NeuralSens: Sensitivity Analysis of Neural Networks

Analysis functions to quantify inputs importance in neural network models. Functions are available for calculating and plotting the inputs importance and obtaining the activation function of each neuron layer and its derivatives. The importance of a given input is defined as the distribution of the derivatives of the output with respect to that input in each training data point <doi:10.18637/jss.v102.i07>.

Version: 1.0.1
Imports: ggplot2, gridExtra, NeuralNetTools, reshape2, caret, fastDummies, stringr, Hmisc, ggforce, scales, ggnewscale, magrittr
Suggests: h2o, RSNNS, nnet, neuralnet, plotly, e1071
Published: 2022-06-21
Author: José Portela González [aut], Antonio Muñoz San Roque [aut], Jaime Pizarroso Gonzalo [aut, ctb, cre]
Maintainer: Jaime Pizarroso Gonzalo <jpizarroso at comillas.edu>
BugReports: https://github.com/JaiPizGon/NeuralSens/issues
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
URL: https://github.com/JaiPizGon/NeuralSens
NeedsCompilation: no
Citation: NeuralSens citation info
CRAN checks: NeuralSens results

Documentation:

Reference manual: NeuralSens.pdf

Downloads:

Package source: NeuralSens_1.0.1.tar.gz
Windows binaries: r-devel: NeuralSens_1.0.1.zip, r-release: NeuralSens_1.0.1.zip, r-oldrel: NeuralSens_1.0.1.zip
macOS binaries: r-release (arm64): NeuralSens_1.0.1.tgz, r-oldrel (arm64): NeuralSens_1.0.1.tgz, r-release (x86_64): NeuralSens_1.0.1.tgz, r-oldrel (x86_64): NeuralSens_1.0.1.tgz
Old sources: NeuralSens archive

Linking:

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