Skip to content

Latest commit

 

History

History
38 lines (28 loc) · 1.13 KB

installation.md

File metadata and controls

38 lines (28 loc) · 1.13 KB

Installation and Usage Instructions

This document contains information on how to install packages and datasets.

Installation

Python

For our experiments, we used Python 3.7. Python 3.6 is possible as well. Furthermore, we require the following Python packages:

pip install sklearn
pip install numpy
pip install shapely
pip install matplotlib
pip install "tensorflow >= 1.13.1, < 2"

AIF360

We are using the AI Fairness 360 toolkit for bias mitigation methods and to compute fairness metrics. More information on AIF360 can be found here: https://github.com/Trusted-AI/AIF360

Download AIF360

AIF360 can be installed as follows:

pip install aif360

Download datasets

We use three datasets for our experiments, which can be retrieved online:

  1. Adult: https://archive.ics.uci.edu/ml/datasets/adult
  2. COMPAS: https://github.com/propublica/compas-analysis
  3. German: https://archive.ics.uci.edu/ml/datasets/Statlog+%28German+Credit+Data%29

For more information on the download of datasets, we refer to: https://github.com/Trusted-AI/AIF360/tree/master/aif360/data.