SimpleModelComparison.add_condition_gain_greater_than#
- SimpleModelComparison.add_condition_gain_greater_than(min_allowed_gain: float = 0.1, max_gain: float = 50, classes: Optional[List[Hashable]] = None, average: bool = False)[source]#
Add condition - require gain between the model and the simple model to be greater than threshold.
- Parameters
- min_allowed_gainfloat , default: 0.1
Minimum allowed gain between the model and the simple model - gain is: difference in performance / (perfect score - simple score)
- max_gainfloat , default: 50
the maximum value for the gain value, limits from both sides [-max_gain, max_gain]
- classesList[Hashable] , default: None
Used in classification models to limit condition only to given classes.
- averagebool , default: False
Used in classification models to flag if to run condition on average of classes, or on each class individually