Installing OpenPNM is done by running the following on your command line:
pip install openpnm
This will install OpenPNM into your Python environment. To use OpenPNM, open a Python console and type:
>>> import OpenPNM
To upgrade your OpenPNM to a newer version, use
pip install --upgrade openpnm.
It is also possible to download the source code directly from Github and work from that. This is not recommended unless you are planning to do ‘development’ work on the framework. The pip install approach places the source code in the Python directory and out of harms way.
Installing the Python-SciPy Stack Environment¶
The above installation instructions assume that you have the Python SciPy Stack (Python including including packages like NumPy, SciPy, matplotlib) installed on your system, and the
pip installer. If you do not, then refer to the sections below for your system. OpenPNM is designed to run with Python 3.
The simplest way to get Python and all the necessary Python packages for scientific computing in Windows is to download the WinPython package. This package also comes with Spyder, which provides an integrated development environment (IDE) that is very similar to Matlab, with an editor, command console, variable explorer and so on combined into the same window.
Once WinPython is installed, navigate to the directory where you chose to install it and open spyder.exe. From the ‘tools’ method, select ‘open command prompt’. This will open a special version of the Window command prompt that is aware of the Python installation. Here you will use
pip install openpnm to have OpenPNM installed on your machine.
As an alternative, you can also use the Anaconda distribution by Continuum Analytics for Windows (cf. below for more details).
The Anaconda distribution by Continuum Analytics is probably the best option for Mac users. Download the installer and follow the install instructions as stated. This will provide you with Python and all the usual scientific packages, as well as Spyder which the OpenPNM developers highly recommend.
Once Anaconda is installed, you can start the Anaconda launcher by double clicking
Launcher.app in your
~/anaconda directory, by which you can start a Spyder session, or also launch Spyder directly from the command line by running
spyder. Anaconda comes with its own package management
conda, which you can run from the command line to install specific packages or to setup virtual environments. OpenPNM is not registed with the
conda repository so you must use
pip install openpnm. Using
conda you can also update your packages, either updating all by running
conda update anaconda, which updates all of your packages, or update individual packages, e.g. updating the Python package itself by running
conda upate python.
Anaconda offers the option of running virtual environments next to each other. This allows to easily switch between different versions of Python packages, i.e. having one Python 2.7 enviroment and one Python 3.3 environment. This is possible by running
conda create -n py3k python=3.4 numpy=1.8.1 scipy=0.14 anaconda from a command line, where you can explicitly specify which package versions should be installed, and now created a new environment called
py3k. In order to activate this virtual environment, run
source activate py3k from a command line, and the run
spyder, which will give you a Spyder session using these specific packages. To run OpenPNM in this environment, you need to first activate the environment with
source activate py3k and then run
pip install openpnm, so that OpenPNM is installed in this specific environment.
If you are using Linux, we also recommend the Anaconda distribution by Continuum Analytics. Download the installer and follow the install instructions as stated. This will provide you with Python and all the usual scientific packages, as well as Spyder which the OpenPNM developers highly recommend. The process of using it is basically identical to Mac environment, so you can look at the instructions for Apple users stated above. The only difference is that the
Launcher.app is adapted to Linux specifics, but also called
Launcher. Your Anaconda distribution is also installed in
~/anaconda and you can use the
conda commands as stated above to set up environments or update packages.