Opensearch

OpensearchVectorStore

Bases: BasePydanticVectorStore

Elasticsearch/Opensearch vector store.

Parameters:

Name	Type	Description	Default
client	OpensearchVectorClient	Vector index client to use for data insertion/querying.	required

Examples:

pip install llama-index-vector-stores-opensearch

from llama_index.vector_stores.opensearch import (
    OpensearchVectorStore,
    OpensearchVectorClient,
)

# http endpoint for your cluster (opensearch required for vector index usage)
endpoint = "http://localhost:9200"
# index to demonstrate the VectorStore impl
idx = "gpt-index-demo"

# OpensearchVectorClient stores text in this field by default
text_field = "content"
# OpensearchVectorClient stores embeddings in this field by default
embedding_field = "embedding"

# OpensearchVectorClient encapsulates logic for a
# single opensearch index with vector search enabled
client = OpensearchVectorClient(
    endpoint, idx, 1536, embedding_field=embedding_field, text_field=text_field
)

# initialize vector store
vector_store = OpensearchVectorStore(client)

Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-opensearch/llama_index/vector_stores/opensearch/base.py

class OpensearchVectorStore(BasePydanticVectorStore):
    """
    Elasticsearch/Opensearch vector store.

    Args:
        client (OpensearchVectorClient): Vector index client to use
            for data insertion/querying.

    Examples:
        `pip install llama-index-vector-stores-opensearch`

        ```python
        from llama_index.vector_stores.opensearch import (
            OpensearchVectorStore,
            OpensearchVectorClient,
        )

        # http endpoint for your cluster (opensearch required for vector index usage)
        endpoint = "http://localhost:9200"
        # index to demonstrate the VectorStore impl
        idx = "gpt-index-demo"

        # OpensearchVectorClient stores text in this field by default
        text_field = "content"
        # OpensearchVectorClient stores embeddings in this field by default
        embedding_field = "embedding"

        # OpensearchVectorClient encapsulates logic for a
        # single opensearch index with vector search enabled
        client = OpensearchVectorClient(
            endpoint, idx, 1536, embedding_field=embedding_field, text_field=text_field
        )

        # initialize vector store
        vector_store = OpensearchVectorStore(client)
        ```
    """

    stores_text: bool = True
    _client: OpensearchVectorClient = PrivateAttr(default=None)

    def __init__(
        self,
        client: OpensearchVectorClient,
    ) -> None:
        """Initialize params."""
        super().__init__()
        self._client = client

    @property
    def client(self) -> Any:
        """Get client."""
        return self._client

    def add(
        self,
        nodes: List[BaseNode],
        **add_kwargs: Any,
    ) -> List[str]:
        """
        Add nodes to index.

        Args:
            nodes: List[BaseNode]: list of nodes with embeddings.

        """
        return asyncio.get_event_loop().run_until_complete(
            self.async_add(nodes, **add_kwargs)
        )

    async def async_add(
        self,
        nodes: List[BaseNode],
        **add_kwargs: Any,
    ) -> List[str]:
        """
        Async add nodes to index.

        Args:
            nodes: List[BaseNode]: list of nodes with embeddings.

        """
        await self._client.index_results(nodes)
        return [result.node_id for result in nodes]

    def delete(self, ref_doc_id: str, **delete_kwargs: Any) -> None:
        """
        Delete nodes using a ref_doc_id.

        Args:
            ref_doc_id (str): The doc_id of the document whose nodes should be deleted.

        """
        asyncio.get_event_loop().run_until_complete(
            self.adelete(ref_doc_id, **delete_kwargs)
        )

    async def adelete(self, ref_doc_id: str, **delete_kwargs: Any) -> None:
        """
        Async delete nodes using a ref_doc_id.

        Args:
            ref_doc_id (str): The doc_id of the document whose nodes should be deleted.

        """
        await self._client.delete_by_doc_id(ref_doc_id)

    def query(self, query: VectorStoreQuery, **kwargs: Any) -> VectorStoreQueryResult:
        """
        Query index for top k most similar nodes.

        Args:
            query (VectorStoreQuery): Store query object.

        """
        return asyncio.get_event_loop().run_until_complete(self.aquery(query, **kwargs))

    async def aquery(
        self, query: VectorStoreQuery, **kwargs: Any
    ) -> VectorStoreQueryResult:
        """
        Async query index for top k most similar nodes.

