classification#

Functions#

classify_using_hessian(result[, dx, cost_tolerance, ...])

A simple check for parameter correlations based on numerical approximation

plot_hessian_eigenvectors(info[, steps])

A function to plot the eigenvectors computed for the Hessian at an optimal point.

Module Contents#

classification.classify_using_hessian(result: pybop.Result, dx=None, cost_tolerance: float = 1e-05, normalise: bool = True)#

A simple check for parameter correlations based on numerical approximation of the Hessian matrix at the optimal point using central finite differences.

Parameters:

  • result (Result) – The optimisation result.

  • dx (array-like, optional) – An array of small positive values used to check proximity to the parameter bounds and as the perturbation distance in the finite difference calculations.

  • cost_tolerance (float, optional) – A small positive tolerance used for cost value comparisons (default: 1e-5).

  • normalise (bool, optional) – If True, the Hessian is scaled by the step size in the parameters so that the Hessian entries are in the same unit as the cost values (default: True).

classification.plot_hessian_eigenvectors(info, steps: int = 10)#

A function to plot the eigenvectors computed for the Hessian at an optimal point.

Parameters:

  • info (dict) – The output from pybop.classify_using_Hessian.

  • steps (int) – Grid resolution per axis.