Metadata-Version: 2.1
Name: ailiga
Version: 0.1.0
Summary: 
Author: Alexander Puck Neuwirth
Author-email: alexander@neuwirth-informatik.de
Requires-Python: >=3.7,<3.11
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.7
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Requires-Dist: PettingZoo (<1.22.0)
Requires-Dist: SuperSuit
Requires-Dist: cfg_load
Requires-Dist: pyglet (==1.5.27)
Requires-Dist: rlcard
Requires-Dist: tianshou (==0.4.10)
Requires-Dist: torch (==1.13.0)
Requires-Dist: tqdm
Description-Content-Type: text/markdown

# AILiga


## Installation

```sh
poerty install
poetry shell
```




## Testing and Training

Currently, training/testing fighters works through the fighter tests.
```sh
python tests/test_dqn_fighter.py
```

## Tensorboard

```sh
tensorboard --logdir log/ --load_fast=false
```


## Limitations

Currently, the implementation through `tianshou.BasePolicy` seems to only support DQNPolicy and also not `Discrete()` observation spaces.

## References

### Frameworks

* https://github.com/Farama-Foundation/PettingZoo
* https://github.com/vwxyzjn/cleanrl
* https://github.com/Farama-Foundation/Gymnasium
* https://github.com/deepmind/open_spiel
* https://github.com/datamllab/rlcard
* https://tianshou.readthedocs.io/en/master/

### Books

* http://incompleteideas.net/book/the-book-2nd.html


## Development

We use black through

### package/python structure:

* https://mathspp.com/blog/how-to-create-a-python-package-in-2022
* https://www.brainsorting.com/posts/publish-a-package-on-pypi-using-poetry/

