MachineShop (version 1.1.0)

MLControl: Resampling Controls

Description

The base MLControl constructor initializes a set of control parameters that are common to all resampling methods.

BootControl constructs an MLControl object for simple bootstrap resampling in which models are fit with bootstrap resampled training sets and used to predict the full data set.

CVControl constructs an MLControl object for repeated K-fold cross-validation. In this procedure, the full data set is repeatedly partitioned into K-folds. Within a partitioning, prediction is performed on each of the K folds with models fit on all remaining folds.

OOBControl constructs an MLControl object for out-of-bootstrap resampling in which models are fit with bootstrap resampled training sets and used to predict the unsampled cases.

SplitControl constructs an MLControl object for splitting data into a seperate trianing and test set.

TrainControl constructs an MLControl object for training and performance evaluation to be performed on the same training set.

Usage

MLControl(surv_times = numeric(), seed = NULL, ...)
BootControl(samples = 25, ...)
CVControl(folds = 10, repeats = 1, ...)
OOBControl(samples = 25, ...)
SplitControl(prop = 2/3, ...)
TrainControl(...)

Arguments

surv_times

numeric vector of follow-up times at which to predict survival events.

seed

integer to set the seed at the start of resampling. This is set to a random integer by default (NULL).

...

arguments to be passed to MLControl.

samples

number of bootstrap samples.

folds

number of cross-validation folds (K).

repeats

number of repeats of the K-fold partitioning.

prop

proportion of cases to include in the training set (0 < prop < 1).

Value

MLControl class object.

See Also

resample

Examples

# NOT RUN {
## 100 bootstrap samples
BootControl(samples = 100)

## 5 repeats of 10-fold cross-validation
CVControl(folds = 10, repeats = 5)

## 100 out-of-bootstrap samples
OOBControl(samples = 100)

## Split sample of 2/3 training and 1/3 testing
SplitControl(prop = 2/3)

## Same training and test set
TrainControl()

# }