Note

This documentation is for a development version. Click here for the latest stable release (v3.0.0).

Ensemble arrayΒΆ

An ensemble array is a group of ensembles that each represent a part of the overall signal.

Ensemble arrays are similar to normal ensembles, but expose a slightly different interface. Additionally, in an ensemble array, the components of the overall signal are not related. As a result, network arrays cannot be used to compute nonlinear functions that mix the dimensions they represent.

[1]:
%matplotlib inline
import matplotlib.pyplot as plt
import numpy as np

import nengo
[2]:
model = nengo.Network(label="Ensemble Array")
with model:
    # Make an input node
    sin = nengo.Node(output=lambda t: [np.cos(t), np.sin(t)])

    # Make ensembles to connect
    A = nengo.networks.EnsembleArray(100, n_ensembles=2)
    B = nengo.Ensemble(100, dimensions=2)
    C = nengo.networks.EnsembleArray(100, n_ensembles=2)

    # Connect the model elements, just feedforward
    nengo.Connection(sin, A.input)
    nengo.Connection(A.output, B)
    nengo.Connection(B, C.input)

    # Setup the probes for plotting
    sin_probe = nengo.Probe(sin)
    A_probe = nengo.Probe(A.output, synapse=0.02)
    B_probe = nengo.Probe(B, synapse=0.02)
    C_probe = nengo.Probe(C.output, synapse=0.02)
[3]:
with nengo.Simulator(model) as sim:
    sim.run(10)
[4]:
# Plot the results
plt.figure()
plt.plot(sim.trange(), sim.data[sin_probe])
plt.plot(sim.trange(), sim.data[A_probe])
plt.plot(sim.trange(), sim.data[B_probe])
plt.plot(sim.trange(), sim.data[C_probe])
[4]:
[<matplotlib.lines.Line2D at 0x7f2a1984c470>,
 <matplotlib.lines.Line2D at 0x7f2a1984cf28>]
../../_images/examples_networks_ensemble-array_4_1.png

These plots demonstrate that the ensemble array works very similarly to a standard N-dimensional population. However, this is not true when it comes to computing functions. Ensemble arrays cannot be used to compute nonlinear functions that mix the dimensions they represent.