predict()

InterpolatedDiscreteModel.predict(parameter, state0, niters, inputs=None)

Step forward the discrete dynamical system niters steps. Essentially, this amounts to the following.

>>> states[:, 0] = state0
>>> states[:, 1] = model.rhs(parameter, states[:, 0], inputs[:, 0])
>>> states[:, 2] = model.rhs(parameter, states[:, 1], inputs[:, 1])
...                                     # Repeat `niters` times.

Parameters

parameter(p,) ndarray

Parameter value \(\bfmu\).

state0(r,) ndarray

Initial state.

nitersint

Number of times to step the system forward.

inputs(m, niters-1) ndarray or None

Inputs for the next niters - 1 time steps.

Returns

states(r, niters) ndarray

Solution to the system, including the initial condition state0.

Notes

For repeated predict() calls with the same parameter value, use evaluate() to first get the nonparametric model corresponding to the parameter value.