mlpy.knowledgerep.cbr.similarity.CosineSimilarity.build_indexing_structure

CosineSimilarity.build_indexing_structure(data, id_map)[source]

Build the indexing structure.

The cosine_similarity function from sklearn.metrics.pairwise takes the raw data as input. Thus the data remains as is and no special indexing structure is implemented.

Parameters:

data : ndarray[ndarray[float]]

The raw data points to be indexed.

id_map : dict[int, int]

The mapping from the data points to their case ids.