Description

This is a repository of the code for the From ℓ1 subgradient to projection: A compact neural network for ℓ1-regularized logistic regression paper.

Getting Started

Dependencies

• Python 3.6+
• PyTorch 1.6+
• sklearn 0.23
• Matplotlib (for graphs and figures)

Installing

• Download repository
• Install Dependencies
• Datasets must be in Libsvm format, download and put them in the datasets folder in the root of the project. Download link available in the Acknowledgments.

Executing program

• Run each file in form of run_(dataset_name).py to obtain corresponding results of the proposed method and sklearn LogisticRegression model

• Run each file in form of script_(figure's_name).py to generate the paper's figures

• Note: Results that acquired from all methods available in .mat format in the results folder.

Authors

• Amir Atashin
• Majid Mohammadi

License

This project is licensed under the MIT License - see the LICENSE file for details

Citation

Please consider referencing the following research paper of this repository if you find it useful or relevant to your research:

@article{MOHAMMADI2023,
    title = {From ℓ1 subgradient to projection: A compact neural network for ℓ1-regularized logistic regression},
    journal = {Neurocomputing},
    year = {2023},
    issn = {0925-2312},
    doi = {https://doi.org/10.1016/j.neucom.2023.01.021},
    url = {https://www.sciencedirect.com/science/article/pii/S0925231223000310},
    author = {Majid Mohammadi and Amir Ahooye Atashin and Damian A. Tamburri},
}

Acknowledgments

Inspiration, code snippets, etc.

• LIBSVM Datasets
• Results of other methods obtained from L1General.zip, a package by Mark Schmidt