in_memory
superduper.vector_search.in_memory
InMemoryVectorSearcher
​
InMemoryVectorSearcher(self,
identifier: str,
dimensions: int,
h: Optional[numpy.ndarray] = None,
index: Optional[List[str]] = None,
measure: Union[str,
Callable] = 'cosine')
Parameter | Description |
---|---|
identifier | Unique string identifier of index |
dimensions | Dimension of the vector embeddings |
h | array/ tensor of vectors |
index | list of IDs |
measure | measure to assess similarity |
Simple hash-set for looking up with vector similarity.