io.prediction.engines.itemrank
Return a list of Query-Actual pair for evaluation.
Return a list of Query-Actual pair for evaluation.
It constructs a list of Query-Actual pair using the list of actions. For each user in the list, it creates a Query instance using all items in actions, and creates an Actual instance with all actions associated with the user. Note that it is the metrics job to decide how to interprete the semantics of the actions.