pyinterp.RTree4DFloat32.packing

RTree4DFloat32.packing(self, coordinates: numpy.ndarray[dtype=float32, shape=(*, *), writable=False], values: numpy.ndarray[dtype=float32, shape=(*), writable=False], sigma2: numpy.ndarray[dtype=float32, shape=(*), writable=False]) → None

Bulk-load observations using STR packing.

Erases any existing data and rebuilds the index from the given (coordinates, values, sigma2) triple.

Parameters:

• coordinates – Matrix of shape (n, 4) — the four Cartesian coordinates of each observation. The user is responsible for unit consistency.

• values – Vector of length n with the observed values.

• sigma2 – Vector of length n with the per-observation error variance. Must be strictly positive.