o CMakeFiles/libshogun.dir/li | 10:22 |
| @besser82 | b/RefCount.cpp.o -c /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.cpp | 10:22 |
| @besser82 | In file included from /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.cpp:1:0: | 10:22 |
| @besser82 | /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.h:47:19: error: 'atomic' in namespace 'std' does not name a template type | 10:22 |
| @besser82 | volatile std::atomic<int> rc; | 10:23 |
| @besser82 | ^ | 10:23 |
| @besser82 | /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.h: In constructor 'shogun::RefCount::RefCount(int32_t)': | 10:23 |
| @besser82 | /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.h:25:34: error: class 'shogun::RefCount' does not have any field named 'rc' | 10:23 |
| @wiking | yeah that was fucked up | 10:23 |
| @besser82 | RefCount(int32_t ref_start=0) : rc(ref_start) {} | 10:23 |
| @besser82 | ^ | 10:23 |
| @besser82 | /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.cpp: In member function 'int32_t shogun::RefCount::ref()': | 10:23 |
| @besser82 | /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.cpp:8:18: error: 'rc' was not declared in this scope | 10:23 |
| @wiking | fixed it in the last commit | 10:23 |
| @besser82 | int32_t count = rc.fetch_add(1)+1; | 10:23 |
| @besser82 | ^ | 10:23 |
| @besser82 | /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.cpp: In member function 'int32_t shogun::RefCount::unref()': | 10:23 |
| @besser82 | /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.cpp:21:18: error: 'rc' was not declared in this scope | 10:23 |
| @besser82 | int32_t count = rc.fetch_sub(1)-1; | 10:23 |
| @besser82 | ^ | 10:23 |
| @besser82 | /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.cpp: In member function 'int32_t shogun::RefCount::ref_count()': | 10:23 |
| @wiking | see d8cea97 | 10:23 |
| @besser82 | /home/besser82/shared/git/shogun/src/shogun/lib/RefCount.cpp:34:18: error: 'rc' was not declared in this scope | 10:23 |
| @besser82 | int32_t count = rc.load(); | 10:23 |
| @wiking | too long copy-past :) | 10:23 |
| @besser82 | ^ | 10:23 |
| @besser82 | src/shogun/CMakeFiles/libshogun.dir/build.make:8119: recipe for target 'src/shogun/CMakeFiles/libshogun.dir/lib/RefCount.cpp.o' failed | 10:23 |
| @besser82 | which? | 10:23 |
| @wiking | *paste | 10:23 |
| @wiking | this refcount trouble | 10:23 |
| @besser82 | I see... | 10:24 |
| @wiking | is fixed in commmit d8cea97 | 10:24 |
| -!- travis-ci [~travis-ci@ec2-54-167-156-211.compute-1.amazonaws.com] has joined #shogun | 10:26 | |
| 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/60931544 | 10:26 |
| -!- travis-ci [~travis-ci@ec2-54-167-156-211.compute-1.amazonaws.com] has left #shogun [] | 10:26 | |
| @besser82 | rebuilding... | 10:30 |
| @besser82 | takes time, because ccaches are invalid again :P | 10:31 |
| -!- travis-ci [~travis-ci@ec2-54-92-244-69.compute-1.amazonaws.com] has joined #shogun | 10:42 | |
| 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/60932021 | 10:42 |
| -!- travis-ci [~travis-ci@ec2-54-92-244-69.compute-1.amazonaws.com] has left #shogun [] | 10:42 | |
| @besser82 | wiking, confirmed... can reproduce... | 10:46 |
| @wiking | besser82: yeah travis can do it as well :) | 10:48 |
| @besser82 | lemme dig into... ;) | 10:49 |
| @besser82 | wiking, ^ | 10:49 |
| -!- travis-ci [~travis-ci@ec2-54-167-156-211.compute-1.amazonaws.com] has joined #shogun | 10:51 | |
| 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/60932021 | 10:51 |
| -!- travis-ci [~travis-ci@ec2-54-167-156-211.compute-1.amazonaws.com] has left #shogun [] | 10:51 | |
| -!- travis-ci [~travis-ci@ec2-54-92-244-69.compute-1.amazonaws.com] has joined #shogun | 10:53 | |
| 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/60932021 | 10:53 |
