def __init__(self, netParams, input_channels=3):
super(SlayerVgg16, self).__init__()
# initialize slayer
slayer = snn.layer(netParams['neuron'], netParams['simulation'])
self.slayer = slayer
# define network functions
self.conv1a = slayer.conv(input_channels, 16, 5, padding=2)
self.conv2a = slayer.conv(16, 16, 3, padding=1)
self.poola = slayer.pool(2)
self.conv1b = slayer.conv(16, 32, 3, padding=1)
self.conv2b = slayer.conv(32, 32, 3, padding=1)
self.poolb = slayer.pool(2)
self.conv1c = slayer.conv(32, 64, 3, padding=1)
self.conv2c = slayer.conv(64, 64, 3, padding=1)
self.conv3c = slayer.conv(64, 64, 3, padding=1)
self.poolc = slayer.pool(2)
self.conv1d = slayer.conv(64, 128, 3, padding=1)
self.conv2d = slayer.conv(128, 128, 3, padding=1)
self.conv3d = slayer.conv(128, 128, 3, padding=1)
self.poold = slayer.pool(2)
self.conv1e = slayer.conv(128, 256, 3, padding=1)
self.conv2e = slayer.conv(256, 256, 3, padding=1)
self.conv3e = slayer.conv(256, 256, 3, padding=1)
self.poole = slayer.pool(2)
self.fc1 = slayer.dense((1, 1, 256), 4096)
self.fc2 = slayer.dense(4096, 4096)
self.fc3 = slayer.dense(4096, 1000)
self.fc4 = slayer.dense(1000, 10)
def __init__(self, netParams):
super(Network, self).__init__()
# Initialize slayer
slayer = snn.layer(netParams['neuron'], netParams['simulation'])
self.slayer = slayer
# Define network functions
# The commented line below should be used if the input spikes were not reshaped
# self.fc1 = slayer.dense((34, 34, 2), 512)
self.fc1 = slayer.dense((34 * 34 * 2), 512)
self.fc2 = slayer.dense(512, 10)
def __init__(self,
netParams,
output_size=28 * 28,
hidden_size=200,
latent_size=2,
vae=False):
super(Decoder, self).__init__()
slayer = snn.layer(netParams['neuron'], netParams['simulation'])
self.vae = vae
self.slayer = slayer
self.fc1 = slayer.dense(latent_size, hidden_size)
self.fc2 = slayer.dense(hidden_size, output_size)
def __init__(self, netParams):
super(Network, self).__init__()
# initialize slayer
slayer = snn.layer(netParams['neuron'], netParams['simulation'])
self.slayer = slayer
# define network functions
self.conv1 = slayer.conv(2, 16, 5, padding=1)
self.conv2 = slayer.conv(16, 32, 3, padding=1)
self.conv3 = slayer.conv(32, 64, 3, padding=1)
self.pool1 = slayer.pool(2)
self.pool2 = slayer.pool(2)
self.fc1 = slayer.dense((8, 8, 64), 10)
def __init__(self, netParams, input_channels=3):
super(SlayerSNN2, self).__init__()
# initialize slayer
slayer = snn.layer(netParams['neuron'], netParams['simulation'])
self.slayer = slayer
# define network functions
self.conv1 = slayer.conv(input_channels, 16, 5, padding=2)
self.conv2 = slayer.conv(16, 32, 3, padding=1)
self.conv3 = slayer.conv(32, 64, 3, padding=1)
self.pool1 = slayer.pool(2)
self.pool2 = slayer.pool(2)
self.fc1 = slayer.dense((7, 7, 64), 10)
def __init__(self, netParams):
super(UpsamplingAutoencoder, self).__init__()
# initialize slayer
slayer = snn.layer(netParams['neuron'], netParams['simulation'])
self.slayer = slayer
# define network functions
self.conv1 = slayer.conv(1, 16, 5, padding=2)
self.pool1 = slayer.pool(2)
self.conv2 = slayer.conv(16, 32, 3, padding=1)
self.pool2 = slayer.pool(2)
self.conv3 = slayer.conv(32, 64, 3, padding=1)
self.unpool3 = slayer.upsampling2d(2, mode='bilinear')
self.conv4 = slayer.conv(64, 32, 3, padding=1)
self.unpool4 = slayer.upsampling2d(2, mode='bilinear')
self.convOut = slayer.conv(32, 1, 1, padding=0)
def __init__(self,
netParams,
input_size=(28, 28, 1),
hidden_size=200,
latent_size=2,
vae=False):
super(Encoder, self).__init__()
slayer = snn.layer(netParams['neuron'], netParams['simulation'])
self.vae = vae
self.slayer = slayer
self.fc1 = slayer.dense(input_size, hidden_size)
if vae:
self.fc_mean = slayer.dense(hidden_size, latent_size)
self.fc_log_var = slayer.dense(hidden_size, latent_size)
else:
self.fc2 = slayer.dense(hidden_size, latent_size)
def __init__(self, netParams, input_channels=3):
super(encoderSpikes, self).__init__()
# initialize slayer
slayer = snn.layer(netParams['neuron'], netParams['simulation'])
self.slayer = slayer