lr = numpy.float64(0.001)
nepochs = int(1)
name = str("mnist_cnet_3l")
input_dims = 1
output_dims = 10
model = Sequential()
for i in range(nlayers_embeded):
if i == 0:
nfilters = input_dims
model.add(CnetConv(nfilters, nfilters_embeded, 3, 1, 0))
model.add(CnetPool(2, 2, 0))
else:
nfilters = nfilters_embeded
model.add(CnetPool(2, 2, 0))
model.add(CnetLin(None, output_dims))
model.build()
chain = ChainRAP()
chain.add_sequence(model)
chain.setup_optimizers('adam', lr)
print("Model define Over ! ")
#################################################Training####################################################
nfilters_embeded = int(32)
nlayers_embeded = int(3)
lr = numpy.float64(0.001)
nepochs = int(1)
name = str("mnist_ebnn")
input_dims = 1
output_dims = 10
model = Sequential()
for i in range(nlayers_embeded):
if i == 0:
nfilters = input_dims
model.add(ConvPoolBNBST(nfilters, nfilters_embeded, 3, 1, 0, 2, 2, 0))
else:
nfilters = nfilters_embeded
model.add(
BinaryConvPoolBNBST(nfilters, nfilters_embeded, 3, 1, 0, 2, 2, 0))
model.add(BinaryLinearBNSoftmax(None, output_dims))
model.build()
chain = ChainRAP()
chain.add_sequence(model)
chain.setup_optimizers('adam', lr)
print("Model define Over ! ")
#################################################Training####################################################