This is code for our work The ARC Toolbox: Artificial Languages with Rhythmicity Control
The following describes how you can set up the software and run the experiments from the paper.
Create a new virtual environment, e.g. using anaconda
conda create -n test_arc python=3.9
... and activate it
conda activate test_arc
The simplest is to clone this repository and install ARC in editable mode:
pip install -e .
If you want to use ARC as a package, you can install it directly from git with
pip install git+https://github.com/milosen/arc.git
Clone this repository. Install jupyter
pip install jupyter
If you use a virtual environement, you also need to install the ipython-kernel:
python -m ipykernel install --user --name=arc
In this case, don't forget to select the arc
kernel in the jupyter session's kernel option (Kernel -> Change kernel -> arc).
Start jupyter
jupyter notebook
and select the notebook you want.
controlled_stream_generation.ipynb
briefly demonstrates ARC usagedata_and_stats_from_the_paper.ipynb
reproduces the data for the figures and the appendices of the paperplots_from_the_paper.ipynb
reproduces the figures of the paper- Optional: If you want to adapt ARC to your own research, you'll probably want to take a look at
arc_types.ipynb
for an intro to ARC's type system.
Please cite our work as
@article {Titone2024arc,
author = {Titone, Lorenzo and Milosevic, Nikola and Meyer, Lars},
title = {The ARC Toolbox: Artificial Languages with Rhythmicity Control},
elocation-id = {2024.05.24.595268},
year = {2024},
doi = {10.1101/2024.05.24.595268},
publisher = {Cold Spring Harbor Laboratory},
URL = {https://www.biorxiv.org/content/early/2024/05/24/2024.05.24.595268},
eprint = {https://www.biorxiv.org/content/early/2024/05/24/2024.05.24.595268.full.pdf},
journal = {bioRxiv}
}