The blockCV package creates spatially or environmentally separated training and testing folds for cross-validation to provide a robust error estimation in spatially structured environments. See
I got an error with the buffering function. It seems that you missed the conversion to SpatialPointsDataFrame if the speciesData is an sf object. I also removed a part of the crs check. Maybe the bug you mentioned in the comments is fixed and the new check is sufficient.
We are currently implementing your package into mlr-org/mlr3spatiotemporal.
I got an error with the
buffering
function. It seems that you missed the conversion toSpatialPointsDataFrame
if thespeciesData
is ansf
object. I also removed a part of the crs check. Maybe the bug you mentioned in the comments is fixed and the new check is sufficient.