Open in new window / Try shogun cloud
--- Log opened Sun Jun 03 00:00:41 2012
KMcQuistenWhere was it getting caught up?00:01
blackburnhmmm something strange here..00:05
blackburnKMcQuisten: funny thing x2 factor comes from the code, but not two feature instances..00:05
KMcQuistenReally?  I wonder if it's pulling them from the initialized kernels00:06
blackburnyeah..00:07
KMcQuistenI tired a CombinedFeature and CombinedKernel with StringChar features and WD kernels of 2 different orders and got the same sort of error as with the StringWordFeatures00:12
KMcQuistenSystemError: [ERROR] assertion char_vec && len>0 failed in file kernel/WeightedDegreeStringKernel.cpp line 52200:12
blackburndifferent error00:13
KMcQuistenYes, but still failing the check to see if the proper number of features are present00:13
KMcQuistenThe same thing happens when I combine a WD kernel on StringChar features with a CommWOrdString kernel on StringWord features.00:15
KMcQuistenLike you, I hae stringChar+Real and StringWord + real working fine00:16
blackburnheh00:17
KMcQuistenSomethign about this doesn't like having multiple kinds of string features in a CombinedKernel00:17
KMcQuistenVery strange00:17
blackburndoes it happen with real+real?00:17
KMcQuistenLet me check quicklty00:18
blackburnhmm it happens with real+string now00:20
KMcQuistenReal and Real worked for me00:21
KMcQuistenReal and StringWord work for me00:21
KMcQuistenLet me check Real and StringChar00:21
KMcQuistenReal and StringChar work for me with a WD kernel00:24
KMcQuistenONly thing that fials for me is combining multibple string features00:24
KMcQuistenman my typing is crap today.00:25
blackburnheh00:25
KMcQuistenNeed more sleep :)00:25
blackburnno problem00:25
blackburnokay but real + string fails now too00:25
blackburnreal+real works00:27
KMcQuistenNot for me.  Real + string still works for me00:28
KMcQuistenreal + real works for me, too00:28
KMcQuistenI set it up real + string and string + real.  Both work00:28
blackburnok more or less it is clear now00:29
blackburnI'll dig into the code again00:29
KMcQuistenwhat's the verdict?00:29
KMcQuistenAllrighty00:29
KMcQuistenOk, i'm going ot head out of here.  When you have news feel free to let me know on the mailing list.00:33
-!- KMcQuisten [d8338942@gateway/web/freenode/ip.216.51.137.66] has quit [Quit: Page closed]00:33
blackburnok00:40
CIA-9shogun: Sergey Lisitsyn master * r4b8c7d0 / (2 files in 2 dirs): Hopefully fixed MKL linadd in SVRLight - http://git.io/uNj-bQ01:18
shogun-buildbotbuild #576 of ruby_modular is complete: Failure [failed test_1]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/ruby_modular/builds/576  blamelist: blackburn91@gmail.com02:36
shogun-buildbotbuild #580 of python_modular is complete: Failure [failed test_1]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/python_modular/builds/580  blamelist: blackburn91@gmail.com03:02
CIA-9shogun: Sergey Lisitsyn master * r17a1306 / examples/undocumented/python_modular/mkl_linadd_regression.py : Commented MKL breaker in example to make test work - http://git.io/cnlkBw03:24
-!- blackburn [~blackburn@31.28.59.65] has quit [Ping timeout: 244 seconds]03:28
shogun-buildbotbuild #577 of ruby_modular is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/ruby_modular/builds/57704:36
shogun-buildbotbuild #581 of python_modular is complete: Success [build successful]  Build details are at http://www.shogun-toolbox.org/buildbot/builders/python_modular/builds/58104:52
-!- pluskid [~pluskid@111.120.67.183] has joined #shogun07:55
-!- pluskid [~pluskid@111.120.67.183] has quit [Ping timeout: 252 seconds]09:30
-!- pluskid [~pluskid@li400-235.members.linode.com] has joined #shogun09:30
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun10:06
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has joined #shogun10:28
