Regression Generalized Linear Model Regression Learner

Calls stats::glm from package stats.

Details

For logistic regression please use mlr_learners_classif.log_reg.

Dictionary

This Learner can be instantiated via the dictionary mlr_learners or with the associated sugar function lrn():

mlr_learners$get("regr.glm")
lrn("regr.glm")

Meta Information

• Task type: “regr”

• Predict Types: “response”, “se”

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

• Required Packages: mlr3extralearners, 'stats'

Parameters

Id	Type	Default	Levels	Range
singular.ok	logical	TRUE	TRUE, FALSE	\((-\infty, \infty)\)
x	logical	FALSE	TRUE, FALSE	\((-\infty, \infty)\)
y	logical	TRUE	TRUE, FALSE	\((-\infty, \infty)\)
model	logical	TRUE	TRUE, FALSE	\((-\infty, \infty)\)
etastart	list	-		\((-\infty, \infty)\)
mustart	list	-		\((-\infty, \infty)\)
start	list	NULL		\((-\infty, \infty)\)
offset	list	-		\((-\infty, \infty)\)
family	character	gaussian	gaussian, poisson, quasipoisson, Gamma, inverse.gaussian	\((-\infty, \infty)\)
na.action	character	-	na.omit, na.pass, na.fail, na.exclude	\((-\infty, \infty)\)
link	character	-	logit, probit, cauchit, cloglog, identity, log, sqrt, 1/mu^2, inverse	\((-\infty, \infty)\)
epsilon	numeric	1e-08		\((-\infty, \infty)\)
maxit	numeric	25		\((-\infty, \infty)\)
trace	logical	FALSE	TRUE, FALSE	\((-\infty, \infty)\)
dispersion	list	NULL		\((-\infty, \infty)\)
type	character	link	response, link, terms	\((-\infty, \infty)\)

Custom mlr3 defaults

• type

  • Actual default: "link"

  • Adjusted default: "response"

  • Reason for change: Response scale more natural for predictions.

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

salauer

Super classes

mlr3::Learner -> mlr3::LearnerRegr -> LearnerRegrGlm

Methods

Public methods

• LearnerRegrGlm$new()

• LearnerRegrGlm$clone()

Inherited methods

• mlr3::Learner$base_learner()

• 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

LearnerRegrGlm$new()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

LearnerRegrGlm$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

if (requireNamespace("stats", quietly = TRUE)) {
  learner = mlr3::lrn("regr.glm")
  print(learner)

  # available parameters:
  learner$param_set$ids()
}
#> <LearnerRegrGlm:regr.glm>
#> * Model: -
#> * Parameters: family=gaussian, type=response
#> * Packages: mlr3, mlr3extralearners, stats
#> * Predict Type: response
#> * Feature types: logical, integer, numeric, character, factor, ordered
#> * Properties: weights
#>  [1] "singular.ok" "x"           "y"           "model"       "etastart"   
#>  [6] "mustart"     "start"       "offset"      "family"      "na.action"  
#> [11] "link"        "epsilon"     "maxit"       "trace"       "dispersion" 
#> [16] "type"