https://github.com/stephane-caron/lpsolvers

Linear programming solvers in Python with a unified API
https://github.com/stephane-caron/lpsolvers

linear-programming numerical-optimization python solver

Last synced: 2 months ago
JSON representation

Linear programming solvers in Python with a unified API

• Host: GitHub
• URL: https://github.com/stephane-caron/lpsolvers
• Owner: stephane-caron
• License: lgpl-3.0
• Created: 2018-07-03T08:58:53.000Z (almost 7 years ago)
• Default Branch: main
• Last Pushed: 2024-09-16T20:27:07.000Z (9 months ago)
• Last Synced: 2024-10-13T16:31:11.290Z (8 months ago)
• Topics: linear-programming, numerical-optimization, python, solver
• Language: Python
• Homepage:
• Size: 6.12 MB
• Stars: 23
• Watchers: 3
• Forks: 4
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# LP Solvers for Python

[![CI](https://img.shields.io/github/actions/workflow/status/stephane-caron/lpsolvers/test.yml?branch=main)](https://github.com/stephane-caron/lpsolvers/actions)

[![Documentation](https://img.shields.io/github/actions/workflow/status/stephane-caron/lpsolvers/docs.yml?branch=main&label=docs)](https://stephane-caron.github.io/lpsolvers/)

[![Coverage](https://coveralls.io/repos/github/stephane-caron/lpsolvers/badge.svg?branch=main)](https://coveralls.io/github/stephane-caron/lpsolvers?branch=main)

[![Conda version](https://img.shields.io/conda/vn/conda-forge/lpsolvers.svg)](https://anaconda.org/conda-forge/lpsolvers)

[![PyPI version](https://img.shields.io/pypi/v/lpsolvers)](https://pypi.org/project/lpsolvers/)

Wrapper around Linear Programming (LP) solvers in Python, with a unified interface.

## Installation

### From conda-forge

```console

conda install -c conda-forge lpsolvers

```

### From PyPI

To install the library and all available LP solvers at the same time:

```console

pip install lpsolvers[open_source_solvers]

```

To install the library only, assuming LP solvers are installed separately: ``pip install lpsolvers``.

## Usage

The function [`solve_lp`](https://stephane-caron.github.io/lpsolvers//linear-programming.html#lpsolvers.solve_lp) is called with the ``solver`` keyword argument to select the backend solver. The linear program it solves is, in standard form:

$$

\begin{split}

\begin{array}{ll}

    \mbox{minimize} &

        c^T x \\

    \mbox{subject to}

        & G x \leq h \\

        & A x = b

\end{array}

\end{split}

$$

Vector inequalities are taken coordinate by coordinate.

## Example

To solve a linear program, build the matrices that define it and call the ``solve_lp`` function:

```python

from numpy import array

from lpsolvers import solve_lp

c = array([1., 2., 3.])

G = array([[1., 2., -1.], [2., 0., 1.], [1., 2., 1.], [-1., -1., -1.]])

h = array([4., 1., 3., 2.])

x = solve_lp(c, G, h, solver="cvxopt")  # select solver here

print(f"LP solution: {x=}")

```

This example outputs the solution ``[2.2, -0.8, -3.4]``.

## Solvers

The list of supported solvers currently includes:

- [cdd](https://github.com/mcmtroffaes/pycddlib)

- [CVXOPT](http://cvxopt.org/)

- [CVXPY](https://www.cvxpy.org/) (interface)

- [PDLP](https://developers.google.com/optimization/lp/pdlp_math)

- [ProxQP](https://github.com/Simple-Robotics/proxsuite#proxqp)