Compute partial dependence — partialDependence • semtree

Compute the partial dependence of a predictor, or set of predictors, on a model parameter.

partialDependence(
  x,
  data,
  reference.var,
  support = 20,
  points = NULL,
  mc = NULL,
  FUN = "median",
  ...
)

Arguments

x: An object for which a method exists
data: Optional data.frame that was used to train the model.
reference.var: Character vector, referring to the (independent) reference variable or variables for which partial dependence is calculated. Providing two (or more) variables allows for probing interactions, but note that this is computationally expensive.
support: Integer. Number of grid points for interpolating the reference.var. Alternatively, use points for one or more variables named in reference.var.
points: Named list, with elements corresponding to reference.var . Use this argument to provide specific points for which to obtain marginal dependence values; for example, the mean and +/- 1SD of reference.var.
mc: Integer. If mc is not NULL, the function will sample mc number of rows from data with replacement, to estimate marginal dependency using Monte Carlo integration. This is less computationally expensive.
FUN: Character string with function used to integrate predictions across all elements of x.
...: Extra arguments passed to FUN.

Caspar J. Van Lissa, , Andreas M. Brandmaier