The output namespace implements an interface for accessing the output channel and SpikeMessage message converters. Kaspersky Neuromorphic Platform can transmit data both synchronously (upon the user request) and asynchronously (upon the call of the handler).
OutputChannel class
The OutputChannel class implements an output channel that receives a set of messages containing spikes from an associated population over several neural network execution cycles. The messages of the SpikeMessage type can be converted to one of the following data formats, defined by message translators:
boolean values.SpikeMessage type.If these data formats are not suitable for you, you can implement your own function to convert the received messages to the desired data format.
ConvertToSet class
The ConvertToSet class implements a data converter that receives messages of the SpikeMessage type from an object of the OutputChannel class and converts them to a vector of neurons that generate spikes.
Each object of the ConvertToSet class contains the size of the neuron vector.
converter_bitwise functor
The converter_bitwise functor implements a data converter that receives the messages of the SpikeMessage type and converts them to a vector of boolean values, where true indicates a spike.
converter_count functor
The converter_count functor implements a data converter that receives the messages of the SpikeMessage type and converts them to a vector of values. Each value of the vector corresponds to the number of spikes generated by the neuron.