LeBauer:Software/Statistics

From OpenWetWare

(Difference between revisions)
Jump to: navigation, search
(Building the PECAn library)
(JAGS)
Line 21: Line 21:
==JAGS==
==JAGS==
 +
=====Installing JAGS=====
 +
1. download the file
 +
<pre>
 +
wget http://www-ice.iarc.fr/~martyn/software/jags/JAGS-<version>.tar.gz
 +
</pre>
 +
2. unpack the gzipped file
 +
<pre> tar xvfz  JAGS-<version>.tar.gz</pre>
 +
3. change directory to folder with unpacked JAGS files
 +
<pre> cd JAGS-<version></pre>
 +
4. run configuration file to check system and set system dependent variables (--prefix=/usr/ specifies where JAGS will install, /usr/ is default on Red Hat, which runs the ebi-forecast server but /usr/local/ is the default installation folder on Ubuntu 
 +
<pre> ./configure --prefix=/usr</pre>
 +
5. build program
 +
<pre> make</pre>
 +
6.  install (requires root access)
 +
<pre> sudo make install</pre>
 +
7. remove files no longer needed after installation
 +
<pre> make clean
 +
cd ..
 +
rm -rf JAGS-<version> </pre>
 +
 +
 +
=====Specific Code for JAGS version 2.1.0=====
 +
<pre>
 +
wget http://www-ice.iarc.fr/~martyn/software/jags/JAGS-2.1.0.tar.gz
 +
tar xvfz  JAGS-2.1.0.tar.gz
 +
cd JAGS-2.1.0/
 +
./configure --prefix=/usr
 +
make
 +
sudo make install
 +
make clean
 +
cd ..
 +
rm -rf JAGS-2.1.0
 +
</pre>
=====Installing Jags packages (rjags and R2jags) in R=====
=====Installing Jags packages (rjags and R2jags) in R=====
<pre>
<pre>

Revision as of 18:00, 7 September 2010


Contents

R

public R packages

  • ggplot2 simplifies graphical data visualization and presentation
  • coda tools for analysis of mcmc chains
  • randtoolbox has function for generating pseudo-random samples; the halton() function produces a pseudo-random over-dispersed matrix that I use to sampling from the distributions of ~20 parameters used in ensemble runs of the ED2 ecosystem model.
  • adimpro for image analysis
##example R code for extracting jpg to N x M x 3 matrix
##much slower than Matlab imread()
a <- read.image('file.jpg')
b <- extract.image(a)
red   <- b[,,1]
green <- b[,,2]
blue  <- b[,,3]

Custom R packages

See the Writing R Extensions guides for more information.


JAGS

Installing JAGS

1. download the file

 wget http://www-ice.iarc.fr/~martyn/software/jags/JAGS-<version>.tar.gz 

2. unpack the gzipped file

 tar xvfz  JAGS-<version>.tar.gz

3. change directory to folder with unpacked JAGS files

 cd JAGS-<version>

4. run configuration file to check system and set system dependent variables (--prefix=/usr/ specifies where JAGS will install, /usr/ is default on Red Hat, which runs the ebi-forecast server but /usr/local/ is the default installation folder on Ubuntu

 ./configure --prefix=/usr

5. build program

 make

6. install (requires root access)

 sudo make install

7. remove files no longer needed after installation

 make clean
cd ..
rm -rf JAGS-<version> 


Specific Code for JAGS version 2.1.0
wget http://www-ice.iarc.fr/~martyn/software/jags/JAGS-2.1.0.tar.gz 
tar xvfz  JAGS-2.1.0.tar.gz
cd JAGS-2.1.0/
./configure --prefix=/usr
make
sudo make install
make clean
cd ..
rm -rf JAGS-2.1.0
Installing Jags packages (rjags and R2jags) in R
wget http://cran.r-project.org/src/contrib/R2jags_0.01-26.tar.gz
sudo R CMD INSTALL R2jags_0.01-26.tar.gz
wget http://cran.r-project.org/src/contrib/rjags_1.0.3-11.tar.gz
sudo R CMD INSTALL rjags_1.0.3-11.tar.gz

Kepler

Kelper project homepage

download Kepler

wget https://code.kepler-project.org/code/kepler/releases/installers/2.0/Kepler-2.0-linux.jar
Kepler-2.0-linux.jar

follow the instructions to install Kepler, if installation does not begin, double-click on the file

after installation, to use mysql with Open Database Connection actor, may need to download jar file for jdbc java connection

wget http://www.mysql.com/get/Downloads/Connector-J/mysql-connector-java-5.1.13.tar.gz/from/ftp://mirror.anl.gov/pub/mysql/
tar -zxf mysql-connector-java-5.1.13.tar.gz
cp mysql-connector-java-5.1.13/mysql-connector-java-5.1.13-bin.jar Kepler-2.0/
Personal tools