+The implementation of kernel density uses the Quartic kernel for it's estimates, with the methodology implemented [following QGIS](https://github.com/qgis/QGIS/blob/master/src/analysis/raster/qgskde.cpp) and as described in ArcGIS' documentation explaining [how Kernel Density works](https://pro.arcgis.com/en/pro-app/latest/tool-reference/spatial-analyst/how-kernel-density-works.htm). There are many alternative kernel density estimate (KDE) functions available in popular python libraries that may offer better performance, for example [scipy](https://docs.scipy.org/doc/scipy/reference/stats.html#univariate-and-multivariate-kernel-density-estimation), [scikit-learn](https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KernelDensity.html), [KDEpy](https://kdepy.readthedocs.io/en/latest/index.html) etc., though these alternatives may not implement the Quartic kernel (with optional weights), as is typically found in GIS software. Additionally, performance with this package will be reduced compared to the native C++ implementaion in QGIS.
0 commit comments