Multinomial log-linear learner via neural networks

Multinomial log-linear models via neural networks. Calls nnet::multinom() from package nnet.

Dictionary

This mlr3::Learner can be instantiated via the dictionary mlr3::mlr_learners or with the associated sugar function mlr3::lrn():

mlr_learners$get("classif.multinom")
lrn("classif.multinom")

Meta Information

• Task type: “classif”

• Predict Types: “response”, “prob”

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

• Required Packages: mlr3, mlr3learners, nnet

Parameters

Id	Type	Default	Levels	Range
Hess	logical	FALSE	TRUE, FALSE	-
abstol	numeric	1e-04		\((-\infty, \infty)\)
censored	logical	FALSE	TRUE, FALSE	-
decay	numeric	0		\((-\infty, \infty)\)
entropy	logical	FALSE	TRUE, FALSE	-
mask	untyped	-		-
maxit	integer	100		\([1, \infty)\)
MaxNWts	integer	1000		\([1, \infty)\)
model	logical	FALSE	TRUE, FALSE	-
linout	logical	FALSE	TRUE, FALSE	-
rang	numeric	0.7		\((-\infty, \infty)\)
reltol	numeric	1e-08		\((-\infty, \infty)\)
size	integer	-		\([1, \infty)\)
skip	logical	FALSE	TRUE, FALSE	-
softmax	logical	FALSE	TRUE, FALSE	-
summ	character	0	0, 1, 2, 3	-
trace	logical	TRUE	TRUE, FALSE	-
Wts	untyped	-		-

See also

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

• Package mlr3extralearners for more learners.

• 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).

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

• Extension packages for additional task types:

  • mlr3proba for probabilistic supervised regression and survival analysis.

  • mlr3cluster for unsupervised clustering.

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

Other Learner: mlr_learners_classif.cv_glmnet, mlr_learners_classif.glmnet, mlr_learners_classif.kknn, mlr_learners_classif.lda, mlr_learners_classif.log_reg, mlr_learners_classif.naive_bayes, mlr_learners_classif.nnet, mlr_learners_classif.qda, mlr_learners_classif.ranger, mlr_learners_classif.svm, mlr_learners_classif.xgboost, mlr_learners_regr.cv_glmnet, mlr_learners_regr.glmnet, mlr_learners_regr.kknn, mlr_learners_regr.km, mlr_learners_regr.lm, mlr_learners_regr.nnet, mlr_learners_regr.ranger, mlr_learners_regr.svm, mlr_learners_regr.xgboost

Super classes

mlr3::Learner -> mlr3::LearnerClassif -> LearnerClassifMultinom

Methods

Public methods

• LearnerClassifMultinom$new()

• LearnerClassifMultinom$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()

---

Method new()

Creates a new instance of this R6 class.

Usage

LearnerClassifMultinom$new()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

LearnerClassifMultinom$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# Define the Learner and set parameter values
learner = lrn("classif.multinom")
print(learner)
#> 
#> ── <LearnerClassifMultinom> (classif.multinom): Multinomial Log-Linear Model ───
#> • Model: -
#> • Parameters: list()
#> • Packages: mlr3, mlr3learners, and nnet
#> • Predict Types: [response] and prob
#> • Feature Types: logical, integer, numeric, and factor
#> • Encapsulation: none (fallback: -)
#> • Properties: multiclass, twoclass, and weights
#> • Other settings: use_weights = 'use', predict_raw = 'FALSE'

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

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

# Train the learner on the training ids
learner$train(task, row_ids = ids$train)
#> # weights:  62 (61 variable)
#> initial  value 96.347458 
#> iter  10 value 28.060727
#> iter  20 value 1.899489
#> iter  30 value 0.003516
#> final  value 0.000058 
#> converged

# Print the model
print(learner$model)
#> Call:
#> nnet::multinom(formula = Class ~ ., data = task$data())
#> 
#> Coefficients:
#> (Intercept)          V1         V10         V11         V12         V13 
#>   606.48499  -292.62216  -142.11791  -588.13233   -88.06081  -314.66330 
#>         V14         V15         V16         V17         V18         V19 
#>  -134.59359   380.41387   -93.35268   203.69452    83.21463    30.36906 
#>          V2         V20         V21         V22         V23         V24 
#>  -208.73362  -213.45852  -335.38329   104.26954   -54.17037  -522.82205 
#>         V25         V26         V27         V28         V29          V3 
#>   700.21783  -201.29489  -279.93755   276.00581  -122.70963   -55.82210 
#>         V30         V31         V32         V33         V34         V35 
#>    39.08240    57.62629   -86.84656  -265.38407   345.24797  -149.54434 
#>         V36         V37         V38         V39          V4         V40 
#>   165.80103   677.99313  -323.04585  -410.21814  -506.70203   468.79101 
#>         V41         V42         V43         V44         V45         V46 
#>   -74.72936  -118.46404  -278.18165  -225.21560  -453.56885  -472.40222 
#>         V47         V48         V49          V5         V50         V51 
#>    29.90197  -336.13828  -343.42369  -451.68864   137.03675  -117.66505 
#>         V52         V53         V54         V55         V56         V57 
#>   -92.13127   -31.72623   -40.72676    44.68332    46.71611    53.03701 
#>         V58         V59          V6         V60          V7          V8 
#>   -34.35565   -51.57852  -168.28882    10.24126   163.11532   640.56455 
#>          V9 
#>   -57.59101 
#> 
#> Residual Deviance: 0.0001159859 
#> AIC: 122.0001 

# Importance method
if ("importance" %in% learner$properties) print(learner$importance())

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

# Score the predictions
predictions$score()
#> classif.ce 
#>  0.2463768