CuCubes: MultiDimensional Feature Selection (MDFS)

Functions for MultiDimensional Feature Selection (MDFS): * calculating multidimensional information gains, * finding interesting tuples for chosen variables, * scoring variables, * finding important variables, * plotting selection results. CuCubes is also known as CUDA Cubes and it is a library that allows fast CUDA-accelerated computation of information gains in binary classification problems. This package wraps CuCubes and provides an alternative CPU version as well as helper functions for building MultiDimensional Feature Selectors.

Version: 0.1.0
Depends: R (≥ 3.2.1)
Published: 2016-12-09
Author: Radosław Piliszek [aut, cre], Andrzej Sułecki [aut], Paweł Tabaszewski [aut], Krzysztof Mnich [aut], Witold Rudnicki [cph]
Maintainer: Radosław Piliszek <r.piliszek at>
License: GPL-3
NeedsCompilation: yes
CRAN checks: CuCubes results


Reference manual: CuCubes.pdf
Package source: CuCubes_0.1.0.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X El Capitan binaries: r-release: CuCubes_0.1.0.tgz
OS X Mavericks binaries: r-oldrel: CuCubes_0.1.0.tgz


Please use the canonical form to link to this page.