--- Log opened Sun Jul 24 00:00:05 2011 | ||
@sonney2k | gtg need to catch the train ;-) | 00:19 |
---|---|---|
-!- blackburn [~blackburn@188.122.253.215] has quit [Quit: Leaving.] | 00:23 | |
-!- in3xes1 [~in3xes@180.149.49.227] has quit [Ping timeout: 240 seconds] | 00:39 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 00:57 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 240 seconds] | 01:44 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 01:44 | |
-!- f-x [~user@117.192.198.14] has joined #shogun | 02:09 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 252 seconds] | 02:34 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 02:34 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 258 seconds] | 02:57 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 03:02 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 240 seconds] | 03:06 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 03:07 | |
-!- f-x [~user@117.192.198.14] has quit [Ping timeout: 260 seconds] | 04:24 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 240 seconds] | 05:14 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 05:21 | |
-!- in3xes_ [~in3xes@180.149.49.227] has joined #shogun | 05:42 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Read error: Operation timed out] | 05:44 | |
-!- in3xes_ [~in3xes@180.149.49.227] has quit [Ping timeout: 258 seconds] | 05:48 | |
-!- in3xes_ [~in3xes@180.149.49.227] has joined #shogun | 05:49 | |
-!- in3xes_ [~in3xes@180.149.49.227] has quit [Ping timeout: 276 seconds] | 05:57 | |
-!- in3xes_ [~in3xes@180.149.49.227] has joined #shogun | 05:57 | |
-!- f-x [~user@117.192.198.14] has joined #shogun | 06:12 | |
-!- f-x [~user@117.192.198.14] has quit [Ping timeout: 260 seconds] | 06:17 | |
-!- in3xes_ [~in3xes@180.149.49.227] has quit [Ping timeout: 258 seconds] | 06:36 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 06:53 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 264 seconds] | 07:25 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 08:00 | |
-!- gsomix [~gsomix@109.169.136.182] has joined #shogun | 08:42 | |
-!- f-x [~user@117.192.220.149] has joined #shogun | 08:50 | |
-!- f-x` [~user@117.192.220.149] has joined #shogun | 08:51 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 252 seconds] | 09:05 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 09:43 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 276 seconds] | 09:50 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 09:55 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 258 seconds] | 10:00 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 10:08 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 264 seconds] | 10:15 | |
-!- f-x` [~user@117.192.220.149] has quit [Ping timeout: 260 seconds] | 10:51 | |
-!- f-x [~user@117.192.220.149] has quit [Ping timeout: 260 seconds] | 10:51 | |
-!- f-x [~user@117.192.201.69] has joined #shogun | 11:00 | |
gsomix | 40C. It's hot. =___= | 11:10 |
-!- blackburn [~blackburn@188.122.253.215] has joined #shogun | 11:28 | |
blackburn | hooray new keyboard | 11:31 |
gsomix | :) | 11:32 |
blackburn | not key6oard! | 11:36 |
-!- blackburn [~blackburn@188.122.253.215] has quit [Quit: Leaving.] | 11:39 | |
@sonney2k | :) | 11:43 |
@sonney2k | congrats | 11:43 |
CIA-87 | shogun: Alesis Novik master * r8af741f / (3 files in 2 dirs): Initial SMEM implementation. With bugs, fix will follow. - https://github.com/shogun-toolbox/shogun/commit/8af741f5fb1f39d97ddbbd4fce3810a2c953f07f | 11:46 |
CIA-87 | shogun: Alesis Novik master * r104f0f0 / src/shogun/clustering/GMM.cpp : Some memory leaks and general fixes. Mostly works now. - https://github.com/shogun-toolbox/shogun/commit/104f0f0ee5e23d95562fa7bd21b69b0da230fd91 | 11:46 |
