Installation
The program has not been released for public consumption yet, please contact the authors if interested

YARM consists of a series of C programs, PERL scripts, and various data files. The C code requires that the LAPACK (Linear Algebra PACKage) be accessible, and naturarally, the PERL scripts require that the PERL software package be installed. The C code must be compiled and all the files must be copied into a commonly accessible directory on your computer.

There are three steps to installing YARM:

  • 1) Install PERL v5 or higher
  • 2) Install the LAPACK and BLAS libraries
  • 3) Compile and install the YARM files
    • Below is a more detailed description of each of these steps.
    1) Install PERL v5 or higher

    Many of the PERL scripts used by YARM require PERL v5 to be installed. You can assertain which version of PERL you are currently running by typing perl -v.

    If you have PERL v5 or higher, goto installation step 2. Hopefully, most modern UNIX computers will have PERL v5 preinstalled. If not, here are a number of sites to help you with this step:

    www.perl.org - Source code... links to all things PERL

    2) Install LAPACK and BLAS

    You will need to install the "LAPACK" (Linear Algebra PACKage) library in order to compile portions of the YARM C code. The source is available from the kind folks at www.netlib.org, visit the lapack home page or just download the whole damn thing.

    Some precompiled libraries may be available for certain OS types:

    IRIX: Newer installations of IRIX come preinstalled with the LAPACK and BLAS routines, usually in the library file /usr/lib/libcomplib.sgimath.a so make sure you have this installed. I believe this comes automatically with IRIX 5.3 and greater.

    LINUX: If you are lucky enough to be running Linux, you can download the precompiled LAPACK libraries from Jakob Schiotz. His version of the LAPACK libraries are very well optimized (using g77), and I would recommend them.

    3) Compile and install the YARM files

    Grab the YARM distribion tarball, which will be named something like yarm_vX.YY.tar.gz where X is the major revision number and YY is the minor revision number. You will want the newest package available.

    Now, decompress (gunzip yarm_vX.YY.tar.gz) and untar (tar xvf gunzip yarm_vX.YY.tar) the file. It will create its own directory called yarm_vX.YY. Enter the distribution directory and C code source directory (cd yarm_vX.YY/src). Edit the Makefile. You may have to change the values of these variables, depending on the type of compiler your computer has, where the LAPACK and BLAS libraries reside, and where you want to install the program. Shown below are standard values used for a Linux installation. 

    CC       = gcc
OPTS     = -O3 -ffast-math
CFLAGS   = 
DEST     = /usr/local
MATHLIBS = -lblas -lapack -lf2c

    Now type make all followed by make install. If all went without a hitch you should have some executables, PERL scripts, and data files in the directory specified by DEST/yarm_vX.YY.

    Finished, you shoul dnow have access to the YARM subroutines.

    Please email one of the authors and let us know how the installation went. Any sggestions on making it easier to install would be appreciated.