README.md

October 8, 2024 ยท View on GitHub

Installation | Examples | Requirements

MLX_BeyondLanguage showcases examples of applying the MLX machine learning framework to Engineering Physics problems.

Some key features of MLX_BeyondLanguage

Feature 1: Uses the Python API of MLX

Feature 2: The included examples showcase some of the advantages of using MLX, e.g. Composable Function Transformations, Lazy Computation, and Unified Memory.

Feature 3: The provided code is designed for execution on Apple M-series Silicon.

Installation

  1. Clone the repository: 
    git clone https://github.com/sck-at-ucy/MLX_BeyondLanguage.git

Examples

Currently, this repo contains a single example which provides the code described in Beyond Language: Applying MLX Transformers to Engineering Physics

Requirements

All requirements can be installed with pip install -r requirements.txt or by installing the packages individually:

  • numpy
  • matplotlib
  • numba
  • seaborn
  • mlx

Citing MLX_BeyondLanguage

If you find MLX_BeyondLanguage useful in your research and wish to cite it, please use the following BibTex entries:

@software{mlxbeyond2024,
  author = {Stavros Kassinos and Alessio Alexiadis},
  title = {{MLX_BeyondLanguage}: Applications of the MLX Machine Learning Framework to Engineering Physics},
  url = {https://github.com/sck-at-ucy/MLX_BeyondLanguage},
  version = {0.0.1},
  year = {2024},
}

@article{mlx_beyond_language2024,
  author = {Stavros Kassinos and Alessio Alexiadis},
  title = {Beyond Language: Applying MLX Transformers to Engineering Physics},
  journal = {arXiv},
  year = {2024},
  url = {https://doi.org/10.48550/arXiv.2410.04167}
}

Citing MLX

The MLX software suite was initially developed with equal contribution by Awni Hannun, Jagrit Digani, Angelos Katharopoulos, and Ronan Collobert. If you also wish to cite MLX, please use the following BibTex entry:

@software{mlx2023,
  author = {Awni Hannun and Jagrit Digani and Angelos Katharopoulos and Ronan Collobert},
  title = {{MLX}: Efficient and flexible machine learning on Apple silicon},
  url = {https://github.com/ml-explore},
  version = {0.0},
  year = {2023},
}