skg.nsphere_fit

skg.nsphere_fit(x, axis=-1, scaling=False)[source]

Fit an n-sphere to ND data.

The center and radius of the n-sphere are optimized using the Coope method. The sphere is described by

\left \lVert \vec{x} - \vec{c} \right \rVert_2 = r

Parameters:
  • x (array-like) – The n-vectors describing the data. Usually this will be a nxm array containing m n-dimensional data points.
  • axis (int) – The axis that determines the number of dimensions of the n-sphere. All other axes are effectively raveled to obtain an (m, n) array.
  • scaling (bool) – If True, scale and offset the data to a bounding box of -1 to +1 during computations for numerical stability. Default is False.
Returns:

  • r (scalar) – The optimal radius of the best-fit n-sphere for x.
  • c (array) – An array of size x.shape[axis] with the optimized center of the best-fit n-sphere.

References