Skip to main content

A set of Python modules for the Partial Label Ranking problem.

Project description

Build Status PyPI version

plr

plr is a Python module for dealing with the Partial Label Ranking problem.

Prerequisites

plr requires:

* Python (>= 3.6)
* Numpy (>= 1.15.2)
* Scipy (>= 1.1.0)

and Linux or Mac OS X operating systems. Windows is not currently supported.

Installation

The easiest way to install plr is using pip package:

pip install plr

Development

Feel free to contribute to the package, but be sure that the standards are followed.

Source code

The latest sources can be obtained with the command:

git clone https://github.com/alfaro96/plr.git

Setting up a development environment

To setup the development environment, it is strongly recommended to use docker tools (from outside the source directory). First, the image must be built.

docker build -t alfaro96/plr:development .

Or:

make docker-build

Then, the docker container is executed (from outside the source directory) with:

docker run -ti -v $(pwd)/:/home/plr/workspace/ --rm alfaro96/plr:development

Or:

make docker-run

In fact, both commands can be executed at once with:

make docker

Alternatively, one can use Python virtual environments (see [https://docs.python.org/3/library/venv.html] for details).

Testing

After installation, the test suite can be executed, from outside the source directory, with:

pytest plr

or

make test-code

Authors

* Alfaro Jiménez, Juan Carlos
* Aledo Sánchez, Juan Ángel
* Gámez Martín, José Antonio

License

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

Project details


Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page