CIA-87 | shogun: Alesis Novik master * rfc4baa4 / (2 files): Added SMEM GMM examples - https://github.com/shogun-toolbox/shogun/commit/fc4baa4e08676a1501b5a244cb3c16dd7392d9a1 | 11:46 |
CIA-87 | shogun: Soeren Sonnenburg master * ra6e9a4a / (5 files in 3 dirs): | 11:46 |
CIA-87 | shogun: Merge pull request #229 from alesis/gmm | 11:46 |
CIA-87 | shogun: SMEM for GMM - https://github.com/shogun-toolbox/shogun/commit/a6e9a4a0b386ac2d9abef5bcbda74d1f943cfdd7 | 11:46 |
-!- blackburn [~blackburn@188.122.253.215] has joined #shogun | 11:46 | |
@sonney2k | 18 C - freezing | 11:47 |
blackburn | sonney2k: you could warm up berlin with your macbook | 11:48 |
gsomix | We are using air conditioning at home. But the outside is very bad to work. | 11:52 |
@sonney2k | blackburn, did you manage to look at the java examples? | 11:53 |
blackburn | sonney2k: going to test all of them today | 11:53 |
@sonney2k | ok | 11:54 |
@sonney2k | and gsomix despite the wheather how is wresting with C5.0 going? | 11:54 |
gsomix | I'm working on some data structures now. Blackburn and I have no idea how to integrate c5.0. =___= | 11:57 |
CIA-87 | shogun: Soeren Sonnenburg master * rcdc1dc6 / README : mention modular interfaces at the top and static ones below - https://github.com/shogun-toolbox/shogun/commit/cdc1dc6ae1467816896f449cb98d871feede8f2f | 11:58 |
@sonney2k | gsomix, I see | 11:58 |
@sonney2k | gsomix, do you perhaps have a C5.0 branch somewhere? | 11:59 |
-!- gsomix [~gsomix@109.169.136.182] has quit [Read error: Connection reset by peer] | 12:02 | |
-!- gsomix [~gsomix@178.45.88.77] has joined #shogun | 12:04 | |
gsomix | sonney2k, only on my hard disk. If I understand the phrase. | 12:04 |
@sonney2k | gsomix, mind pushing it to github such that I can have a look? | 12:05 |
blackburn | I guess he have nothing to push yet | 12:05 |
blackburn | with C5.0 I mean | 12:05 |
blackburn | DORGQR! | 12:07 |
gsomix | Only source code of C5.0. No more. | 12:07 |
gsomix | sonney2k, send the code of C5.0 to github? Specify. | 12:13 |
@sonney2k | gsomix, I meant you have a local branch w/ c5.0 that you could just push - but it seems not. | 12:19 |
@sonney2k | so forget it | 12:19 |
gsomix | Ok. | 12:20 |
blackburn | sonney2k: ok to merge? https://github.com/shogun-toolbox/shogun/pull/230/files | 13:12 |
@sonney2k | blackburn, you have the OK from the author or where does this come from? | 13:14 |
blackburn | sonney2k: he rewrote it ;) | 13:14 |
@sonney2k | seriously? | 13:15 |
blackburn | gsomix: did you? | 13:15 |
gsomix | yep | 13:15 |
blackburn | sonney2k: so I guess no license issues | 13:15 |
@sonney2k | true then | 13:15 |
@sonney2k | lots of work ... | 13:15 |
blackburn | it could speed up isomap when I use it for dijkstra | 13:16 |
blackburn | sonney2k: could you merge it? I'm messed with some QR now | 13:16 |
blackburn | sonney2k: ah, nevermind, I'll do | 13:18 |
blackburn | gtg | 13:36 |
gsomix | sonney2k, all ok? could you merge it? | 13:39 |
-!- blackburn [~blackburn@188.122.253.215] has quit [Ping timeout: 255 seconds] | 13:42 | |
-!- in3xes [~in3xes@180.149.49.230] has joined #shogun | 13:45 | |
-!- in3xes [~in3xes@180.149.49.230] has quit [Quit: Leaving] | 14:39 | |
-!- blackburn [~blackburn@188.122.253.215] has joined #shogun | 14:45 | |
@sonney2k | I did add some comments | 15:00 |
blackburn | sonney2k: they all are fixed | 15:01 |
gsomix | yep | 15:02 |
CIA-87 | shogun: Evgeniy Andreev master * r194468a / (2 files): Rewritten and added CFibonacciHeap class. - https://github.com/shogun-toolbox/shogun/commit/194468a2e9867b9aa5645edaabe70c951f2e6da0 | 15:06 |
