Given two feature vectors and a response vector, produces surface maps
of partial dependence function. Uses R package rgl
partialPlot2var(x1, x2, y, gridlength=NULL, x1_grid=NULL, x2_grid=NULL,
x1lab='v1', x2lab='v2', ylab=NA, range.color=NULL, col.palette=c('blue',
'yellow'), plot_quantile_scale = TRUE, plot.colorbar=TRUE, ...)
numeric vectors containing the first, second and the response variable
If either x1_grid or x2_grid is missin, a scalar indicating length of grids against which to make the plot
grids of points between which the surface map will be constant
Labels of x, y and z-axis respectively
range of values which should be colored. input in the form (min, max). If not provided, the program automatically sets it to the range of the partial dependence function calculated on grid points
Color palette, typically a vector of length 2 indicating the range of colors
If TRUE, plots against quantiles of x1 and x2 instead of their raw values
produce a separate colorbar for the partial dependence plot?
additional arguments to pass to rgl::persp3d
A RGL plot of type persp3d() of the package rgl