Re: [R] use "caret" to rank predictors by random forest model

f important predictors given a series of sizes of the subsets. I'm wondering if "caret" can automatically give the best size of the selected subset without user providing the candidate sizes. Thanks, Best, Xiaoqi - Original Message - From: "Max Kuhn" To: "Xiaoqi

[R] use "caret" to rank predictors by random forest model

Hi, I'm using package "caret" to rank predictors using random forest model and draw predictors importance plot. I used below commands: rf.fit<-randomForest(x,y,ntree=500,importance=TRUE) ## "x" is matrix whose columns are predictors, "y" is a binary resonse vector ## Then I got the ranked predi

[R] R package "kernlab" can not be properly loaded

Hi, I tried to load the package "kernlab" under R-v11 and R-v10, however it gave error message: Error in library.dynam(lib, package, package.lib) : shared library 'kernlab' not found In addition: Warning message: package 'kernlab' was built under R version 2.12.0 Error: package/namespace loa

[R] can not produce graph using "splom"

Hi, I wrote a function basically to first read an input data file, then open an pdf file and draw graph using "splom". When testing, I ran the function line by line, it can produce nice plot, but with like 50 warnings. However, whenever I ran this function as a whole, it can not produce any plo

[R] change axis labels and text size in "splom"

Hi everyone: I'm using "splom" to draw scatterplot matrix. I'm wondering how can I change the axis labels to c(1,10,100,1000,...) instead of c(1,2,3,...), and also how can I change the text size (for labels)? Thanks a lot! xcui __ R-help@r-project.or