| -!- travis-ci [~travis-ci@ec2-54-92-244-69.compute-1.amazonaws.com] has left #shogun [] | 10:53 | |
| @wiking | besser82: ok this whole thing got introduced in b6e297c8c4a64c558d30d496df5be084c06686a3 | 11:07 |
| @besser82 | wiking, thx for the info :D | 11:21 |
| @wiking | just sent a mail to the mailing list | 11:21 |
| @wiking | because that patch actually doesn't make any fucking sense :) | 11:21 |
| @besser82 | wiking, let's bully yorkelin to refix :P | 11:21 |
| -!- nszceta [~nszceta@5ED062E5.cm-7-1b.dynamic.ziggo.nl] has joined #shogun | 12:07 | |
| -!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 265 seconds] | 12:09 | |
| -!- besser82 [~besser82@fedora/besser82] has joined #shogun | 12:16 | |
| -!- mode/#shogun [+o besser82] by ChanServ | 12:16 | |
| -!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 265 seconds] | 12:21 | |
| -!- besser82 [~besser82@2003:a:372:5500:f2de:f1ff:fe89:42d4] has joined #shogun | 12:29 | |
| -!- besser82 [~besser82@2003:a:372:5500:f2de:f1ff:fe89:42d4] has quit [Changing host] | 12:29 | |
| -!- besser82 [~besser82@fedora/besser82] has joined #shogun | 12:29 | |
| -!- mode/#shogun [+o besser82] by ChanServ | 12:29 | |
| -!- lambday [6a33f00b@gateway/web/freenode/ip.106.51.240.11] has joined #shogun | 12:56 | |
| -!- mode/#shogun [+o lambday] by ChanServ | 12:56 | |
| shogun-notifier- | shogun: lambday :develop * 7bbf957 / / (10 files): https://github.com/shogun-toolbox/shogun/commit/7bbf957df313bbf5f25188422e60c24747c017ee | 12:58 |
| shogun-notifier- | shogun: Added elementwise operators framework in linalg | 12:58 |
| shogun-notifier- | shogun: lambday :develop * 35c609b / / (10 files): https://github.com/shogun-toolbox/shogun/commit/35c609bce2c1d1a7357a5eac96404d69892b391d | 12:58 |
| shogun-notifier- | shogun: Added a benchmark for elementwise operations | 12:58 |
| shogun-notifier- | shogun: Soumyajit De :develop * 7f15f5e / / (12 files): https://github.com/shogun-toolbox/shogun/commit/7f15f5e1ca1fef6507e2a1ce800b9a21366d0240 | 12:58 |
| shogun-notifier- | shogun: Merge pull request #2830 from lambday/develop | 12:58 |
| shogun-notifier- | shogun: | 12:58 |
| shogun-notifier- | shogun: Added element-wise unary operators in linalg | 12:58 |
| -!- travis-ci [~travis-ci@ec2-54-167-156-211.compute-1.amazonaws.com] has joined #shogun | 13:08 | |
| travis-ci | it's Soumyajit De'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/60939812 | 13:08 |
| -!- travis-ci [~travis-ci@ec2-54-167-156-211.compute-1.amazonaws.com] has left #shogun [] | 13:08 | |
| @lambday | wiking: lisitsyn: hey | 13:29 |
| lisitsyn | lambday: hey | 13:29 |
| @lambday | do you guys know why index_t is signed int? | 13:29 |
| shogun-buildbot | build #2685 of bsd1 - libshogun is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/bsd1%20-%20libshogun/builds/2685 blamelist: lambday <heavensdevil6909@gmail.com>, Soumyajit De <heavensdevil6909@gmail.com> | 13:30 |
| lisitsyn | soeren had some negative opinion on signed | 13:30 |
| lisitsyn | unisgned | 13:30 |
| lisitsyn | sorry | 13:30 |
| lisitsyn | that's a holywar actually hah | 13:30 |
| lisitsyn | pretty common | 13:30 |
| @lambday | umm.. is there any case when we expect index_t to be -ve? :/ | 13:31 |
| lisitsyn | not really | 13:31 |
| lisitsyn | but unsigned can overflow pretty funny | 13:31 |
| lisitsyn | I mean when it overflows you get huge positive | 13:31 |
| lisitsyn | and you don't really detect that | 13:32 |
| lisitsyn | that was one point I think | 13:32 |
| @lambday | lisitsyn: actually I added size() method to all matrix and vector types... and for matrices I am returning uint64_t ... | 13:32 |
| @lambday | nothing too serious but I am getting a bunch of warnings because of that.. | 13:33 |
| lisitsyn | yeah | 13:34 |
