https://github.com/datapuzzler/multilevel_py

A library that simplifies the construction of deep classification hierarchies
https://github.com/datapuzzler/multilevel_py

Last synced: 12 months ago
JSON representation

A library that simplifies the construction of deep classification hierarchies

• Host: GitHub
• URL: https://github.com/datapuzzler/multilevel_py
• Owner: dataPuzzler
• License: bsd-2-clause
• Created: 2020-12-30T09:57:24.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2021-08-13T12:47:52.000Z (almost 5 years ago)
• Last Synced: 2025-02-18T01:48:07.708Z (over 1 year ago)
• Language: Python
• Size: 2.71 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.rst
  • Changelog: CHANGES.rst
  • License: LICENSE

Awesome Lists containing this project

README

          
.. image:: https://github.com/dataPuzzler/multilevel_py/workflows/Run%20unittests/badge.svg

	:target: https://github.com/dataPuzzler/multilevel_py/workflows/Run%20unittests/badge.svg

	:alt: Unittests Badge

.. image:: https://readthedocs.org/projects/multilevel-py/badge/?version=latest

    :target: https://readthedocs.org/projects/multilevel-py/badge/?version=latest

    :alt: Docs Badge

multilevel_py

=============

*Multilevel_py* is a library that simplifies the construction of classification hierarchies over more than two levels.

The framework depends on python3 only and implements a "deep instantiation" mechanism using pythons metaprogramming

facilities. In academia, the addressed topic is also discussed under the term "Multilevel (Meta-) Modelling". Since

there is no corresponding framework in the python community until this point, multilevel_py was built to fill this gap.

Installing

----------

Install and update using `pip`_:

.. code-block:: text

    # Python only

    pip install multilevel_py 

    

    # with graphical syntax

    pip install multilevel_py[viz] 

Note that for using the graphical syntax, an installation of the underlying visualisation engine `graphviz`_ is required. 

A Simple Example

----------------

 

 The following code constructs a classification structure that spans three levels.

.. code-block:: python

    from multilevel_py.constraints import is_int_constraint, is_str_constraint

    from multilevel_py.core import create_clabject_prop, Clabject

    

    Breed = Clabject(name="Breed")

    yearReg = create_clabject_prop(n="yearReg", t=1, f=0, i_f=True, c=[is_int_constraint])

    age = create_clabject_prop(n="age", t=2, f=0, i_f=True, c=[is_int_constraint])

    Breed.define_props([yearReg, age])

    

    Collie = Breed(name="Collie", init_props={"yearReg": 1888})

    lassie = Collie(name="Lassie", init_props={"age": 7}, declare_as_instance=True)

Visualisation

-------------

Using the viz module, the following graph can be rendered for the previous

example:

.. image:: https://github.com/dataPuzzler/multilevel_py/blob/master/docs/images/collie_chain_initial_example.png

    :width: 400

    :alt: Visulisation of the collie example

Links

-----

* Github Repository: https://github.com/dataPuzzler/multilevel_py

* Documentation: https://multilevel-py.readthedocs.io/en/latest/

* Releases: https://pypi.org/project/multilevel-py/#description

.. _pip: https://pip.pypa.io/en/stable/quickstart/

.. _graphviz: https://graphviz.org/