dymad.sako.adapter

Typed adapter seam for checkpoint-backed spectral analysis.

Classes

SpectralAnalysisAdapter(*, snapshot, eigensystem)	Adapter that delegates spectral operations to SAKO and RALowRank kernels.
SpectralEigensystem(discrete_eigs, ...[, dt])	Eigensystem terms consumed by the spectral adapter kernels.
SpectralRuntime(*args, **kwargs)	Runtime hooks needed by measure and Jacobian compatibility methods.

class dymad.sako.adapter.SpectralAnalysisAdapter(*, snapshot, eigensystem, runtime=None, reps=1e-10, etol=1e-13)

Bases: object

Adapter that delegates spectral operations to SAKO and RALowRank kernels.

estimate_measure(fobs, order, eps, thetas=101)

Estimate measure using runtime observable mapping plus SAKO kernel.

estimate_ps(grid=None, return_vec=False, mode='cont', method='standard')

Estimate pseudospectrum over a grid via the current adapter kernels.

eval_eigfunc_jac(ref=None, rng=None, **kwargs)

Evaluate Jacobian of eigenfunctions via runtime forward-mode hooks.

Return type:

ndarray

eval_eigmode_jac(ref=None, rng=None, **kwargs)

Evaluate Jacobian of eigenmodes via runtime backward-mode hooks.

Return type:

ndarray

property rals: RALowRank

resolvent_analysis(z, return_vec, mode, method)

Resolve one pseudospectrum point through SAKO or standard low-rank RA.

property sako: SAKO

class dymad.sako.adapter.SpectralEigensystem(discrete_eigs, left_eigvecs, right_eigvecs, projector, dt=1.0)

Bases: object

Eigensystem terms consumed by the spectral adapter kernels.

discrete_eigs: ndarray

dt: float = 1.0

left_eigvecs: ndarray

projector: ndarray

right_eigvecs: ndarray

class dymad.sako.adapter.SpectralRuntime(*args, **kwargs)

Bases: Protocol

Runtime hooks needed by measure and Jacobian compatibility methods.

apply_obs(fobs)

Return type:

ndarray

get_backward_modes(ref, rng=None, **kwargs)

Return type:

ndarray

get_forward_modes(ref, rng=None, **kwargs)

Return type:

ndarray