HybridLeastSquares

class pysit.objective_functions.HybridLeastSquares(solver, parallel_wrap_shot=<pysit.util.parallel.ParallelWrapShotNull object>, modeling_kwargs={})[source]

Bases: pysit.objective_functions.objective_function.ObjectiveFunctionBase

Methods Summary

apply_hessian(self, shots, m0, m1[, …])
compute_gradient(self, shots, m0[, …]) Compute the gradient for a set of shots.
evaluate(self, shots, m0[, frequencies, …]) Evaluate the least squares objective function over a list of shots.

Methods Documentation

apply_hessian(self, shots, m0, m1, frequencies=None, frequency_weights=None, hessian_mode='approximate', levenberg_mu=0.0, **kwargs)[source]
compute_gradient(self, shots, m0, frequencies=None, frequency_weights=None, aux_info={}, **kwargs)[source]

Compute the gradient for a set of shots.

Computes the gradient as
-F*(d - scriptF[m0]) = -sum(F*_s(d - scriptF_s[m0])) for s in shots
Parameters:
shots : list of pysit.Shot

List of Shots for which to compute the gradient.

evaluate(self, shots, m0, frequencies=None, frequency_weights=None, **kwargs)[source]

Evaluate the least squares objective function over a list of shots.