ElasticsearchEmbeddingsCache

ElasticsearchEmbeddingsCache(
  self,
  index_name: str,
  *,
  store_input: bool = True,
  metadata: Optional[Dict[str, Any]] = None,
  namespace: Optional[str] = None,
  maximum_duplicates_allowed: int = 1,
  client: Optional[Elasticsearch] = None,
  es_url: Optional[str] = None,
  es_cloud_id: Optional[str] = None,
  es_user: Optional[str] = None,
  es_api_key: Optional[str] = None,
  es_password: Optional[str] = None
)

Setup:

Install langchain_elasticsearch and start Elasticsearch locally using the start-local script.

pip install -qU langchain_elasticsearch
curl -fsSL https://elastic.co/start-local | sh

This will create an elastic-start-local folder. To start Elasticsearch and Kibana:

cd elastic-start-local
./start.sh

Elasticsearch will be available at http://localhost:9200. The password for the elastic user and API key are stored in the .env file in the elastic-start-local folder.

Initialize the Elasticsearch embeddings cache store.

Instantiate:

from langchain_elasticsearch import ElasticsearchEmbeddingsCache

cache = ElasticsearchEmbeddingsCache(
    index_name="embeddings-cache",
    es_url="http://localhost:9200"
)

Instantiate with API key:

from langchain_elasticsearch import ElasticsearchEmbeddingsCache

cache = ElasticsearchEmbeddingsCache(
    index_name="embeddings-cache",
    es_url="http://localhost:9200",
    es_api_key="your-api-key"
)

Instantiate from cloud:

from langchain_elasticsearch import ElasticsearchEmbeddingsCache

cache = ElasticsearchEmbeddingsCache(
    index_name="embeddings-cache",
    es_cloud_id="<cloud_id>",
    es_api_key="your-api-key"
)

Instantiate from existing connection:

from langchain_elasticsearch import ElasticsearchEmbeddingsCache
from elasticsearch import Elasticsearch

client = Elasticsearch("http://localhost:9200")
cache = ElasticsearchEmbeddingsCache(
    index_name="embeddings-cache",
    client=client
)

Use with CacheBackedEmbeddings:

from langchain.embeddings import CacheBackedEmbeddings
from langchain_openai import OpenAIEmbeddings
from langchain_elasticsearch import ElasticsearchEmbeddingsCache

underlying_embeddings = OpenAIEmbeddings()
cache = ElasticsearchEmbeddingsCache(
    index_name="embeddings-cache",
    es_url="http://localhost:9200"
)
cached_embeddings = CacheBackedEmbeddings.from_bytes_store(
    underlying_embeddings,
    cache,
    namespace=underlying_embeddings.model
)

For synchronous applications, use the ElasticsearchEmbeddingsCache class. For asynchronous applications, use the AsyncElasticsearchEmbeddingsCache class.