R is a language and environment for statistical computing and graphics. It provides a wide variety of statistical (linear and nonlinear modeling, classical statistical tests, time-series analysis, classification, clustering, ...) and graphical techniques.
altix: not available on altix
inQuiry: not available at inQuiry
pcluster:Running Program | Documentation | Installation | System
Running Program:
Example of usage:
To run R in a batch queue:
* create a script (called runR.sh in this example) with the following lines:
Script filename (runR.sh):
#!/bin/csh
cd working_directory
time R CMD BATCH R_program_name
* submit this script runR.sh to a serial queue (e.g., using up to 24h of wallclock time):
pcluster> ugsub T1-i1-t1-24h runR.sh
For more information on how to run batch jobs on pcluster, please refer to Running jobs on pcluster
Documentation: http://www.r-project.org/
Instal lation:
* Installed in /usr/local/bin/
* Compiled without Lapack support.
* Version 1.9.1 is installed in /usr/local/R-1.9.1/
* Add-on packages fields, mcmc, and sna installed in /usr/local/R/lib/R/library/
Syst em: Unix (AIX)
rcluster,IOB:Running Program | Documentation | Installation | System
Running Program:
Example of usage:
To run R in a batch queue:
* create a script (called runR.sh in this example) with the following lines:
Script filename (runR.sh):
To use R version 2.4.0:
#!/bin/csh
cd working_directory
time R CMD BATCH R_program_name
To use R version 2.5.1:
#!/bin/csh
cd working_directory
time /usr/local/R-2.5.1/bin/R CMD BATCH R_program_name
* submit this script runR.sh to a serial queue (e.g., using up to 24h of wallclock time):
rcluster> chmod u+x runR.sh
rcluster> bsub -q r1-24h -o out.%J -e err.%J ./runR.sh
For more information on how to run batch jobs on rcluster, please refer to Running jobs on rcluster
Documentation: http://www.r-project.org/
Instal lation:
* Version 2.4.0 was compiled with PGI 6.1 compilers, and it is installed in /usr/local/R-2.4.0/bin/, a directory that is on users default path (that is, if you invoke R with no explicit path, this is the version that you will access).
* Add-on packages for 2.4.0 installed in /usr/local/R-2.4.0/lib64/R/library/ (please check this directory to see which packages are installed). Some examples:
o dr, elasticnet, far, fields, lars, lasso2, mcmc, reldist, sna, ....
o Bioconductor package, which consists of
affy, affydata, affyio, affyPLM, annaffy, annotate, Biobase, Biostrings, DynDoc, edd, gcrma, genefilter, geneplotter, globaltest, GO, golubEsets, hgu95av2, KEGG, limma, makecdfenv, marray, matchprobes, multtest, pamr, RColorBrewer, ROC, siggenes, sma, statmod, tkWidgets, vsn, widgetTools, xtable
NOTE: ROC was compiled with a maximum length of truth set to 100000. This value can be increased upon request.
* Version 2.5.1 was compiled with PGI 6.2 compilers, and it is installed in /usr/local/R-2.5.1/bin/, a directory that is not on users default path. To use this version, start it with the full path, that is, type /usr/local/R-2.5.1/bin/R
* Add-on packages for 2.5.1 are installed in /usr/local/R-2.5.1/lib64/R/library/ (please check this directory to see which packages are installed). Note that the Bioconductor package installed with version 2.5.1 does not have the restriction in ROC that was implemented in the installation with version 2.4.0.
* Both versions were compiled with PGI compilers, with Blas and Lapack support.
Answered by
Sharad Singh
, an ibibo Master,
at
3:15 PM on September 27, 2008