CIA-87 | shogun: Evgeniy Andreev master * rbb39c2f / (2 files): Some fixes. - https://github.com/shogun-toolbox/shogun/commit/bb39c2f33cd6e1d8301fd98e097581236a12de07 | 15:06 |
CIA-87 | shogun: Evgeniy Andreev master * r1ba5573 / (2 files): Some fixes. - https://github.com/shogun-toolbox/shogun/commit/1ba5573b2f2c3d06af937d0da92a3804156c423d | 15:06 |
CIA-87 | shogun: Soeren Sonnenburg master * r96ba02b / (2 files): | 15:06 |
CIA-87 | shogun: Merge pull request #230 from gsomix/master | 15:06 |
CIA-87 | shogun: Implemented Fibonacci heap. - https://github.com/shogun-toolbox/shogun/commit/96ba02b9a7073d8a4b8acd943b447a1bd37c9ca9 | 15:06 |
gsomix | Yahoo. My first real code in the project. :) | 15:08 |
blackburn | okay let the great java test begin | 15:08 |
@sonney2k | gsomix, thanks | 15:15 |
@sonney2k | gtg | 15:15 |
blackburn | PCA differs | 15:47 |
blackburn | okay LoadMatrix is definitely wrong | 15:48 |
-!- f-x [~user@117.192.201.69] has quit [Ping timeout: 260 seconds] | 15:53 | |
blackburn | hmm no, println produces flat matrix | 15:58 |
-!- blackburn [~blackburn@188.122.253.215] has quit [Quit: Leaving.] | 16:40 | |
-!- f-x [~user@117.192.199.217] has joined #shogun | 17:00 | |
-!- blackburn [~blackburn@188.122.239.253] has joined #shogun | 17:23 | |
-!- in3xes [~in3xes@180.149.49.230] has joined #shogun | 18:18 | |
-!- blackburn [~blackburn@188.122.239.253] has quit [Ping timeout: 255 seconds] | 18:32 | |
-!- f-x [~user@117.192.199.217] has quit [Ping timeout: 260 seconds] | 18:52 | |
-!- in3xes [~in3xes@180.149.49.230] has quit [Read error: Connection reset by peer] | 19:41 | |
-!- in3xes [~in3xes@180.149.49.230] has joined #shogun | 19:57 | |
-!- blackburn [~blackburn@188.122.239.253] has joined #shogun | 20:08 | |
-!- [1]warpy [~warpy@bzq-79-181-43-167.red.bezeqint.net] has joined #shogun | 20:16 | |
blackburn | sonney2k: I got rid of wi-fi lost connection. the solution is to use another kernel :D | 20:22 |
CIA-87 | shogun: Alesis Novik master * r81c0c6d / examples/undocumented/python_modular/graphical/em_1d_gmm.py : Small example changes - https://github.com/shogun-toolbox/shogun/commit/81c0c6de0de7fbf69ef6e16ce6aba4a62a094bb3 | 20:35 |
CIA-87 | shogun: Alesis Novik master * r220e8bf / (2 files in 2 dirs): Bug fixes for Gaussian and example - https://github.com/shogun-toolbox/shogun/commit/220e8bf4829ddc1203abd613281d01d4ea9c7a7a | 20:35 |
CIA-87 | shogun: Alesis Novik master * r74f3665 / examples/undocumented/python_modular/graphical/smem_2d_gmm.py : A more illustrative example of SMEMs effect. - https://github.com/shogun-toolbox/shogun/commit/74f3665bb02bf1ad2240471acd3d76994023746b | 20:35 |
CIA-87 | shogun: Soeren Sonnenburg master * ra75874e / (3 files in 2 dirs): | 20:35 |
CIA-87 | shogun: Merge pull request #231 from alesis/gmm | 20:35 |
CIA-87 | shogun: Important EM bug fix and nice examples - https://github.com/shogun-toolbox/shogun/commit/a75874e61e7719575310acc46bb68df5134a26e3 | 20:35 |
@sonney2k | blackburn, congrats | 20:36 |
@sonney2k | blackburn, is java loadmatrix working or not? | 20:38 |
blackburn | sonney2k: it is ok, I was wrong | 20:39 |
-!- serialhex_ is now known as serialhex | 20:39 | |
@sonney2k | k | 20:39 |
blackburn | when printing | 20:39 |
blackburn | it produces strange output | 20:39 |
@sonney2k | you mean toString() | 20:40 |
@sonney2k | ? | 20:40 |
blackburn | println, yes | 21:08 |
blackburn | sonney2k: difficult to compare randomized algo like landmarkisomap :D | 21:09 |
@sonney2k | blackburn, why? | 21:09 |
blackburn | how can I fix landmarks if they are all random | 21:09 |
