Hivemind documentation¶
Description¶
The game engine logic is loosely inspired from Mzinga Engine.
The engine comes with different AI agent configurations. More on this below.
This projects also provides:
🔹 Releases - Prebuilt executables for Linux and Windows.
🔹 Documentation - Detailed codebase reference.
Setup¶
Setting up the environment is pretty easy:
Set up Python 3.12.7 (you can use any environment manager or none).
Install the dependencies from the file
requirements.txt
.
The suggested IDE is Visual Studio Code, and settings for it are included.
Usage¶
There are two ways to use this Hive engine:
- Run
engine.py
from the command line or with your IDE and start using the console to interact with it.The engine will be fully functional, but there won’t be any graphical interface. - Use the released executables (or build one yourself) along with MzingaViewer.To do this, move the
HivemindEngine
executable into the same directory asMzingaViewer
and then follow the instructions here, specificallystep 2 > iii
.
To build the HivemindEngine
executable yourself, simply run the following command in the project root:
pyinstaller ./src/engine.py --name HivemindEngine --noconsole --onefile
This will create an executable for your platform.
AI¶
There are currently 2 implemented AI strategies:
Random: the agent plays random moves.
Minmax: the agent plays moves following a Minmax policy with alpha-beta pruning and a custom node (game state) evaluation.
A third implementation will come in the future that will leverage machine learning.