Neuron Models

This is a repository of simulating Neuron Models:

• Leaky integrate and fire model (LIF)
• Izhekevich Taken from https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1257420

Given an input current value the neuron gives the corresponding spikes. The data obtained can be used in various Spiking Neural Networks (SNN) architectures.

All parameters of the neuron can be changed in the configs.py. Note that the units of time is milliseconds.

Usage:

From neuron.py, call the function runLIF().

def runNeuron(model, t_span, dt, I):
    v = np.zeros_like(t_span)

    for i, t in enumerate(t_span):
        v[i] = model.generateSpiking(I[i], t, dt)

    
    if model.isPlot:
        plt.plot(t_span,v, label = 'V')
        plt.plot(t_span,I, label = 'I')
        plt.title('Neuron Model')
        plt.ylabel('Membrane Potential (V) and input current(I)')
        plt.xlabel('Time (msec)')
        plt.grid()
        plt.legend(loc="upper right")
        plt.show()
    return v

Here, t_span is the the overall time duration, dt being the small time steps and I being the current function w.r.t time.

To get an idea just run the neuron.py, and see the default example given at the end:

t_tot = 1000
dt = 0.01
t_span = np.arange(0, t_tot+dt, dt)
I = [1 if 200/dt <= i <= 600/dt  else 10 for i in range(len(t_span))] # defining the current(time)
neuron = Izhikevich() #defining the neuron model
# passing the current through the neuron.
v = runNeuronSimple(neuron, t_span, dt, I)

NOTE: To plot the voltage and current data, turn ON the isPLot argument in the configs.py

Test Case:

• LIF
  
• Izhikevich
  

Some links:

• MIT open courseware
• https://neuronaldynamics.epfl.ch/index.html
• https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1257420

To-Do:

• Add random salt-pepper noise to the output voltage.