maxResNN is a function that can be used to compute a reasonable grid resolution for nearest neighbor based uniform sampling. The core idea behind its working principle is that we want to expect a grid cell to contain points if it overlaps with the environment. This implementation looks at the low density regions using distance to n neighbors as a proxy. The approach assumes that both coordinates have similar range, as the axes are not weighted when computing Neighbors and converting from distances to number of grid cells If multiple points from the same grid cell should be sampled, the number of neighbors included in the computation should be set accordingly
paSamplingMcmc is a near drop in replacement for paSampling from the original USE package, that allows to perform a Gaussian mixture based pseudo absence sampling using a markov. In a first step a density function is constructed using a GMM fitted to the environment as a limit to the sampling space and a GMM fitted on the target species as a way to evade regions associated with the presence.
plotDensityLines enables the plotting of density function as well as a trace of a chain made of points in a dataframe on this density surface. In addition the species that was used to generate the density model can be supplied to verify the performance of the model. If that is the case the supplied dataset will be plotted as points instead of as a chain. This function has some issues with the check of devtools