Open in new window / Try shogun cloud
--- Log opened Sun Feb 24 00:00:50 2013
-!- alexlovesdata [] has joined #shogun01:14
-!- alexlovesdata [] has left #shogun []01:22
shogun-buildbot_build #292 of nightly_default is complete: Failure [failed test]  Build details are at
-!- sumit_ [75ea48a9@gateway/web/freenode/ip.] has joined #shogun06:45
-!- sumit_ [75ea48a9@gateway/web/freenode/ip.] has quit [Ping timeout: 245 seconds]08:06
-!- hoijui [] has joined #shogun08:33
-!- lambday [0e8b614f@gateway/web/freenode/ip.] has joined #shogun13:42
-!- wiking [~wiking@huwico/staff/wiking] has quit [Read error: Connection reset by peer]14:43
-!- n4nd0 [] has joined #shogun15:06
-!- zxtx [] has quit [Ping timeout: 256 seconds]15:39
-!- zxtx [] has joined #shogun15:51
-!- n4nd0 [] has quit [Quit: leaving]16:09
lambdayhello.. this is soumyajit.. I have been trying with string kernels... was using fisher's protein homology detection data (
lambdaygot an error "ALPHABET does not contain all symbols in histogram"... even after removing the sequences which contained something other than [A-Z]...16:19
lambdaywas using PROTEIN alphabet...16:19
lambdaywhen I ran it with ALPHANUM, it worked... but none of the sequences had any digits16:20
lambdayand also, is the string subsequence kernel ( or something similar) implemented in shogun? I was looking at the string kernels but couldn't find out :(16:28
-!- sumit [73f91219@gateway/web/freenode/ip.] has joined #shogun19:12
-!- sumit [73f91219@gateway/web/freenode/ip.] has quit [Ping timeout: 245 seconds]19:16
-!- lambday [0e8b614f@gateway/web/freenode/ip.] has quit [Quit: Page closed]19:17
-!- n4nd0 [] has joined #shogun20:13
blackburnhey n4nd020:25
n4nd0blackburn: hi20:28
n4nd0blackburn: how is it going?20:28
blackburnpretty fine20:28
-!- nona994 [] has joined #shogun21:18
nona994Hi, I am desprate to work with WeightedDegreeRBFKernel but I can't figure its variables out. I went through its source code but it seems rather impossible to make a connection.21:25
blackburnnona994: hi nona21:29
blackburnnona994: what are your features?21:30
blackburnnona994: weighted degree rbf works with dense real-valued features only21:30
nona994can you please explain what is a dense real-valued feature. I guess it's a matrix right?21:34
blackburnnona994: exactly21:34
blackburnjust dense matrix21:34
nona994n by n matrix right?21:35
nona994these are arguaments for wd RBF 'int32_t size: is cache size, float64_t w: is ???, int32_t d: is degree for wd kernel, int32_t nof_prop: ??'21:37
blackburnnona994: no, n_features by n_vectors21:37
blackburnnona994: w is width (in means of exponential parameter coefficient)21:38
blackburnline 102 in .cpp21:38
blackburnnof_prop - 'number of properties' but not sure what it means21:39
blackburnI am afraid it is quite experimental21:40
blackburnif sonney2k was here he could tell us21:40
nona994I see How can I pass a dense features21:40
nona994Yes sure,  he is the Oracle here21:42
blackburnnona994: is that a question about dense features?21:43
nona994I mean since you said it works with dense features I want to know how can I pass this dens feature21:45
blackburnI am confused21:45
blackburnwell if you have real valued features just use it21:45
blackburnwhat is the data you are working with?21:45
nona994I have protein sequences21:45
nona994say they are strings21:46
blackburnso you have no real-valued features I guess21:46
blackburnthen it looks quite impossible to use that kernel21:47
nona994weighted degree (wd) works with strings and  wd rbf include some specific measurement (real value) for each character or in my case amino acid21:48
blackburnin current implementation it just uses something similar to weighted degree kernel but for real values so I am unsure21:49
nona994you are saying instead of strings it uses real-valued features and does the same computations and everything?21:53
blackburnnona994: yes21:53
nona994so if I initiate it with a real-valued matrix as TRAINing set it has to work21:54
nona994what is the difference between line 24 and 31 why dense features have l and r?21:56
blackburnnona994: treat it as rows and cols21:57
blackburnnona994: say l has 5 features21:57
blackburn5 vectors21:58
blackburnand r has 3 vectors21:58
blackburnthen kernel matrix is 5 x 321:58
nona994you mean dense features is 5 x 321:59
blackburnnona994: no, matrix of kernel between l and r is 5 x 321:59
nona994i have got no idea about dense features :(22:10
blackburnnona994: well images, etc22:21
blackburnthey are examples of dense features22:21
nona994Are off to find them22:26
nona994many thanks22:26
nona994I am off to find them22:26
-!- nona994 [] has quit [Quit: irc2go]22:58
-!- hoijui [] has quit [Ping timeout: 244 seconds]23:41
--- Log closed Mon Feb 25 00:00:50 2013