Skip to content

Latest commit

 

History

History
25 lines (15 loc) · 1.32 KB

README.md

File metadata and controls

25 lines (15 loc) · 1.32 KB
Raw

Restarts subject to approximate sharpness

MATLAB code for the numerical experiments in the preprint Restarts subject to approximate sharpness: A parameter-free and optimal scheme for first-order methods. This repository may be updated as the article undergoes review.

Requirements

The experiments should run on MATLAB R2020b (or a later version) without issue.

  • linspecer function (link) (simply include linspecer.m in your MATLAB userpath)
  • CVX (link)
  • Statistics and Machine Learning Toolbox

Running the experiments

Clone or download the repository, and set the MATLAB path to be from the repository root. The experiments are located in the experiments/ folder, organized by the subsections in Section 5 of the paper.

Attributions

The datasets in data/libsvm-data.tar.bz2 are obtained from LIBSVM. The wine data in data/winequality.tar.bz2 is obtained from the UCI machine learning repository.

Issues

Pertaining to the code, post questions, requests, and bugs in Issues.