Enabling ML explainability with SageMaker Clarify

In the previous two recipes, we used SageMaker Clarify to detect pre-training and post-training bias. In this recipe, we will take a closer look at ML explainability and how we can use SageMaker Clarify to generate an ML explainability report.

We will see the importance of ML explainability as we deal with ethical and legal concerns. For example, customers will want a better idea of how their information is used by a machine learning system to perform recommendations or predictions. In addition to this, ML explainability empowers data scientists and machine learning practitioners to make more accurate and fair models.

Note

It is important to distinguish model interpretability from model explainability. Model interpretability focuses on understanding what a machine learning model is doing internally. On the other hand, model explainability involves understanding how a machine learning model performed a prediction using certain...