Example Notebooks

Here’s a list of examples on how to use the library. We will be adding more examples soon. If you’re interested in contributing to existing notebooks or adding new ones please consult the guide on Contributing example notebooks.

Note

The Fairlearn API is still evolving, so if you want to run these on your local Fairlearn installation, make sure to match versions.

• GridSearch with Census Data
  • Load and preprocess the data set
  • Training a fairness-unaware predictor
  • Mitigation with GridSearch
• Metrics with Multiple Features
  • Getting the Data
  • Analysing the Model with Metrics
  • Quantifying Disparities
  • Intersections of Features
  • Control Features
• Making Derived Metrics
  • Getting the Data
  • Creating a derived metric
  • Pregenerated Metrics

Selection rates in census dataset

GridSearch with Census Data

Making Derived Metrics

Metrics with Multiple Features