@sonney2k | blackburn, there is no randomness in this world | 21:11 |
@sonney2k | just init the rnd generator with the same seed | 21:11 |
@sonney2k | CMath::init_random | 21:11 |
blackburn | sonney2k: I just ran all of examples without > /dev/null -- looks pathetic | 21:12 |
@sonney2k | blackburn, which means? | 21:13 |
blackburn | many asserts fail | 21:14 |
@sonney2k | great | 21:14 |
blackburn | will check regression now | 21:14 |
@sonney2k | blackburn, so lets concentrate on simple examples first | 21:14 |
@sonney2k | features_* | 21:14 |
blackburn | hm okay | 21:14 |
@sonney2k | you should check if you can get/set features and result is the OK | 21:14 |
* sonney2k was afraid of that | 21:16 | |
blackburn | sonney2k: ufff just take a look at features_simple_modular.java | 21:17 |
@sonney2k | blackburn, yes? | 21:19 |
blackburn | aren't you getting confused with adding a matrix asList? | 21:20 |
@sonney2k | blackburn, you mean this Arrays.asList(A) ? | 21:21 |
@sonney2k | looks weird indeed | 21:21 |
blackburn | sonney2k: yes | 21:21 |
blackburn | uh oh | 21:22 |
@sonney2k | better use an ArrayList of DoubleMatrix | 21:22 |
blackburn | I guess Arrays.asList(A) doesn't produce doublematrix | 21:23 |
blackburn | okay I'll just try to set/get features | 21:25 |
blackburn | sonney2k: any way to map from "get_feature_matrix" style to "getFeatureMatrix" style automagically? | 21:29 |
@sonney2k | blackburn, no | 21:30 |
blackburn | sad | 21:30 |
@sonney2k | I don't thinks os | 21:30 |
@sonney2k | so | 21:30 |
blackburn | sonney2k:ok, set/get RealFeatures - works | 21:30 |
@sonney2k | I know this is troublesome for pure java people but at least for me having to program in python java etc etc - I will die | 21:31 |
@sonney2k | ok - at least | 21:31 |
@sonney2k | blackburn, next level would be a kernel matrix / distance matrix | 21:31 |
blackburn | I guess I have to rewrite simple features test | 21:32 |
blackburn | and then distance,yes | 21:32 |
blackburn | sonney2k: is it ok to use DoubleMatrix everywhere? | 21:39 |
@sonney2k | blackburn, jblas simply doesn't support other data types than float/double | 21:42 |
@sonney2k | so it is the only way to 'emulate' other types - so yes | 21:42 |
blackburn | so, e.g. new bytefeatures(doublematrix) is ok? | 21:44 |
@sonney2k | yes | 21:51 |
@sonney2k | other things will only work with ujmp | 21:51 |
@sonney2k | s/other things/native types/ | 21:51 |
blackburn | I see | 21:52 |
@sonney2k | blackburn, so do kernels work? | 22:09 |
blackburn | not tested yet | 22:09 |
CIA-87 | shogun: Sergey Lisitsyn master * r4a479cc / (3 files in 2 dirs): Merge branch 'master' of github.com:shogun-toolbox/shogun - https://github.com/shogun-toolbox/shogun/commit/4a479cce25911d1de3477babd7c9d288e7e0b6c9 | 22:15 |
CIA-87 | shogun: Sergey Lisitsyn master * r32df164 / (4 files): Replaced simple features java test with one - https://github.com/shogun-toolbox/shogun/commit/32df164cd0a76b87248fef5e9cca2bcdcd15d4a0 | 22:15 |
CIA-87 | shogun: Sergey Lisitsyn master * rc1bbc5d / (3 files): Split isomap java example - https://github.com/shogun-toolbox/shogun/commit/c1bbc5df97c100500f386a85c40b4e5976bec7f3 | 22:18 |
CIA-87 | shogun: Sergey Lisitsyn master * ra844225 / examples/undocumented/java_modular/preprocessor_landmarkisomap_modular.java : Removed unnecessary println at landmark isomap java example - https://github.com/shogun-toolbox/shogun/commit/a8442259d8ac93013e54eb86148d94b76fa159fa | 22:20 |
blackburn | sonney2k: looks like it is ok with gaussian kernel | 22:23 |
@sonney2k | hurray - at least sth | 22:24 |
@sonney2k | blackburn, next level classifier svm + kernel | 22:25 |
