A shortcut function for install.packages() and library()

September 11, 2011 6 comments

I enjoy trying difference kind of R packages. Since I have more than 1 computers (1 at home, 1 at office and a laptop)
it is troublesome to check whether I have installed some new packages for each computer. Therefore i wrote a function to load and install packages at once. If the package does not exist, then the it will be downloaded from CRAN and be loaded it.

packages<-function(x, repos="http://cran.r-project.org", ...){
   x <- deparse(substitute(x))
   if (!require(x,character.only=TRUE)){
      install.packages(pkgs=x, repos=repos, ...)
      require(x,character.only=TRUE)
   }
}
packages(Hmisc)

Thanks richierocks for the suggestion of using deparse(substitute(x)) in the code.

richierocks’s

Advertisements
Categories: Custom Function

A quick way to do row repeat and col repeat (rep.row, rep.col)

September 2, 2011 2 comments

Today I worked on a simulation program which require me to create a matrix by repeating the vector n times (both by row and by col).

Even the task is extremely simple and only take 1 line to finish(10sec), I have to think about should the argument in rep be each or times and should the argument in matrix is nrow or ncol. It distracted me from the original task i am working on.

Just now, I wrote a function rep.row and rep.col to do what I really want to do. Next time, i don’t have to worry about how to use the matrix and rep command to repeat an vector to form a matrix!

Code

rep.row<-function(x,n){
   matrix(rep(x,each=n),nrow=n)
}
rep.col<-function(x,n){
   matrix(rep(x,each=n), ncol=n, byrow=TRUE)
}

x is the vector to be repeated and n is the number of replication. Example:

> rep.row(1:3,5)
      [,1] [,2] [,3]
[1,]    1    2    3
[2,]    1    2    3
[3,]    1    2    3
[4,]    1    2    3
[5,]    1    2    3
> rep.col(1:3,5)
      [,1] [,2] [,3] [,4] [,5]
[1,]    1    1    2    2    3
[2,]    1    1    2    3    3
[3,]    1    2    2    3    3

I am sure it should appear in some packages, but it would be faster for me to write it out than find it out!

Categories: Custom Function