[R] SOLVED: Repeated failures to install "caret" package (of Max Kuhn)

Ronald Wyllys wyllys at ischool.utexas.edu
Sat Apr 4 20:00:56 CEST 2015

These are notes on what I did to be able to install package "caret".

0.    The initial action was to re-install R v. 3.1.3 from this source 
file: r-base-core_3.1.3-1trusty_amd64.deb (available on CRAN Mirrors). 
This is the source file for Ubuntu 14.04, known as "trusty", upon which 
LinuxMint 17.1 is based.  (LinuxMint 17.1 is the operating system that I 
use in my desktop and laptop.)

1.    got llapack and lbas
      (This starting point was suggested in a stackoverflow query: 
"cannot find -llapack", as answered by janneb.  I got to stackoverflow 
after Uwe Liggis raised [in his response to my post on the 
R-helper at r-project.org mailing list] the question of whether my R 
installation was complete.  I am also grateful for Sarah Goslee's and 
John Kane's thought-provoking comments in their responses to my post.)
     This step required my using the Synaptic Package Manager to install
     These four packages together supplied llapack and lbas

2.    used the SPM to install

Note:  Steps 1 and 2 together enabled me to use install.packages to 
install BradleyTerry2.

3.    used SPM to install

4.    used R's install.packages to install quantreg, nloptr

5.    found via the SPM that minqa was already installed as r-cran-minqa

6.    used SPM to install

7.     found via the SPM that stringr was already installed as 

8.    Succeeded in using install.packages to install caret

I wish to express my sincere gratitude to the people named above who 
responded to my initial post concerning my failures to install "caret".

This experience has taught me some useful things about working with R in 
Linux: in particular, I learned that one needs to work patiently and 
perseveringly through the process of figuring out what dependencies are 
impeding the desired installation and how these dependencies can be 
found and installed in R or in Linux.

Ronald Wyllys

More information about the R-help mailing list