probnum.filtsmooth

Bayesian Filtering and Smoothing.

This package provides different kinds of Bayesian filters and smoothers which estimate the distribution over observed and hidden variables in a sequential model. The two operations differ by what information they use. Filtering considers all observations up to a given point, while smoothing takes the entire set of observations into account.

Functions

effective_number_of_events(categ_rv)

Approximate effective number of events in the support of a categorical random variable.

Classes

BayesFiltSmooth(dynamics_model, …)

Bayesian filtering and smoothing.

Kalman(dynamics_model, measurement_model, initrv)

Gaussian filtering and smoothing, i.e. Kalman-like filters and smoothers.

EKFComponent(non_linear_model)

Interface for extended Kalman filtering components.

ContinuousEKFComponent(non_linear_model[, …])

Continuous-time extended Kalman filter transition.

DiscreteEKFComponent(non_linear_model[, …])

Discrete extended Kalman filter transition.

UKFComponent(non_linear_model[, spread, …])

Interface for unscented Kalman filtering components.

ContinuousUKFComponent(non_linear_model[, …])

Continuous-time unscented Kalman filter transition.

DiscreteUKFComponent(non_linear_model[, …])

Discrete unscented Kalman filter transition.

UnscentedTransform(dimension[, spread, …])

Used for unscented Kalman filter.

TimeSeriesPosterior(locations, states)

Posterior Distribution over States after time-series algorithms such as filtering/smoothing or solving ODEs.

KalmanPosterior(locations, states, transition)

Posterior distribution after approximate Gaussian filtering and smoothing.

FilteringPosterior(locations, states, transition)

Filtering posterior.

SmoothingPosterior(locations, states, …)

Smoothing posterior.

StoppingCriterion([atol, rtol, maxit])

Stop iteration if absolute and relative tolerance are reached.

IteratedDiscreteComponent(component[, stopcrit])

Iterated updates.

ParticleFilter(dynamics_model, …[, …])

Particle filter (PF).

ParticleFilterPosterior(locations, states)

Posterior distribution of a particle filter..

Class Inheritance Diagram

Inheritance diagram of probnum.filtsmooth.bayesfiltsmooth.BayesFiltSmooth, probnum.filtsmooth.gaussfiltsmooth.kalman.Kalman, probnum.filtsmooth.gaussfiltsmooth.extendedkalman.EKFComponent, probnum.filtsmooth.gaussfiltsmooth.extendedkalman.ContinuousEKFComponent, probnum.filtsmooth.gaussfiltsmooth.extendedkalman.DiscreteEKFComponent, probnum.filtsmooth.gaussfiltsmooth.unscentedkalman.UKFComponent, probnum.filtsmooth.gaussfiltsmooth.unscentedkalman.ContinuousUKFComponent, probnum.filtsmooth.gaussfiltsmooth.unscentedkalman.DiscreteUKFComponent, probnum.filtsmooth.gaussfiltsmooth.unscentedtransform.UnscentedTransform, probnum.filtsmooth.timeseriesposterior.TimeSeriesPosterior, probnum.filtsmooth.gaussfiltsmooth.kalmanposterior.KalmanPosterior, probnum.filtsmooth.gaussfiltsmooth.kalmanposterior.FilteringPosterior, probnum.filtsmooth.gaussfiltsmooth.kalmanposterior.SmoothingPosterior, probnum.filtsmooth.gaussfiltsmooth.stoppingcriterion.StoppingCriterion, probnum.filtsmooth.gaussfiltsmooth.iterated_component.IteratedDiscreteComponent, probnum.filtsmooth.particlefiltsmooth._particle_filter.ParticleFilter, probnum.filtsmooth.particlefiltsmooth._particle_filter_posterior.ParticleFilterPosterior