Entropy Triangle Package

A Python package to work with entropic coordinates and the entropy triangles defined in:

• 100% Classification Accuracy Considered Harmful: The Normalized Information Transfer Factor Explains the Accuracy Paradox

• The Evaluation of Data Sources using Multivariate Entropy Tools

• Assessing Information Transmission in Data Transformations with the Channel Multivariate Entropy Triangle

The packages uses NumPy, Pandas, SciPy, Scikit-learn and Matplotlib

You may find the theoretical framework and further support in

• Evaluation of Machine Learning Techniques by Entropic Means

Package Installation

Pip

The package is stored in PyPI

Using pip you can install the current release (0.1.1) using the command:

pip install entropytriangle

Note that with pip you may need to use sudo.

sudo pip install entropytriangle

With setup.py

Alternatively you can clone the repository and run setup.py in the usual manner:

git clone https://github.com/Jaimedlrm/entropytriangle.git
cd entropytriangle
python setup.py install

Note that you may need

sudo python setup.py install

Anaconda

The package is stored in Anaconda Cloud

It is just available at the moment for osx-64

You can install entropytriangle with the conda command:

conda config --add channels Jaimedlrm
conda install entropytriangle

or :

conda install -c jaimedlrm entropytriangle

Usage and Examples

You can see here some use cases in this link, where you can find some examples of real case scenarios.