This is my implementation of the CBGM. It was designed for testing and changing the various algorithms and is not (therefore) the fastest user-facing package. The idea was that everything be calculated from scratch each time - although in later development I added a Genealogical Coherence Cache for convenience (and speed). If you change the input data then remember to delete the old cache...
- You need these packages installed (these are the Ubuntu names):
- openmpi-bin
- libopenmpi-dev
- graphviz
- python3-dev
- libgraphviz-dev
- Then you can install using pip straight from github, like this:
- pip install git+https://github.com/edmondac/CBGM
- OR clone the repository and do one of the following:
- python setup.py develop # if you want to edit the code
- python setup.py install # if you just want to use it
The main program is cbgm.
Run cbgm --help
for info, and for info about the subcommands do, for example,
cbgm combanc --help
.
Similarly, see the help for the other programs. They all start 'cbgm_'.
To create the code documentation, run doxygen doxygen.conf
To run the unit tests, run ./test.sh
or just python -m unittest discover CBGM
Also see the "examples" folder, including the README.sh in there, which can be executed with bash to test the common CBGM features.