Trevor Irons b8ca9c2cee Set theme jekyll-theme-leap-day | 7 years ago | |
---|---|---|
akvo | 7 years ago | |
INSTALL.txt | 7 years ago | |
LICENSE.txt | 7 years ago | |
README.md | 7 years ago | |
_config.yml | 7 years ago | |
akvologo.svg | 7 years ago | |
bir.sh | 7 years ago | |
pyuic.json | 7 years ago | |
setup.py | 7 years ago |
Akvo provides processing of surface NMR data. It aims to be simple to use yet flexible for accommodating changes to processing flow. Akvo is written primarily in Python 3 with a small amount of R as well. The application is written around a Qt GUI with plotting provided by Matplotlib.
The bleeding-edge code may be accesed using the git client
git clone https://git.lemmasoftware.org/akvo.git
or, using our GitHub mirror
git clone https://github.com/LemmaSoftware/akvo.git
Installation is straightforward. The only prerequisite that is sometimes not properly handled is PyQt5 which needs to be manually installed.
python3 setup.py build
python3 setup.py install
Akvo is developed by several teams including the University of Utah.
Akvo currently has preprocessing capabilities for VistaClara GMR data.
Processing steps are retained and logged in the processed file header, which is written in YAML. This allows data processing to be repeatible, which is a major benefit.
Akvo is written primarily in Python 3, but has components written in R as well. The graphical user unterface is written in PyQt5. An interface to modelling software written in C++ is in development.