-!- blackburn [~blackburn@31.28.59.65] has joined #shogun11:24
pluskidblackburn: do you know how to clone a VowpalWabbitMachine?11:29
blackburna=b.clone()?11:29
pluskidis there such a method?11:29
blackburnI don't think so11:29
pluskidhmm, too bad11:30
pluskidI want to have a copy of a VWMachine somewhere11:30
blackburnadd a copy constructor :)11:30
pluskidseems I have to construct the copy manually11:30
pluskidto do that then I should have to understand the implementation of VW first11:30
pluskidbtw: since shogun has a generic serialization component, I guess it should not be too difficult to build a generic clone component based on that11:31
blackburnvirtual clone method?11:32
pluskidI'm not sure, but I guess it could be implemented in a similar way to serialization11:34
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving]11:34
pluskidfor example, a *very naive* way is to serialize it to some where and then de-serialize, we get a clone ...11:34
blackburnI don't mind actually if somebody do that :)11:36
pluskid:D11:37
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun12:36
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 244 seconds]13:43
-!- blackburn [~blackburn@31.28.59.65] has left #shogun []14:47
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun15:02
-!- pluskid [~pluskid@li400-235.members.linode.com] has quit [Quit: Leaving]15:59
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has joined #shogun18:13
-!- wiking [~wiking@78-23-189-112.access.telenet.be] has quit [Changing host]18:13
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun18:13
-!- puffin444 [62e3926e@gateway/web/freenode/ip.98.227.146.110] has joined #shogun18:47
-!- puffin444 [62e3926e@gateway/web/freenode/ip.98.227.146.110] has quit [Ping timeout: 245 seconds]19:42
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 245 seconds]19:51
-!- wiking_ [~wiking@208.76.55.198] has joined #shogun20:11
-!- wiking_ [~wiking@208.76.55.198] has quit [Changing host]20:11
-!- wiking_ [~wiking@huwico/staff/wiking] has joined #shogun20:11
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 252 seconds]20:12
-!- wiking_ is now known as wiking20:12
-!- puffin444 [62e3926e@gateway/web/freenode/ip.98.227.146.110] has joined #shogun20:12
CIA-9shogun: Soeren Sonnenburg master * r7fb034c / (4 files in 2 dirs): add liblinear's L2R_LR - http://git.io/7cx9Ew21:04
-!- heiko [~heiko@host86-185-9-122.range86-185.btcentralplus.com] has joined #shogun21:07
-!- heiko [~heiko@host86-185-9-122.range86-185.btcentralplus.com] has quit [Ping timeout: 260 seconds]21:23
-!- heiko [~heiko@host86-183-73-53.range86-183.btcentralplus.com] has joined #shogun21:36
-!- heiko [~heiko@host86-183-73-53.range86-183.btcentralplus.com] has quit [Ping timeout: 260 seconds]21:45
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun22:05
CIA-9shogun: Soeren Sonnenburg master * r9b380e8 / (7 files in 5 dirs): Integrate liblinears SVR - http://git.io/n9UVCg22:36
n4nd0hey wiking22:49
n4nd0how did it go in the conference?22:50
n4nd0still there?22:50
wikingarrived today22:50
wikingi mean i came back today22:50
n4nd0good22:50
n4nd0did you enjoy it?22:50
-!- blackburn [~blackburn@31.28.59.65] has joined #shogun22:55
blackburnwiking: heya22:55
wikingyeah it was great22:55
wikingbut today i still need to recover22:56
wikingblackburn: yo22:56
blackburnwhen do you think you can get back on the track?22:57
-!- KMcQuisten [d8338942@gateway/web/freenode/ip.216.51.137.66] has joined #shogun23:00
KMcQuistenHello all23:00
blackburnhey23:00
KMcQuistenGet my email?23:00
blackburnwhich one?23:00
KMcQuistenmost recent.  sent today23:01
KMcQuistenmoments ago, inf act23:01
blackburnah right I just sent an answer23:01
blackburn10 mins ago I guess23:01
KMcQuistenwell then let me look23:01
blackburnok I suggested to turn off linadd at all23:01
KMcQuistenok let me check23:02
