CDTK.Tools.NumericalDerivatives module

CDTK.Tools.NumericalDerivatives.gradient(f, x, ms=None, step=0.1, fast=True)[source]

Return an array with the gradient in x along the directions in modes

f – function with respect the gradient is taken x – coordinates to take the gradient ms – list of arrays with orthonormal directions step – step to take in the numerical differentiation

CDTK.Tools.NumericalDerivatives.hessian(f, x, ms=None, step=0.1, fast=True)[source]

Return an array with the hessian in x along the directions in modes

f – function with respect the hessian taken x – coordinates to take the hessian ms – list of arrays with orthonormal directions step – step to take in the numerical differentiation