MongoDB Atlas
MongoDB Atlas is a fully-managed clouddatabase available in AWS, Azure, and GCP. It now has support for nativeVector Search on the MongoDB document data.
Installation and Setup
Seedetail configuration instructions.
We need to installlangchain-mongodb python package.
pip install langchain-mongodb
Vector Store
See ausage example.
from langchain_mongodbimport MongoDBAtlasVectorSearch
Retrievers
Full Text Search Retriever
Hybrid Search Retrieverperforms full-text searches usingLucene’s standard (BM25) analyzer.
from langchain_mongodb.retrieversimport MongoDBAtlasFullTextSearchRetriever
Hybrid Search Retriever
Hybrid Search Retrievercombines vector and full-text searches weightingthem the viaReciprocal Rank Fusion(RRF) algorithm.
from langchain_mongodb.retrieversimport MongoDBAtlasHybridSearchRetriever
Model Caches
MongoDBCache
An abstraction to store a simple cache in MongoDB. This does not use Semantic Caching, nor does it require an index to be made on the collection before generation.
To import this cache:
from langchain_mongodb.cacheimport MongoDBCache
To use this cache with your LLMs:
from langchain_core.globalsimport set_llm_cache
# use any embedding provider...
from tests.integration_tests.vectorstores.fake_embeddingsimport FakeEmbeddings
mongodb_atlas_uri="<YOUR_CONNECTION_STRING>"
COLLECTION_NAME="<YOUR_CACHE_COLLECTION_NAME>"
DATABASE_NAME="<YOUR_DATABASE_NAME>"
set_llm_cache(MongoDBCache(
connection_string=mongodb_atlas_uri,
collection_name=COLLECTION_NAME,
database_name=DATABASE_NAME,
))
MongoDBAtlasSemanticCache
Semantic caching allows users to retrieve cached prompts based on semantic similarity between the user input and previously cached results. Under the hood it blends MongoDBAtlas as both a cache and a vectorstore.The MongoDBAtlasSemanticCache inherits fromMongoDBAtlasVectorSearch and needs an Atlas Vector Search Index defined to work. Please look at theusage example on how to set up the index.
To import this cache:
from langchain_mongodb.cacheimport MongoDBAtlasSemanticCache
To use this cache with your LLMs:
from langchain_core.globalsimport set_llm_cache
# use any embedding provider...
from tests.integration_tests.vectorstores.fake_embeddingsimport FakeEmbeddings
mongodb_atlas_uri="<YOUR_CONNECTION_STRING>"
COLLECTION_NAME="<YOUR_CACHE_COLLECTION_NAME>"
DATABASE_NAME="<YOUR_DATABASE_NAME>"
set_llm_cache(MongoDBAtlasSemanticCache(
embedding=FakeEmbeddings(),
connection_string=mongodb_atlas_uri,
collection_name=COLLECTION_NAME,
database_name=DATABASE_NAME,
))