Optimal Interpolation

Optimal Interpolation#

Optimal Interpolation (OI / BLUE) for scattered 4D observations, with an anisotropic covariance kernel and a per-observation error variance. The estimator is backed by a 4D R-tree that stores each observation together with its measurement-error variance.

Estimator#

OptimalInterpolation(obs_coords, obs_values, ...)

Optimal Interpolation with anisotropic covariance and per-obs error.

OIResult(value, error, neighbors)

Result of an Optimal Interpolation analysis at a set of query points.

Spatial Index (4D R-Tree)#

The 4D R-tree is the indexing primitive feeding the estimator. Unlike RTree3D(), it does not provide inverse-distance weighting, kriging, RBF or window-function methods; it exposes k-nearest neighbour queries in 4D space-time and the built-in optimal_interpolation backend.

RTree4D([dtype])

Spatial index for 4D point data with per-observation error variance.

RTree4DFloat32(self)

Spatial index for 4D point data with per-observation error variance.

RTree4DFloat64(self)

Spatial index for 4D point data with per-observation error variance.