Greetings:
I am trying to use your R code for R-SVM as follows. Why it dosen't print
the LOO.error and the list of features?
http://www.stanford.edu/group/wonglab/RSVMpage/R-SVM.html
My training data as follows contains 142 cases and 264 features. instead I
get en error as below "invalid 'digits' argument". Is a
working example
available anywhere using R-code of R-SVM?
Thanks very much.
> source("R-SVM.R")
> data<-ReadSVMdata("train.txt")
> len<-length(data$y)
> print(data$y)
[1] -1 -1 1 -1 -1 1 1 1 -1 -1 -1 -1 -1 -1 1 1 1 -1 -1 1 -1 1 1
1 1
[26] 1 1 1 1 -1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 -1 -1 1 1 1 1 -1
-1 -1
[51] -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 1 -1 -1 -1 -1 1
-1 1
[76] -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 1
1 -1
[101] -1 1 -1 -1 -1 -1 -1 1 -1 1 1 -1 -1 -1 -1 1 1 -1 -1 1 -1 -1 -1
-1 -1
[126] 1 -1 1 -1 1 1 -1 -1 -1 1 1 1 -1 -1 -1 -1 1> print(dim(data$x))
[1] 142 264> x<-data$x
> y<-data$y
> ladder<-CreatLadder(len)
> RSVMsum <- RSVM(x,y,ladder,"LOO")
> SummaryRSVM(RSVMsum)
*Error in print.default("MinCV error of", min(RSVMres$Error),
"at",
MinLevel, :
invalid 'digits' argument
Calls: SummaryRSVM -> print -> print.default
Execution halted*
----
I only modified the ReadSVMdata function
ReadSVMdata <- function(filename)
{
dd <- read.table( filename, header=TRUE,sep="\t")
x <- as.matrix( dd[, 14:(ncol(dd))] )
class <- as.numeric(dd[,2])
class[class==2] <- -1
y <- class
ret <- list(x=x, y=y)
}
and un-commented this line " print("MinCV error of",
min(RSVMres$Error),
"at", MinLevel, "genes" )" in function Summary(RSVM).
[[alternative HTML version deleted]]