466ef0befceaad2be3244efd724c76c90869b17d,tensorrec/eval.py,,ndcg_at_k,#Any#Any#Any#Any#Any#Any#,69

Before Change

    ranks_of_relevant = sp.csr_matrix(predicted_ranks *
                                      positive_test_interactions.A)

    relevance = sp.csr_matrix(
        test_interactions.A *
        positive_test_interactions.A
    )

    k_mask = np.less(ranks_of_relevant.data, k + 1)
    ror_at_k = np.maximum(np.multiply(ranks_of_relevant.data, k_mask), 1)

    relevance_at_k = (2**np.multiply(relevance.data, k_mask)) - 1
    ranks_of_relevant.data = relevance_at_k/np.log2(ror_at_k + 1)  // ranks at 1

    dcg = ranks_of_relevant.sum(axis=1).flatten()
    idcg = np.apply_along_axis(_idcg, 1, relevance.A)

After Change

    predicted_ranks = model.predict_rank(user_features=user_features,
                                         item_features=item_features)

    relevance, k_mask, ranks_of_relevant, ror_at_k = _setup_ndcg(predicted_ranks,
                                                                 test_interactions,
                                                                 k)

Instances