def __init__(self, config, queue=None):
AbstractSNN.__init__(self, config, queue)
self.snn = None
self._spiking_layers = {}
self._input_images = None
self._binary_activation = None
def __init__(self, config, queue=None):
AbstractSNN.__init__(self, config, queue)
self.layers = []
self._conns = [] # Temporary container for layer connections.
self._biases = [] # Temporary container for layer biases.
self.connections = [] # Final container for all layers.
self.cellparams = {key: config.getfloat('cell', key) for key in
cellparams_pyNN}
def __init__(self, config, queue=None):
AbstractSNN.__init__(self, config, queue)
self.snn = None
self._spiking_layers = {}
self._input_images = None
self._binary_activation = None
self.avg_rate = None
self._input_spikecount = None
self.latency = None
self.acc_at_t = []
def __init__(self, config, queue=None):
AbstractSNN.__init__(self, config, queue)
self.layers = []
self.connections = []
self.cellparams = {key: config.getfloat('cell', key) for key in
config_string_to_set_of_strings(config.get(
'restrictions', 'cellparams_pyNN'))}
if 'i_offset' in self.cellparams.keys():
print("SNN toolbox WARNING: The cell parameter 'i_offset' is "
"reserved for the biases and should not be set globally.")
self.cellparams.pop('i_offset')
self.change_padding = False
def __init__(self, config, queue=None):
AbstractSNN.__init__(self, config, queue)
self.layers = []
self.connections = [] # Final container for all layers.
self.threshold = 'v >= v_thresh'
self.v_reset = 'v = v_reset'
self.eqs = 'v : 1'
self.spikemonitors = []
self.statemonitors = []
self.snn = None
self._input_layer = None
self._cell_params = None
# Track the output layer spikes.
self.output_spikemonitor = None
def __init__(self, config, queue=None):
AbstractSNN.__init__(self, config, queue)
self.layers = []
self._conns = [
] # Temporary container for layer connections (excitatory neuron).
self._inconns = [
] # Temporary container for layer connections (inhibitory neuron).
self._con_exconns = [
] # Temporary container for convolution layer connections (excitatory neuron).
self._con_inconns = [
] # Temporary container for convolution layer connections (inhibitory neuron).
self._biases = [] # Temporary container for layer biases.
self.connections = [] # Final container for all layers.
self.cellparams = {
key: config.getfloat('cell', key)
for key in cellparams_pyNN
}
def __init__(self, config, queue=None):
AbstractSNN.__init__(self, config, queue)
self.layers = []
self.connections = [] # Final container for all layers.
self.threshold = 'v >= v_thresh'
if 'subtraction' in config.get('cell', 'reset'):
self.v_reset = 'v = v - v_thresh'
else:
self.v_reset = 'v = v_reset'
self.eqs = '''dv/dt = bias : 1
bias : hertz'''
self.spikemonitors = []
self.statemonitors = []
self.snn = None
self._input_layer = None
self._cell_params = None
# Track the output layer spikes.
self.output_spikemonitor = None
def set_spiketrain_stats_input(self):
AbstractSNN.set_spiketrain_stats_input(self)
def get_spiketrains_input(self):
# Added this here because PyCharm complains about not all abstract
# methods being implemented (even though this is not abstract).
AbstractSNN.get_spiketrains_input(self)