Skip to content

Included support for lava-nc LIF.

Toeter, A.H.H. (Akke) requested to merge Akke.Toeter/simsnn:master into master

Included support for bias and du as used in lava-nc.org LIF_neurons:

class LIF(AbstractProcess):
    """Leaky-Integrate-and-Fire neural process with activation input and spike
    output ports a_in and s_out.

    Realizes the following abstract behavior:
    u[t] = u[t-1] * (1-du) + a_in
    v[t] = v[t-1] * (1-dv) + u[t] + bias
    s_out = v[t] > vth
    v[t] = v[t] - s_out*vth
    """

And allowed for expanding multimeter data with voltage per timestep whilst running the simulator 1 timestep at a time. This is because, some simulations may require more advanced checks inbetween simulation steps, whilst still requiring the full network history. Same for the raster (spikes).

Also allowed users option to name neurons and give them an xy position. This is used to visualise the network interactively in plotly-Dash.

The original usage has been unaltered.

Pre-commit is included, to automatically improve formatting and improve the quality of contributions.

Typingchecking is not (yet) included.

Pip publishing code is included.

The main manual differences have been commented in the commit-changes, the rest is mainly auto-formatting and pip release files.

Visualisation (not included here)

Below is an example of the network visualisation, where one can hover over the nodes to see their properties, and loop over the network over time with the slider. image

Edited by Toeter, A.H.H. (Akke)

Merge request reports