Get the Insights of Your Neural Network R package innsight version 0.3.2 Interpretation methods for analyzing the behavior and individual predictions of modern neural networks in a three-step procedure: Converting the model, running the interpretation method, and visualizing the results. Implemented methods are, e.g., 'Connection Weights' described by Olden et al. (2004) <doi:10.1016/j.ecolmodel.2004.03.013>, layer-wise relevance propagation ('LRP') described by Bach et al. (2015) <doi:10.1371/journal.pone.0130140>, deep learning important features ('DeepLIFT') described by Shrikumar et al. (2017) <doi:10.48550/arXiv.1704.02685> and gradient-based methods like 'SmoothGrad' described by Smilkov et al. (2017) <doi:10.48550/arXiv.1706.03825>, 'Gradient x Input' or 'Vanilla Gradient'. Details can be found in the accompanying scientific paper: Koenen & Wright (2024, Journal of Statistical Software, <doi:10.18637/jss.v111.i08>). Software Depends: R (>= 3.5.0) Imports: checkmate, cli, ggplot2, methods, R6, torch Suggests: covr, fastshap, GGally, grid, gridExtra, gtable, keras, knitr, lime, luz, neuralnet, palmerpenguins, plotly, rmarkdown, ranger, spelling, tensorflow, testthat (>= 3.0.0) Niklas Koenen <niklas.koenen@gmail.com> Comprehensive R Archive Network (CRAN) Niklas Koenen [aut, cre] (ORCID: <https://orcid.org/0000-0002-4623-8271>), Raphael Baudeu [ctb] MIT + file LICENSE (https://CRAN.R-project.org/package=innsight/LICENSE) 2025-03-30 application/tgz https://CRAN.R-project.org/package=innsight doi:10.32614/CRAN.package.innsight en-US