AsyncRetrievalStrategy#

class langchain_elasticsearch.vectorstores.AsyncRetrievalStrategy[source]#

Methods

before_index_creation(*, client, text_field, ...)

Executes before the index is created.

es_mappings_settings(*, text_field, ...)

Create the required index and do necessary preliminary work, like creating inference pipelines or checking if a required model was deployed.

es_query(*, query, query_vector, text_field, ...)

Returns the Elasticsearch query body for the given parameters.

needs_inference()

Some retrieval strategies index embedding vectors and allow search by embedding vector, for example the DenseVectorStrategy strategy.
async before_index_creation(*, client: AsyncElasticsearch, text_field: str, vector_field: str) → None[source]#

Executes before the index is created. Used for setting up any required Elasticsearch resources like a pipeline. Defaults to a no-op.

Parameters:

  • client (AsyncElasticsearch) – The Elasticsearch client.

  • text_field (str) – The field containing the text data in the index.

  • vector_field (str) – The field containing the vector representations in the index.

Return type:

None

abstractmethod es_mappings_settings(*, text_field: str, vector_field: str, num_dimensions: int | None) → Tuple[Dict[str, Any], Dict[str, Any]][source]#

Create the required index and do necessary preliminary work, like creating inference pipelines or checking if a required model was deployed.

Parameters:

  • client – Elasticsearch client connection.

  • text_field (str) – The field containing the text data in the index.

  • vector_field (str) – The field containing the vector representations in the index.

  • num_dimensions (int | None) – If vectors are indexed, how many dimensions do they have.

Returns:

Dictionary with field and field type pairs that describe the schema.

Return type:

Tuple[Dict[str, Any], Dict[str, Any]]

abstractmethod es_query(*, query: str | None, query_vector: List[float] | None, text_field: str, vector_field: str, k: int, num_candidates: int, filter: List[Dict[str, Any]] = []) → Dict[str, Any][source]#

Returns the Elasticsearch query body for the given parameters. The store will execute the query.

Parameters:

  • query (str | None) – The text query. Can be None if query_vector is given.

  • k (int) – The total number of results to retrieve.

  • num_candidates (int) – The number of results to fetch initially in knn search.

  • filter (List[Dict[str, Any]]) – List of filter clauses to apply to the query.

  • query_vector (List[float] | None) – The query vector. Can be None if a query string is given.

  • text_field (str)

  • vector_field (str)

Returns:

The Elasticsearch query body.

Return type:

Dict[str, Any]

needs_inference() → bool[source]#

Some retrieval strategies index embedding vectors and allow search by embedding vector, for example the DenseVectorStrategy strategy. Mapping a user input query string to an embedding vector is called inference. Inference can be applied in Elasticsearch (using a model_id) or outside of Elasticsearch (using an EmbeddingService defined on the VectorStore). In the latter case, this method has to return True.

Return type:

bool