evalml.objectives.BinaryClassificationObjective

class evalml.objectives.BinaryClassificationObjective[source]

Base class for all binary classification objectives.

problem_types (list(ProblemType)): List of problem types that this objective is defined for. Set to [ProblemTypes.BINARY]. can_optimize_threshold (bool): Determines if threshold used by objective can be optimized or not.

Methods

__init__

Initialize self.

calculate_percent_difference

Calculate the percent difference between scores.

decision_function

Apply a learned threshold to predicted probabilities to get predicted classes.

is_defined_for_problem_type

objective_function

Computes the relative value of the provided predictions compared to the actual labels, according a specified metric

optimize_threshold

Learn a binary classification threshold which optimizes the current objective.

score

Returns a numerical score indicating performance based on the differences between the predicted and actual values.

validate_inputs

Validates the input based on a few simple checks.

Class Inheritance

Inheritance diagram of BinaryClassificationObjective