Model interpretation has been an important area of study nowadays and because of that some different techniques have risen to help on this task. Maybe the two most famous ones are LIME and SHAP.
How would you explain the main differences between the two methods in a simple way?
And in which situations one method would maybe be better suited than the other?