| --- Log opened Fri Jun 03 00:00:52 2011 | ||
| -!- blackburn [~qdrgsm@188.122.237.128] has quit [Quit: Leaving.] | 01:36 | |
| -!- sploving [~root@124.16.139.196] has left #shogun [] | 03:02 | |
| -!- CIA-32 [cia@cia.atheme.org] has quit [Read error: Operation timed out] | 05:20 | |
| -!- CIA-18 [cia@cia.atheme.org] has joined #shogun | 05:23 | |
| -!- blackburn [~qdrgsm@188.122.237.128] has joined #shogun | 09:33 | |
| -!- blackburn [~qdrgsm@188.122.237.128] has quit [Ping timeout: 250 seconds] | 09:51 | |
| -!- sploving [~root@124.16.139.196] has joined #shogun | 10:05 | |
| sploving | sonney2k, around? | 10:05 |
|---|---|---|
| sploving | anyone? | 10:05 |
| sploving | it still can not work. :( | 10:05 |
| -!- sploving [~root@124.16.139.196] has left #shogun [] | 10:24 | |
| -!- sploving [~root@124.16.139.196] has joined #shogun | 10:39 | |
| sploving | thx sonney2k, all my work dir is in root dir. It is not easy for me to move it to another user | 10:43 |
| @sonney2k | sploving, look at my email | 10:44 |
| @sonney2k | it works here | 10:44 |
| @sonney2k | gtg | 10:44 |
| sploving | I saw. maybe my swig_type_map.i. I write the ujmp typemap. so it is still have the error. sonney2k, i need have a check my typemap | 10:45 |
| sploving | hmm, not work | 11:10 |
| sploving | I think, i need to dive swig java doc more deeply | 11:10 |
| -!- sploving [~root@124.16.139.196] has left #shogun [] | 11:10 | |
| -!- f-x [~shashwat@117.192.209.39] has joined #shogun | 15:40 | |
| -!- f-x [~shashwat@117.192.209.39] has quit [Ping timeout: 258 seconds] | 17:08 | |
| -!- blackburn [~qdrgsm@188.122.237.128] has joined #shogun | 18:05 | |
| -!- epps [~epps@84.18.157.200] has joined #shogun | 18:59 | |
| -!- epps [~epps@84.18.157.200] has quit [Changing host] | 18:59 | |
| -!- epps [~epps@unaffiliated/epps] has joined #shogun | 18:59 | |
| -!- epps [~epps@unaffiliated/epps] has left #shogun ["Leaving"] | 19:29 | |
| serialhex | hey blackburn, whats going on? | 19:38 |
| blackburn | hey, going fine! just get classic mds to work | 19:38 |
| blackburn | will paste a link in a minute | 19:38 |
| blackburn | what's up with you? | 19:38 |
| serialhex | not too much, about to get to work wrtn some ruby examples | 19:38 |
| blackburn | how is your work going on? | 19:39 |
| blackburn | sold many ones? :) | 19:39 |
| blackburn | well, data for mds: http://dl.dropbox.com/u/10139213/share/mds_data.png | 19:40 |
| serialhex | ehh, i've sold ~$400 worth of knives, i need more appts | 19:40 |
| blackburn | and result: http://dl.dropbox.com/u/10139213/share/mds_data_transformed.png | 19:40 |
| serialhex | right now i'm below average :( | 19:40 |
| serialhex | cool cool, it's spiraly!!! | 19:41 |
| blackburn | oh it is more fast than LLE | 19:48 |
| blackburn | ~65s for 2000 3d examples | 19:52 |
| serialhex | cool cool | 20:33 |
| serialhex | so blackburn, how does the dim reduction work for cases where i might need that third dimension?? | 20:35 |
| blackburn | just the same | 20:36 |
| blackburn | they all are hmmm | 20:36 |
| blackburn | eh.. how to say that :) | 20:36 |
| serialhex | so like if the spiral didnt contract/expand, it sould simply look like a circle | 20:36 |
| blackburn | well if given data X | 20:36 |
| blackburn | and we got x,y,z coordinates of embedding | 20:37 |
| blackburn | so x,y will be 2d spiral | 20:37 |
| blackburn | and z will be just like z in X | 20:37 |
| blackburn | i.e. if map is 3d->3d it will stay the same | 20:37 |
| serialhex | true | 20:37 |
| serialhex | but if - as in your example - the 3d spiral dosnt shrink, the resulting 2d image will simply be a circle | 20:38 |
| blackburn | i dk | 20:38 |
| blackburn | 'll test it right now | 20:38 |
| serialhex | i'm not saying what you're doing is wrong, but the overall theory is kind of wonky imo... :P | 20:38 |
| blackburn | not it should be a circle | 20:38 |
| blackburn | it preserves distances | 20:39 |
| serialhex | oh, ok | 20:39 |
| blackburn | the key idea of MDS is to preserve distance | 20:39 |
| blackburn | *pairwise distances | 20:39 |
