jaxsnn.event.to_nir_data.ValuedEventData

class jaxsnn.event.to_nir_data.ValuedEventData(idx: numpy.ndarray, time: numpy.ndarray, n_neurons: int, t_max: float, values: numpy.ndarray)

Bases: nir.data_ir.graph.EventData

Valued event-based data as a list of event indices, event times and event values.

idxnp.ndarray[int], shape (n_samples, n_events)

Event indices. If there is no event, the index is -1.

timenp.ndarray[float], shape (n_samples, n_events)

Event times. If there is no event, the time is np.inf.

valuesnp.ndarray[float], shape (n_samples, n_events)

Event values.

n_neuronsint

Total number of neurons in the layer.

t_maxfloat

Maximum time of the recording.

__init__(idx: numpy.ndarray, time: numpy.ndarray, n_neurons: int, t_max: float, values: numpy.ndarray) → None

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__(idx, time, n_neurons, t_max, values)	Initialize self.
to_time_gridded(dt[, interpolation])	Parameters

Attributes

to_time_gridded(dt: float, interpolation: nir.data_ir.graph.Interpolation = <Interpolation.NONE: 'None'>) → nir.data_ir.graph.TimeGriddedData

dtfloat

Time step size.

interpolationstr, optional

Interpolation method to use when converting to time-gridded data. Currently, only “None” is supported, which means that the values are assigned directly to the corresponding time steps without any interpolation.

values: numpy.ndarray