blackburn | hmm | 22:26 |
blackburn | not, sth is wrong | 22:26 |
blackburn | I replaced data with | 22:28 |
blackburn | 2.0 2.0 | 22:28 |
blackburn | 3.0 3.0 | 22:28 |
blackburn | 2.5 2.5 | 22:28 |
blackburn | in python got | 22:28 |
blackburn | [[ 1. 0.82505297 0.38230427] | 22:28 |
blackburn | [ 0.82505297 1. 0.46336937] | 22:28 |
blackburn | [ 0.38230427 0.46336937 1. ]] | 22:28 |
blackburn | in java | 22:28 |
blackburn | [1.0, 0.3823042728920807, 0.3823042728920807; | 22:28 |
blackburn | 0.3823042728920807, 1.0, 0.6807123983233854; | 22:28 |
blackburn | 0.3823042728920807, 0.6807123983233854, 1.0] | 22:28 |
-!- blackburn was kicked from #shogun by bettyboo [flood] | 22:28 | |
-!- blackburn [~blackburn@188.122.239.253] has joined #shogun | 22:29 | |
blackburn | sonney2k: ^ | 22:29 |
@sonney2k | blackburn, please check if features are indeed the same after you got the kernel matrix | 22:31 |
@sonney2k | sth like | 22:31 |
@sonney2k | kernel.get_lhs().get_feature_matrix() | 22:31 |
blackburn | sonney2k: trouble is get_lhs return features.. | 22:40 |
@sonney2k | blackburn, ahh yes true - that is troublesome under python | 22:41 |
@sonney2k | then just check if the features you create still contain the values | 22:41 |
blackburn | sonney2k: hmm it is ok | 22:43 |
blackburn | I guess 'reshape' works different in java and python | 22:43 |
@sonney2k | both in java and python? | 22:43 |
@sonney2k | reshape? | 22:43 |
@sonney2k | what for? | 22:43 |
blackburn | load uses it | 22:44 |
blackburn | yes, different way of reshape | 22:45 |
blackburn | *very* strange way of reshape in java | 22:45 |
blackburn | 2.0 2.0 | 22:45 |
blackburn | 3.0 3.0 | 22:45 |
blackburn | 2.5 2.5 | 22:45 |
blackburn | in python: | 22:45 |
blackburn | [[ 2. 2. 3. ] | 22:45 |
blackburn | [ 3. 2.5 2.5]] | 22:45 |
blackburn | in java | 22:45 |
blackburn | [2.0, 2.5, 3.0; 3.0, 2.0, 2.5] | 22:46 |
@sonney2k | what is the input? | 22:46 |
blackburn | 2.0 2.0 | 22:47 |
blackburn | 3.0 3.0 | 22:47 |
blackburn | 2.5 2.5 | 22:47 |
blackburn | I modified fm_train_real.dat | 22:47 |
@sonney2k | how can that be reshaped? | 22:47 |
@sonney2k | it is a matrix already?! | 22:47 |
@sonney2k | shouldn't this be transposed? | 22:47 |
blackburn | no, it reads not matrix but vector I guess | 22:48 |
blackburn | just take a look how it works in python and java | 22:48 |
blackburn | Load.java and load.py or so | 22:49 |
@sonney2k | blackburn, looks like the python thing is totally broken | 23:00 |
blackburn | :D | 23:00 |
blackburn | surprise! | 23:00 |
@sonney2k | it should be numpy.loadtxt('../data/fm_train_real.dat').T | 23:00 |
@sonney2k | instead of whatever is done in load_numbers | 23:00 |
blackburn | wonderful | 23:01 |
@sonney2k | I now understand why this happened | 23:02 |
blackburn | why? | 23:02 |
@sonney2k | at some point we had data in matlab compatible format | 23:02 |
@sonney2k | ascii format | 23:02 |
@sonney2k | that means one example == one column | 23:03 |
@sonney2k | but this is stupid for ascii | 23:03 |
@sonney2k | so we transposed this | 23:03 |
@sonney2k | but the python function was never changed obviously | 23:03 |
blackburn | sonney2k: so, what should I/you do? | 23:07 |
@sonney2k | blackburn, please go on | 23:08 |
blackburn | sonney2k: with java? | 23:08 |
@sonney2k | sure | 23:09 |
CIA-87 | shogun: Soeren Sonnenburg master * rd8c70f9 / examples/undocumented/python_modular/tools/load.py : | 23:09 |
CIA-87 | shogun: fix loading of matrices (we now have one example per line - no longer | 23:09 |
CIA-87 | shogun: one per column) - https://github.com/shogun-toolbox/shogun/commit/d8c70f99ee09ace4bca8c8d192997e60ada519d6 | 23:09 |
