innsight

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).

Last 30 days

This package has been downloaded 682 times in the last 30 days. Not bad! The download count is somewhere between 'small-town buzz' and 'moderate academic conference'. The following heatmap shows the distribution of downloads per day. Yesterday, it was downloaded 71 times.

The following line graph shows the downloads per day. You can hover over the graph to see the exact number of downloads per day.

Last 365 days

This package has been downloaded 9,437 times in the last 365 days. A solid achievement! Enough downloads to get noticed at department meetings. The day with the most downloads was Sep 11, 2024 with 84 downloads.

The following line graph shows the downloads per day. You can hover over the graph to see the exact number of downloads per day.