Source code for docarray.array.memory

from docarray.array.document import DocumentArray
from import StorageMixins

[docs]class DocumentArrayInMemory(StorageMixins, DocumentArray): """ Default DocumentArray that stores Documents in memory. With this implementation, :meth:`match` and :meth:`find` perform exact (exhaustive) vector search. Example usage: .. code-block:: python from docarray import Document, DocumentArray da = DocumentArray( [Document(text='The cake is a lie'), Document(text='Do a barrel roll!')] ) da.apply(Document.embed_feature_hashing) query = Document(text='Can i have some cake?').embed_feature_hashing() query.match(da, metric='jaccard', use_scipy=True) print(query.matches[:, ('text', 'scores__jaccard__value')]) .. code-block:: bash [['The cake is a lie', 'Do a barrel roll!'], [0.9, 1.0]] .. seealso:: For further details, see our :ref:`user guide <documentarray>`. """ def __new__(cls, *args, **kwargs): return super().__new__(cls)