IntraListDiversity

class rectools.metrics.diversity.IntraListDiversity(k: int, distance_calculator: PairwiseDistanceCalculator)[source]

Bases: MetricAtK

Intra-List Diversity metric.

Estimate average pairwise distance between items in user recommendations.

\[ILD@k = (\sum_{i=1}^{k+1} \sum_{j=1}^{k+1} d(i, j)) / (k * (k-1))\]

where - d(i, j) is distance between recommended items with rank i and rank j.

Parameters

k (int) – Number of items at the top of recommendations list that will be used to calculate metric.
distance_calculator (PairwiseDistanceCalculator) – Distance calculator, object that returns distance between any item pair.

Examples

>>> from rectools.metrics.distances import PairwiseHammingDistanceCalculator
>>> reco = pd.DataFrame(
...     {
...         Columns.User: [1, 1, 1, 2, 2],
...         Columns.Item: [1, 2, 3, 1, 4],
...         Columns.Rank: [1, 2, 3, 1, 2],
...     }
... )
>>> features_df = pd.DataFrame(
...     [
...         [1, 0, 0],
...         [2, 0, 1],
...         [3, 1, 1],
...         [4, 0, 0],
...     ],
...     columns=[Columns.Item, "feature_1", "feature_2"]
... ).set_index(Columns.Item)
>>> calculator = PairwiseHammingDistanceCalculator(features_df)
>>> IntraListDiversity(k=1, distance_calculator=calculator).calc_per_user(reco).values
array([0, 0])
>>> IntraListDiversity(k=2, distance_calculator=calculator).calc_per_user(reco).values
array([1., 0.])
>>> IntraListDiversity(k=3, distance_calculator=calculator).calc_per_user(reco).values
array([1.33333333, 0.        ])

Inherited-members

Parameters

k (int) –
distance_calculator (PairwiseDistanceCalculator) –

Methods

`calc`(reco)	Calculate metric value.
`calc_from_fitted`(fitted)	Calculate metric value from fitted data.
`calc_per_user`(reco)	Calculate metric values for all users.
`calc_per_user_from_fitted`(fitted)	Calculate metric values for all users from fitted data.
`fit`(reco, k_max)	Prepare intermediate data for effective calculation.

Attributes

distance_calculator

calc(reco: DataFrame) → float[source]

Calculate metric value.

Parameters: reco (pd.DataFrame) – Recommendations table with columns Columns.User, Columns.Item, Columns.Rank.
Returns: Value of metric (average between users).
Return type: float

calc_from_fitted(fitted: ILDFitted) → float[source]

Calculate metric value from fitted data.

For parameters used result of fit method.

Parameters: fitted (ILDFitted) – Meta data that got from .fit method.
Returns: Value of metric (average between users).
Return type: float

calc_per_user(reco: DataFrame) → Series[source]

Calculate metric values for all users.

Parameters: reco (pd.DataFrame) – Recommendations table with columns Columns.User, Columns.Item, Columns.Rank.
Returns: Values of metric (index - user id, values - metric value for every user).
Return type: pd.Series

calc_per_user_from_fitted(fitted: ILDFitted) → Series[source]

Calculate metric values for all users from fitted data.

For parameters used result of fit method.

Parameters: fitted (ILDFitted) – Meta data that got from .fit method.
Returns: Values of metric (index - user id, values - metric value for every user).
Return type: pd.Series

classmethod fit(reco: DataFrame, k_max: int) → ILDFitted[source]

Prepare intermediate data for effective calculation.

You can use this method to prepare some intermediate data for later calculation. It can optimize calculations if you want calculate metric value for different k or distance_calculator.

Parameters

reco (pd.DataFrame) – Recommendations table with columns Columns.User, Columns.Item, Columns.Rank.
k_max (int) – k is number of items at the top of recommendations list that will be used to calculate metric. So k_max is maximum value of k metric for that you want to calculate.

Return type

ILDFitted