dodiscover is a Python library for causal discovery, or structure learning. This is generally considered a "first step" in the causal inference pipeline, if one does not have access to a hypothesized causal graph for their situation.
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, dodiscover requires:
* Python (>=3.8)
* numpy
* scipy
* networkx
* pywhy-graphs
If you already have a working installation of numpy, scipy and networkx, the easiest way to install dodiscover is using pip:
# doesn't work until we make an official release :p
pip install -U dodiscover
# If you are a developer and would like to install the developer dependencies
pip install dodiscover[doc,style,test]
# If you would like full functionality, which installs all of the above
pip install dodiscover[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 corresponding algorithms. However, I believe it is technically feasible based on the design of how we use networkx.