        Args:
            query (VectorStoreQuery): Store query object.

        """
        query_embedding = cast(List[float], query.query_embedding)

        return await self._client.aquery(
            query.mode,
            query.query_str,
            query_embedding,
            query.similarity_top_k,
            filters=query.filters,
        )

client property

client: Any

Get client.

add

add(nodes: List[BaseNode], **add_kwargs: Any) -> List[str]

Add nodes to index.

Parameters:

Name	Type	Description	Default
nodes	List[BaseNode]	List[BaseNode]: list of nodes with embeddings.	required

Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-opensearch/llama_index/vector_stores/opensearch/base.py

def add(
    self,
    nodes: List[BaseNode],
    **add_kwargs: Any,
) -> List[str]:
    """
    Add nodes to index.

    Args:
        nodes: List[BaseNode]: list of nodes with embeddings.

    """
    return asyncio.get_event_loop().run_until_complete(
        self.async_add(nodes, **add_kwargs)
    )

async_add async

async_add(nodes: List[BaseNode], **add_kwargs: Any) -> List[str]

Async add nodes to index.

Parameters:

Name	Type	Description	Default
nodes	List[BaseNode]	List[BaseNode]: list of nodes with embeddings.	required

Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-opensearch/llama_index/vector_stores/opensearch/base.py

async def async_add(
    self,
    nodes: List[BaseNode],
    **add_kwargs: Any,
) -> List[str]:
    """
    Async add nodes to index.

    Args:
        nodes: List[BaseNode]: list of nodes with embeddings.

    """
    await self._client.index_results(nodes)
    return [result.node_id for result in nodes]

delete

delete(ref_doc_id: str, **delete_kwargs: Any) -> None

Delete nodes using a ref_doc_id.

Parameters:

Name	Type	Description	Default
ref_doc_id	str	The doc_id of the document whose nodes should be deleted.	required

Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-opensearch/llama_index/vector_stores/opensearch/base.py

def delete(self, ref_doc_id: str, **delete_kwargs: Any) -> None:
    """
    Delete nodes using a ref_doc_id.

    Args:
        ref_doc_id (str): The doc_id of the document whose nodes should be deleted.

    """
    asyncio.get_event_loop().run_until_complete(
        self.adelete(ref_doc_id, **delete_kwargs)
    )

adelete async

adelete(ref_doc_id: str, **delete_kwargs: Any) -> None

Async delete nodes using a ref_doc_id.

Parameters:

Name	Type	Description	Default
ref_doc_id	str	The doc_id of the document whose nodes should be deleted.	required

Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-opensearch/llama_index/vector_stores/opensearch/base.py

async def adelete(self, ref_doc_id: str, **delete_kwargs: Any) -> None:
    """
    Async delete nodes using a ref_doc_id.

    Args:
        ref_doc_id (str): The doc_id of the document whose nodes should be deleted.

    """
    await self._client.delete_by_doc_id(ref_doc_id)

query

query(query: VectorStoreQuery, **kwargs: Any) -> VectorStoreQueryResult

Query index for top k most similar nodes.

Parameters:

Name	Type	Description	Default
query	VectorStoreQuery	Store query object.	required

Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-opensearch/llama_index/vector_stores/opensearch/base.py

def query(self, query: VectorStoreQuery, **kwargs: Any) -> VectorStoreQueryResult:
    """
    Query index for top k most similar nodes.

    Args:
        query (VectorStoreQuery): Store query object.

    """
    return asyncio.get_event_loop().run_until_complete(self.aquery(query, **kwargs))

aquery async

aquery(query: VectorStoreQuery, **kwargs: Any) -> VectorStoreQueryResult

Async query index for top k most similar nodes.

Parameters:

Name	Type	Description	Default
query	VectorStoreQuery	Store query object.	required

Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-opensearch/llama_index/vector_stores/opensearch/base.py

async def aquery(
    self, query: VectorStoreQuery, **kwargs: Any
) -> VectorStoreQueryResult:
    """
    Async query index for top k most similar nodes.

    Args:
        query (VectorStoreQuery): Store query object.

    """
    query_embedding = cast(List[float], query.query_embedding)

    return await self._client.aquery(
        query.mode,
        query.query_str,
        query_embedding,
        query.similarity_top_k,
        filters=query.filters,
    )