Reference

API reference for bartz. Each module page lists its public objects in summary tables organized by topic; follow a link for the dedicated page of an object.

High-level interface

bartz.Bart(x_train, y_train, *[, ...])	Nonparametric regression with Bayesian Additive Regression Trees (BART).
bartz.SparseConfig([theta, a, b, rho, ...])	Configuration of a sparsity-inducing variable selection prior.
bartz.PredictKind(*values)	Kind of output of Bart.predict.
bartz.DataFrame(*args, **kwargs)	DataFrame duck-type for Bart.
bartz.Series(*args, **kwargs)	Series duck-type for Bart.

OutcomeType(*values)

Likelihood types for each outcome component in the regression.

R BART3-compatible interface

bartz.BART

Implement classes mc_gbart and gbart that mimic the R BART3 package.

stochtree-compatible interface

bartz.stochtree

stochtree-compatible interface to bartz.

MCMC and trees

bartz.mcmcstep	Functions that implement the BART posterior MCMC initialization and update step.
bartz.mcmcloop	Functions that implement the full BART posterior MCMC loop.
bartz.grove	Functions to create, manipulate, and check binary decision trees.
bartz.prepcovars	Functions and classes to preprocess data.

Debugging and testing

bartz.debug	Debugging utilities.
bartz.testing	Testing utilities to generate synthetic datasets.