nntrf: Supervised Data Transformation by Means of Neural Network Hidden Layer

A supervised transformation of datasets is performed. The aim is similar to that of Principal Component Analysis (PCA), that is, to carry out data transformation and dimensionality reduction, but in a supervised way. This is achieved by first training a 3-layer Multi-Layer Perceptron and then using the activations of the hidden layer as a transformation of the input features. In fact, it takes advantage of the change of representation provided by the hidden layer of a neural network. This can be useful as data pre-processing for Machine Learning methods in general, specially for those that do not work well with many irrelevant or redundant features. It uses the nnet package under the hood. Rumelhart, D.E., Hinton, G.E. and Williams, R.J. (1986) "Learning representations by back-propagating errors" <doi:10.1038/323533a0>.

Version: 0.1.3
Depends: R (≥ 3.2.4)
Imports: nnet, NeuralNetTools, FNN, pracma
Suggests: dplyr, knitr, rmarkdown, ggplot2, ggridges, tidyr, forcats, mlr, mlrCPO
Published: 2020-08-06
Author: Ricardo Aler [aut, cre], Jose Valls [aut], Ines Galvan [aut], David Camacho [aut]
Maintainer: Ricardo Aler <ricardo.aler at uc3m.es>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: no
Materials: README NEWS
CRAN checks: nntrf results


Reference manual: nntrf.pdf
Vignettes: nntrf
nntrf for regression
nntrf hyper-parameter tuning
Package source: nntrf_0.1.3.tar.gz
Windows binaries: r-devel: nntrf_0.1.3.zip, r-release: nntrf_0.1.3.zip, r-oldrel: nntrf_0.1.3.zip
macOS binaries: r-release: nntrf_0.1.3.tgz, r-oldrel: nntrf_0.1.3.tgz
Old sources: nntrf archive


Please use the canonical form https://CRAN.R-project.org/package=nntrf to link to this page.