You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@madlib.apache.org by "Frank McQuillan (JIRA)" <ji...@apache.org> on 2016/03/09 01:18:40 UTC
[jira] [Closed] (MADLIB-604) SVM Regression Performance : Several
data sets handling is much slower than libsvm
[ https://issues.apache.org/jira/browse/MADLIB-604?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Frank McQuillan closed MADLIB-604.
----------------------------------
> SVM Regression Performance : Several data sets handling is much slower than libsvm
> ----------------------------------------------------------------------------------
>
> Key: MADLIB-604
> URL: https://issues.apache.org/jira/browse/MADLIB-604
> Project: Apache MADlib
> Issue Type: Bug
> Reporter: Jiali Yao
> Assignee: Rahul Iyer
> Fix For: v1.9
>
> Attachments: svm-regression-benchmarking.jpg
>
>
> For several data sets, MADlib is slower than libsvm
> 1. Time differnce
> {code}
> Kernel is dot
> Data Sets MADlib(Para=true) MADlib(Para=false) libsvm MADlib/libsvm
> cadata 874.15 277.35 2 138.68
> etfidf 501.61 1844.26 32 15.68
> kernel is Polymial
> cadata 932.13 8979.85 2761 0.34
> etfidf 2269.23 3175.87 33 68.76
> space 139.12 238.26 1 139.12
> kernel is Gaussian
> cadata 900.83 9130.2 1 900.83
> cpusmall 390.57 196.13 1 196.13
> 2. Test case example:
> SELECT madlib.svm_regression
> ( 'madlibtestdata.svm_cadata'::text --input_table
> , 'madlibtestresult.reg_model_table'::text --model_table
> , 'false'::boolean --parallel
> , 'madlibtestdata.svm_polynomial'::text --kernel_func
> , 'false'::boolean --verbose
> , '0.1'::float8 --eta
> , '0.005'::float8 --nu
> , '0.05'::float8 --slambda
> ) AS q;
> {code}
> 3. Data sets
> 4. Parameter seting
> MADlib parameter: default value
> R parameter:
> {code}
> svm-train -s 4 -t 0 -c $cost -n 0.005
> eunite2001 0.0595
> E2006 0.0012
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)