Trains a Project-Pursuit oblique decision tree.

This function trains a Project-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. 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

pptr(
  formula = NULL,
  data = NULL,
  x = NULL,
  y = NULL,
  lambda = 0,
  seed = NULL,
  pp = NULL,
  sr = 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.

lambda

A regularization 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). Cannot be used together with pp.

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.

pp

A projection pursuit strategy object created by pp_pda. Cannot be used together with lambda.

sr

A split rule strategy object created by sr_mean_of_means (default).

Value

A pptr model trained on x and y.

See also

predict.pptr, formula.pptr, summary.pptr, print.pptr, save_json, load_json, pp_tree for parsnip integration, pp_pda, sr_mean_of_means, vignette("introduction") for a tutorial

Examples

# Example 1: formula interface with the `iris` dataset
pptr(Type ~ ., data = iris)
#> 
#> Project-Pursuit Oblique Decision Tree:
#> If ([ 0.01 0.04 -0.04 -0.01 ] * x) < 0.06660754:
#>  If ([ 0.04 0.07 -0.09 -0.15 ] * x) < -0.2075133:
#>    Predict: virginica 
#>  Else:
#>    Predict: versicolor 
#> Else:
#>   Predict: setosa 
#> 

# Example 2: formula interface with the `iris` dataset with regularization
pptr(Type ~ ., data = iris, lambda = 0.5)
#> 
#> Project-Pursuit Oblique Decision Tree:
#> If ([ 0 -0.04 0.03 0.03 ] * x) < 0.01580044:
#>   Predict: setosa 
#> Else:
#>  If ([ 0 0.03 -0.06 -0.15 ] * x) < -0.4503323:
#>    Predict: virginica 
#>  Else:
#>    Predict: versicolor 
#> 

# Example 3: matrix interface with the `iris` dataset
pptr(x = iris[, 1:4], y = iris[, 5])
#> If ([ 0.01 0.04 -0.04 -0.01 ] * x) < 0.06660754:
#>  If ([ 0.04 0.07 -0.09 -0.15 ] * x) < -0.2075133:
#>    Predict: virginica 
#>  Else:
#>    Predict: versicolor 
#> Else:
#>   Predict: setosa 
#> 

# Example 4: matrix interface with the `iris` dataset with regularization
pptr(x = iris[, 1:4], y = iris[, 5], lambda = 0.5)
#> If ([ 0 -0.04 0.03 0.03 ] * x) < 0.01580044:
#>   Predict: setosa 
#> Else:
#>  If ([ 0 0.03 -0.06 -0.15 ] * x) < -0.4503323:
#>    Predict: virginica 
#>  Else:
#>    Predict: versicolor 
#> 

# Example 5: formula interface with the `crabs` dataset
pptr(Type ~ ., data = crabs)
#> 
#> Project-Pursuit Oblique Decision Tree:
#> If ([ 0 0 0 0 0 0 0 ] * x) < 0.004743028:
#>   Predict: B 
#> Else:
#>   Predict: O 
#> 

# Example 6: formula interface with the `crabs` dataset with regularization
pptr(Type ~ ., data = crabs, lambda = 0.5)
#> 
#> Project-Pursuit Oblique Decision Tree:
#> If ([ 0 0 0.01 0 0 0 0.01 ] * x) < 0.324472:
#>   Predict: B 
#> Else:
#>   Predict: O 
#>