https://github.com/ankostis/evilometer

a human-tunable classifier for names, how evil they sound - coded as a toy
https://github.com/ankostis/evilometer

Last synced: 21 days ago
JSON representation

a human-tunable classifier for names, how evil they sound - coded as a toy

• Host: GitHub
• URL: https://github.com/ankostis/evilometer
• Owner: ankostis
• License: agpl-3.0
• Created: 2019-06-16T21:01:49.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2021-02-23T11:28:10.000Z (over 3 years ago)
• Last Synced: 2024-06-13T15:35:08.802Z (5 months ago)
• Language: Python
• Homepage:
• Size: 31.3 KB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.rst
  • Changelog: CHANGES.txt
  • License: COPYING.txt

Awesome Lists containing this project

README

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

evilometer: Rates arbitrary names based on a pre-rated list of names on some characteristic (ie "evilness")

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

  * Copyright : July-2014, (c) AGPLv3 or later

  * Developed : by [email protected] & `Nikifors Zacharoff `_

  * License   : GNU Affero General Public License v3 or later (AGPLv3+)

Overview

========

Given a pre-rated list of names on some characteristic,

it decomposes them using n_grams and applies information retrieval rating[inv_index]_

to estimate the rating of any other name on that characteristic.

Install:

========

To install it, assuming you have download the sources,

do the usual::

    python setup.py install

Or get it directly from the PIP repository::

    pip3 install evilometer

Tested with Python 3.4.

Usage:

======

Fuefit accepts as input 2 vectors of names, the pre-rated "training" set of and

the set of names to be rated, like that::

    >> import evilometer

    >> train_names = {'trendy': 1, 'good':2, 'better':2, 'talon':-5, 'bad_ass':-10}

    >> asked_names = {'kolon':1, 'trekking':2, 'trepper':-10}

    >> name_scores = evilometer(train_names, asked_names)

    >> print_scored_names_sorted(name_scores)

cmdline

-------

    >> python -m evilometer prerated.csv asked.txt

Contributors

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

* Nikifors Zacharoff & Nikias Fontaras:  Original idea of the game

.. rubric:: Footnotes

.. [inv_index] http://nlp.stanford.edu/IR-book/html/htmledition/inverse-document-frequency-1.html