MonteCarlo-class {performanceEstimation}

Class "MonteCarlo"

Description

This class of objects contains the information describing a monte carlo experiment, i.e. its settings.

Objects from the Class

Objects can be created by calls of the form MonteCarlo(...) providing the values for the class slots. These objects contain information on the number of repetitions of the experiments, the data used for training the models on each repetition, the data used for testing these models, the random number generator seed and optionally the concrete data splits to use on each iteration of the Monte Carlo experiment. Note that most of the times you will not supply these data splits as the Monte Carlo routines in this infra-structure will take care of building them. Still, this allows you to replicate some experiment carried out with specific train/test splits.

Slots

nReps:

Object of class numeric indicating the number of repetitions of the Monte Carlo experiment (defaulting to 10).

szTrain:

Object of class numeric. If it is a value between 0 and 1 it is interpreted as a percentage of the available data set, otherwise it is interpreted as the number of cases to use. It defaults to 0.25.

szTest:

Object of class numeric If it is a value between 0 and 1 it is interpreted as a percentage of the available data set, otherwise it is interpreted as the number of cases to use. It defaults to 0.25.

seed:

Object of class numeric with the random number generator seed (defaulting to 1234).

dataSplits:

Object of class list containing the data splits to use on each Monte Carlo repetition. Each element should be a list with two components: test and train, on this order. Each of these is a vector with the row ids to use as test and train sets of each repetition of the Monte Carlo experiment.

Extends

Class EstCommon, directly. Class EstimationMethod, directly.

Methods

show

signature(object = "MonteCarlo"): method used to show the contents of a MonteCarlo object.

Author(s)

Luis Torgo [e-mail]

References

Torgo, L. (2014) An Infra-Structure for Performance Estimation and Experimental Comparison of Predictive Models in R. arXiv:1412.0436 [cs.MS] http://arxiv.org/abs/1412.0436

See Also

LOOCV, CV, Bootstrap, Holdout, EstimationMethod, EstimationTask

Examples

showClass("MonteCarlo")

m1 <- MonteCarlo(nReps=10,szTrain=0.3,szTest=0.2)
m1

## Small example illustrating the format of user supplied data splits
## it assumes that the source data is formed by 10 cases and that each
## model is trainined with 3 cases and tested in the following case.
## This is obviously a unrealistic example in terms of size but
## illustrates the format of the data splits
m2 <- MonteCarlo(dataSplits=list(list(test=sample(1:150,50),train=sample(1:150,50)),
                                   list(test=sample(1:150,50),train=sample(1:150,50)),
                                   list(test=sample(1:150,50),train=sample(1:150,50))
                                  ))
m2