Transition¶
-
class
probnum.filtsmooth.statespace.
Transition
[source]¶ Bases:
abc.ABC
Markov transition rules in discrete or continuous time.
In continuous time, this is a Markov process and described by a stochastic differential equation (SDE)
\[d x_t = f(t, x_t) d t + d w_t\]driven by a Wiener process \(w\). In discrete time, it is defined by a transformation
\[x_{t + \Delta t} \sim p(x_{t + \Delta t} | x_t).\]Sometimes, these can be equivalent. For example: mild solutions to linear, time-invariant SDEs have an equivalent, discretised form that can be written as a transformation.
See also
ContinuousModel
Continuously indexed transitions (SDEs)
DiscreteModel
Discretely indexed transitions (transformations)
Attributes Summary
Dimension of the transition model.
Methods Summary
transition_realization
(real, start[, stop, …])Transition a realization of a random variable from time \(t\) to time \(t+\Delta t\).
Applies the transition, assuming that the state is already preconditioned.
transition_rv
(rv, start[, stop, step, …])Transition a random variable from time \(t\) to time \(t+\Delta t\).
transition_rv_preconditioned
(rv, start[, …])Applies the transition, assuming that the state is already preconditioned.
Attributes Documentation
-
dimension
¶ Dimension of the transition model.
Not all transition models have a unique dimension. Some turn a state (x, y) into a scalar z and it is not clear whether the dimension should be 2 or 1.
- Return type
Methods Documentation
-
abstract
transition_realization
(real, start, stop=None, step=None, linearise_at=None)[source]¶ Transition a realization of a random variable from time \(t\) to time \(t+\Delta t\).
For random variable \(x_t\), it returns the random variable defined by
\[x_{t + \Delta t} \sim p(x_{t + \Delta t} | x_t = r) .\]This is different to
transition_rv()
which computes the parametrization of \(x_{t + \Delta t}\) based on the parametrization of \(x_t\).Nb: Think of transition as a verb, i.e. this method “transitions” a realization of a random variable.
- Parameters
real (
ndarray
) – Realization of the random variable.start (
float
) – Starting point \(t\).step (
Optional
[float
]) – Intermediate step-size. Optional, default is None.linearise_at (
Optional
[RandomVariable
]) – For approximate transitions , for instance ContinuousEKFComponent, this argument overloads the state at which the Jacobian is computed.
- Return type
- Returns
RandomVariable – Random variable, describing the state at time \(t + \Delta t\) based on realization at time \(t\).
dict – Additional information in form of a dictionary, for instance the cross-covariance in the prediction step, access to which is useful in smoothing.
See also
transition_rv()
Apply transition to a random variable.
-
transition_realization_preconditioned
(real, start, stop=None, step=None, linearise_at=None)[source]¶ Applies the transition, assuming that the state is already preconditioned.
This is useful for numerically stable implementation of Kalman smoothing steps and Kalman updates.
- Return type
-
abstract
transition_rv
(rv, start, stop=None, step=None, linearise_at=None)[source]¶ Transition a random variable from time \(t\) to time \(t+\Delta t\).
For random variable \(x_t\), it returns the random variable defined by
\[x_{t + \Delta t} \sim p(x_{t + \Delta t} | x_t) .\]This returns a random variable where the parametrization depends on the paramtrization of \(x_t\). This is different to
transition_rv()
which computes the parametrization of \(x_{t + \Delta t}\) based on a realization of \(x_t\).Nb: Think of transition as a verb, i.e. this method “transitions” a random variable.
- Parameters
rv (
RandomVariable
) – Realization of the random variable.start (
float
) – Starting point \(t\).step (
Optional
[float
]) – Intermediate step-size. Optional, default is None.linearise_at (
Optional
[RandomVariable
]) – For approximate transitions , for instance ContinuousEKFComponent, this argument overloads the state at which the Jacobian is computed.
- Return type
- Returns
RandomVariable – Random variable, describing the state at time \(t + \Delta t\) based on realization at time \(t\).
dict – Additional information in form of a dictionary, for instance the cross-covariance in the prediction step, access to which is useful in smoothing.
See also
transition_realization()
Apply transition to a realization of a random variable.