| lisitsyn | don't know :) | 13:34 |
| lisitsyn | keep it maybe | 13:34 |
| @lambday | lisitsyn: okay.. | 13:35 |
| @lambday | lisitsyn: btw if you have a few mins could you please have just a quick look at https://github.com/shogun-toolbox/shogun/pull/2830/files | 13:35 |
| lisitsyn | yeah sure | 13:36 |
| @lambday | element-wise operations for linalg | 13:36 |
| lisitsyn | lambday: what about these strings? | 13:37 |
| @lambday | lisitsyn: thanks.. could really use some useful inputs on the design | 13:37 |
| @lambday | lisitsyn: those are for OpenCL kernel | 13:37 |
| lisitsyn | lambday: I have some idea | 13:37 |
| @lambday | there are a bunch of predefined element-wise operations... and users can provide custom as well | 13:38 |
| @lambday | lisitsyn: yeah tell me | 13:38 |
| lisitsyn | can we provide supported operations in some namespace? | 13:38 |
| lisitsyn | so they are really objects | 13:39 |
| @lambday | lisitsyn: shogun::linalg::operations | 13:39 |
| @lambday | yep - they are | 13:39 |
| lisitsyn | no | 13:39 |
| lisitsyn | I mean in this strings | 13:39 |
| lisitsyn | lambday: are these "fdsfd"+dsadsa just for test? | 13:39 |
| @lambday | lisitsyn: yes... | 13:39 |
| @lambday | check the Sin.h file | 13:39 |
| lisitsyn | operation.append(data_type+" outer_factor=-2*"+s_pi+"*sqrt(element)*pow("+s_weights+", 2);\n"); | 13:41 |
| lisitsyn | lambday: ^ will this happen only in test? | 13:41 |
| @lambday | that string is just an illustration to show that if we want, we can specify complex, custom element-wise operations as well | 13:41 |
| lisitsyn | yes | 13:41 |
| @lambday | lisitsyn: for predefined operations, users don't need to do anything | 13:41 |
| lisitsyn | so can we avoid this string thing a bit? | 13:42 |
| lisitsyn | ok let me comment | 13:42 |
| lisitsyn | in github | 13:42 |
| @lambday | lisitsyn: yeah... | 13:42 |
| shogun-buildbot | build #2643 of deb3 - modular_interfaces is complete: Failure [failed csharp modular] Build details are at http://buildbot.shogun-toolbox.org/builders/deb3%20-%20modular_interfaces/builds/2643 blamelist: lambday <heavensdevil6909@gmail.com>, Soumyajit De <heavensdevil6909@gmail.com> | 13:45 |
| shogun-buildbot | build #1020 of FCRH - libshogun is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/FCRH%20-%20libshogun/builds/1020 blamelist: lambday <heavensdevil6909@gmail.com>, Soumyajit De <heavensdevil6909@gmail.com> | 13:50 |
| shogun-buildbot | build #5 of FC22 - libshogun is complete: Failure [failed test] Build details are at http://buildbot.shogun-toolbox.org/builders/FC22%20-%20libshogun/builds/5 blamelist: lambday <heavensdevil6909@gmail.com>, Soumyajit De <heavensdevil6909@gmail.com> | 13:50 |
| -!- dataVinci [~Akash@123.201.110.122] has joined #shogun | 13:51 | |
| lisitsyn | lambday: https://github.com/shogun-toolbox/shogun/pull/2830/files#r29546970 | 13:51 |
| lisitsyn | just some ideas | 13:52 |
| shogun-buildbot | build #2460 of deb2 - static_interfaces is complete: Failure [failed octave static] Build details are at http://buildbot.shogun-toolbox.org/builders/deb2%20-%20static_interfaces/builds/2460 blamelist: lambday <heavensdevil6909@gmail.com>, Soumyajit De <heavensdevil6909@gmail.com> | 13:52 |
| @lambday | lisitsyn: ah this would be much better | 13:52 |
| lisitsyn | you can implement 2nd thing in a hour I think | 13:52 |
| lisitsyn | first thing is tougher I think | 13:53 |
| @lambday | lisitsyn: thinking about that only | 13:54 |
| lisitsyn | lambday: was there some trouble with elementwise<sin,..> | 13:55 |
| lisitsyn | instead of elementwise_sin? | 13:55 |
| @lambday | lisitsyn: well, that's just more of a wrapper.. sin can be done using elementwise<operand_type, sin> as well | 13:56 |
| lisitsyn | ahh | 13:57 |
| lisitsyn | cool | 13:57 |
