[B! svm] takc923ã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

takc923 id:takc923

svmã«é–¢ã™ã‚‹takc923ã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (9)

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

${{author_name}}$
{{author_name}}{{created}}
{{ #comment }}{{ comment }}{{ /comment }}
- {{ label }}

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

SVM-Light(ã‚ªãƒ—ã‚·ãƒ§ãƒ³ã€ä½¿ã„æ–¹)
ã“ã“ã§ã¯ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢SVM-Lightã®ä½¿ã„æ–¹ã‚’èª¬æ˜Žã—ã¾ã™ã€‚SVMè‡ªä½“ã®åŽŸç†ã‚’çŸ¥ã‚ŠãŸã„æ–¹ã¯ã€ŒChris Burgeæ°ã®ãƒãƒ¥ãƒ¼ãƒˆãƒªã‚¢ãƒ«ã€ãŒè‰¯ã„æ•™ç§‘æ›¸ã¨ãªã‚‹ã§ã—ã‚‡ã†ã€‚ SVM-Lightã¯å¦ç¿’ãƒ¢ã‚¸ãƒ¥ãƒ¼ãƒ«(svm_learn)ã¨è˜åˆ¥ãƒ¢ã‚¸ãƒ¥ãƒ¼ãƒ«(svm_classify)ã§æ§‹æˆã•ã‚Œã¦ã„ã¾ã™ã€‚ã“ã®è˜åˆ¥ãƒ¢ã‚¸ãƒ¥ãƒ¼ãƒ«ã¯ã€å¦ç¿’ã•ã‚ŒãŸãƒ¢ãƒ‡ãƒ«ã‚’æ–°ãŸãªã‚µãƒ³ãƒ—ãƒ«ã«é©ç”¨ã™ã‚‹ã“ã¨ãŒå‡ºæ¥ã¾ã™ã€‚è©³ã—ãã¯ä»¥ä¸‹ã«ç¤ºã™ã€Œsvm_learnã€ã¨ã€Œsvm_classifyã€ã®ä½¿ã„æ–¹ã‚’å‚ç…§ã—ã¦ãã ã•ã„ã€‚ svm_learn svm_learnã¯æ¬¡ã®ã‚ˆã†ãªãƒ‘ãƒ©ãƒ¡ãƒ¼ã‚¿ã«ã‚ˆã£ã¦å‘¼ã³å‡ºã•ã‚Œã¾ã™ã€‚
takc923 2009/10/28
svm

å’è«–
ãƒªãƒ³ã‚¯
svm light user defined kernels
takc923 2009/10/28
å’è«–

svm
ãƒªãƒ³ã‚¯
Support Vector Machine - PukiWiki
ã‚¤ãƒ³ã‚¹ãƒˆãƒ¼ãƒ« â€ $ mkdir svm_light; cd svm_light $ wget http://download.joachims.org/svm_light/current/svm_light.tar.gz $ tar zxf svm_light.tar.gz $ make â†‘ ä»•æ§˜ãªã© â€ svm_learn $ ./svm_learn [options] example_file model_file example_file ãƒˆãƒ¬ãƒ¼ãƒ‹ãƒ³ã‚°ãƒ‡ãƒ¼ã‚¿ (å…¥åŠ›) <line> .=. <target> <feature>:<value> <feature>:<value> ... <feature>:<value> # <info> <target> .=. +1 | -1 | 0 | <float> <feature> .=. <integer> | "qid" <valu
takc923 2009/10/28
å’è«–

svm
ãƒªãƒ³ã‚¯
Momma's Wiki: SupportVectorMachine/SVMlight - æ‰‹è»½ã«è©¦ã›ã‚‹Support Vector Machine SVM-Ligh...
svm_learn svm_learn [options] example_file model_file ã§å¦ç¿’ã™ã‚‹ã€‚ example_fileã¯å¾Œã§èª¬æ˜Žã™ã‚‹æ›¸å¼ã«å¾“ã£ã¦è¨˜è¿°ã—ãŸãƒ†ã‚ã‚¹ãƒˆãƒ•ã‚¡ã‚¤ãƒ«ã§ã€model_fileã¯å¦ç¿’çµæžœã‚’åŽã‚ã‚‹ãƒ•ã‚¡ã‚¤ãƒ«ã§svm_learnãŒä½œã‚‹ã‚‚ã®ãªã®ã§é©å½“ãªåå‰ã‚’æŒ‡å®šã™ã‚‹ã ã‘ã§è‰¯ã„ã€‚ ã¤ã¾ã‚Šmodel_fileã‚’ä¿å˜ã—ã¦ãŠã‘ã°å¦ç¿’çŠ¶æ…‹ã‚’ç‹¬ç«‹ã—ã¦ç®¡ç†ã§ãã‚‹ã€‚ ã‚ªãƒ—ã‚·ãƒ§ãƒ³ã«é–¢ã—ã¦ã¯åŸºæœ¬çš„ã«-cã‚ªãƒ—ã‚·ãƒ§ãƒ³ã—ã‹å¼„ã‚‰ãªã„ã§è‰¯ã•ãã†ã€‚-cã«æ¸¡ã™æ•°å€¤ãŒå¤§ãããªã‚‹ã»ã©ãƒžãƒ¼ã‚¸ãƒ³ã‚’åºƒãã¨ã‚‹ãŒå¦ç¿’ã«æ™‚é–“ã‚’è¦ã™ã‚‹ã€‚ ã¡ãªã¿ã«-zã‚ªãƒ—ã‚·ãƒ§ãƒ³ã¯å‹•ä½œãƒ¢ãƒ¼ãƒ‰ã®é¸æŠžã§ã€åˆ†é¡žã€å›žå¸°ã€ãƒ©ãƒ³ã‚ãƒ³ã‚°ãŒé¸ã¹ã‚‹ã€‚ example_fileã®æ›¸å¼ <target> <feature>:<value> <feature>:<value> ... <feature>:<value> # <info> <targ
takc923 2009/10/27
svm

