In the sparkling water pipeline examples, when there is a prediction for binary classification there is a column named "prediction" with the predicted label. Example:
def isSpam(smsText, model):
smsTextDF = spark.createDataFrame([(smsText,)], ["text"]) # create one element tuple
prediction = model.transform(smsTextDF)
return prediction.select("prediction").first() == "spam"
However, I'm running an AutoML estimator on a spark pipeline and the output column is named "prediction_output" that is a struct with the probability of p0 and p1. There is not a "prediction" column with the predicted label.
Sparkling Water Version: 2.4.13
H2O cluster version: 126.96.36.199