Skip to content

Latest commit

 

History

History
51 lines (37 loc) · 2.57 KB

README.md

File metadata and controls

51 lines (37 loc) · 2.57 KB
Raw

ConnN: A framework for playing Connect Four tournaments

I wrote this project to run tournaments for game-playing agents written by my students. The idea came from an undergraduate AI course I took (many years ago now). In contrast to that course, the objective of the course I taught was not just to introduce them to AI, but also good software development practices. My students came from various backgrounds, and most had limited software development experience. As such, I created a few lessons to try and get them up to speed as quickly as possible. The content of those lessons can be found in the Wiki.

Requirements

All users need the following Python packages:

  • NumPy
  • Stockings

Anyone that want to use the connectn.results module needs:

  • PyTables

Both the client and the server need to have Stunnel installed.

Server

Due to compatibility constraints (for GridMap), the server needs to be running in an environment with Python 3.6. The following Python packages also need to be installed in that environment:

  • NumPy
  • Stockings
  • PyTables
  • Numba

If you want to run on a compute cluster, you'll also need:

  • GridMap

Tournament Participants

To participate in a running tournament, you'll need to use the client script to upload your agent, and eventually download results. You'll need to provide your group name, password, and the location of your agent. These can be passed as command line arguments (run python client.py -h for details), or hardcoded at the top of client.py.

Tournament Administrators

If you want to run your own tournament, just run server.py. That script will start a process to run the actual tournament, and then will begin to handle requests from clients (to upload new agents, or download results). If the tournament is running on a remote machine, use screen so it doesn't shut down when you disconnect (if you later reconnect to the remote machine, just run screen -r to re-attach to the screen session).

While the server is running, if you send an interrupt signal (by pressing Ctrl-C in the terminal or the stop button if you're running in an IDE) it will ask if you want to shut down the server (it does this by telling the tournament process to shutdown and then exiting). If you say no, it will ask you if you want to run an "all-against-all" tournament. This is not normally necessary though, since by default the tournament plays "all-against-all" automatically every 6 hours.

For details on the various configuration options for the server, just run python server.py -h on the command line.