See the Spark ML Documentation BinaryClassificationEvaluator
ml_binary_classification_eval(predicted_tbl_spark, label, score,
metric = "areaUnderROC")
The result of running sdf_predict
Name of column string specifying which column contains the true, indexed labels (ie 0 / 1)
Name of column contains the scored probability of a success (ie 1)
The classification metric - one of: areaUnderRoc (default) or areaUnderPR
area under the specified curve