Tutorial: Spike and learn

In this short tutorial we combine both the spiking and the learning model.

Setting everything up

As in the last two tutorials, we

• import the needed packages (lines 1 – 2),
• set up the SRM and STDP models (lines 4 & 6),
• set up the initial spiketrain and weights (lines 8 & 10).

import numpy as np
from neurons import spiking, learning

srm_model = spiking.SRM(neurons=3, threshold=1, t_current=0.3, t_membrane=20, eta_reset=5)

stdp_model = learning.STDP(eta=0.05, w_in=0.5, w_out=0.5, tau=10.0, window_size=5)

weights = np.array([[0, 0, 1.], [0, 0, 1.], [0, 0, 0]])

spiketrain = np.array([[0, 0, 1, 0, 0, 0, 1, 1, 0, 0],
                       [1, 0, 0, 0, 0, 0, 1, 1, 0, 0],
                       [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]], dtype=bool)

Simulate

The simulation is again similar. But now we calculate in each timestep first the spike change (line 2) and immediately afterwards the weight change (line 3):

for time in range(10):
    srm_model.check_spikes(spiketrain, weights, time)
    stdp_model.weight_change(spiketrain, weights, time)

That was the major step :) Not too huge, right?

Result

Print out the final spiketrain and the weight changes to the console:

print("Final Spiketrain:")
print(spiketrain)

print("Final weights:")
print(weights)

Which gives us:

Final Spiketrain:
[[0 0 1 0 0 0 1 1 0 0]
 [1 0 0 0 0 0 1 1 0 0]
 [0 0 0 1 0 0 0 0 1 1]]
Final weights:
[[ 0.          0.          1.28884081]
 [ 0.          0.          1.28063985]
 [ 0.          0.          0.        ]]

Notice, that the result is slightly different to our previous results:

First, neuron 3 also spikes at time 10 now, because when checking for new spikes at time 10 the weights have been bigger, and thus forcing the neuron to spike another time.

Second, the weights have gone up slightly more, because neuron 3 spikes also at time 10.

Conclusion

In this part of the tutorial we have seen that

In the next part, we want to visualize the neuron currents and weight changes.