This is based on my project on firing statistics of heterogenous cluster sizes in cortical networks. Written in C++ (also works with C) and plotting is done with gnuplot.
This project has 3 main C++/C codes:
- nocluster_network.cpp - Produces datafile for plotting a raster plot with no clusters.
- partial_balance.cpp - Produces partially balanced network and its firing statistics. (read comments to provide external stimulation to certain neurons) This file can be modified to create the raster plots of Liwin-Kumar, Doiron paper. The input file should be 50 clusters with same size.
- hierarchy.cpp - Produces a partially balanced network with hierarchy.
The community sizes for the exciters are defined in the popexp.dat file. The sizes are exponentially distributed with mean = 80 (on average 50 clusters).
ran2.cpp is the random number generator (RAN2) used in all the scripts.
plot.plt plots the raster plot for the spiking time of each neuron with each community distinguished by alternative coloring scheme. This shows the variability/uniformity of firing of the neurons. Change the name of the data file that is going to be plotted (same as the one used to write the spike data in the c++ program).