Feature importances are measures that indicate how much individual input features contribute to a model’s predictions or performance. They are used to interpret models, compare variables, and identify influential signals in a dataset.
Synonym for variable importances.