mahalanobis_distance_cpp: (Generalized) Mahalanobis distance

In shapr: Prediction Explanation with Dependence-Aware Shapley Values

(Generalized) Mahalanobis distance

Description

Used to get the Euclidean distance as well by setting mcov = diag(m).

Usage

mahalanobis_distance_cpp(
  featureList,
  Xtrain_mat,
  Xtest_mat,
  mcov,
  S_scale_dist
)

Arguments

featureList	List of vectors indicating all factor combinations that should be included in the computations. Assumes that the first one is empty.
Xtrain_mat	Matrix
Xtest_mat	Matrix
mcov	Matrix. The Sigma-matrix in the Mahalanobis distance formula (stats::cov(Xtrain_mat)) gives Mahalanobis distance, diag(m) gives the Euclidean distance.
S_scale_dist	Logical indicating

Value

Array of three dimensions. Contains the squared distance for between all training and test observations for all feature combinations passed to the function.

Author(s)

Martin Jullum

shapr documentation built on May 4, 2023, 5:10 p.m.