Connecting Nengo to FPGA

NengoFPGA is an extension for Nengo that connects to FPGA. The current implementation supports a single ensemble of neurons with the PES learning rule.

You will need a supported FPGA device, then simply replace a standard ensemble with the augmented FPGA ensemble:

import nengo
from nengo_fpga.networks import FpgaPesEnsembleNetwork

# This standard Nengo ensemble...
ens = nengo.Ensemble(50, 1)

# ...is replaced with an FPGA ensemble
fpga_ens = FpgaPesEnsembleNetwork(
     'de1', n_neurons=50, dimensions=1, learning_rate=0)

Check out the rest of the documentation for more information.