Regression BART (Bayesian Additive Regression Trees) Learner

Bayesian Additive Regression Trees are similar to gradient boosting algorithms. Calls dbarts::bart() from dbarts.

Dictionary

This Learner can be instantiated via lrn():

lrn("regr.bart")

Meta Information

• Task type: “regr”

• Predict Types: “response”

• Feature Types: “integer”, “numeric”, “factor”, “ordered”

• Required Packages: mlr3, mlr3extralearners, dbarts

Parameters

Id	Type	Default	Levels	Range
ntree	integer	200		\([1, \infty)\)
sigest	untyped	NULL		-
sigdf	integer	3		\([1, \infty)\)
sigquant	numeric	0.9		\([0, 1]\)
k	numeric	2		\([0, \infty)\)
power	numeric	2		\([0, \infty)\)
base	numeric	0.95		\([0, 1]\)
ndpost	integer	1000		\([1, \infty)\)
nskip	integer	100		\([0, \infty)\)
printevery	integer	100		\([0, \infty)\)
keepevery	integer	1		\([1, \infty)\)
keeptrainfits	logical	TRUE	TRUE, FALSE	-
usequants	logical	FALSE	TRUE, FALSE	-
numcut	integer	100		\([1, \infty)\)
printcutoffs	integer	0		\((-\infty, \infty)\)
verbose	logical	FALSE	TRUE, FALSE	-
nthread	integer	1		\((-\infty, \infty)\)
keeptrees	logical	FALSE	TRUE, FALSE	-
keepcall	logical	TRUE	TRUE, FALSE	-
sampleronly	logical	FALSE	TRUE, FALSE	-
seed	integer	NA		\((-\infty, \infty)\)
proposalprobs	untyped	NULL		-
splitprobs	untyped	NULL		-
keepsampler	logical	-	TRUE, FALSE	-

Custom mlr3 parameters

• Parameter: offset

  • The parameter is removed, because only dbarts::bart2 allows an offset during training, and therefore the offset parameter in dbarts:::predict.bart is irrelevant for dbarts::dbart.

• Parameter: nchain, combineChains, combinechains

  • The parameters are removed as parallelization of multiple models is handled by future.

Initial parameter values

• keeptrees is initialized to TRUE because it is required for prediction.

References

Sparapani, Rodney, Spanbauer, Charles, McCulloch, Robert (2021). “Nonparametric machine learning and efficient computation with bayesian additive regression trees: the BART R package.” Journal of Statistical Software, 97, 1–66.

Chipman, A H, George, I E, McCulloch, E R (2010). “BART: Bayesian additive regression trees.” The Annals of Applied Statistics, 4(1), 266–298.

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/basics.html#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

ck37

Super classes

mlr3::Learner -> mlr3::LearnerRegr -> LearnerRegrBart

Methods

Public methods

• LearnerRegrBart$new()

• LearnerRegrBart$clone()

Inherited methods

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

---

Method new()

Creates a new instance of this R6 class.

Usage

LearnerRegrBart$new()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

LearnerRegrBart$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# Define the Learner
learner = mlr3::lrn("regr.bart")
print(learner)
#> <LearnerRegrBart:regr.bart>: Bayesian Additive Regression Trees
#> * Model: -
#> * Parameters: keeptrees=TRUE
#> * Packages: mlr3, mlr3extralearners, dbarts
#> * Predict Types:  [response]
#> * Feature Types: integer, numeric, factor, ordered
#> * Properties: weights

# Define a Task
task = mlr3::tsk("mtcars")

# Create train and test set
ids = mlr3::partition(task)

# Train the learner on the training ids
learner$train(task, row_ids = ids$train)
#> 
#> Running BART with numeric y
#> 
#> number of trees: 200
#> number of chains: 1, number of threads 1
#> tree thinning rate: 1
#> Prior:
#> 	k prior fixed to 2.000000
#> 	degrees of freedom in sigma prior: 3.000000
#> 	quantile in sigma prior: 0.900000
#> 	scale in sigma prior: 0.002230
#> 	power and base for tree prior: 2.000000 0.950000
#> 	use quantiles for rule cut points: false
#> 	proposal probabilities: birth/death 0.50, swap 0.10, change 0.40; birth 0.50
#> data:
#> 	number of training observations: 21
#> 	number of test observations: 0
#> 	number of explanatory variables: 10
#> 	init sigma: 1.829733, curr sigma: 1.829733
#> 
#> Cutoff rules c in x<=c vs x>c
#> Number of cutoffs: (var: number of possible c):
#> (1: 100) (2: 100) (3: 100) (4: 100) (5: 100) 
#> (6: 100) (7: 100) (8: 100) (9: 100) (10: 100) 
#> 
#> Running mcmc loop:
#> iteration: 100 (of 1000)
#> iteration: 200 (of 1000)
#> iteration: 300 (of 1000)
#> iteration: 400 (of 1000)
#> iteration: 500 (of 1000)
#> iteration: 600 (of 1000)
#> iteration: 700 (of 1000)
#> iteration: 800 (of 1000)
#> iteration: 900 (of 1000)
#> iteration: 1000 (of 1000)
#> total seconds in loop: 0.266385
#> 
#> Tree sizes, last iteration:
#> [1] 2 2 3 3 2 3 2 2 2 2 2 2 3 2 2 2 2 2 
#> 2 3 2 2 1 2 2 2 4 3 3 2 2 4 1 2 2 2 2 3 
#> 1 4 2 2 2 2 2 2 2 2 3 1 2 1 2 2 3 2 3 2 
#> 2 2 2 3 2 2 2 2 2 4 2 2 4 2 1 2 3 2 3 3 
#> 2 2 2 2 2 2 2 2 2 4 2 2 2 2 2 2 2 2 2 2 
#> 3 2 2 2 3 2 1 2 2 2 2 2 3 2 3 2 2 3 2 3 
#> 2 3 2 2 2 2 3 4 2 3 1 3 3 2 2 2 4 2 3 2 
#> 2 2 2 2 3 2 2 2 2 2 2 2 2 4 2 2 2 2 3 3 
#> 2 3 2 2 2 3 2 3 2 2 2 2 2 2 3 3 2 2 2 2 
#> 2 2 3 2 2 2 4 4 4 2 3 4 4 3 3 2 2 3 2 1 
#> 2 3 
#> 
#> Variable Usage, last iteration (var:count):
#> (1: 30) (2: 27) (3: 16) (4: 20) (5: 32) 
#> (6: 23) (7: 22) (8: 21) (9: 30) (10: 39) 
#> 
#> DONE BART
#> 

print(learner$model)
#> 
#> Call:
#> dbarts::bart(x.train = data, y.train = outcome, keeptrees = TRUE)
#> 


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

# Score the predictions
predictions$score()
#> regr.mse 
#> 15.86164