Skip to content

Latest commit

 

History

History
156 lines (97 loc) · 5 KB

README.md

File metadata and controls

156 lines (97 loc) · 5 KB
Raw

DOI

gol_rule_switching

Implementation of the Game of Life with rule switching.

About

This repo contains model and scripts used to run numerical experiments for Game of Life with rule switching mechanisms. The code was used in

J.A. Miszczak, Rule switching mechanisms in the Game of Life with synchronous and asynchronous updating policy, Physica Scripta, 98, 115210 (2023). DOI:10.1088/1402-4896/acfc6c arXiv:2310.05979

@article{miszczak2023rule,
	author = {Miszczak, J.A.},
	title = {Rule switching mechanisms in the Game of Life with synchronous and asynchronous updating policy},
	doi = {10.1088/1402-4896/acfc6c},
	journal = {Physica Scripta}, 
	volume = {98},
	number = {11},
	pages = {115210},
	year = {2023},
}

Reproducing data

Scripts in this repo can be used to reproduce data illustrating the behaviour of the Game of Live (GoL) cellular automata extended with the mechanisms for rule switching. Additionally, the model can be used in asynchronous and synchronous mode.

Code compilation and running is controlled by file Makefile in the main directory. This file should be modified to suit our NetLogo installation. In the provided Makefile it is assumed that NetLogo is installed in

~/Local/NetLogo/6.3.0

directory. It is also assumed that java and javac are in you PATH environment variable.

Makefile also uses GNU parallel for running Java programmes.

Final configurations

Sample configurations are produced using GenerateExample.java script. To compile this file run

make

in the main directory. This will also compile CalculateMeanLiving.java and GenerateConfigurations.java.

Examples are generated by issuing

make genex

in the main directory. This will produce files world*.csv files, named according to the parameters defined in CalculateMeanLiving.java file.

Plotting of the resulting data is handled by plot_world_example.py script. Running

python plot_world_example.py word_data.csv

will produce PDF file for visualizing data from the CSV file world_data.csv.

To plot all data from the generated examples run

for f in world*.csv ; do python plot_world_example.py $f ; done

Note: By default, the world size is set to 2^6. Details are defined in *experiements.xml file.

Note: Plots are saved in plots subdirectory.

Realizations

Exemplary realizations can be obtained by using run.sh script and experiments.xml file. For each experiment defined in experiments.xml file, make a symbolic link to run.sh

ln -s run.sh experiment-name.sh

Next, run

bash experiment-name.sh

The resulting data will be saved in experiment-name.csv In the included experiments.xml file there are defined two experiments: deterministic_realizations and random_realizations

Note: The above commands can by run using

make realizations

To plot data from the realizations run

python plot_realizations.py

Note: Configure name of the data files in the plotting script.

Average growth

Data for the average number of living cells are calculated by CalculateMeanLiving.java and can be obtained by running

make calcmean

You need to compile CalculateMeanLiving.java first

Note: BY default, 200 realizations are used fo averaging. This value can be modified in CalculateMeanLiving.java file by altering steps variable.

Plotting of the data is handled by plot_living_average.py, which accepts one argument with name of the data file.

Note: Data from this experiment are saved in *living_data_250steps_200runs.csv file. This file is included in the *repository. The name of the file used to save data should be configured in *Makefile and in the plotting script plot_living_average.py.

Mutual entropy

For the purpose of analysing the mutual entropy of the generated patters, it is necessary to generate a number of final configurations. This data is generated using genfc target in the Makefile,

make gencf

You need to compile GenerateConfigurations.java before running this command.

Resulting csv files in configurations subdirectory. For each combination of parameters 1000 files representing final configurations are generated. The names of the resulting configurations can be saved in a shell variable using

golTypes=$(for i in *_0001.csv ; do echo `echo $i | sed s/_0001.csv//g` ; done)

in the configurations directory.

The calculation of the average mutual entropy is implemented in analyze_mutent.py script. Using GNU parallel, the analysis results can be obtained by running

parallel ./analyze.py {} ::: $golTypes

in configurations subdirectory.

This will result in one *.dat file for each combination of parameters. To obtain the file with all data run

bash postprocess_dat.sh

script. This will produce anayze_mutent.csv, used by the plotting script plot_mutual_entropy.py located in the main directory.