RankProduct

class rsdiv.aggregation.RankProduct(multi_scores: DataFrame)[source]

Bases: object

Rank product is a biologically motivated test to combine the lists to a comprehensive rank.

get_rp_values(weights: Optional[ndarray] = None) ndarray[source]

Get the list of rank product values for each item.

Parameters

weights (Optional[np.ndarray], optional) – Weights for each column. The larger the more important. Defaults to None. If not given, a uniformed weight will be set.

Returns

The rank product scores for each item, the larger the more preferable. Laplacian smoothing is applied to avoid zeros.

Return type

np.ndarray