Skip to content

Machine learning for a FOSSology server: rigel is for mining of data from conclusions, clearing expert corrections and bulk scans, create a model, use this model for providing a new classifier for licenses.

License

Notifications You must be signed in to change notification settings

ag4ums/FOSSologyML

 
 

Repository files navigation

FOSSologyML

Open Source License Classifier, driven by Machine Learning. FOSSologyML is a toolkit which introduces a new agent for fossology: rigel. This new agent can be used as a standalone rigel-cli or started as a simple server with rigel-server.

Installation

pip install git+https://github.com/fossology/FOSSologyML.git

Or in a develop mode after downloading a zip or cloning the git repository

git clone https://github.com/fossology/FOSSologyML.git
cd
pip install -e .

Once installed you need to download default model and language preprocessing data for english by running

rigel-download-data

Then you can run

rigel-cli --help

or

rigel-server --help

Development

To start all tests run

python setup.py test

To generate documentation with Sphinx run

cd docs
sphinx-apidoc ../FOSSologyML/ -f -o .
make html

To package make sure you have the following installed

pip install --user --upgrade setuptools wheel

and run

python setup.py sdist bdist_wheel

How to build the model

See Documentation <https://>_ for info on how to build your own model and more.

License

SPDX-License-Identifier: GPL-2.0-only

See the file LICENSE.rst

About

Machine learning for a FOSSology server: rigel is for mining of data from conclusions, clearing expert corrections and bulk scans, create a model, use this model for providing a new classifier for licenses.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 96.7%
  • Shell 2.5%
  • Java 0.5%
  • PHP 0.1%
  • C++ 0.1%
  • C 0.1%