get_centroid_weighted: Get weighted centroid of hypervolume or hypervolume list

In bblonder/hypervolume: High Dimensional Geometry, Set Operations, Projection, and Inference Using Kernel Density Estimation, Support Vector Machines, and Convex Hulls

Get weighted centroid of hypervolume or hypervolume list

Description

Returns the column weighted mean of the random points in each hypervolume. Useful for hypervolumes generated with hypervolume_n_occupancy() or hypervolume_n_occupancy_test().

Usage

get_centroid_weighted(hv)

Arguments

hv	A Hypervolume or HypervolumeList object.

Details

The function get_centroid_weighted() differs from get_centroid() because it uses occupancy values to weight random points for evaluating centroids position.

Value

Either a vector or a matrix of column of centroid values along each axis.

See Also

hypervolume_n_occupancy, hypervolume_n_occupancy_test

Examples

## Not run: 
data(penguins,package='palmerpenguins')
penguins_no_na = as.data.frame(na.omit(penguins))


penguins_no_na_split = split(penguins_no_na, 
paste(penguins_no_na$species, penguins_no_na$sex, sep = "_"))


hv_list = lapply(penguins_no_na_split, function(x) 
  hypervolume_gaussian(x[, c("bill_length_mm","bill_depth_mm","flipper_length_mm")],
  samples.per.point=100))

hv_list = hypervolume_join(hv_list)
hv_occupancy = hypervolume_n_occupancy(hv_list)

# unweighted centroids
get_centroid(hv_occupancy)

# weighted centroids
get_centroid_weighted(hv_occupancy)


## End(Not run)

bblonder/hypervolume documentation built on Feb. 1, 2024, 8:01 p.m.