ggInteract_3D: ggInteract_3D

In JBjouffray/ggBRT: Explore and Visualise the Results of Boosted Regression Trees

Description

Function to draw 3D interaction plot showing predicted values for two predictors from a boosted regression trees obtained with the gbm.step routine in the dismo package. Values for all other variables are set at their mean by default. Possibility to export an interactive version of the plot in html format.

Usage

ggInteract_3D(gbm.object, x = 1, y = 2, col.gradient = "Blues",
  legend = T, main = "", x.label = NULL, y.label = NULL,
  x.range = NULL, y.range = NULL, z.label = "Fitted value",
  z.range = NULL, pred.means = NULL, smooth = "none", ...)

Arguments

gbm.object	a gbm.step object (object of S3 class gbm)
x	the first predictor. Can be either a character name (use "") or a number indicating its index
y	the second predictor. Can be either a character name (use "") or a number indicating its index
col.gradient	a vector of two colors to define the color gradient of the plot or a predefined palette
legend	Logical. If TRUE (default) shows the legend
main	title for the plot
x.label	title for the x-axis
y.label	title for the y-axis
x.range	manual range for the x-axis
y.range	manual range for the y-axis
z.label	title for the z-axis
z.range	manual range for the z-axis
pred.means	allows specification of values for other variables (see gbm.perspec in dismo)
smooth	controls smoothing of the predicted surface (default= "none")

JBjouffray/ggBRT documentation built on Jan. 30, 2021, 8:13 a.m.