Updating README

Author: ajosh0504

README.md

@@ -0,0 +1 @@
+Javascript and Python apps and demos showcasing how to use MongoDB in GenAI applications.

apps/README.md

@@ -1,55 +1,66 @@
-from langchain_core.documents import Document
-from langchain_community.graphs.graph_document import GraphDocument, Node, Relationship
-from pymongo import MongoClient
-from dotenv import load_dotenv
-from pprint import pprint
 import os
-import json
-from nodes_relationships import nodes,links
+
+from dotenv import load_dotenv
+from nodes_relationships import links, nodes
+from pymongo import MongoClient
+
 
 def build_lookup_map():
     quick_lookup = {}
     for key in links.keys():
         relationship = links[key]
         source_node = relationship.source
-        lookup_key = str(source_node.id)+":"+str(source_node.type)
-        lookup_content = quick_lookup.get(lookup_key,"empty")
+        lookup_key = str(source_node.id) + ":" + str(source_node.type)
+        lookup_content = quick_lookup.get(lookup_key, "empty")
         if lookup_content != "empty":
             quick_lookup.get(lookup_key).append(relationship)
         else:
             quick_lookup[lookup_key] = [relationship]
     return quick_lookup
 
+
 def create_mongo_documents():
     mongo_documents = []
     quick_lookup = build_lookup_map()
     for key in nodes.keys():
         node = nodes[key]
-        id = str(node.id)+":"+str(node.type)
+        id = str(node.id) + ":" + str(node.type)
         type = node.type
-        rel = quick_lookup.get(id,None)
+        rel = quick_lookup.get(id, None)
         relationships = set()
         targets = {}
-        if rel!=None:
+        if rel is not None:
             for relationship in rel:
-                target_id = str(relationship.target.id)+":"+str(relationship.target.type)
+                target_id = (
+                    str(relationship.target.id) + ":" + str(relationship.target.type)
+                )
                 relationships.add(target_id)
-                target_type = targets.get(target_id,None)
-                if target_type != None:
+                target_type = targets.get(target_id, None)
+                if target_type is not None:
                     targets[target_id].append(relationship.type)
                 else:
                     targets[target_id] = [relationship.type]
-            mongo_documents.append({"_id":id,"type":type,"relationships":list(relationships),"targets":targets})
+            mongo_documents.append(
+                {
+                    "_id": id,
+                    "type": type,
+                    "relationships": list(relationships),
+                    "targets": targets,
+                }
+            )
         else:
-            mongo_documents.append({"_id":id,"type":type,"relationships":[],"targets":{}})
+            mongo_documents.append(
+                {"_id": id, "type": type, "relationships": [], "targets": {}}
+            )
     return mongo_documents
 
+
 def mongo_insert():
     mongo_documents = create_mongo_documents()
     try:
         uri = os.getenv("ATLAS_CONNECTION_STRING")
         print(uri)
-        client = MongoClient(uri)
+        client = MongoClient(uri, appname="devrel.showcase.graph_rag_app")
         database = client["langchain_db"]
         collection = database["nodes_relationships"]
         for doc in mongo_documents:
@@ -58,8 +69,10 @@ def mongo_insert():
         print(e)
     finally:
         client.close()
-if __name__=="__main__":
+
+
+if __name__ == "__main__":
     load_dotenv()
     print("Inserting Documents")
     mongo_insert()
-    print("Successfully Inserted Documents")
+    print("Successfully Inserted Documents")

apps/graph_rag_demo/data_insert.py

@@ -1,16 +1,17 @@
-from langchain_core.globals import set_verbose, set_debug
-from langchain_ollama import ChatOllama, OllamaEmbeddings
+import logging
+from typing import Optional
+
+import yaml
 from langchain.schema.output_parser import StrOutputParser
-from langchain_mongodb.vectorstores import MongoDBAtlasVectorSearch
-from pymongo import MongoClient
-from langchain_community.document_loaders import PyPDFLoader
-from langchain.text_splitter import RecursiveCharacterTextSplitter
 from langchain.schema.runnable import RunnablePassthrough
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.document_loaders import PyPDFLoader
 from langchain_community.vectorstores.utils import filter_complex_metadata
+from langchain_core.globals import set_debug, set_verbose
 from langchain_core.prompts import ChatPromptTemplate
-import logging
-import yaml
-
+from langchain_mongodb.vectorstores import MongoDBAtlasVectorSearch
+from langchain_ollama import ChatOllama, OllamaEmbeddings
+from pymongo import MongoClient
 
 # Enable verbose debugging
 set_debug(True)
@@ -20,11 +21,13 @@
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
 
+
 def load_config(config_file: str = "config.yaml"):
     """Load configuration from a YAML file."""
-    with open(config_file, "r") as file:
+    with open(config_file) as file:
         return yaml.safe_load(file)
 
+
 class ChatPDF:
     """A class designed for PDF ingestion and question answering using RAG with detailed debugging logs."""
 
