Regression MultilayerPerceptron Learner

Regressor that uses backpropagation to learn a multi-layer perceptron. Calls RWeka::make_Weka_classifier() from RWeka.

Custom mlr3 parameters

• output_debug_info:

  • original id: output-debug-info

• do_not_check_capabilities:

  • original id: do-not-check-capabilities

• num_decimal_places:

  • original id: num-decimal-places

• batch_size:

  • original id: batch-size

• Reason for change: This learner contains changed ids of the following control arguments since their ids contain irregular pattern

• G removed:

  • GUI will be opened

• Reason for change: The parameter is removed because we don't want to launch GUI.

Dictionary

This Learner can be instantiated via lrn():

lrn("regr.multilayer_perceptron")

Meta Information

• Task type: “regr”

• Predict Types: “response”

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

• Required Packages: mlr3, RWeka

Parameters

Id	Type	Default	Levels	Range
subset	untyped	-		-
na.action	untyped	-		-
L	numeric	0.3		\([0, 1]\)
M	numeric	0.2		\([0, 1]\)
N	integer	500		\([1, \infty)\)
V	numeric	0		\([0, 100]\)
S	integer	0		\([0, \infty)\)
E	integer	20		\([1, \infty)\)
A	logical	FALSE	TRUE, FALSE	-
B	logical	FALSE	TRUE, FALSE	-
H	untyped	"a"		-
C	logical	FALSE	TRUE, FALSE	-
I	logical	FALSE	TRUE, FALSE	-
R	logical	FALSE	TRUE, FALSE	-
D	logical	FALSE	TRUE, FALSE	-
output_debug_info	logical	FALSE	TRUE, FALSE	-
do_not_check_capabilities	logical	FALSE	TRUE, FALSE	-
num_decimal_places	integer	2		\([1, \infty)\)
batch_size	integer	100		\([1, \infty)\)
options	untyped	NULL		-

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

damirpolat

Super classes

mlr3::Learner -> mlr3::LearnerRegr -> LearnerRegrMultilayerPerceptron

Methods

Public methods

• LearnerRegrMultilayerPerceptron$new()

• LearnerRegrMultilayerPerceptron$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

LearnerRegrMultilayerPerceptron$new()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

LearnerRegrMultilayerPerceptron$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# Define the Learner
learner = mlr3::lrn("regr.multilayer_perceptron")
print(learner)
#> <LearnerRegrMultilayerPerceptron:regr.multilayer_perceptron>: MultilayerPerceptron
#> * Model: -
#> * Parameters: list()
#> * Packages: mlr3, RWeka
#> * Predict Types:  [response]
#> * Feature Types: logical, integer, numeric, factor, ordered
#> * Properties: missings

# 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)

print(learner$model)
#> Linear Node 0
#>     Inputs    Weights
#>     Threshold    -0.015969456983696718
#>     Node 1    -2.1692573407088287
#>     Node 2    -0.7077242121733289
#>     Node 3    -0.056176684028666796
#>     Node 4    2.313727999857748
#>     Node 5    -0.21335592133631895
#> Sigmoid Node 1
#>     Inputs    Weights
#>     Threshold    -0.7072445353781918
#>     Attrib am    -0.26861614997115696
#>     Attrib carb    1.4100171814201048
#>     Attrib cyl    -1.4667773623774722
#>     Attrib disp    0.30450622306476643
#>     Attrib drat    1.8974012731963632
#>     Attrib gear    0.5132688608174519
#>     Attrib hp    0.31504501028227233
#>     Attrib qsec    0.1935797559055674
#>     Attrib vs    0.4949141508934376
#>     Attrib wt    1.393935809370832
#> Sigmoid Node 2
#>     Inputs    Weights
#>     Threshold    -0.8705441818471156
#>     Attrib am    -1.052634951740239
#>     Attrib carb    0.37205223850545815
#>     Attrib cyl    0.46171164922471153
#>     Attrib disp    -1.1835072014639807
#>     Attrib drat    0.6455188662127403
#>     Attrib gear    0.39804849268592507
#>     Attrib hp    -0.7515700098993566
#>     Attrib qsec    0.9366048570938219
#>     Attrib vs    -0.8449987839804968
#>     Attrib wt    1.2813095314089125
#> Sigmoid Node 3
#>     Inputs    Weights
#>     Threshold    -1.1213607302328183
#>     Attrib am    -3.6863237825343045E-4
#>     Attrib carb    0.3122576252781201
#>     Attrib cyl    0.03665997343280893
#>     Attrib disp    0.29014712719618335
#>     Attrib drat    0.3486960480062039
#>     Attrib gear    0.42589743686606885
#>     Attrib hp    0.18510099768406033
#>     Attrib qsec    0.18320151815703709
#>     Attrib vs    0.05182113293971315
#>     Attrib wt    0.1484389963888983
#> Sigmoid Node 4
#>     Inputs    Weights
#>     Threshold    -1.6030663892764494
#>     Attrib am    -0.8959326661963962
#>     Attrib carb    1.218057815178789
#>     Attrib cyl    0.9181654399975391
#>     Attrib disp    -1.3009661488139976
#>     Attrib drat    1.9220835984816766
#>     Attrib gear    1.624804291498969
#>     Attrib hp    -2.099043341286964
#>     Attrib qsec    1.9073677697245148
#>     Attrib vs    0.3220234155699061
#>     Attrib wt    -1.9038503269752183
#> Sigmoid Node 5
#>     Inputs    Weights
#>     Threshold    -1.0088414329428468
#>     Attrib am    0.10080355003748682
#>     Attrib carb    0.352546195281757
#>     Attrib cyl    -0.01593998751734075
#>     Attrib disp    0.23533827032751004
#>     Attrib drat    0.5951664578281444
#>     Attrib gear    0.41035972092086925
#>     Attrib hp    0.14466892990570607
#>     Attrib qsec    0.2441574857622681
#>     Attrib vs    -4.844292321345325E-4
#>     Attrib wt    0.29363241842889115
#> Class 
#>     Input
#>     Node 0
#> 


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

# Score the predictions
predictions$score()
#> regr.mse 
#> 46.28459