AzureAISearchEmbeddingRetriever

Retrieve documents from an Azure AI Search document store using vector similarity.

Basic Information

• Type: haystack_integrations.components.retrievers.azure_ai_search.embedding_retriever.AzureAISearchEmbeddingRetriever
• Components it can connect with:
  • TextEmbedder: AzureAISearchEmbeddingRetriever receives query embeddings from a text embedder.
  • Ranker: AzureAISearchEmbeddingRetriever sends retrieved documents to a Ranker for reranking.

Inputs

Parameter	Type	Default	Description
query_embedding	List[float]		A list of floats representing the query embedding.
filters	Optional[Dict[str, Any]]	None	Filters applied to the retrieved documents. The way runtime filters are applied depends on the filter_policy chosen at retriever initialization.
top_k	Optional[int]	None	The maximum number of documents to retrieve.

Outputs

Parameter	Type	Default	Description
documents	List[Document]		A list of documents retrieved from the AzureAISearchDocumentStore.

Overview

Use AzureAISearchEmbeddingRetriever to retrieve documents from an Azure AI Search document store using vector similarity. This retriever must be connected to an AzureAISearchDocumentStore to run.

Vector (embedding) retrieval is useful when you want to find semantically similar documents based on the meaning of the query rather than exact keyword matches.

Azure AI Search Setup

To use this retriever, you need an Azure AI Search service with a properly configured vector index. For more information, see the Azure AI Search documentation.

Usage Example

Initializing the Component

components:
  AzureAISearchEmbeddingRetriever:
    type: haystack_integrations.components.retrievers.azure_ai_search.embedding_retriever.AzureAISearchEmbeddingRetriever
    init_parameters:
      document_store:
        type: haystack_integrations.document_stores.azure_ai_search.document_store.AzureAISearchDocumentStore
        init_parameters:
          api_key:
            type: env_var
            env_vars:
            - AZURE_SEARCH_API_KEY
            strict: false
          azure_endpoint:
            type: env_var
            env_vars:
            - AZURE_SEARCH_ENDPOINT
            strict: false
          index_name: my-index
      top_k: 10

Parameters

Init Parameters

These are the parameters you can configure in Pipeline Builder:

Parameter	Type	Default	Description
document_store	AzureAISearchDocumentStore		An instance of AzureAISearchDocumentStore to use with the retriever.
filters	Optional[Dict[str, Any]]	None	Filters applied when fetching documents from the document store.
top_k	int	10	Maximum number of documents to return.
filter_policy	Union[str, FilterPolicy]	FilterPolicy.REPLACE	Policy to determine how filters are applied. Can be REPLACE (runtime filters replace initialization filters) or MERGE (runtime filters are merged with initialization filters).
query_type	Optional[str]	None	A string indicating the type of query to perform. Possible values are simple, full, and semantic.
semantic_configuration_name	Optional[str]	None	The name of semantic configuration to be used when processing semantic queries.

Run Method Parameters

These are the parameters you can configure for the component's run() method. You can pass these parameters at query time through the API, in Playground, or when running a job.

Parameter	Type	Default	Description
query_embedding	List[float]		A list of floats representing the query embedding.
filters	Optional[Dict[str, Any]]	None	Filters applied to the retrieved documents. The way runtime filters are applied depends on the filter_policy chosen at retriever initialization.
top_k	Optional[int]	None	The maximum number of documents to retrieve.