| serialhex | ok, that makes sense... because in your current example, if looked at from the top it might look like a couple of zig-zag lines... tho a spiffy alg would change that :D | 20:41 |
| blackburn | hmm interesting thing with just flat spiral | 20:42 |
| blackburn | http://dl.dropbox.com/u/10139213/share/spiral.png | 20:43 |
| blackburn | http://dl.dropbox.com/u/10139213/share/spiral_transformed.png | 20:43 |
| blackburn | serialhex: well these pictures show that idea more clearly | 20:44 |
| blackburn | 3d->2d is just the same spiral | 20:44 |
| blackburn | in 2d* | 20:44 |
| serialhex | hmm, ok | 20:45 |
| blackburn | idk exactly how it will be proceed on different data | 20:46 |
| blackburn | in fact it presents some suprises sometimes | 20:47 |
| blackburn | for example when flat 3d data is given it have only 2 non-zero eigenvalues | 20:47 |
| blackburn | if it is not flat - 3 non-zero eigs | 20:48 |
| blackburn | looks like it reflects initial dimension somehow | 20:48 |
| serialhex | hmm | 20:48 |
| blackburn | I speak some broken way now :) | 20:49 |
| serialhex | heh, well it's ok, you're trying to explain funky technical things, which is hard sometimes in ones native language :D | 20:50 |
| @bettyboo | funny? | 20:50 |
| serialhex | yes it is funny bettyboo | 20:50 |
| @bettyboo | serialhex: for static interfaces like matlab you have to modify libshogunui/GUIKernel.cpp / SGInterface.cpp - not as nice but will work :) | 20:50 |
| blackburn | serialhex: just tried for 3d parabola | 20:52 |
| serialhex | cool | 20:52 |
| blackburn | .. and it creates 3d parabola | 20:52 |
| blackburn | :D | 20:52 |
| @bettyboo | ;D blackburn | 20:52 |
| blackburn | huh how useless it is for some data | 20:52 |
| serialhex | you cant squinch to 2d?? | 20:52 |
| blackburn | i can't but it is done not that way I want | 20:53 |
| blackburn | from side or so | 20:53 |
| blackburn | right mapping should be just like circle | 20:53 |
| serialhex | how about the top? | 20:53 |
| serialhex | looking into the parabola... | 20:53 |
| blackburn | aha | 20:53 |
| blackburn | should be this way | 20:53 |
| blackburn | I think the difference lies in parabola height | 20:54 |
| blackburn | btw MDS for euclidean distance is equal to PCA :D | 20:54 |
| blackburn | serialhex: well multiplying Z by 0.005 gave right mapping to 2d | 20:56 |
| blackburn | haha | 21:30 |
| blackburn | I had been using *deleted* array | 21:31 |
| blackburn | serialhex: around? | 21:34 |
| blackburn | ah, nevermind I found :) | 21:36 |
| -!- alesis-novik [~alesis@188.74.87.84] has joined #shogun | 21:47 | |
| @sonney2k | wtf?! | 22:00 |
| @sonney2k | now I am also getting baozengs error | 22:00 |
| blackburn | with C++ you can not only make errors but also inherit it hehe | 22:13 |
| @bettyboo | hrhr | 22:13 |
| serialhex | hey blackburn i'm here now, whats up?? | 22:32 |
| blackburn | serialhex: just english issue :) | 22:32 |
| serialhex | oh, ok :) | 22:33 |
| blackburn | forgot how to say about eigenvector's eigenvalue | 22:33 |
| serialhex | sonney2k, baozeng wanted me to rub the java tests, how do i do that?? (i have 0 java expierence) | 22:33 |
| blackburn | associated one | 22:34 |
| serialhex | oh, ok... eigenvalues & such still kind of confuse me, mostly cause i havnt needed/studied them at all | 22:34 |
| blackburn | serialhex: just wanted how to say that they are associated - now I know :) | 22:35 |
| blackburn | wanted to know8 | 22:35 |
| blackburn | * | 22:35 |
| serialhex | ok, cool | 22:35 |
| serialhex | hmm... apparently the documentation is out of date, even when you generate it :( | 22:39 |
| serialhex | STEAK!!! | 22:40 |
| -!- blackburn [~qdrgsm@188.122.237.128] has quit [Remote host closed the connection] | 22:59 | |
| -!- blackburn [~qdrgsm@188.122.237.128] has joined #shogun | 23:00 | |
| -!- blackburn1 [~qdrgsm@188.122.237.128] has joined #shogun | 23:04 | |
| -!- blackburn [~qdrgsm@188.122.237.128] has quit [Read error: Connection reset by peer] | 23:04 | |
| -!- blackburn1 [~qdrgsm@188.122.237.128] has quit [Ping timeout: 246 seconds] | 23:11 | |
| --- Log closed Sat Jun 04 00:00:54 2011 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!
