Investigating the phenomenon of empire fragmentation on a grid using agent based modelling. The goal of this project is to investigate the conditions under which an empire is overthrown due to a revolution. For more details, definitions, and the exact model description please check the report in docs/empire_fragmentation_report.pdf.
The easiest way play around with the simulations produced by the code is to install all dependencies within an anaconda environment, running on a linux device. After cloning the repository locally, you would need to run:
conda env create -f environment.yml
using the environment.yml file provided in the repo.
After installing and activating the conda environment you can start an interactive session of the simulation by running
python src/visualisation_server.py
You can play with the model parameters of interest, visualise the interaction of the agents on the grid, and extract insights from the corresponding plots.
 |
|---|
| The interactive simulation session. |
One can also configure the parameters and run many simulations in parallel using src/batch_runner.py. This script scans the defined parameter space and outputs ther final state of the simulation in pandas dataframe format. From this file we can get further insights about the model as it can be seen in the notebook src/results_analysis.ipynb.