NVAR

class brainpy.dyn.NVAR(num_in, delay, order=None, stride=1, constant=False, mode=None, name=None)

Nonlinear vector auto-regression (NVAR) node.

This class has the following features:

• it supports batch size,

• it supports multiple orders,

Parameters:

• delay (int) – The number of delay step.

• order (int, sequence of int) – The nonlinear order.

• stride (int) – The stride to sample linear part vector in the delays.

• constant (optional, float) – The constant value.

References

[1]

Gauthier, D.J., Bollt, E., Griffith, A. et al. Next generation reservoir computing. Nat Commun 12, 5564 (2021). https://doi.org/10.1038/s41467-021-25801-2

get_feature_names(for_plot=False)

Get output feature names for transformation.

Parameters:

for_plot (bool) – Use the feature names for plotting or not? (Default False)

Return type:

List[str]

reset_state(batch_or_mode=None, **kwargs)

Reset the node state which depends on batch size.

update(x)

The function to specify the updating rule.