protesting on graphs : the pog database

I: Introduction

To further automate the simulation, there are different Python script that will run the protesting_on_graphs binary with different parameters and store the results in a directory. These script rely on the autorun.py script. Which itself relies on the main protesting_on_graphs C program.

II: How to create a base of simulations

To use these script you need to create a directory with :

• a copy of the protesting_on_graphs binary (type "make protesting_on_graphs" in the protesting_on_graphs directory)
• a directory called "base", a file called index_base.csv
• a copy of the batch_launch.sh script (which can be found in the graph_treatment/) subdirectory

II.1: The create_db.py script

You will then be able to use the functions defined in the create_db.py python file. You might want to adjust the generation functions to your needs and your machine. The function currently used will produce results for 10001 different parameters of the simulations. Each result is obtained by doing the mean of 10 simulations of 1000 steps with the same parameters. I t ran on one of the lab's server and I was only interrested in the last 10 steps of the simulation so I used a different version of the program that only dumped it's last iterations.

However is you were to use it now it would take ages as it would analyze every iteration of the simulations.

II.2: The db_analysis.py script

The db_analysis.py script takes one or more directories containing simulation bases as argument. It will then create a custom scatterplot of these bases.

Like the create_db.py script, you might want to tweek it because it relies on matplotlib and a bunch of dodgy code that is somewhat specific to the simulations I ran.