SPlit-package: SPlit

In SPlit: Split a Dataset for Training and Testing

SPlit

Description

Split a dataset for training and testing

Details

The package SPlit provides the function SPlit() to optimally split a dataset for training and testing using the method of support points (Mak and Joseph, 2018). Support points is a model-independent method for finding optimal representative points of a distribution. SPlit() attempts to obtain a split in which the distribution of both the training and testing sets resemble the distribution of the dataset. The benefits of SPlit over existing data splitting procedures are detailed in Joseph and Vakayil (2021).

Author(s)

Akhil Vakayil, V. Roshan Joseph, Simon Mak

Maintainer: Akhil Vakayil <[e-mail]>

References

Joseph, V. R., & Vakayil, A. (2021). SPlit: An Optimal Method for Data Splitting. Technometrics, 1-11. doi:10.1080/00401706.2021.1921037.

Mak, S., & Joseph, V. R. (2018). Support points. The Annals of Statistics, 46(6A), 2562-2592.

SPlit documentation built on March 22, 2022, 9:06 a.m.