alexanderbrenning/wiml: Interpreting machine-learning models in transformed space

Main effects plots (such as partial dependence and ALE plots) can be confusing and even misleading when dealing with large numbers of highly correlated features. Example applications include land cover classification using multitemporal satellite remote-sensing data or texture features derived from such imagery. This package introduces a simple and pragmatic approach to dealing with this problem. This approach can be especially beneficial in situations where features tend to be linearly dependent, or in other words, where principal component analysis seems like a reasonable approach. Please refer to the paper on arXiv for details.

install.packages("remotes")
remotes::install_github("alexanderbrenning/wiml")