TrainFunction {D2MCS}

Control parameters for train stage.

Description

Abstract class used as template to define customized functions to control the computational nuances of train function.

Methods

Public methods

• TrainFunction$new()

• TrainFunction$create()

• TrainFunction$getResamplingMethod()

• TrainFunction$getNumberFolds()

• TrainFunction$getSavePredictions()

• TrainFunction$getClassProbs()

• TrainFunction$getAllowParallel()

• TrainFunction$getVerboseIter()

• TrainFunction$getTrFunction()

• TrainFunction$getMeasures()

• TrainFunction$getType()

• TrainFunction$getSeed()

• TrainFunction$setSummaryFunction()

• TrainFunction$setClassProbs()

• TrainFunction$clone()

Method new()

Function used to initialize the object parameters during execution time.

Usage

TrainFunction$new(
  method,
  number,
  savePredictions,
  classProbs,
  allowParallel,
  verboseIter,
  seed
)

Arguments

method

The resampling method: "boot", "boot632", "optimism_boot", "boot_all", "cv", "repeatedcv", "LOOCV", "LGOCV" (for repeated training/test splits), "none" (only fits one model to the entire training set), "oob" (only for random forest, bagged trees, bagged earth, bagged flexible discriminant analysis, or conditional tree forest models), timeslice, "adaptive_cv", "adaptive_boot" or "adaptive_LGOCV"

number

Either the number of folds or number of resampling iterations

savePredictions

An indicator of how much of the hold-out predictions for each resample should be saved. Values can be either "all", "final", or "none". A logical value can also be used that convert to "all" (for true) or "none" (for false). "final" saves the predictions for the optimal tuning parameters.

classProbs

A logical value. Should class probabilities be computed for classification models (along with predicted values) in each resample?

allowParallel

A logical value. If a parallel backend is loaded and available, should the function use it?

verboseIter

A logical for printing a training log.

seed

An optional integer that will be used to set the seed during model training stage.

Method create()

Creates a trainControl requires for the training stage.

Usage

TrainFunction$create(summaryFunction, search.method = "grid", class.probs)

Arguments

summaryFunction

An object inherited from SummaryFunction class.

search.method

Either "grid" or "random", describing how the tuning parameter grid is determined.

class.probs

A logical indicating if class probabilities should be computed for classification models (along with predicted values) in each resample.

Method getResamplingMethod()

Returns the resampling method used during training staged.

Usage

TrainFunction$getResamplingMethod()

Returns

A character vector or length 1 or NULL if not defined.

Method getNumberFolds()

Returns the number or folds or number of iterations used during training.

Usage

TrainFunction$getNumberFolds()

Returns

An integer vector or length 1 or NULL if not defined.

Method getSavePredictions()

Indicates if the predictions for each resample should be saved.

Usage

TrainFunction$getSavePredictions()

Returns

A logical value or NULL if not defined.

Method getClassProbs()

Indicates if class probabilities should be computed for classification models in each resample.

Usage

TrainFunction$getClassProbs()

Returns

A logical value.

Method getAllowParallel()

Determines if model training is performed in parallel.

Usage

TrainFunction$getAllowParallel()

Returns

A logical value. TRUE indicates parallelization is enabled and FALSE otherwise.

Method getVerboseIter()

Determines if training log should be printed.

Usage

TrainFunction$getVerboseIter()

Returns

A logical value. TRUE indicates training log is enabled and FALSE otherwise.

Method getTrFunction()

Function used to return the trainControl object.

Usage

TrainFunction$getTrFunction()

Returns

A trainControl object.

Method getMeasures()

Returns the measures used to optimize model hyperparameters.

Usage

TrainFunction$getMeasures()

Returns

A character vector.

Method getType()

Obtains the type of classification problem ("Bi-class" or "Multi-class").

Usage

TrainFunction$getType()

Returns

A character vector with length 1. Either "Bi-class" or "Multi-class".

Method getSeed()

Indicates seed used during model training stage.

Usage

TrainFunction$getSeed()

Returns

An integer value or NULL if not defined.

Method setSummaryFunction()

Function used to change the SummaryFunction used in the training stage.

Usage

TrainFunction$setSummaryFunction(summaryFunction)

Arguments

summaryFunction

An object inherited from SummaryFunction class.

Method setClassProbs()

The function allows changing the class computation capabilities.

Usage

TrainFunction$setClassProbs(class.probs)

Arguments

class.probs

A logical indicating if class probabilities should be computed for classification models (along with predicted values) in each resample

Method clone()

The objects of this class are cloneable with this method.

Usage

TrainFunction$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

See Also

TwoClass