# turicreate.recommender.ranking_factorization_recommender.RankingFactorizationRecommender.evaluate_rmse¶

RankingFactorizationRecommender.evaluate_rmse(dataset, target)

Evaluate the prediction error for each user-item pair in the given data set.

Parameters: dataset : SFrame An SFrame in the same format as the one used during training. target : str The name of the target rating column in dataset. out : dict A dictionary with three items: ‘rmse_by_user’ and ‘rmse_by_item’, which are SFrames containing the average rmse for each user and item, respectively; and ‘rmse_overall’, which is a float.

Examples

>>> import turicreate as tc
>>> sf = tc.SFrame('https://static.turi.com/datasets/audioscrobbler')
>>> train, test = tc.recommender.util.random_split_by_user(sf)
>>> m = tc.recommender.create(train, target='target')
>>> m.evaluate_rmse(test, target='target')