Metadata-Version: 2.4
Name: aigs
Version: 0.1.0
Summary: Package the Artificial Intelligence in Games and Simulation course
Requires-Python: <3.11,>=3.10
Description-Content-Type: text/markdown
Requires-Dist: cma>=4.3.0
Requires-Dist: equinox>=0.12.2
Requires-Dist: flax>=0.10.7
Requires-Dist: grpcio>=1.73.0
Requires-Dist: gymnasium>=1.1.1
Requires-Dist: jax>=0.5.3
Requires-Dist: jaxlib>=0.5.3
Requires-Dist: jumanji>=1.1.1
Requires-Dist: mctx>=0.0.5
Requires-Dist: mlxp>=1.0.4
Requires-Dist: pettingzoo>=1.25.0
Requires-Dist: pgx>=2.6.0
Requires-Dist: pygame>=2.6.1
Requires-Dist: qdax>=0.5.0
Requires-Dist: tictactoe-gym>=1.0.1
Requires-Dist: torch>=2.7.1
Requires-Dist: transformers>=4.54.1

# aigs

To get up and running:

1. Install `uv` so you can get our environment up and running (https://docs.astral.sh/uv/)
2. Clone this repo `git clone https://github.com/syrkis/aigs.git`

Open the repo in your IDE. You can run the code with our dependencies using `uv run python`

## labs

- MCTS. Find a simple game (easier than chess or go) that can be played in the temrinal.
  1. Implement the game (do it in unity if you want) (connect four or checkers)
  2. Impement MCTS
  3. play with params and have a competation
- DRL (getting a good player)
  1. get unity ml-agent to run
  2. pick game. Use PPO. Finetune.
  3.
- quality diversity (finding good levels)
  1. implement map elite for levels
  2. create a dataset of different solvable levels
  3.

## exam

1. Make your own game that is more awesome than default unity, and train ppo on it
