Fo-XAI library for explainable AI
Explainable AI (XAI) refers to a collection of techniques and methods that enable human users to understand and have confidence in the outcomes and decisions generated by machine learning algorithms.
In some cases, it's not enough for a machine learning model to produce accurate predictions; it's also essential to interpret and explain each result. Want to explain the reasoning behind the models you're training?
Try Fo-XAI, our new library that puts focus on the Explainable AI (XAI) algorithms.
Fo-XAI operates as a conglomeration of already established libraries that encompass diverse XAI algorithms. The goal is to make using XAI easier in machine learning projects. Also to encourage more people to use XAI in their work.
At present, Fo-XAI only caters to computer vision problems, however, there are plans to expand the scope of supported algorithms to include text, tabular, and multimodal data issues in the future.
Advantages of using Fo-XAI
- developers can be sure that the training model is working as expected,
- developers focus on training the right models and are able to stop and iterate their assumptions more accurately,
- a comprehensive understanding and interpretation of the behaviour and predictions of the machine learning systems under development is shared among all project participants.