I have this code:
import nengo
with nengo.Network() as model:
stim = nengo.Node([-1, -0.75, -0.5, -0.25, 0, 0.25, 0.5, 0.75, 1])
bg = nengo.networks.BasalGanglia(9)
thal = nengo.networks.Thalamus(9)
nengo.Connection(stim, bg.input)
nengo.Connection(bg.output, thal.input)
Although there is a clear maximum, the thalamus outputs 0 for all signals:
What is the reason for that? And what are the limitations of using the Basal Ganglia and Thalamus modules?
After another test, I found that my results were much worse than what Terry demonstrated in the summer school:
My Code:
Summer School:
The Basal Ganglia’s output looks much different, but the Thalamus’ output is way less accurate in my code, as it should return 1 (approximate) for the selected action… But I can’t distinguish the better action, while in the summer school example, the best action is very clear.
Even a clear winner doesn’t even close to 1 (actually ~0.31), in my code:
What is the reason for that?
