Classification Fast Generalized Additive Model Learner

Fast generalized additive models for large datasets. Calls mgcv::bam() from package mgcv with a binomial family. For smaller datasets, use classif.gam instead.

Formula

A gam formula specific to the task at hand is required for the formula parameter (see example and ?mgcv::formula.gam). Beware, if no formula is provided, a fallback formula is used that will make the model behave like a glm (this behavior is required for the unit tests). Only features specified in the formula will be used, superseding columns with col_roles "feature" in the task.

Offset

If a Task contains a column with the offset role, it is automatically incorporated during training via the offset argument in mgcv::bam(). No offset is applied during prediction for this learner.

Custom mlr3 parameters

• family is always binomial, so it is not exposed as a hyperparameter.

• control is passed to mgcv::gam.control() as a list. Unlike classif.gam, the individual control options are not exposed as hyperparameters, because nthreads is both a control option and an argument of mgcv::bam().

• discrete is a parameter of both mgcv::bam() and mgcv::predict.bam(), with different defaults. Only the training parameter is exposed, and prediction uses the default of mgcv::predict.bam().

• Only the training parameter nthreads is tagged with "threads", so mlr3::set_threads() does not change the prediction parameter n.threads.

Dictionary

This Learner can be instantiated via lrn():

lrn("classif.bam")

Meta Information

• Task type: “classif”

• Predict Types: “response”, “prob”

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

• Required Packages: mlr3, mlr3extralearners, mgcv

Parameters

Id	Type	Default	Levels	Range
formula	untyped	-		-
method	character	fREML	fREML, GCV.Cp, GACV.Cp, REML, P-REML, ML, P-ML	-
control	untyped	list()		-
select	logical	FALSE	TRUE, FALSE	-
scale	numeric	0		\((-\infty, \infty)\)
gamma	numeric	1		\([1, \infty)\)
knots	untyped	NULL		-
sp	untyped	NULL		-
min.sp	untyped	NULL		-
paraPen	untyped	NULL		-
chunk.size	integer	10000		\([1, \infty)\)
rho	numeric	0		\((-\infty, \infty)\)
AR.start	untyped	NULL		-
discrete	logical	FALSE	TRUE, FALSE	-
nthreads	integer	1		\([1, \infty)\)
use.chol	logical	FALSE	TRUE, FALSE	-
samfrac	numeric	1		\([0, 1]\)
coef	untyped	NULL		-
drop.unused.levels	logical	TRUE	TRUE, FALSE	-
G	untyped	NULL		-
drop.intercept	logical	NULL	TRUE, FALSE	-
in.out	untyped	NULL		-
nei	untyped	NULL		-
cluster	untyped	NULL		-
gc.level	integer	0		\([0, \infty)\)
terms	untyped	NULL		-
exclude	untyped	NULL		-
block.size	integer	50000		\((-\infty, \infty)\)
n.threads	integer	1		\([1, \infty)\)

References

Hastie, J T, Tibshirani, J R (2017). Generalized additive models. Routledge.

Wood, Simon (2012). “mgcv: Mixed GAM Computation Vehicle with GCV/AIC/REML smoothness estimation.”

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

Tino-Rg

Super classes

mlr3::Learner -> mlr3::LearnerClassif -> LearnerClassifBam

Methods

Public methods

• LearnerClassifBam$new()

• LearnerClassifBam$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$train()
• mlr3::LearnerClassif$predict_newdata_fast()

---

LearnerClassifBam$new()

Creates a new instance of this R6 class.

Usage

LearnerClassifBam$new()

---

LearnerClassifBam$clone()

The objects of this class are cloneable with this method.

Usage

LearnerClassifBam$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# simple example
t = tsk("spam")$filter(1:1000)
l = lrn("classif.bam")
l$param_set$set_values(
  formula = type ~ s(george, k = 3) + s(charDollar, k = 3) + s(edu)
)
l$train(t)
l$model
#> 
#> Family: binomial 
#> Link function: logit 
#> 
#> Formula:
#> type ~ s(george, k = 3) + s(charDollar, k = 3) + s(edu)
#> 
#> Estimated degrees of freedom:
#> 0.368 0.368 0.368  total = 1.47 
#> 
#> fREML score: -40.99272