A visualisation library for event data analysed using the pyISC anomaly detection framework.
##Prerequisite
Install pyisc (including all prerequisites) : https://github.com/STREAM3/pyisc/
Notice, pyisc must be placed in the same folder as visisc so that visisc can find the source code. The pyisc folder must be named "pyisc".
Install Mayavi and wxPython (version 2.8.1 is preferred, otherwise if it does not exists as for Mac OS X choose 3.0)
Windows:
>> conda install --channel https://conda.anaconda.org/krisvanneste wxpython==2.8.12
Other:
>> conda install wxpython
All:
>> conda install mayavi
To undo the downgrading of the numpy libraries in previous step:
>> conda install numpy
##Installation
>> git clone https://github.com/STREAM3/visisc
>> cd visisc
>> python setup.py install
##Run tutorial
>> cd docs
>> jupyter notebook visISC_tutorial.ipynb
If not opened automatically, click on visISC_tutorial.ipynb
in the web page that was opened in a web browser.