This function takes a matrix as input and transforms it into the specified format. The matrix in input can be either sparse or ndarray. I

Checks whether the URM in the list have an empty intersection, therefore there is no data point contained in more than one URM at a time

:param epochs_max: max number of epochs the training will last :param epochs_min: min number of ep

Load config, equivalent to loading json and updating this classes' dict

Checks whether the URM in the list only contain implicit data in the form 1 or 0 :param URM_list: :return:

Clean temporary folder only if the default one :return:

Evaluate the performance (Hit_Ratio, NDCG) of top-K recommendation Return: score of each test rating.

The function selects the TopK most similar elements, column-wise :param item_weights: :param forceSparseOutput: :param k: :param

:param recommender_object: the trained recommender object, a BaseRecommender subclass :param URM_test_list: list of URMs to test the

Reciprocal rank of the FIRST relevant item in the ranked list (0 if none) :param is_relevant: boolean array :return:

Evaluate the performance (Hit_Ratio, NDCG) of top-K recommendation Return: score of each test rating.

Convert a dictionary to a sparse matrix

Evaluate the performance (Hit_Ratio, NDCG) of top-K recommendation Return: score of each test rating.

normalized discounted cumulative gain@k for binary relevance ASSUMPTIONS: all the 0's in heldout_data indicate 0 relevance

Items are assumed to be on rows :param dataMatrix: :return:

Evaluate the performance (Hit_Ratio, NDCG) of top-K recommendation Return: score of each test rating.