Calculate which features the forest split on at each depth.

split_frequencies(forest, max.depth = 4)

Arguments

forest	The trained forest.
max.depth	Maximum depth of splits to consider.

Value

A matrix of split depth by feature index, where each value is the number of times the feature was split on at that depth.

Examples

# \donttest{
# Train a quantile forest.
n <- 250
p <- 10
X <- matrix(rnorm(n * p), n, p)
Y <- X[, 1] * rnorm(n)
q.forest <- quantile_forest(X, Y, quantiles = c(0.1, 0.5, 0.9))

# Calculate the split frequencies for this forest.
split_frequencies(q.forest)
#>      [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10]
#> [1,]  952  122   96  158  100  155  126   90   96   105
#> [2,] 1074  304  274  313  239  344  307  276  279   255
#> [3,]  949  492  495  440  473  468  506  429  432   460
#> [4,]  858  619  564  595  520  541  548  511  499   611
# }