Materials, code, datasets, etc., for the BOSS project
The contents of this repository are for this paper:
Files related to Bryan Andrews, Joseph Ramsey, Ruben Sanchez-Romero, Jazmin Camchong, and Erich Kummerfeld (2023), Fast Scalable and Accurate Discovery of DAGs Using the Best Order Score Search and Grow-Shrink Trees, Neurips.
Contact information:
Bryan Andrews: [email protected], Joseph Ramsey: [email protected], Ruben Sanchez-Romero: [email protected], Jazmin Camchong: [email protected], Erich Kummerfeld: [email protected]
https://github.com/cmu-phil/causal-cmd-python-wrapper
All of the simulation datasets and true graphical models used in our paper are in the directory 'files' in this repository. This includes all datasets in tabular form plus all true graphical models for the dataset to compare results to. We include tabular data (instead of covariance matrices) because we realize some tools require it.