spd_learn.functional.nearest_sym_pos_def#

spd_learn.functional.nearest_sym_pos_def(X, reg=1e-06)[source]#

Find the nearest SPD matrices.

A NumPy port of John D’Errico’s nearestSPD MATLAB code [1], which credits [2].

Parameters:

  • X (ndarray, shape (..., n, n)) – Square matrices.

  • reg (float, default=1e-6) – Regularization parameter.

Returns:

P – Nearest SPD matrices.

Return type:

ndarray, shape (…, n, n)

Notes

Added in version 0.4.

Changed in version 0.11: Add broadcasting.

Changed in version 0.12: Add support for NumPy and PyTorch.

References

[1]

nearestSPD J. D’Errico, MATLAB Central File Exchange

[2]

Computing a nearest symmetric positive semidefinite matrix N.J. Higham, Linear Algebra and its Applications, vol 103, 1988