When distributing requests (in the form of queries or seed documents) between experimental systems, their counters will be increased upon delivering results. Make sure pre-computed/live systems as well as ranking/recommender systems are handled separately.
breuert
commented
3 years ago
When distributing requests (in the form of queries or seed documents) between experimental systems, their counters will be increased upon delivering results. Make sure pre-computed/live systems as well as ranking/recommender systems are handled separately.