CIA-87 | shogun: Soeren Sonnenburg master * r020d909 / (7 files): Merge branch 'master' of github.com:shogun-toolbox/shogun - https://github.com/shogun-toolbox/shogun/commit/020d90962a1ff71ab4a67e7b689f4c50dbae1fce | 23:09 |
blackburn | ah | 23:09 |
blackburn | okay | 23:09 |
blackburn | the most lame weekly report ever | 23:18 |
blackburn | I should raise my efficiency, I feel sorry for such low productivity | 23:18 |
-!- gsomix [~gsomix@178.45.88.77] has quit [Ping timeout: 260 seconds] | 23:18 | |
@sonney2k | blackburn, do kernels now work as they should? | 23:28 |
blackburn | sonney2k: will test in a min | 23:28 |
CIA-87 | shogun: Sergey Lisitsyn master * r480b717 / examples/undocumented/python_modular/tools/load.py : Fixed load.py - https://github.com/shogun-toolbox/shogun/commit/480b7177dab18f8844fdd44a2c28c8023aa49549 | 23:31 |
blackburn | sonney2k: ok, different matrices again | 23:31 |
@sonney2k | different in which respect? | 23:31 |
blackburn | .dat is | 23:32 |
blackburn | 2.0 2.0 | 23:32 |
blackburn | 3.0 3.0 | 23:32 |
blackburn | 2.5 2.5 | 23:32 |
blackburn | python: | 23:32 |
blackburn | [[ 2. 3. 2.5] | 23:32 |
blackburn | [ 2. 3. 2.5]] | 23:32 |
blackburn | java: | 23:32 |
blackburn | [2.0, 2.5, 3.0; 3.0, 2.0, 2.5] | 23:32 |
blackburn | sonney2k: java one is wrong I guess? | 23:36 |
@sonney2k | yup | 23:36 |
@sonney2k | blackburn, my guess is you need to transpose not reshape | 23:38 |
* sonney2k wonders if blackburn needs some more fuel | 23:39 | |
@sonney2k | aka the V-word | 23:40 |
blackburn | why I need? :( | 23:40 |
blackburn | remember I work not on my project and I don't know much about it | 23:40 |
blackburn | ;) | 23:40 |
@sonney2k | I think you know pretty well what a matrix is :) | 23:42 |
blackburn | but don't know much about jblas? | 23:43 |
CIA-87 | shogun: Sergey Lisitsyn master * rf70a405 / examples/undocumented/java_modular/Load.java : Fixed load matrix for java modular - https://github.com/shogun-toolbox/shogun/commit/f70a4056edbfef3e3070a99a787331c591b7d321 | 23:43 |
@sonney2k | blackburn, me neither but it is trivial... | 23:43 |
blackburn | okay fixed | 23:43 |
@sonney2k | at least google finds me everything I need | 23:43 |
blackburn | yes yes | 23:43 |
@sonney2k | and? | 23:43 |
@sonney2k | does it now return the same thing? | 23:43 |
blackburn | I'm such a stupid russian okay :D | 23:43 |
blackburn | yes, kernel matrix is the same now | 23:44 |
@sonney2k | yeaaah | 23:44 |
@sonney2k | more | 23:44 |
@sonney2k | more | 23:44 |
@sonney2k | classifier | 23:44 |
blackburn | sonney2k: toString in println! such a useless thing. everywhere | 23:47 |
@sonney2k | blackburn, is this automagically called? | 23:47 |
blackburn | of course it is | 23:48 |
@sonney2k | I know it is in python when __str__ or __repr__ is there | 23:48 |
@sonney2k | ok | 23:48 |
blackburn | yes, just like in python | 23:48 |
@sonney2k | anyway, please call some libsvm classifier thingy | 23:49 |
@sonney2k | that should be easy enough to see if things work at all | 23:49 |
blackburn | now try gaussian naive bayes, then libsvm | 23:49 |
blackburn | hmm | 23:51 |
blackburn | hm gaussian naive bayes is broken I gues | 23:53 |
blackburn | s | 23:53 |
blackburn | sonney2k: libsvm minimal produces ~0.08 mean error both for java and python | 23:57 |
@sonney2k | blackburn, please compare label - outputs not just accuracy | 23:57 |
@sonney2k | label outputs must match exactly | 23:58 |
blackburn | damn that day I say I could work on java examples | 23:58 |
--- Log closed Mon Jul 25 00:00:12 2011 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!