python-weka-wrapper allows you to use Weka from within Python.
The library uses the javabridge library for starting up, communicating with and shutting down the Java Virtual Machine in which the Weka processes get executed.
python-weka-wrapper provides a thin wrapper around the basic (non-GUI) functionality of Weka. You can automatically add all your Weka packages to the classpath. Additional jars can be added as well.
Project homepage: https://github.com/fracpete/python-weka-wrapper
MLOSS project: https://mloss.org/software/view/548/
The library has the following requirements:
- javabridge (>=1.0.1, required)
- pygraphviz (optional)
- PIL (optional)
- matplotlib (optional)
The Python libraries you can either install using pip install <name> or use pre-built packages available for your platform.
For Ubuntu this could look as follows:
$ sudo apt-get install python-numpy python-imaging python-matplotlib python-pygraphviz $ sudo pip install javabridge
A build environment is required to build libraries, like javabridge, from source. For Ubuntu that would be the build-essential meta-package (general Linux instructions), Xcode for Mac OSX and MinGW for Windows.