recpack.metrics.DiscountedGainK

class recpack.metrics.DiscountedGainK(K)

Computes the discounted gain of every item in the Top-K recommendations of a user.

Relevant items that are ranked higher in the Top-K recommendations have a higher gain.

Detailed results show the gain of each item in the list of Top-K recommended items for every user.

For each item \(i \in \text{TopK}(u)\) the discounted gain is computed as

\[\text{DiscountedGain(u,i)} = \frac{y^{true}_{u,i}}{\log_2(\text{rank}(u,i) + 1)}\]
Parameters

K (int) – Size of the recommendation list consisting of the Top-K item predictions.

Methods

calculate(y_true, y_pred)

Computes metric given true labels y_true and predicted scores y_pred.

Attributes

col_names

The names of the columns in the results DataFrame.

name

Name of the metric.

num_items

Dimension of the item-space in both y_true and y_pred

num_users

Dimension of the user-space in both y_true and y_pred after elimination of users without interactions in y_true.

results

Get the detailed results for this metric.

value

Global metric value obtained by summing up scores for every user then taking the average over all users.
calculate(y_true: scipy.sparse._csr.csr_matrix, y_pred: scipy.sparse._csr.csr_matrix) → None

Computes metric given true labels y_true and predicted scores y_pred. Only Top-K recommendations are considered.

Detailed metric results can be retrieved with results. Global aggregate metric value is retrieved as value.

Parameters

  • y_true (csr_matrix) – True user-item interactions.

  • y_pred (csr_matrix) – Predicted affinity of users for items.

property col_names

The names of the columns in the results DataFrame.

property name

Name of the metric.

property num_items: int

Dimension of the item-space in both y_true and y_pred

property num_users: int

Dimension of the user-space in both y_true and y_pred after elimination of users without interactions in y_true.

property results: pandas.core.frame.DataFrame

Get the detailed results for this metric.

Contains an entry for every user-item pair in the Top-K recommendations list of every user.

If there is a user with no recommendations, the results DataFrame will contain K rows for that user with item_id = NaN and score = 0.

Returns

The results DataFrame with columns: user_id, item_id, score

Return type

pd.DataFrame

property value

Global metric value obtained by summing up scores for every user then taking the average over all users.