evalml.objectives.FraudCost.objective_function

FraudCost.objective_function(y_true, y_predicted, X)[source]

Calculate amount lost to fraud per transaction given predictions, true values, and dataframe with transaction amount.

Parameters

  • y_predicted (ww.DataColumn, pd.Series) – Predicted fraud labels

  • y_true (ww.DataColumn, pd.Series) – True fraud labels

  • X (ww.DataTable, pd.DataFrame) – Data with transaction amounts

Returns

Amount lost to fraud per transaction

Return type

float

evalml.objectives.FraudCost.is_defined_for_problem_type evalml.objectives.FraudCost.optimize_threshold