| lisitsyn | missed that | 13:57 |
| lisitsyn | lambday: I don't have other comments for now then :) | 13:57 |
| @lambday | lisitsyn: the first thing you suggested is so cool!!.. | 13:57 |
| @lambday | lisitsyn: will think a bit | 13:57 |
| @lambday | no wonder it left Heiko puzzled... all those strings are really messy | 13:58 |
| lisitsyn | lambday: about the formatted thing | 13:58 |
| lisitsyn | you know how to implement that? | 13:58 |
| lisitsyn | if not I can outline a bit | 13:58 |
| @lambday | lisitsyn: haven't thought about that yet... I am stuck with the first... | 13:59 |
| @lambday | but you suggest... | 13:59 |
| lisitsyn | I think 2nd is easier | 13:59 |
| lisitsyn | :) | 13:59 |
| @lambday | lisitsyn: I think I can make it work | 14:00 |
| lisitsyn | yeah | 14:00 |
| @lambday | lisitsyn: but the first one looks more elegant.. | 14:00 |
| lisitsyn | I mean you probably spend more time debugging it | 14:00 |
| lisitsyn | while it is not that important | 14:01 |
| @lambday | lisitsyn: with the first one? | 14:01 |
| lisitsyn | yeah I expect it to be a bit buggier | 14:01 |
| lisitsyn | as it is more complex | 14:01 |
| @lambday | all the ops should ve overloaded | 14:01 |
| lisitsyn | yeah | 14:02 |
| @lambday | lisitsyn: let me think a bit.. I'll get back | 14:02 |
| @lambday | thanks for my food for thought for today :D | 14:02 |
| lisitsyn | hah you're welcome | 14:05 |
| shogun-buildbot | build #630 of deb4 - python3 is complete: Failure [failed test python modular] Build details are at http://buildbot.shogun-toolbox.org/builders/deb4%20-%20python3/builds/630 blamelist: lambday <heavensdevil6909@gmail.com>, Soumyajit De <heavensdevil6909@gmail.com> | 14:06 |
| -!- lambday [6a33f00b@gateway/web/freenode/ip.106.51.240.11] has quit [Quit: Page closed] | 14:52 | |
| -!- nszceta [~nszceta@5ED062E5.cm-7-1b.dynamic.ziggo.nl] has quit [Max SendQ exceeded] | 14:53 | |
| -!- nszceta [~nszceta@5ED062E5.cm-7-1b.dynamic.ziggo.nl] has joined #shogun | 15:05 | |
| -!- dataVinci [~Akash@123.201.110.122] has quit [Ping timeout: 272 seconds] | 15:22 | |
| -!- nszceta [~nszceta@5ED062E5.cm-7-1b.dynamic.ziggo.nl] has quit [Max SendQ exceeded] | 15:44 | |
| -!- nszceta [~nszceta@5ED062E5.cm-7-1b.dynamic.ziggo.nl] has joined #shogun | 15:52 | |
| -!- besser82 [~besser82@fedora/besser82] has quit [Read error: Permission denied] | 15:54 | |
| -!- besser82 [~besser82@2003:a:372:5500:f2de:f1ff:fe89:42d4] has joined #shogun | 15:54 | |
| -!- besser82 [~besser82@2003:a:372:5500:f2de:f1ff:fe89:42d4] has quit [Changing host] | 15:54 | |
| -!- besser82 [~besser82@fedora/besser82] has joined #shogun | 15:54 | |
| -!- mode/#shogun [+o besser82] by ChanServ | 15:55 | |
| -!- nszceta [~nszceta@5ED062E5.cm-7-1b.dynamic.ziggo.nl] has quit [Max SendQ exceeded] | 15:56 | |
| -!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 15:58 | |
| -!- besser82 [~besser82@fedora/besser82] has quit [Read error: Permission denied] | 16:25 | |
| -!- besser82 [~besser82@fedora/besser82] has joined #shogun | 16:25 | |
| -!- mode/#shogun [+o besser82] by ChanServ | 16:25 | |
| -!- pickle27 [~pickle27@192-0-134-235.cpe.teksavvy.com] has joined #shogun | 16:27 | |
| -!- pickle27 [~pickle27@192-0-134-235.cpe.teksavvy.com] has quit [Remote host closed the connection] | 16:35 | |
| -!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 265 seconds] | 19:28 | |
| -!- pickle27 [~pickle27@192-0-134-235.cpe.teksavvy.com] has joined #shogun | 21:32 | |
| -!- pickle27 [~pickle27@192-0-134-235.cpe.teksavvy.com] has quit [Remote host closed the connection] | 21:38 | |
| -!- pickle27 [~pickle27@192-0-134-235.cpe.teksavvy.com] has joined #shogun | 23:59 | |
| --- Log closed Sun May 03 00:00:09 2015 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!