github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-06-21 06:14:37 +00:00
Code Issues Packages Projects Releases Wiki Activity

docs: Update MongoDB information in llm_caching (#22708)

Browse Source 
**Description:**: Update MongoDB information in llm_caching
This commit is contained in:
Prakul 2024-06-10 08:05:55 -07:00 committed by GitHub
parent c1fced9269
commit 3f76c9e908
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 79 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

81
docs/docs/integrations/llm_caching.ipynb
View File

@ -724,6 +724,83 @@
"llm(\"Tell me joke\")"
]
},
{
"cell_type": "markdown",
"id": "9b2b2777",
"metadata": {},
"source": [
"## `MongoDB Atlas` Cache\n",
"\n",
"[MongoDB Atlas](https://www.mongodb.com/docs/atlas/) is a fully-managed cloud database available in AWS, Azure, and GCP. It has native support for \n",
"Vector Search on the MongoDB document data.\n",
"Use [MongoDB Atlas Vector Search](/docs/integrations/providers/mongodb_atlas) to semantically cache prompts and responses."
]
},
{
"cell_type": "markdown",
"id": "ecdc2a0a",
"metadata": {},
"source": [
"### `MongoDBCache`\n",
"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.\n",
"\n",
"To import this cache:\n",
"\n",
"```python\n",
"from langchain_mongodb.cache import MongoDBCache\n",
"```\n",
"\n",
"\n",
"To use this cache with your LLMs:\n",
"```python\n",
"from langchain_core.globals import set_llm_cache\n",
"\n",
"# use any embedding provider...\n",
"from tests.integration_tests.vectorstores.fake_embeddings import FakeEmbeddings\n",
"\n",
"mongodb_atlas_uri = \"<YOUR_CONNECTION_STRING>\"\n",
"COLLECTION_NAME=\"<YOUR_CACHE_COLLECTION_NAME>\"\n",
"DATABASE_NAME=\"<YOUR_DATABASE_NAME>\"\n",
"\n",
"set_llm_cache(MongoDBCache(\n",
" connection_string=mongodb_atlas_uri,\n",
" collection_name=COLLECTION_NAME,\n",
" database_name=DATABASE_NAME,\n",
"))\n",
"```\n",
"\n",
"\n",
"### `MongoDBAtlasSemanticCache`\n",
"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.\n",
"The MongoDBAtlasSemanticCache inherits from `MongoDBAtlasVectorSearch` and needs an Atlas Vector Search Index defined to work. Please look at the [usage example](/docs/integrations/vectorstores/mongodb_atlas) on how to set up the index.\n",
"\n",
"To import this cache:\n",
"```python\n",
"from langchain_mongodb.cache import MongoDBAtlasSemanticCache\n",
"```\n",
"\n",
"To use this cache with your LLMs:\n",
"```python\n",
"from langchain_core.globals import set_llm_cache\n",
"\n",
"# use any embedding provider...\n",
"from tests.integration_tests.vectorstores.fake_embeddings import FakeEmbeddings\n",
"\n",
"mongodb_atlas_uri = \"<YOUR_CONNECTION_STRING>\"\n",
"COLLECTION_NAME=\"<YOUR_CACHE_COLLECTION_NAME>\"\n",
"DATABASE_NAME=\"<YOUR_DATABASE_NAME>\"\n",
"\n",
"set_llm_cache(MongoDBAtlasSemanticCache(\n",
" embedding=FakeEmbeddings(),\n",
" connection_string=mongodb_atlas_uri,\n",
" collection_name=COLLECTION_NAME,\n",
" database_name=DATABASE_NAME,\n",
"))\n",
"```\n",
"\n",
"To find more resources about using MongoDBSemanticCache visit [here](https://www.mongodb.com/blog/post/introducing-semantic-caching-dedicated-mongodb-lang-chain-package-gen-ai-apps)"
]
},
{
"cell_type": "markdown",
"id": "726fe754",
@ -993,7 +1070,7 @@
"metadata": {},
"outputs": [
{
"name": "stdin",
"name": "stdout",
"output_type": "stream",
"text": [
"CASSANDRA_KEYSPACE = demo_keyspace\n"
@ -1029,7 +1106,7 @@
"metadata": {},
"outputs": [
{
"name": "stdin",
"name": "stdout",
"output_type": "stream",
"text": [
"ASTRA_DB_ID = 01234567-89ab-cdef-0123-456789abcdef\n",