[BioC] number of variable in svm plot
nofe [guest]
guest at bioconductor.org
Tue Jun 18 01:50:34 CEST 2013
hi all,
the data set i have consists of
x which is a matrix of 39 cancer patients [rows] and 2000 gene names [colmns]. each cell is the value of the gene for a particular patient. there are two types of cancer people representedas factor y.
here is the code:
library(e1071)
#load database
db <- read.csv(file="databases\\colon-cancer\\colon-cancer.csv",head=FALSE,sep=",")
x = as.matrix(db[,1:(ncol(db)-1)])
y = as.factor(db[,ncol(db)])
svmModel = svm(x, y, cost = 10, cachesize=500, scale=F,
type="C-classification", kernel="linear" )
my question is :to plot the svm model i need to plot 2 vaiables as y axis and x axis....but according to the nature of my dataset i have about 2000 variables?? does that mean i have to plot only two?? or is there a way to plot all of the 2000 and beable to show how the svm model classify the dataset.
-- output of sessionInfo():
sessionInfo()
R version 3.0.1 (2013-05-16)
Platform: x86_64-w64-mingw32/x64 (64-bit)
locale:
[1] LC_COLLATE=Arabic_Saudi Arabia.1256 LC_CTYPE=Arabic_Saudi Arabia.1256
[3] LC_MONETARY=Arabic_Saudi Arabia.1256 LC_NUMERIC=C
[5] LC_TIME=Arabic_Saudi Arabia.1256
attached base packages:
[1] parallel stats graphics grDevices utils datasets methods
[8] base
other attached packages:
[1] e1071_1.6-1 class_7.3-7 genefilter_1.42.0
[4] ALL_1.4.14 Biobase_2.20.0 BiocGenerics_0.6.0
[7] BiocInstaller_1.10.2
loaded via a namespace (and not attached):
[1] annotate_1.38.0 AnnotationDbi_1.22.5 DBI_0.2-7
[4] IRanges_1.18.1 RSQLite_0.11.3 splines_3.0.1
[7] stats4_3.0.1 survival_2.37-4 tools_3.0.1
[10] XML_3.96-1.1 xtable_1.7-1
--
Sent via the guest posting facility at bioconductor.org.
More information about the Bioconductor
mailing list