pprf: Trains a Random Forest of Projection-Pursuit oblique decision...

In ppforest2: Projection Pursuit Oblique Decision Trees and Random Forests

Trains a Random Forest of Projection-Pursuit oblique decision trees.

Description

This function trains a Random Forest of Projection-Pursuit oblique decision tree using either a formula and data frame interface or a matrix-based interface. When using the formula interface, specify the model formula and the data frame containing the variables. For the matrix-based interface, provide matrices for the features and labels directly. The number of trees is controlled by the size parameter. Each tree is trained on a stratified bootstrap sample drawn from the data. The number of variables to consider at each split is controlled by the n_vars parameter. If lambda = 0, the model is trained using Linear Discriminant Analysis (LDA). If lambda > 0, the model is trained using Penalized Discriminant Analysis (PDA).

Usage

pprf(
  formula = NULL,
  data = NULL,
  x = NULL,
  y = NULL,
  mode = NULL,
  size = 100,
  lambda = 0.5,
  n_vars = NULL,
  p_vars = NULL,
  seed = NULL,
  max_retries = 3L,
  threads = NULL,
  pp = NULL,
  vars = NULL,
  cutpoint = NULL,
  stop = NULL,
  binarize = NULL,
  grouping = NULL,
  leaf = NULL
)

Arguments

formula	A formula of the form y ~ x1 + x2 + ..., where y is a vector of labels and x1, x2, ... are the features.
data	A data frame containing the variables in the formula.
x	A matrix containing the features for each observation.
y	A matrix containing the labels for each observation.
mode	Training mode: either "classification" or "regression". When NULL (default), mode is auto-detected from y's type — factor or character vectors trigger classification, numeric vectors trigger regression. Setting it explicitly is useful for the binary-integer-labels case (mode = "classification" with integer 0/1 labels) and for failing fast on a type mismatch (mode = "regression" with a factor y errors immediately).
size	The number of trees in the forest (default: 100).
lambda	A regularization parameter (default: 0.5). If lambda = 0, the model is trained using Linear Discriminant Analysis (LDA). If lambda > 0, the model is trained using Penalized Discriminant Analysis (PDA). The default uses PDA because pure LDA (lambda = 0) is ill-conditioned when there are more variables than effective observations (see the "Known limitations" section of the README). Cannot be used together with pp.
n_vars	The number of variables to consider at each split (integer). These are chosen uniformly in each split. By default, half of the variables are used (p_vars = 0.5). Cannot be used together with p_vars or dr.
p_vars	The proportion of variables to consider at each split (number between 0 and 1, exclusive). For example, p_vars = 0.5 uses half the features. Cannot be used together with n_vars or dr.
seed	An optional integer seed for reproducibility. If NULL (default), a seed is drawn from R's RNG, so set.seed() controls reproducibility. If an integer is provided, that value is used directly. The same seed is used for training and for computing permuted variable importance.
max_retries	Maximum number of retries for degenerate trees (default: 3). When a bootstrap sample yields a singular covariance matrix, the tree is retrained with a different seed up to this many times.
threads	The number of threads to use. The default is the number of cores available.
pp	A projection pursuit strategy object created by pp_pda. Cannot be used together with lambda.
vars	A variable selection strategy object created by vars_uniform or vars_all. Cannot be used together with n_vars or p_vars.
cutpoint	A split cutpoint strategy object created by cutpoint_mean_of_means (default).
stop	A stopping rule object. Default depends on mode: stop_pure_node() for classification, and stop_any(stop_min_size(5), stop_min_variance(0.01)) for regression.
binarize	A binarization strategy object. Default depends on mode: binarize_largest_gap() for classification, and binarize_disabled() for regression (regression's default grouping always yields a 2-group partition, so no binarization is needed).
grouping	A grouping strategy object. Default depends on mode: grouping_by_label() for classification, and grouping_by_cutpoint() for regression.
leaf	A leaf strategy object. Default depends on mode: leaf_majority_vote() for classification, and leaf_mean_response() for regression.

Details

Mode is taken from the mode argument when explicit, and otherwise auto-detected from 'y' (factor/character → classification, numeric → regression). Pass mode = "classification" to force classification on integer labels (e.g. binary 0/1), or mode = "regression" to assert intent on numeric responses.

OOB error, OOB predictions, permuted variable importance, and weighted variable importance are computed lazily on first access via the accessor functions ('oob_error()', 'oob_predictions()', 'permuted_importance()', 'weighted_importance()'). Training itself is fast because these OOB-based computations are deferred.

Value

A pprf model. Its S3 class vector is c("pprf_classification", "pprf", "ppmodel") or c("pprf_regression", "pprf", "ppmodel") depending on the mode.

See Also

predict.pprf_classification, predict.pprf_regression, formula.ppmodel, oob_error, save_json, load_json, pp_rand_forest for parsnip integration, vignette("introduction") for a tutorial

Examples

# Example 1: formula interface with the `iris` dataset
pprf(Species ~ ., data = iris)

# Example 2: formula interface with the `iris` dataset with regularization
pprf(Species ~ ., data = iris, lambda = 0.5)

# Example 3: matrix interface with the `iris` dataset
pprf(x = iris[, 1:4], y = iris[, 5])

# Example 4: matrix interface with the `iris` dataset with regularization
pprf(x = iris[, 1:4], y = iris[, 5], lambda = 0.5)

# Example 5: formula interface with the `crabs` dataset
pprf(Type ~ ., data = crabs)

# Example 6: formula interface with the `crabs` dataset with regularization
pprf(Type ~ ., data = crabs, lambda = 0.5)

ppforest2 documentation built on July 21, 2026, 9:07 a.m.