rbartpackages.dbarts.dbartsData

class rbartpackages.dbarts.dbartsData(formula, data=None, *, test=None, subset=None, weights=None, offset=None, offset_test=None)[source]

Bundle the data of a dbarts sampler.

Python interface to R’s dbarts::dbartsData. A string formula argument is converted to an R formula; the backwards-compatible matrix form (formula/data as the x_train/y_train pair) passes through unchanged. Wraps an R S4 object with no components exposed; pass it to dbarts.setData or in place of the formula argument of the fitting interfaces. Arguments left to None are omitted from the R call.

Parameters:

  • formula (str | Float64[ndarray, 'n p'] | DataFrame) – A model formula (as a string), or, in matrix mode, the x_train predictor matrix.

  • data (Float64[ndarray, 'n'] | DataFrame | None, default: None) – The data frame the formula refers to, or, in matrix mode, the y_train response vector.

  • test (Float64[ndarray, 'm p'] | DataFrame | None, default: None) – Test predictors, with the same columns as the training data.

  • subset (Integer[ndarray, 'k'] | None, default: None) – Subset of observations to keep.

  • weights (Float64[ndarray, 'n'] | None, default: None) – Per-observation weights; the model becomes y | x ~ N(f(x), sigma^2 / w).

  • offset (Float64[ndarray, 'n'] | float | None, default: None) – Offset added to f(x); useful for binary responses, where P(Y = 1 | x) = Phi(f(x) + offset).

  • offset_test (Float64[ndarray, 'm'] | float | None, default: None) – The offset for the test data; defaults to offset when applicable.

R documentation

title
-----

Discrete Bayesian Additive Regression Trees Sampler Data

name
----

dbartsData

alias
-----

dbartsData

description
-----------

   Convenience function to create a data object for use with a  dbarts  sampler.


usage
-----


 dbartsData(
     formula, data, test, subset, weights,
     offset, offset.test = offset)


arguments
---------


     formula, data, test, subset, weights, offset, offset.test
      As in  dbarts . Retains backwards compatibility with  bart , so that  formula / data  can be a  formula / data.frame  pair, or a pair of  x.train / y.train  matrices/vector.



value
-----


   An object of class  dbartData .


seealso
-------


    dbarts