CatalogCoverage

class rectools.metrics.catalog.CatalogCoverage(k: int, normalize: bool = False)[source]

Bases: MetricAtK

Count (or share) of items from catalog that is present in recommendations for all users.

Parameters

  • k (int) – Number of items at the top of recommendations list that will be used to calculate metric.

  • normalize (bool, default False) – Flag, which says whether to normalize metric or not.

Inherited-members

Methods

calc(reco, catalog)

Calculate metric value.

Attributes

normalize

calc(reco: DataFrame, catalog: Collection[Union[str, int]]) → float[source]

Calculate metric value.

Parameters

  • reco (pd.DataFrame) – Recommendations table with columns Columns.User, Columns.Item, Columns.Rank.

  • catalog (collection) – Collection of unique item ids that could be used for recommendations.

Returns

Value of metric (aggregated for all users).

Return type

float