navicto / Discretization-MDLPC

Discretization with Fayyad and Irani's minimum description length principle criterion (MDLPC)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Discretization MDLP

Python implementation of Fayyad and Irani's MDLP criterion discretiation algorithm

Reference:

Irani, Keki B. "Multi-interval discretization of continuous-valued attributes for classification learning." (1993).

Changes: This new release is compatible with Scikit-learn, and may be used as a part of a Scikit-learn pipeline. The old version is still available in release v0.9.

Dependencies:

  1. Numpy: Tested on version 1.14.5
  2. Scikit-learn: Tested on version 0.19.1

Python version Tested on python 2.7.5

About

Discretization with Fayyad and Irani's minimum description length principle criterion (MDLPC)

License:GNU General Public License v3.0


Languages

Language:Python 100.0%