README.rst

Template for Benchopt Benchmark repositories

This repo should be used with the following steps:

1. Hit the Use this template button on the top of this page,
2. Use the form to create a new github repository with your benchmark name,
3. Clone the newly created repository on your computer,
4. Run python clean_template.py script that will remove instruction relative to the template in README.rst and update it with your repo and org name.
5. Edit the problem description in the README.rst.
6. Update objective.py and the files in datasets and solvers to create the benchmark.

My Benchopt Benchmark

Benchopt is a package to simplify and make more transparent and reproducible comparisons of optimization methods. This benchmark is dedicated to solvers of describe your problem:

$$\min_{\beta} f(X, \beta),$$

where $X$ is the matrix of data and $\beta$ is the optimization variable.

Install

This benchmark can be run using the following commands:

$ pip install -U benchopt
$ git clone https://github.com/#ORG/#BENCHMARK_NAME
$ benchopt run #BENCHMARK_NAME

Apart from the problem, options can be passed to benchopt run, to restrict the benchmarks to some solvers or datasets, e.g.:

$ benchopt run #BENCHMARK_NAME -s solver1 -d dataset2 --max-runs 10 --n-repetitions 10

Use benchopt run -h for more details about these options, or visit https://benchopt.github.io/api.html.