The direction NAs are sent are indicated with the arrow fill. An empty arrow indicates that NAs are sent that way. If trained without missing values, both arrows are filled.

# S3 method for grf_tree
plot(x, include.na.path = NULL, ...)

Arguments

x

The tree to plot

include.na.path

A boolean toggling whether to include the path of missing values or not. It defaults to whether the forest was trained with NAs.

...

Additional arguments (currently ignored).

Examples

if (FALSE) { # Plot a tree in the forest (requires the `DiagrammeR` package). n <- 500 p <- 10 X <- matrix(rnorm(n * p), n, p) W <- rbinom(n, 1, 0.5) Y <- pmax(X[, 1], 0) * W + X[, 2] + pmin(X[, 3], 0) + rnorm(n) c.forest <- causal_forest(X, Y, W) plot(tree <- get_tree(c.forest, 1)) # Compute the leaf nodes the first five samples falls into. leaf.nodes <- get_leaf_node(tree, X[1:5, ]) # Saving a plot in .svg can be done with the `DiagrammeRsvg` package. install.packages("DiagrammeRsvg") tree.plot = plot(tree) cat(DiagrammeRsvg::export_svg(tree.plot), file = 'plot.svg') }