Linux instructions

Instructions below are for Circuitscape version 3.5; for version 4.0 see user guide


We only recommend Linux for users who must run large grids (> 2-6 million cells) and have access to computers with large amounts of RAM.  Typically, grids smaller than 2 million cells or so run fine on standard PC and Mac computers with 2GB of RAM.  


Before installing Linux just to run a large grid, we recommend you try coarsening your grid first.  Typically this yields results that are very similar to those obtained at finer resolutions (see McRae et al. 2008).


Unfortunately, installing Circuitscape on Linux is not an exact science yet, and your success will depend on your system.  Below are three suggested methods.


All methods require you to install the Python packages Circuitscape depends on first. You'll need root access (su) or sudo access.



Brad’s method for Ubuntu 9.04 (We strongly recommend Ubuntu, which I installed on my mac using VMware.  This works with Sun’s free VirtualBox as well, but VirtualBox limits RAM to 3584 MB, at least on the Mac.  Using VMware, I’m able to access 8 GB).


Open a terminal window, and run these commands:

sudo apt-get update

sudo apt-get upgrade

sudo apt-get install python-setuptools

sudo apt-get install python-dev

sudo apt-get install python-wxgtk2.8 python-scipy python-pythoncard

sudo easy_install -U pyamg


Then download and unzip the Circuitscape sources.  From the terminal window in the Circuitscape directory, try running the GUI:


python cs_gui.py


If you have problems with the GUI (which hasn’t been extensively tested on Linux), or if you are running remotely, you can always run Circuitscape from the command line:


python cs_run.py file.ini


where file.ini is your configuration file specifying your input files, etc (see user manual).  You can create this file using the GUI on a Windows or Mac computer, or modify one of the configuration files in the examples directory.



Viral’s method:


easy_install -U numpy

easy_install -U scipy

easy_install -U pyamg


Then download the Circuitscape sources, and use the command line interface. You do not need to be root to run Circuitscape.


python cs_run.py file.ini


where file.ini is your configuration file specifying your input files, etc (see user manual).  You can create this file using the GUI on a Windows or Mac computer, or modify one of the configuration files in the examples directory.


If you want to run the GUI as well, then as root install the following. Note that as of right now, easy_install seems to be having trouble installing wxpython and PythonCard. So you will need to download sources and compile them yourself if you want the GUI on Linux. Needless to say, we recommend using the command line on Linux.


easy_install -U wxpython

easy_install -U PythonCard


And then, launch the GUI from Circuitscape sources with:


python cs_gui.py



Here’s how Milton got Circuitscape running on Ubuntu 8.10 on an AMD64 platform:


To installl numpy, scipy, and pyamg, follow:

http://code.google.com/p/pyamg/wiki/CompilingOnUbuntu


To install wxPython download the source from SourceForge and install using the codes below:


dpkg --install [filename].deb

dpkg -i [filename].deb


To install PythonCard follow:

http://ubuntuforums.org/showthread.php?t=535314


Then download the Circuitscape sources and launch the GUI with:


python cs_gui.py


Good luck!  Please let us know if one of these methods works for you.