-!- heiko [~heiko@109.144.207.254] has joined #shogun23:05
blackburnit can be that linadd is a little broken23:06
KMcQuistenLooks like tha'ts the case.  It works with LINADD disables23:07
KMcQuistend23:07
wikingblackburn: tomorrow23:10
blackburncool23:12
blackburnKMcQuisten: cool too :)23:12
KMcQuistenLooks like we got this squared away for now.23:12
KMcQuistenThanks for all your helop23:12
blackburnno problem! ok then I know something is broken in linadd here23:12
KMcQuistenGlad I could help root these out.23:13
KMcQuistenI think I may have found another one, though.23:14
KMcQuistenif I do mkl.get_kernel().get_subkernel_weights() to extract the weights after training, I get this:23:14
KMcQuistenTraceback (most recent call last):   File "mkl_linadd_regression.py", line 76, in <module>     mkl_binclass_modular (*parameter_list[0])   File "mkl_linadd_regression.py", line 71, in mkl_binclass_modular     print mkl.get_kernel().get_subkernel_weights()23:16
KMcQuistenTypeError: Kernel_get_subkernel_weights expected 1 arguments, got 023:16
blackburnhmm23:17
KMcQuistenIf I just use the method on a frech CombinedKernel, everything works fine, but if I pass a kernel to a machine and train it, I get that error when I try to extract the weights23:17
blackburnahh I think I know23:17
KMcQuistenrighto23:18
blackburndifferent types23:18
KMcQuistenYes.  when I extract the kernel, it's just type "Kernel",23:18
blackburnget_kernel() returns CKernel23:18
blackburnso rather use old pointer or cast it somehow23:18
KMcQuistenThen how should I get my subkernel weights after I train an MKL?23:18
blackburnlet me check why CKernel has that method23:18
blackburnwell you have combined kernel somewhere in code23:19
blackburnjust call get_subkernel_weights on that23:19
blackburnit will be the same kernel still after train too23:19
KMcQuistenif I call the method from the variable I used to set up the combined kernel, it segfaults23:19
blackburnoh23:19
blackburncool23:19
blackburnlet me fix it then23:19
-!- heiko [~heiko@109.144.207.254] has quit [Ping timeout: 252 seconds]23:20
blackburnshould be easy23:20
-!- heiko [~heiko@host86-181-155-21.range86-181.btcentralplus.com] has joined #shogun23:27
CIA-9shogun: Heiko Strathmann master * r4343cf8 / (2 files): some initial work on Gram-Eigen-Spectrum based Threshold for MMD^2 - http://git.io/h73zWg23:27
CIA-9shogun: Heiko Strathmann master * re04a0e6 / (3 files in 2 dirs): -added convienience methods for display_matrix/vector based on SGVector - http://git.io/wmDC7Q23:27
CIA-9shogun: Heiko Strathmann master * r96dc418 / (2 files): -added create_identity_matrix based on SGMatrix - http://git.io/trXx2A23:27
CIA-9shogun: Heiko Strathmann master * r525eb66 / examples/undocumented/libshogun/mathematics_lapack.cpp : usage of create_identity_matrix - http://git.io/ekx0Ow23:27
CIA-9shogun: Heiko Strathmann master * rf91331a / (5 files in 3 dirs): Merge pull request #563 from karlnapf/master - http://git.io/mMTKXw23:27
blackburnKMcQuisten: I'll commit in a min23:28
blackburnor two or three :D23:28
KMcQuistenRighto23:28
blackburnargh23:37
KMcQuisten?23:37
blackburnfixed in my fork instead of main repo :D23:37
KMcQuistenHa!  Whoops23:37
blackburnKMcQuisten: done - update your repo and recompile - should work23:39
blackburnargh23:39
blackburnnot yet23:39
KMcQuistenok23:39
KMcQuistenI'll wiat23:39
KMcQuisten:)23:39
blackburnnow yes23:39
CIA-9shogun: Sergey Lisitsyn master * r71ddf75 / src/shogun/kernel/CombinedKernel.cpp : Fixed get subkernel weights in CombinedKernel - http://git.io/cGIC9w23:40
KMcQuistenWorks fine.23:48
KMcQuistenThanks23:48
blackburnwelcome!23:50
KMcQuistenI"ll definitely let you know if anything else pops up.23:51
blackburnheh ok23:51
KMcQuistenThanks so much , again23:51
-!- KMcQuisten [d8338942@gateway/web/freenode/ip.216.51.137.66] has quit [Quit: Page closed]23:51
--- Log closed Mon Jun 04 00:00:41 2012