plot.xspliner: Plot method for 'xspliner' model

In krystian8207/xspliner: Assisted Model Building, using Surrogate Black-Box Models to Train Interpretable Spline Based Additive Models

Description

The method provides all plotting methods offered by 'xspliner' package. See plot_variable_transition and plot_model_comparison for more details.

Usage

## S3 method for class 'xspliner'
plot(x, variable_names = NULL, model = NULL,
  plot_response = TRUE, plot_approx = TRUE, data = NULL,
  plot_data = FALSE, plot_deriv = FALSE, n_plots = 6,
  sort_by = NULL, use_coeff = TRUE, compare_with = list(),
  prediction_funs = list(function(object, newdata) predict(object,
  newdata)), ...)

Arguments

x	Object of class 'xspliner'.
variable_names	Names of predictors which transitions should be plotted.
model	Base model that xspliner is based on.
plot_response	If TRUE black box model response is drawn.
plot_approx	If TRUE black box model response approximation is drawn.
data	Training data used for building x model. Required for plot_data option and model comparing.
plot_data	If TRUE raw data is drawn.
plot_deriv	If TRUE derivative of approximation is showed on plot.
n_plots	Threshold for number of plots when plotting all variables.
sort_by	When comparing models determines according to which model should observations be ordered.
use_coeff	If TRUE both PDP function and its approximation is scaled with corresponding surrogate model coefficient.
compare_with	Named list. Other models that should be compared with xspliner and model.
prediction_funs	Prediction functions that should be used in model comparison.
...	Another arguments passed into model specific method.

krystian8207/xspliner documentation built on Oct. 6, 2019, 11:02 p.m.