What’s new?

Here we list a changelog of dodiscover.

Version 0.1

In Development

Changelog

• Feature Implement continuous integration and repository docs, by Adam Li (#15)

• Feature Implement conditional independence tests under the dodiscover.ci submodule, by Adam Li (#16)

• Feature Implement skeleton learning method, dodiscover.constraint.LearnSkeleton under the dodiscover.constraint submodule, by Adam Li (#20)

• Feature Add ContextBuilder class and make_context factory function for creating context objects. by Chris Trevino (#42)

• Feature Implement confusion matrix method for comparing networkx-like graphs, dodiscover.metrics.confusion_matrix_networks(), by Adam Li (#48)

• Feature Implement classification-based CI test (CCIT), dodiscover.ci.ClassifierCITest under the dodiscover.ci submodule, by Adam Li (#28)

• Feature Implement PC algorithm, dodiscover.constraint.PC for learning causal structure from observational data under the dodiscover.constraint submodule, by Adam Li (#30)

• Feature Implement algorithm to learn skeleton using potentially d-separated sets (PDS), dodiscover.constraint.LearnSemiMarkovianSkeleton for learning causal structure from observational data with latent confounders, by Adam Li (#50)

• Feature Implement FCI algorithm, dodiscover.constraint.FCI for learning causal structure from observational data with latent confounders under the dodiscover.constraint submodule, by Adam Li (#52)

• Feature Implement Structural Hamming Distance metric to compare directed graphs, dodiscover.metrics.structure_hamming_dist(), by Adam Li (#55)

• Fix Update dependency on networkx, which removes a PR branch dependency with pywhy-graphs having the MixedEdgeGraph class that was causing a dependency conflict, by Adam Li (#74)

• Enhancement Add tutorial for PC algorithm with Asia data, by Robert Osazuwa Ness (#67)

• Enhancement Add wrapper for GIN algorithm from causal-learn, by Robert Osazuwa Ness (#94)

• Feature Add conditional k-sample (discrepancy) test, dodiscover.cd.KernelCDTest, by Adam Li (#81)

• Feature Add conditional k-sample (discrepancy) test, dodiscover.cd.BregmanCDTest, by Adam Li (#82)

• Feature Add conditional mutual information test, dodiscover.ci.CMITest, by Adam Li (#83)

• Feature Add classifier conditional mutual information test, dodiscover.ci.ClassifierCMITest, by Adam Li (#85)

• Feature Add rules for orientation under selection bias in FCI algorithm, dodiscover.constraint.FCI, by Jaron Lee (#106)

• Feature Add pre-commit hooks for linting, type-checking, and code formatting, by Jaron Lee (#117)

• Feature Adds parallelization via joblib to the skeleton learners, by Adam Li (#127)

• Feature Add a suite of general categorical data CI tests, by Adam Li (#128)

• Feature Implement CAM, SCORE, DAS, NoGAM algorithms in dodiscover.toporder submodule (#129)

• Feature Add Psi-FCI and I-FCI algorithm for handling soft-interventional data, dodiscover.constraint.PsiFCI by Adam Li (#111)

Code and Documentation Contributors

Thanks to everyone who has contributed to the maintenance and improvement of the project since version inception, including:

• Adam Li

• Chris Trevino

• Robert Osazuwa Ness

• Jaron Lee