Interpretation methods for analyzing the behavior and individual predictions of modern neural networks in a threestep 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>, layerwise 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) <arXiv:1704.02685> and gradientbased methods like 'SmoothGrad' described by Smilkov et al. (2017) <arXiv:1706.03825>, 'Gradient x Input' described by Baehrens et al. (2009) <arXiv:0912.1128> or 'Vanilla Gradient'.
Package details 


Author  Niklas Koenen [aut, cre] (<https://orcid.org/0000000246238271>), Raphael Baudeu [ctb] 
Maintainer  Niklas Koenen <niklas.koenen@gmail.com> 
License  MIT + file LICENSE 
Version  0.3.0 
URL  https://bipshb.github.io/innsight/ https://github.com/bipshb/innsight/ 
Package repository  View on CRAN 
Installation 
Install the latest version of this package by entering the following in R:

Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.