iterative.stratification.holdout {mldr.datasets}

Hold-out partitioning of an mldr object

Description

Iterative stratification

Implemented from the algorithm explained in: Konstantinos Sechidis, Grigorios Tsoumakas, and Ioannis Vlahavas. 2011. On the stratification of multi-label data. In Proceedings of the 2011 European conference on Machine learning and knowledge discovery in databases - Volume Part III (ECML PKDD'11), Dimitrios Gunopulos, Thomas Hofmann, Donato Malerba, and Michalis Vazirgiannis (Eds.), Vol. Part III. Springer-Verlag, Berlin, Heidelberg, 145-158.

Usage

iterative.stratification.holdout(mld, p = 60, seed = 10,
  get.indices = FALSE)

Arguments

mld

The mldr object to be partitioned

p

The percentage of instances to be selected for the training partition

seed

The seed to initialize the random number generator. By default is 10. Change it if you want to obtain partitions containing different samples, for instance to use a 2x5 fcv strategy

get.indices

A logical value indicating whether to return lists of indices or lists of "mldr" objects

Value

An mldr.folds object. This is a list containing k elements, one for each fold. Each element is made up of two mldr objects, called train and test

Examples

## Not run: 
library(mldr.datasets)
library(mldr)
parts.emotions <- iterative.stratification.holdout(emotions, p = 70)
summary(parts.emotions$train)
summary(parts.emotions$test)

## End(Not run)