dlt.destinations.impl.lancedb.factory
lancedb Objects
class lancedb(Destination[LanceDBClientConfiguration, "LanceDBClient"])
__init__
def __init__(credentials: Union["DBConnection", LanceDBCredentials,
Dict[str, Any]] = None,
lance_uri: Optional[str] = None,
embedding_model_provider: TEmbeddingProvider = None,
embedding_model: str = None,
vector_field_name: str = None,
destination_name: str = None,
environment: str = None,
**kwargs: Any) -> None
Configure the LanceDB destination to use in a pipeline.
All arguments provided here supersede other configuration sources such as environment variables and dlt config files.
Arguments:
credentials
Union["DBConnection", LanceDBCredentials, Dict[str, Any]] - Credentials to connect to the LanceDB database. Can be an instance ofLanceDBCredentials
or an instance of native LanceDB client or a dictionary with the credentials parameters.lance_uri
Optional[str] - LanceDB database URI. Defaults to local, on-disk instance. The available schemas are:/path/to/database
- local database.db://host:port
- remote database (LanceDB cloud).
embedding_model_provider
TEmbeddingProvider, optional - Embedding provider used for generating embeddings. Default is "cohere". See LanceDB documentation for the full list of available providers.embedding_model
str, optional - The model used by the embedding provider for generating embeddings. Default is "embed-english-v3.0". Check with the embedding provider which options are available.vector_field_name
str, optional - Name of the special field to store the vector embeddings. Default is "vector".destination_name
str, optional - Name of the destination, can be used in config section to differentiate between multiple of the same typeenvironment
str, optional - Environment of the destination**kwargs
Any, optional - Additional arguments forwarded to the destination config