Downloads · 30 days
0
alphazero-quoridor
mishasamin/alphazero-quoridor
alphazero-quoridor is a reinforcement learning model from mishasamin. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product.
Access
Public
Updated Aug 25, 2023
Repo size
2.8 GB
Likes
2
Public
Repo makeup
Click a slice to open those files.
.tar2.8 GB · 100%
At a glance
- Task
- Reinforcement Learning
- Access
- Public
- Created
- Aug 11, 2023
- Updated
- Aug 25, 2023
- SHA
- bdf164cf
Model card
From the Hugging Face model README
AlphaZero, Quoridor Version!
Based on the framework provided here.
To start training, modify parameters in main.py and then start using
python main.py
Playing against it
Once you're done training, you need to modify pit.py to create one NN player, pointing it to your best.pth.tar and a human player.
During the game, you have a choice of ten actions:
u(up)d(down)r(right)l(left)- plus four diagonal move
ur,ul,dr,dl
In order to place walls, you type h (for horizontal wall) or v (for vertical wall), press enter followed by x y of where you want the wall to be placed.
- Task
- Reinforcement Learning
- Created
- Aug 11, 2023
- Updated
- Aug 25, 2023