Recording examples

This script demonstrates how to select different recording backends and read the result data back in. The simulated network itself is rather boring with only a single poisson generator stimulating a single neuron, so we get some data.

import nest
import numpy as np


def setup(record_to, time_in_steps):
    """Set up the network with the given parameters."""

    nest.ResetKernel()
    nest.SetKernelStatus({'overwrite_files': True})

    pg_params = {'rate': 1000000.}
    sr_params = {'record_to': record_to, 'time_in_steps': time_in_steps}

    n = nest.Create('iaf_psc_exp')
    pg = nest.Create('poisson_generator', 1, pg_params)
    sr = nest.Create('spike_recorder', 1, sr_params)

    nest.Connect(pg, n, syn_spec={'weight': 10.})
    nest.Connect(n, sr)

    return sr


def get_data(sr):
    """Get recorded data from the spike_recorder."""

    if sr.record_to == 'ascii':
        return np.loadtxt(f'{sr.filenames[0]}', dtype=object)
    if sr.record_to == 'memory':
        return sr.get('events')


# Just loop through some recording backends and settings
for time_in_steps in (True, False):
    for record_to in ('ascii', 'memory'):
        sr = setup(record_to, time_in_steps)
        nest.Simulate(30.0)
        data = get_data(sr)
        print(f"simulation resolution in ms: {nest.GetKernelStatus('resolution')}")
        print(f"data recorded by recording backend {record_to} (time_in_steps={time_in_steps})")
        print(data)