Merge pull request #21 from davidhou17/DOCSP-50370

(DOCSP-50370): Create new LangChain self-query retrieval notebook

Author: davidhou17

README.md

@@ -14,11 +14,11 @@ The following table summarizes the contents of the notebooks in each directory:
 |--------------------|--------------------------------------------------|
 | [/create-embeddings](https://github.com/mongodb/docs-notebooks/tree/main/create-embeddings) | Learn how to generate embeddings for vector search |
 | [/get-started](https://github.com/mongodb/docs-notebooks/tree/main/get-started) | Complete our quick start tutorial |
-| [/ai-integrations](https://github.com/mongodb/docs-notebooks/tree/main/ai-integrations) | Implement RAG with popular AI frameworks that integrate with MongoDB |
+| [/ai-integrations](https://github.com/mongodb/docs-notebooks/tree/main/ai-integrations) | Build AI applications and agents with popular AI frameworks that integrate with MongoDB |
 | [/manage-indexes](https://github.com/mongodb/docs-notebooks/tree/main/manage-indexes) | Create, view, edit, and delete vector search indexes |
 | [/quantization](https://github.com/mongodb/docs-notebooks/tree/main/quantization) | Quantize your vector embeddings for efficient processing |
 | [/run-queries](https://github.com/mongodb/docs-notebooks/tree/main/run-queries) | Learn how to run vector search queries (ANN and ENN) |
-| [/use-cases](https://github.com/mongodb/docs-notebooks/tree/main/use-cases) | Implement RAG using a MongoDB-native retrieval system |
+| [/use-cases](https://github.com/mongodb/docs-notebooks/tree/main/use-cases) | Implement RAG and build AI agents using a MongoDB-native retrieval system |
 
 ## Other Resources
 

ai-integrations/README.md

@@ -0,0 +1,18 @@
+# AI Integrations
+
+This folder contains Jupyter Notebooks that demonstrate how to integrate various AI frameworks with MongoDB. These notebooks show you how to implement RAG and other features for your AI-powered and agentic applications by leveraging MongoDB Atlas as both a vector database and document database.
+
+| Notebook | Description |
+|----------|-------------|
+| [langchain](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/langchain.ipynb) | Implement basic RAG with LangChain and MongoDB Atlas Vector Search |
+| [langchain-memory-semantic-cache](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/langchain-memory-semantic-cache.ipynb) | Implement RAG with memory with LangChain and MongoDB |
+| [langchain-hybrid-search](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/langchain-hybrid-search.ipynb) | Combine vector search with full-text search using LangChain and MongoDB |
+| [langchain-parent-document-retrieval](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/langchain-parent-document-retrieval.ipynb) | Perform parent-document retrieval with LangChain and MongoDB |
+| [langchain-self-query-retrieval](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/langchain-self-query-retrieval.ipynb) | Perform self-querying retrieval with LangChain and MongoDB |
+| [langchain-local-rag](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/langchain-local-rag.ipynb) | Implement RAG with local models with LangChain and MongoDB |
+| [langchain-graphrag](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/langchain-graphrag.ipynb) | Implement graph-based RAG with LangChain and MongoDB |
+| [langchain-natural-language](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/langchain-natural-language.ipynb) | Perform natural language querying with LangChain and MongoDB |
+| [langgraph](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/langgraph.ipynb) | Build an AI agent with LangGraph and MongoDB |
+| [llamaindex](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/llamaindex.ipynb) | Implement basic RAG with LlamaIndex and MongoDB Atlas |
+| [haystack](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/haystack.ipynb) | Implement basic RAG with Haystack and MongoDB Atlas |
+| [semantic-kernel](https://github.com/mongodb/docs-notebooks/blob/main/ai-integrations/semantic-kernel.ipynb) | Implement basic RAG with Microsoft Semantic Kernel and MongoDB Atlas |

ai-integrations/langchain-parent-document-retrieval.ipynb

@@ -11,7 +11,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "This notebook is a companion to the [LangChain Parent Document Retrieval](https://www.mongodb.com/docs/atlas/atlas-vector-search/ai-integrations/langchain/parent-document-retrieval/) page. Refer to the page for set-up instructions and detailed explanations.\n",
+    "This notebook is a companion to the [Parent Document Retrieval](https://www.mongodb.com/docs/atlas/atlas-vector-search/ai-integrations/langchain/parent-document-retrieval/) page. Refer to the page for set-up instructions and detailed explanations.\n",
     "\n",
     "<a target=\"_blank\" href=\"https://colab.research.google.com/github/mongodb/docs-notebooks/blob/main/ai-integrations/langchain-hybrid-search.ipynb\">\n",
     "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
@@ -28,7 +28,7 @@
    },
    "outputs": [],
    "source": [
-    "pip install --quiet --upgrade langchain langchain-community langchain-core langchain-mongodb langchain-openai pymongo pypdf"
+    "pip install --quiet --upgrade langchain langchain-community langchain-core langchain-mongodb langchain-voyageai langchain-openai pymongo pypdf"
    ]
   },
   {
@@ -39,7 +39,8 @@
    "source": [
     "import os\n",
     "\n",
-    "os.environ[\"OPENAI_API_KEY\"] = \"<api-key>\"\n",
+    "os.environ[\"VOYAGE_API_KEY\"] = \"<voyage-api-key>\"\n",
+    "os.environ[\"OPENAI_API_KEY\"] = \"<openai-api-key>\"\n",
     "ATLAS_CONNECTION_STRING = \"<connection-string>\""
    ]
   },
@@ -71,15 +72,15 @@
    "outputs": [],
    "source": [
     "from langchain_mongodb.retrievers import MongoDBAtlasParentDocumentRetriever\n",
-    "from langchain_openai import OpenAIEmbeddings\n",
+    "from langchain_voyageai import VoyageAIEmbeddings\n",
     "\n",
     "# Define the embedding model to use\n",
-    "embedding_model = OpenAIEmbeddings(model=\"text-embedding-3-small\")\n",
+    "embedding_model = VoyageAIEmbeddings(model=\"voyage-3-large\")\n",
     "\n",
     "# Define the chunking method for the child documents\n",
     "child_splitter = RecursiveCharacterTextSplitter(chunk_size=200, chunk_overlap=20)\n",
     "\n",
-    "# Database and collection name\n",
+    "# Specify the database and collection name\n",
     "database_name = \"langchain_db\"\n",
     "collection_name = \"parent_document\"\n",
     "\n",
@@ -117,7 +118,7 @@
     "\n",
     "# Use helper method to create the vector search index\n",
     "vector_store.create_vector_search_index(\n",
-    "   dimensions = 1536,       # The dimensions of the vector embeddings to be indexed\n",
+    "   dimensions = 1024,       # The dimensions of the vector embeddings to be indexed\n",
     "   wait_until_complete = 60 # Number of seconds to wait for the index to build (can take around a minute)\n",
     ")\n"
    ]