/oteann3

Primary LanguageJupyter NotebookMIT LicenseMIT

OTEANN

This repository is the official implementation of OTEANN: Estimating the Transparency of Orthographies with an Artificial Neural Network.

Requirements

  • Require Ubuntu 18.04, Python 3.6+.

  • Download files on your machine

  • Go to the oteann3 main directory

    • cd oteann3

  • Create a virtual environment

    • python3 -m venv oteann

  • Activate virtual environment

    • source oteann/bin/activate

  • Load the python librairies needed for OTEANN (e.g. numpy, pandas, torch...) from the requirements file

    • python -m pip install -r requirements.txt

  • Extract the subdatasets (required local free spacesize=334Mo)

    • python extract_subdatasets.py

Evaluation and Training

In order to run this code, you need to:

  • Run the oteann.ipynb in order to create the ANN model, train it, test it and display the results of the paper.

Results

OTEANN achieves the following performance:

orthography write score read score
ent 99.8 ± 0.1 99.8 ± 0.2
eno 00.0 ± 0.0 00.0 ± 0.0
ar 83.8 ± 1.4 99.4 ± 0.2
br 79.4 ± 1.7 77.1 ± 1.1
de 69.1 ± 1.7 76.9 ± 1.0
en 36.4 ± 1.3 30.6 ± 1.0
eo 99.5 ± 0.3 99.7 ± 0.2
es 67.2 ± 1.4 85.8 ± 0.7
fi 97.5 ± 0.5 92.9 ± 0.7
fr 28.1 ± 1.7 78.6 ± 0.8
fro 99.2 ± 0.2 89.0 ± 0.9
it 94.6 ± 1.0 72.0 ± 1.9
ko 81.9 ± 1.1 97.7 ± 0.4
nl 72.6 ± 1.5 56.6 ± 1.8
pt 75.2 ± 1.6 82.2 ± 1.2
ru 41.9 ± 1.3 97.4 ± 0.7
sh 99.1 ± 0.3 54.8 ± 3.3
tr 95.4 ± 0.6 95.6 ± 0.8
zh 20.4 ± 1.3 78.4 ± 1.1

License

OTEANN uses minGTPT (https://github.com/karpathy/minGPT) which is released under the MIT licence.

Citation

This code was used for the following paper:

@misc{marjou2020oteann,
      title={OTEANN: Estimating the Transparency of Orthographies with an Artificial Neural Network}, 
      author={Xavier Marjou},
      year={2020},
      eprint={1912.13321},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}