I am glad to announce the first official release of Fairness.jl : A Comprehensive Bias Audit and Mitigation Toolkit in Julia designed with the aim to solve practical problems faced by practitioners with existing toolkits.
It is also the first Fairness Toolkit of Julia!!
Blog Post on Introduction to Fairness.jl: https://nextjournal.com/ashryaagr/fairness
Github Repository: https://github.com/ashryaagr/Fairness.jl
As of writing, the package is registered in Julia General registry, with version 0.1.1
This Toolkit offers a number of unique features over the fairness toolkits in other languages such as Python, R, etc. :
As of writing, it is the only bias audit and mitigation toolkit to support data with multi-valued protected attributes. For eg. If the protected attribute, say race has more than 2 values: “Asian”, “African”, “American” and so on, then Fairness.jl can easily handle it with normal workflow.
Multiple Fairness algorithms can be applied at the same time by wrapping the wrapped Model.
Due to the support for multi-valued protected attributes, intersectional fairness can also be dealt with this toolkit. For eg. If the data has 2 protected attributes, say race and gender, then Fairness.jl can be used to handle it by combining attributes like “female_american”, “male_asian” and so on.
Extensive support and functionality provided by MLJ can be leveraged when using Fairness.jl
It leverages the flexibility and speed of Julia to make it more efficient and easy-to-use at the same time
The concept of Fairness Tensor has been used to optimize the package.
Requests/suggestions, PRs are more than welcome