A Final Year Project in CUHK, Autumn 2021

Network Dynaimcs Simulation

Files

param.h

• edit all the variables & settings here

simulate.c

• the main program to run the network dynaimcs

How to use

1. edit variables in param.h
2. place param.h and simulate.c in the same folder
3. compile simulate.c: gcc -O3 simulate.c -o simulate
4. wait for results

Output

export up to 4 files

OUT_SPIK

• stores all the spiking data
  • column 1: index of nodes, starting from 1
  • column 2: number of spikes of the corresponding node
  • remining columns: time-stamps of each spikes

OUT_POTV

• stores the time series of membrane potential v(t) for the network dynamics

OUT_INFO

• stores all the variables and settings as well as execution time for a simulation, for later reference

INI_CNFG

• same as OUT_INFO, designated for easy computer program importation

Notes

1. results will be output in the same folder as the codes, i.e., next to them

Optimization

Choice of compiler

After compiling the source code with several C compilerson Windows system, MinGW TDM-GCC 64 seems to be a good choice. Its running time is lesser than Cygwin64, the attached terminal of Visual Studio Code and MinGW 64/32. You can find MinGW TDM-GCC 64 here: https://jmeubank.github.io/tdm-gcc/

Compiling flag

I recommend using the -O3 flag when compiling, e.g., >gcc -O3 simulate.c -o simulate It turns on all the -O3 optimization flags, which reduce the running time significantly. Visit here for more details: https://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html

Notes

This program creates multiple 1-/2-dimensional arrays when running. It accesses the array elements in the tightest loops. Fast memory is essential as the program freqently reads from / writes into RAM. Also, if you enable output for time series, try to write the file on a fast drive, such as SSD, it will be substantially faster. You can change the output path for time series data file in 'param.h'.

Analysing Network and Their Dynamics

Files

coupling.py

• calculate
  • connection probability
  • statistics of synaptic weight
  • average synaptic weight
  • ratio of suppression & enhancement
• plot
  • average synaptic weight distribution

spiking.py

• calculate
  • average firing rate and its statistics
  • statistics of ISI (inter-spike interval)
  • identifying bursting nodes (work in progess)
  • statistics of synaptic weight of a network
  • average synaptic weight
  • ratio of suppression & enhancement
• plot
  • reformat spiking data
  • spike raster plot
  • firing rate distribution
  • ISI distribution

spiking_compare.py

• calculate
  • changes in firing rate
  • ratio of change in firing rate
• plot
  • firing rate distribution (compared)
  • ISI distribution (compared)
  • change in firing rate distribution (&combined)
  • Ratio of suppression / enhancement vs ratio of increase in firing rate (&combined)