å’è«–
ãƒªãƒ³ã‚¯
MySVM ã«é–¢ã™ã‚‹ Tips
MySVM ã«é–¢ã™ã‚‹ Tips [æˆ»ã‚‹] ã¡ã‚‡ã£ã¨ã‚µãƒãƒ¼ãƒˆãƒ™ã‚¯ã‚¿ãƒ¼ãƒžã‚·ãƒ³ã¨ã„ã†ã®ã«èˆˆå‘³ãŒæ²¸ã„ã¦ãã¾ã—ãŸ. ã‚µãƒãƒ¼ãƒˆãƒ™ã‚¯ã‚¿ãƒ¼ãƒžã‚·ãƒ³ã£ã¦ä½•ã‹ã¨ã„ã†ã¨ ãƒ‘ã‚¿ãƒ¼ãƒ³ x ãŒã‚ã‚‹ã‚¯ãƒ©ã‚¹ã«å±žã™ã‚‹ã‹å±žã•ãªã„ã‹ã‚’ çŸ¥ã‚‹ãŸã‚ã®ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã®ã‚ˆã†ã§ ã‚ã‚‰ã‹ã˜ã‚å¦ç¿’ç”¨ã‚µãƒ³ãƒ—ãƒ« x1, x2, x3... xn ã‚’ ç”¨æ„ã—ã¦ãŠããã‚Œãžã‚Œã‚¯ãƒ©ã‚¹ A ã«å±žã™ã‚‹ã‹å±žã•ãªã„ã‹ã‚’ y1, y2, y3,...yn ã« 1, -1 ã§å…¥ã‚Œã¦ãŠã. ã™ã‚‹ã¨ f(x)=Î£Î±i yi, K(x, xi) + b ã¨ã„ã†å¼ã‚’ã‹ã¾ã›ã°f ã®æ£è² ã«ã‚ˆã£ã¦æ–°ã—ã„ãƒ‘ã‚¿ãƒ¼ãƒ³ x ãŒã‚¯ãƒ©ã‚¹ A ã«å±žã™ã‚‹ã‹å±žã•ãªã„ã‹ãŒåˆ¤åˆ¥ã§ãã‚‹. ã¨ã„ã†ã‚‚ã®ã‚‰ã—ã„. ã“ã“ã§ K(x, x') ã¯ã‚«ãƒ¼ãƒãƒ«ã¨å‘¼ã°ã‚Œã‚‹ã‚‚ã®ã§ã„ã‚ã‚“ãªã‚„ã‚Šæ–¹ãŒ ã‚ã‚Šã¾ãŸå¦ç¿’æ™‚ã« Î±i ãŒã©ã†ã„ã†æ•°å€¤ãªã®ã‹ã‚ã‚“ã©ã†ãã•ã„è¨ˆç®—ã§æ±‚ã‚ãªãã¦ã¯ ã„ã‘ãªã„. ã“ã®ã¸ã‚“ãŒã‚„ã‚„ã“ã—ã„ã¨ã¨ã‚‚ã«ã‚µãƒãƒ¼ãƒˆãƒ™ã‚¯
takc923 2009/10/27
svm

å’è«–
ãƒªãƒ³ã‚¯
Kernel-Machines.Org â€” Kernel Machines
This page is devoted to learning methods building on kernels, such as the support vector machine. It grew out of earlier pages at the Max Planck Institute for Biological Cybernetics and at GMD FIRST, snapshots of which can be found here and here. In those days, information about kernel methods was sparse and nontrivial to find, and the kernel machines web site acted as a central repository for the
takc923 2009/10/27
svm

å’è«–
ãƒªãƒ³ã‚¯
winSVM
Split the data into three sets (in the ratio 50%, 25%, 25%). Save the training set as train.txt in the following format: @parameters #training set @examples format xy 7 0 7 9 1 10 4 6 10 3 8 11 9 4 13 3 8 11 5 4 9 6 0 6 3 8 11 4 5 9 Click on "Input file..." and select 'train.txt'. Enter '100' for the number of runs. Click on "Optimize". Inspect input-opt.csv, and note which paramete
takc923 2009/10/27
svm

å’è«–
ãƒªãƒ³ã‚¯
http://www.bi.a.u-tokyo.ac.jp/~tak/index.html
takc923 2009/10/27
svm

å’è«–
ãƒªãƒ³ã‚¯
SVM-Light: Support Vector Machine
SVMlight Support Vector Machine Author: Thorsten Joachims <[email protected]> Cornell University Department of Computer Science Developed at: University of Dortmund, Informatik, AI-Unit Collaborative Research Center on 'Complexity Reduction in Multivariate Data' (SFB475) Version: 6.02 Date: 14.08.2008 Overview SVMlight is an implementation of Support Vector Machines (SVMs) in C. The main featu
takc923 2009/10/27
svm

å’è«–
ãƒªãƒ³ã‚¯
1