Class KoopmanDynamics
Bases: DiscreteDynamics
Koopman Dynamics of the form \(\Psi_{n+1} = K \Psi_n\). The states are updated from the dictionary's output.
__init__(koopman, dictionary, state_pos, state_dim, param_dim=0)
¶
Initializes the Dynamics class with the given parameters.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
koopman
|
Koopman or ParamKoopman
|
An instance representing the Koopman operator. |
required |
dictionary
|
Dictionary or TrainableDictionary
|
A dictionary or function used for state transition. |
required |
state_pos
|
list[int]
|
The position or index of the state in the input data. |
required |
state_dim
|
int
|
The dimensionality of the state space. |
required |
param_dim
|
int
|
The dimensionality of the parameter space. Defaults to 0. |
0
|
step(x, u)
¶
Advances the system dynamics by one time step.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
x
|
Tensor
|
The current state of the system. |
required |
u
|
Tensor
|
The control input applied to the system. |
required |
Returns:
| Type | Description |
|---|---|
Tensor
|
The next state of the system after applying the control input. |