https://github.com/thowell/achtung
https://github.com/thowell/achtung
Last synced: 9 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/thowell/achtung
- Owner: thowell
- Created: 2020-10-06T05:30:27.000Z (over 5 years ago)
- Default Branch: main
- Last Pushed: 2020-11-16T23:47:46.000Z (over 5 years ago)
- Last Synced: 2025-09-07T22:39:18.519Z (10 months ago)
- Language: Jupyter Notebook
- Size: 7.3 MB
- Stars: 4
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Achtung Die Kurve!
Experiment to train agent to play Achtung Die Kurve! using reinforcement learning.
(1-4 players)
Try the game yourself:
```
python achtung.py
```
Or play with a friend:
```
python achtung.py 2
```
(achtung.py) Environment - Pygame implementation. Initially based on: https://github.com/janowskipio/FarBy, but extensively modified.
(rl.py) Stochastic policy gradient - 2-layer fully-connect neural network policy: based on: http://karpathy.github.io/2016/05/31/rl/
(mu_zero_achtung.py) MuZero - interface to open-source MuZero implementation: https://github.com/werner-duvaud/muzero-general
(train.py) Stochastic policy gradient - simple ResNet policy and stochastic policy gradient implemented in PyTorch, based on: https://github.com/pytorch/examples/tree/master/reinforcement_learning
(a2c/ppo/dqn.ipynb) Actor Critic - Proximal Policy Optimization - Deep Q-Learning - using Stable Baselines3: https://github.com/DLR-RM/stable-baselines3