Choosing the kernel for SVM (regression) using rminer package
1
0
Entering edit mode
Guest User ★ 12k
@guest-user-4897
Last seen 7.7 years ago
In the rminer package by Paulo Cortez, using the mining function, it is possible to do a SVM regression.Using the script from the documentation SV=mining(V26~.,d,model="svm",Runs=10,method=v,mpar=m,search=s,feat="s ") Is it possible to choose a kernel for the regression other than the default gaussian kernel? I would like to apply the same to a non- linear data and prefer to use a spline or other types of kernel for the same function. -- output of sessionInfo(): R version 3.0.3 (2014-03-06) Platform: x86_64-w64-mingw32/x64 (64-bit) locale: [1] LC_COLLATE=French_France.1252 LC_CTYPE=French_France.1252 LC_MONETARY=French_France.1252 [4] LC_NUMERIC=C LC_TIME=French_France.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] rminer_1.3.1 BiocInstaller_1.12.0 randomForest_4.6-7 loaded via a namespace (and not attached): [1] grid_3.0.3 igraph_0.7.0 kernlab_0.9-19 kknn_1.2-5 lattice_0.20-27 Matrix_1.1-3 nnet_7.3-8 [8] plotrix_3.5-5 rpart_4.1-8 tools_3.0.3 -- Sent via the guest posting facility at bioconductor.org.
Regression Regression • 1.0k views
ADD COMMENT
0
Entering edit mode
@steve-lianoglou-2771
Last seen 7 hours ago
United States
Hi, On Tue, Apr 1, 2014 at 6:32 AM, Paul [guest] <guest at="" bioconductor.org=""> wrote: > > In the rminer package by Paulo Cortez, using the mining function, it is possible to do a SVM regression.Using the script from the documentation > > SV=mining(V26~.,d,model="svm",Runs=10,method=v,mpar=m,search=s,feat= "s") > > Is it possible to choose a kernel for the regression other than the default gaussian kernel? I would like to apply the same to a non- linear data and prefer to use a spline or other types of kernel for the same function. It looks like this wraps the `ksvm` function from kernlab, and it looks like the `...` arguments in the `mining` passes those params through to functions it delegates to, so you should look at how to specify kernels and what not in `ksvm` Also: this isn't really a bioconductor related question and is probably best directed to R-help (or maybe SO, even), but CCing the maintainer was a good idea. -steve -- Steve Lianoglou Computational Biologist Genentech
ADD COMMENT

Login before adding your answer.

Traffic: 260 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6