pywhy-graphs is a Python graph library that extends MixedEdgeGraph in networkx to implement a light-weight API for causal graphical structures.
Representation of causal inference models in Python are severely lacking. Moreover, sampling from causal models is non-trivial. However, sampling from simulations is a requirement to benchmark different structural learning, causal ID, or other causal related algorithms.
PyWhy-Graphs implements a graphical API layer for ADMG, CPDAG and PAG. For causal DAGs, we recommend using the networkx.DiGraph class and
ensuring acylicity via networkx.is_directed_acyclic_graph function.
See the development version documentation.
Or see stable version documentation
Installation is best done via pip or conda. For developers, they can also install from source using pip. See installation page for full details.
Minimally, pywhy-graphs requires:
* Python (>=3.8)
* numpy
* scipy
* networkx
If you already have a working installation of numpy, scipy and networkx, the easiest way to install pywhy-graphs is using pip:
# doesn't work until we make an official release :p
pip install -U pywhy-graphs
# If you are a developer and would like to install the developer dependencies
pip install pywhy-graphs[doc,style,test]
# If you would like full functionality, which installs all of the above
pip install pywhy-graphs[all]
To install the package from github, clone the repository and then cd into the directory:
pip install -e .
# One can also add the different identifiers, such as '[doc]' to install
# extra dependencies
Currently, selection bias representation is not implemented in the graphs and corresponding algorithms. However, I believe it is technically feasible based on the design of how we use networkx.