Dear all,
I am trying to iterate an elastic net regression method on a matrix
1) Each column of the matrix ( nxp) will act as a response (y) and rest of
the variables (columns) which means p-1 will act as predictor set
2) i want to store selected variables as matrix
I wrote a code. Please help me to automate this procedure
# install packages for analysis
rm(list = ls())
library(caret)
library(glmnet)
X<-matrix(rnorm(100*500),nrow=100)
y<-X[,1] # I want automatically it will take next column as a response
means untill 500 columns
X1<-X[,-1] # If I use first column as a response it should delete first
column from the matrix
dim(X1)
### Applicarion of the Elastic net for selecting the genes
con<-trainControl(method="cv",number=10)
fit_data<-train(X1,
y,method="glmnet",metric="RMSE",trControl=con,tuneLength = 10)
glmnetcalc<-glmnet(X1,y,alpha=fit_data$finalModel$tuneValue$.alpha)
fitcoef<-predict(glmnetcalc,s=fit_data$finalModel$tuneValue$.alpha,type="coefficients")
CoefEN_1<-as.matrix(fitcoef)
write.table(CoefEN_1, "varSelect_1.txt") # I want to store the non zero
values in a matrix
thanks
Nico
[[alternative HTML version deleted]]
______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
