Plotting Partial Dependence Functions

Plots partial dependence functions (i.e., marginal effects) using ggplot2 graphics.

# S3 method for partial
autoplot(object, center = FALSE, plot.pdp = TRUE,
  pdp.color = "red", pdp.size = 1, pdp.linetype = 1, rug = FALSE,
  smooth = FALSE, smooth.method = "auto", smooth.formula = y ~ x,
  smooth.span = 0.75, smooth.method.args = list(), contour = FALSE,
  contour.color = "white", palette = c("viridis", "magma", "inferno",
  "plasma", "cividis"), train = NULL, xlab = NULL, ylab = NULL,
  main = NULL, legend.title = "yhat", ...)

# S3 method for ice
autoplot(object, center = FALSE, plot.pdp = TRUE,
  pdp.color = "red", pdp.size = 1, pdp.linetype = 1, rug = FALSE,
  train = NULL, xlab = NULL, ylab = NULL, main = NULL, ...)

# S3 method for cice
autoplot(object, plot.pdp = TRUE, pdp.color = "red",
  pdp.size = 1, pdp.linetype = 1, rug = FALSE, train = NULL,
  xlab = NULL, ylab = NULL, main = NULL, ...)

Arguments

object	An object that inherits from the "partial" class.
center	Logical indicating whether or not to produce centered ICE curves (c-ICE curves). Only useful when object represents a set of ICE curves; see partial for details. Default is FALSE.
plot.pdp	Logical indicating whether or not to plot the partial dependence function on top of the ICE curves. Default is TRUE.
pdp.color	Character string specifying the color to use for the partial dependence function when plot.pdp = TRUE. Default is "red".
pdp.size	Positive number specifying the line width to use for the partial dependence function when plot.pdp = TRUE. Default is 1.
pdp.linetype	Positive number specifying the line type to use for the partial dependence function when plot.pdp = TRUE. Default is 1.
rug	Logical indicating whether or not to include rug marks on the predictor axes. Default is FALSE.
smooth	Logical indicating whether or not to overlay a LOESS smooth. Default is FALSE.
smooth.method	Character string specifying the smoothing method (function) to use (e.g., "auto", "lm", "glm", "gam", "loess", or "rlm"). Default is "auto". See geom_smooth for details.
smooth.formula	Formula to use in smoothing function (e.g., y ~ x, y ~ poly(x, 2), or y ~ log(x)).
smooth.span	Controls the amount of smoothing for the default loess smoother. Smaller numbers produce wigglier lines, larger numbers produce smoother lines. Default is 0.75.
smooth.method.args	List containing additional arguments to be passed on to the modeling function defined by smooth.method.
contour	Logical indicating whether or not to add contour lines to the level plot.
contour.color	Character string specifying the color to use for the contour lines when contour = TRUE. Default is "white".
palette	Character string indicating the colormap option to use. Five options are available: "viridis" (the default), "magma", "inferno", "plasma", and "cividis".
train	Data frame containing the original training data. Only required if rug = TRUE or chull = TRUE.
xlab	Character string specifying the text for the x-axis label.
ylab	Character string specifying the text for the y-axis label.
main	Character string specifying the text for the main title of the plot.
legend.title	Character string specifying the text for the legend title. Default is "yhat".
...	Additional optional arguments to be passed onto geom_line.

Value

A "ggplot" object.

Examples

if (FALSE) {
#
# Regression example (requires randomForest package to run)
#

# Load required packages
library(ggplot2)  # required to use autoplot
library(randomForest)

# Fit a random forest to the Boston housing data
data (boston)  # load the boston housing data
set.seed(101)  # for reproducibility
boston.rf <- randomForest(cmedv ~ ., data = boston)

# Partial dependence of cmedv on lstat
boston.rf %>%
  partial(pred.var = "lstat") %>%
  autoplot(rug = TRUE, train = boston)

# Partial dependence of cmedv on lstat and rm
boston.rf %>%
  partial(pred.var = c("lstat", "rm"), chull = TRUE, progress = "text") %>%
  autoplot(contour = TRUE, legend.title = "rm")

# ICE curves and c-ICE curves
age.ice <- partial(boston.rf, pred.var = "lstat", ice = TRUE)
grid.arrange(
  autoplot(age.ice, alpha = 0.5),                 # ICE curves
  autoplot(age.ice, center = TRUE, alpha = 0.5),  # c-ICE curves
  ncol = 2
)
}