The goal of spatialexplain is to provide model agnostic tools for exploring and explaining spatial machine learning models. It is based on the DALEX package and uses the terra package for spatial raster data handling.
You can install the development version of spatialexplain from GitHub with:
# install.packages("pak")
pak::pak("Nowosad/spatialexplain")A set of vignettes that show how to use the package to explain spatial machine learning models:
- An introduction to spatial explanations: this one is recommended as a starting point
- Spatial break-down: example of the Break Down method usage
- Spatial oscillations: example of the Oscillations method usage
- Spatial SHAP: example of the SHAP method usage
- Spatial LIME: example of the LIME method usage
Contributions to this package are welcome - let us know if you have any suggestions or spotted a bug. The preferred method of contribution is through a GitHub pull request. Feel also free to contact us by creating an issue.
The initial development of this R package was made possible through the financial support of the European Union’s Horizon Europe research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101147446.
