mlrCPO0.3.8 package

Composable Preprocessing Operators and Pipelines for Machine Learning

applyCPO

Apply a CPO to Data

as.list.CPO

Split a Pipeline into Its Constituents

attachCPO

Attach a CPO to a Learner

clearRI

Clear Retrafo and Inverter Attributes

composeCPO

CPO Composition

covrTraceCPOs

Add 'covr' coverage to CPOs

CPO

Composable Preprocessing Operators

cpoApplyFun

Apply a Function Element-Wise

cpoApplyFunRegrTarget

Transform a Regression Target Variable

cpoAsNumeric

Convert All Features to Numerics

cpoCache

Caches the Result of CPO Transformations

cpoCbind

cbind the Result of Multiple CPOs

cpoCollapseFact

Compine Rare Factors

CPOConstructor

Constructor for CPO Objects

cpoDropConstants

Drop Constant or Near-Constant Features

cpoDropMostlyConstants

Drop Constant or Near-Constant Features

cpoDummyEncode

CPO Dummy Encoder

cpoFilterAnova

Filter Features: anova.test

cpoFilterCarscore

Filter Features: carscore

cpoFilterChiSquared

Filter Features: chi.squared

cpoFilterFeatures

Filter Features by Thresholding Filter Values

cpoFilterGainRatio

Filter Features: gain.ratio

cpoFilterInformationGain

Filter Features: information.gain

cpoFilterKruskal

Filter Features: kruskal.test

cpoFilterLinearCorrelation

Filter Features: linear.correlation

cpoFilterMrmr

Filter Features: mrmr

cpoFilterOneR

Filter Features: oneR

cpoFilterPermutationImportance

Filter Features: permutation.importance

cpoFilterRankCorrelation

Filter Features: rank.correlation

cpoFilterRelief

Filter Features: relief

cpoFilterRfCImportance

Filter Features: cforest.importance

cpoFilterRfImportance

Filter Features: randomForest.importance

cpoFilterRfSRCImportance

Filter Features: randomForestSRC.rfsrc

cpoFilterSymmetricalUncertainty

Filter Features: symmetrical.uncertainty

cpoFilterUnivariate

Filter Features: univariate.model.score

cpoFilterVariance

Filter Features: variance

cpoFixFactors

Clean Up Factorial Features

cpoIca

Construct a CPO for ICA Preprocessing

cpoImpactEncodeClassif

Impact Encoding

cpoImpactEncodeRegr

Impact Encoding

cpoImpute

Impute and Re-Impute Data

cpoImputeConstant

Perform Imputation with Constant Value

cpoImputeHist

Perform Imputation with Random Values

cpoImputeLearner

Perform Imputation with an mlr Learner

cpoImputeMax

Perform Imputation with Multiple of Minimum

cpoImputeMean

Perform Imputation with Mean Value

cpoImputeMedian

Perform Imputation with Median Value

cpoImputeMin

Perform Imputation with Multiple of Minimum

cpoImputeMode

Perform Imputation with Mode Value

cpoImputeNormal

Perform Imputation with Normally Distributed Random Values

cpoImputeUniform

Perform Imputation with Uniformly Random Values

CPOLearner

CPO Learner Object

cpoLogTrafoRegr

Log-Transform a Regression Target Variable.

cpoMakeCols

Create Columns from Expressions

cpoMissingIndicators

Convert Data into Factors Indicating Missing Data

cpoModelMatrix

Create a Model Matrix from the Data Given a Formula

cpoOversample

Over- or Undersample Binary Classification Tasks

cpoPca

Construct a CPO for PCA Preprocessing

cpoProbEncode

Probability Encoding

cpoQuantileBinNumerics

Split Numeric Features into Quantile Bins

cpoRegrResiduals

Train a Model on a Task and Return the Residual Task

cpoResponseFromSE

Use the se predict.type for response Prediction

cpoSample

Sample Data from a Task

cpoScale

Construct a CPO for Scaling / Centering

cpoScaleMaxAbs

Max Abs Scaling CPO

cpoScaleRange

Range Scaling CPO

cpoSelect

Drop All Columns Except Certain Selected Ones from Data

cpoSmote

Perform SMOTE Oversampling for Binary Classification

cpoSpatialSign

Scale Rows to Unit Length

cpoTemplate

Dummy Function for Documentation Purposes

CPOTrained

Get the Retransformation or Inversion Function from a Resulting Object

cpoTransformParams

Transform CPO Hyperparameters

cpoWrap

CPO Wrapper

discrete

defined to avoid problems with the static type checker

funct

defined to avoid problems with the static type checker

getCPOAffect

Get the Selection Arguments for Affected CPOs

getCPOClass

Get the CPO Class

getCPOConstructor

Get the CPOConstructor Used to Create a CPO Object

getCPOId

Get the ID of a CPO Object

getCPOName

Get the CPO Object's Name

getCPOOperatingType

Determine the Operating Type of the CPO

getCPOPredictType

Get the CPO predict.type

getCPOProperties

Get the Properties of the Given CPO Object

getCPOTrainedCapability

Get the CPOTrained's Capabilities

getCPOTrainedCPO

Get CPO Used to Train a Retrafo / Inverter

getCPOTrainedState

Get the Internal State of a CPORetrafo Object

getLearnerBare

Get the Learner with the CPOs Removed

getLearnerCPO

Get the CPO Associated with a Learner

grapes-greater-than-greater-than-grapes

CPO Composition / Attachment / Application Operator

identicalCPO

Check Whether Two CPO are Fundamentally the Same

internal-grapes-greater-than-greater-than-grapes

Internally Used %>>% Operators

invert

Invert Target Preprocessing

is.inverter

Check CPOInverter

is.nullcpo

Check for NULLCPO

is.retrafo

Check CPORetrafo

listCPO

List all Built-in CPOs

makeCPO

Create a Custom CPO Constructor

makeCPOCase

Build Data-Dependent CPOs

makeCPOMultiplex

CPO Multiplexer

makeCPOTrainedFromState

Create a CPOTrained with Given Internal State

mlrCPO-package

Composable Preprocessing Operators

NULLCPO

CPO Composition Neutral Element

nullcpoToNull

NULLCPO to NULL

nullToNullcpo

NULL to NULLCPO

pipeCPO

Turn a list of CPOs into a Single Chained One

print.CPOConstructor

Print CPO Objects

pSS

Turn the argument list into a ParamSet

randomForestSRC_filters

Filter randomForestSRC_importance computes the importance of random fo...

setCPOId

Set the ID of a CPO Object

untyped

defined to avoid problems with the static type checker

Toolset that enriches 'mlr' with a diverse set of preprocessing operators. Composable Preprocessing Operators ("CPO"s) are first-class R objects that can be applied to data.frames and 'mlr' "Task"s to modify data, can be attached to 'mlr' "Learner"s to add preprocessing to machine learning algorithms, and can be composed to form preprocessing pipelines.

  • Maintainer: Martin Binder
  • License: BSD_2_clause + file LICENSE
  • Last published: 2025-06-16