ahvn.utils.vdb package

ahvn.utils.vdb.parse_encoder_embedder(encoder=None, embedder=None)

Parse encoder and embedder parameters into standardized tuples and detect dimensions.

Parameters:

• encoder (Union[Callable[[Any], str], Tuple[Callable[[Any], str], Callable[[Any], str]]]) – Required encoder function or tuple (k_encoder, q_encoder).

• embedder (Union[str, Callable[[str], List[float]], Tuple[Callable[[str], List[float]], Callable[[str], List[float]]], LLM, None]) – Optional embedder function, tuple (k_embedder, q_embedder), or LLM instance.

Returns:

• encoder_tuple: (k_encoder, q_encoder) functions

• embedder_tuple: (k_embedder, q_embedder) functions or LLM instances

• k_dim: dimension of key embedder

• q_dim: dimension of query embedder

Return type:

Tuple of (encoder_tuple, embedder_tuple, k_dim, q_dim) where

Raises:

ValueError – If encoder is None or dimensions cannot be determined.

ahvn.utils.vdb.resolve_vdb_config(collection=None, provider=None, **kwargs)

Resolve vector database configuration from parameters and environment variables.

Parameters:

• collection (str) – Optional collection name.

• provider (str) – Optional vector database provider name.

• **kwargs – Additional configuration parameters.

Returns:

Resolved configuration dictionary.

Submodules

• ahvn.utils.vdb.base module
  • VectorDatabase
    • VectorDatabase.__init__()
    • VectorDatabase.connect()
    • VectorDatabase.close()
    • VectorDatabase.k_encode()
    • VectorDatabase.k_embed()
    • VectorDatabase.batch_k_encode()
    • VectorDatabase.batch_k_embed()
    • VectorDatabase.q_encode()
    • VectorDatabase.q_embed()
    • VectorDatabase.batch_q_encode()
    • VectorDatabase.batch_q_embed()
    • VectorDatabase.k_encode_embed()
    • VectorDatabase.batch_k_encode_embed()
    • VectorDatabase.q_encode_embed()
    • VectorDatabase.batch_q_encode_embed()
    • VectorDatabase.search()
    • VectorDatabase.insert()
    • VectorDatabase.delete()
    • VectorDatabase.batch_insert()
    • VectorDatabase.clear()
    • VectorDatabase.flush()
• ahvn.utils.vdb.compiler module
  • VectorCompiler
    • VectorCompiler.compile()
• ahvn.utils.vdb.types module
  • BaseVdbType
    • BaseVdbType.__init__()
    • BaseVdbType.from_ukf()
    • BaseVdbType.to_ukf()
    • BaseVdbType.pyarrow_type()
  • VdbIdType
    • VdbIdType.from_ukf()
    • VdbIdType.to_ukf()
    • VdbIdType.pyarrow_type()
  • VdbTextType
    • VdbTextType.__init__()
    • VdbTextType.from_ukf()
    • VdbTextType.to_ukf()
    • VdbTextType.pyarrow_type()
  • VdbIntegerType
    • VdbIntegerType.from_ukf()
    • VdbIntegerType.to_ukf()
    • VdbIntegerType.pyarrow_type()
  • VdbBooleanType
    • VdbBooleanType.from_ukf()
    • VdbBooleanType.to_ukf()
    • VdbBooleanType.pyarrow_type()
  • VdbDurationType
    • VdbDurationType.from_ukf()
    • VdbDurationType.to_ukf()
    • VdbDurationType.pyarrow_type()
  • VdbTimestampType
    • VdbTimestampType.from_ukf()
    • VdbTimestampType.to_ukf()
    • VdbTimestampType.pyarrow_type()
  • VdbJsonType
    • VdbJsonType.from_ukf()
    • VdbJsonType.to_ukf()
    • VdbJsonType.pyarrow_type()
  • VdbVectorType
    • VdbVectorType.__init__()
    • VdbVectorType.from_ukf()
    • VdbVectorType.to_ukf()
    • VdbVectorType.pyarrow_type()
  • VdbTagsType
    • VdbTagsType.from_ukf()
    • VdbTagsType.to_ukf()
    • VdbTagsType.pyarrow_type()
  • VdbSynonymsType
    • VdbSynonymsType.from_ukf()
    • VdbSynonymsType.to_ukf()
    • VdbSynonymsType.pyarrow_type()
  • VdbRelatedType
    • VdbRelatedType.from_ukf()
    • VdbRelatedType.to_ukf()
    • VdbRelatedType.pyarrow_type()
  • VdbAuthsType
    • VdbAuthsType.from_ukf()
    • VdbAuthsType.to_ukf()
    • VdbAuthsType.pyarrow_type()
• ahvn.utils.vdb.vdb_utils module
  • parse_encoder_embedder()
  • resolve_vdb_config()