oedokumaci / gale-shapley

Python implementation of the Gale-Shapley Algorithm.

 

This is a Python implementation of the celebrated Gale-Shapley (a.k.a. the Deferred Acceptance) Algorithm.

Time complexity is O(n^2), space complexity is O(n).

 

 

User Guide

Requirements

• Python >= 3.9
• OS: Ubuntu, MacOS, Windows

Installation

There are three easy ways to install the package.

Using Docker

An easy way to run the project is to use Docker. First, install Docker. Then, run the following command in the project directory.

docker build -t gale-shapley .

This will build the Docker image. After the image is built, the following are some examples of how to run the project.

docker run --rm -it \
-v $(pwd)/config/example_config_custom_input.yaml:/usr/src/app/config/config.yaml \
-v $(pwd)/logs:/usr/src/app/logs \
-e number_of_simulations=$(number_of_simulations) \
gale-shapley

These commands will run the project with the specified config file and number of simulations. The output can be seen in the terminal. The -v option mounts the specified config file and logs directory to the container. The -e option sets the environment variable number_of_simulations to the specified value. The --rm option removes the container after it exits. The -it option is for interactive mode. You can also use Makefile recipes to run the project with Docker. See the Makefile section for more information.

Download Zip File

Pip installing the package from PyPI is not yet available. Instead, download from this link and unzip. You will also need to change the folder name from gale-shapley-main to gale-shapley (or cd into gale-shapley-main in step 2 below).

Using Git

If you have git installed, simply run

git clone https://github.com/oedokumaci/gale-shapley

to install the package locally. After downloading, here are the steps to install the dependencies in a virtual environment using PDM:

1. pip install pdm
2. cd gale-shapley
3. pdm install --prod

Usage

Configuration

First edit the ./config/config.yaml to your liking. Example config files can be found at ./config/example_config_*.

Quick Start

After configuring the ./config/config.yaml, simply run the following command in the project directory.

pdm run python -m gale_shapley

Detailed Usage

For a list of all the CLI arguments and options, run

pdm run python -m gale_shapley --help

A sample output with currently implemented CLI arguments and options is shown below.

 

Developer Guide

Makefile

There is a Makefile in the project directory. You can run make help to see the available commands as below. The Makefile is also used in the CI/CD pipeline.

Setup

This project is PDM-managed, which is compatible with PEP 621 (also compatible with the rejected PEP 582). If you are a developer, first pip install pdm and then git clone the project. Next you can pdm install in the project directory, which will install all the dependencies in a virtual environment.

Development

Pre-commit Hooks

The project also uses pre-commit hooks. Because the project uses PDM, you do not need to pip install pre-commit. Instead, run directly

pdm run pre-commit install

in the project directory to install hooks to your local .git. Alternatively, you can also activate the virtual environment and run

pre-commit install