Generalized Random Forest Regression Learner

Generalized random forest for regression. Calls grf::regression_forest() from grf.

Custom mlr3 parameters

• sample.weights is not exposed as a hyperparameter. Observation weights are taken from the task instead, see the weights property.

• estimate.variance is not exposed as a hyperparameter. It is enabled automatically when the predict_type is "se", which requires ci.group.size to be at least 2.

Dictionary

This Learner can be instantiated via lrn():

lrn("regr.grf")

Meta Information

• Task type: “regr”

• Predict Types: “response”, “se”

• Feature Types: “integer”, “numeric”

• Required Packages: mlr3, mlr3extralearners, grf

Parameters

Id	Type	Default	Levels	Range
num.trees	integer	2000		\([1, \infty)\)
clusters	untyped	NULL		-
equalize.cluster.weights	logical	FALSE	TRUE, FALSE	-
sample.fraction	numeric	0.5		\([0, 1]\)
mtry	integer	-		\([1, \infty)\)
min.node.size	integer	5		\([1, \infty)\)
honesty	logical	TRUE	TRUE, FALSE	-
honesty.fraction	numeric	0.5		\([0, 1]\)
honesty.prune.leaves	logical	TRUE	TRUE, FALSE	-
alpha	numeric	0.05		\([0, 0.25]\)
imbalance.penalty	numeric	0		\([0, \infty)\)
ci.group.size	integer	2		\([1, \infty)\)
tune.parameters	untyped	"none"		-
tune.num.trees	integer	50		\([1, \infty)\)
tune.num.reps	integer	100		\([1, \infty)\)
tune.num.draws	integer	1000		\([1, \infty)\)
compute.oob.predictions	logical	TRUE	TRUE, FALSE	-
seed	integer	-		\((-\infty, \infty)\)
linear.correction.variables	untyped	NULL		-
ll.lambda	numeric	NULL		\([0, \infty)\)
ll.weight.penalty	logical	FALSE	TRUE, FALSE	-
num.threads	integer	-		\([1, \infty)\)

References

Athey, Susan, Tibshirani, Julie, Wager, Stefan (2019). “Generalized random forests.” The Annals of Statistics, 47(2), 1148–1178. doi:10.1214/18-AOS1709 .

See also

• Dictionary of Learners: mlr3::mlr_learners.

• as.data.table(mlr_learners) for a table of available Learners in the running session (depending on the loaded packages).

• Chapter in the mlr3book: https://mlr3book.mlr-org.com/chapters/chapter2/data_and_basic_modeling.html#sec-learners

• mlr3learners for a selection of recommended learners.

• mlr3cluster for unsupervised clustering learners.

• mlr3pipelines to combine learners with pre- and postprocessing steps.

• mlr3tuning for tuning of hyperparameters, mlr3tuningspaces for established default tuning spaces.

Author

jesuisantoine

Super classes

mlr3::Learner -> mlr3::LearnerRegr -> LearnerRegrGRF

Methods

Public methods

• LearnerRegrGRF$new()

• LearnerRegrGRF$clone()

Inherited methods

• mlr3::Learner$base_learner()
• mlr3::Learner$configure()
• mlr3::Learner$encapsulate()
• mlr3::Learner$format()
• mlr3::Learner$help()
• mlr3::Learner$predict()
• mlr3::Learner$predict_newdata()
• mlr3::Learner$print()
• mlr3::Learner$reset()
• mlr3::Learner$selected_features()
• mlr3::Learner$train()
• mlr3::LearnerRegr$predict_newdata_fast()

---

LearnerRegrGRF$new()

Creates a new instance of this R6 class.

Usage

LearnerRegrGRF$new()

---

LearnerRegrGRF$clone()

The objects of this class are cloneable with this method.

Usage

LearnerRegrGRF$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# Define the Learner
learner = lrn("regr.grf")
print(learner)
#> 
#> ── <LearnerRegrGRF> (regr.grf): Generalized Random Forest ──────────────────────
#> • Model: -
#> • Parameters: list()
#> • Packages: mlr3, mlr3extralearners, and grf
#> • Predict Types: [response] and se
#> • Feature Types: integer and numeric
#> • Encapsulation: none (fallback: -)
#> • Properties: missings and weights
#> • Other settings: use_weights = 'use', predict_raw = 'FALSE'

# Define a Task
task = tsk("mtcars")

# Create train and test set
ids = partition(task)

# Train the learner on the training ids
learner$train(task, row_ids = ids$train)

print(learner$model)
#> GRF forest object of type regression_forest 
#> Number of trees: 2000 
#> Number of training samples: 21 
#> Variable importance: 
#>  1  2  3  4  5  6  7  8  9 10 
#>  0  0  0  0  0  0  0  0  0  0 


# Make predictions for the test rows
predictions = learner$predict(task, row_ids = ids$test)

# Score the predictions
predictions$score()
#> regr.mse 
#> 67.41058