Learning Quoridor

In this repository, open_spiel is used to train an agent to play Quoridor.

My agent is still very bad (Qlearning), how would you approach it?

Install tensorflow with gpu support

You must have python installed as well as cmake and a cpp compiler.

sudo apt install cmake clang

Best is to install via poetry as it automatically sets up a virtual environment.

poetry install

poetry shell
python3 multi_player.py

Trained policies are not uploaded.

Name		Name	Last commit message	Last commit date
Latest commit History 13 Commits
alpha_zero		alpha_zero
mcts		mcts
playground		playground
policies		policies
testing		testing
training		training
.gitignore		.gitignore
README.md		README.md
multi_player.py		multi_player.py
multi_player_random_move.py		multi_player_random_move.py
poetry.lock		poetry.lock
pyproject.toml		pyproject.toml
test_tensorflow.py		test_tensorflow.py