Tutorials

The FAT Forensics tutorials aim to get you up to speed and build up your confidence in using the package. They are intended for the beginners and each tutorials aims to teach you how to solve a particular FAT-related issue of a predictive modeling task: data, models and/or predictions.

Before continuing, please make sure that you have installed the FAT Forensics package and you are ready to follow the tutorials. To check whether fatf is installed, launch a Python interpreter and check the version of the package:

$ python
>>> import fatf
>>> fatf.__version__
'0.1.2'

Note

Doctest Mode

The code-examples in the Tutorials and How-to Guides are written in a python-console format. If you wish to easily execute these examples in IPython, use the:

%doctest_mode

magic command in the IPython-console. You can then simply copy and paste the examples directly into IPython without having to worry about removing the >>> manually.

Tutorials Content

• Exploring the Grouping Concept – Defining Sub-Populations
• Using Grouping to Evaluate Fairness of Data and Models – Group-Based Fairness
• Using Grouping to Evaluate Robustness of Data and Models
• Explaining a Machine Learning Model: ICE and PD
• Explaining Machine Learning Predictions: LIME and Counterfactuals

Note

Additional learning resources are available on the FAT Forensics events website.