@@ -40,29 +43,33 @@ def __init__(self, config_file: str = "config.yaml"):
         mongo_connection_str = config["mongo_connection_str"]
         database_name = config["database_name"]
         collection_name = config["collection_name"]
-        
+
         self.model = ChatOllama(model=llm_model)
         self.embeddings = OllamaEmbeddings(model=embedding_model)
-        self.text_splitter = RecursiveCharacterTextSplitter(chunk_size=1024, chunk_overlap=100)
+        self.text_splitter = RecursiveCharacterTextSplitter(
+            chunk_size=1024, chunk_overlap=100
+        )
         self.prompt = ChatPromptTemplate.from_template(
             """
             You are a helpful assistant answering questions based on the uploaded document and the conversation.
-            
+
             Conversation History:
             {conversation_history}
-            
+
             Context from Documents:
             {context}
-            
+
             Question:
             {question}
-            
+
             Provide a concise, accurate answer (preferably within three sentences), ensuring it directly addresses the question.
             """
         )
-        
+
         # Setup MongoDB connection
-        self.client = MongoClient(mongo_connection_str)
+        self.client = MongoClient(
+            mongo_connection_str, appname="devrel.showcase.local_rag_pdf_app"
+        )
         self.collection = self.client[database_name][collection_name]
 
         # Verbose connection check
@@ -74,7 +81,7 @@ def __init__(self, config_file: str = "config.yaml"):
             collection=self.collection,
             embedding=self.embeddings,
             index_name="vector_index",
-            relevance_score_fn="cosine"
+            relevance_score_fn="cosine",
         )
 
         # Create vector search index on the collection
@@ -107,7 +114,13 @@ def upload_and_index_pdf(self, pdf_file_path: str):
         self.vector_store.add_documents(documents=chunks)
         logger.info("Document embeddings stored successfully in MongoDB Atlas.")
 
-    def query_with_context(self, query: str, conversation_history: list = None, k: int = 5, score_threshold: float = 0.2):
+    def query_with_context(
+        self,
+        query: str,
+        conversation_history: Optional[list] = None,
+        k: int = 5,
+        score_threshold: float = 0.2,
+    ):
         """
         Answer a query using the RAG pipeline with verbose debugging and conversation history.
 
@@ -132,7 +145,9 @@ def query_with_context(self, query: str, conversation_history: list = None, k: i
         # Generate and log query embeddings
         query_embedding = self.embeddings.embed_query(query)
         logger.info(f"User Query: {query}")
-        logger.debug(f"Query Embedding (sample values): {query_embedding[:10]}... [Total Length: {len(query_embedding)}]")
+        logger.debug(
+            f"Query Embedding (sample values): {query_embedding[:10]}... [Total Length: {len(query_embedding)}]"
+        )
 
         logger.info(f"Retrieving context for query: {query}")
         retrieved_docs = self.retriever.invoke(query)
@@ -147,17 +162,19 @@ def query_with_context(self, query: str, conversation_history: list = None, k: i
 
         # Format the input for the LLM, including conversation history
         formatted_input = {
-            "conversation_history": "\n".join(conversation_history) if conversation_history else "",
+            "conversation_history": (
+                "\n".join(conversation_history) if conversation_history else ""
+            ),
             "context": "\n\n".join(doc.page_content for doc in retrieved_docs),
             "question": query,
         }
 
         # Build the RAG chain
         chain = (
             RunnablePassthrough()  # Passes the input as-is
-            | self.prompt           # Formats the input for the LLM
-            | self.model            # Queries the LLM
-            | StrOutputParser()     # Parses the LLM's output
+            | self.prompt  # Formats the input for the LLM
+            | self.model  # Queries the LLM
+            | StrOutputParser()  # Parses the LLM's output
         )
 
         logger.info("Generating response using the LLM.")

apps/local-rag-pdf/rag_module.py

@@ -0,0 +1 @@
+Miscellaneous collection of guides, code snippets etc.

misc/README.md

@@ -0,0 +1,5 @@
+Step-by-step Jupyter Notebook examples on how to use MongoDB as a vector database, data store, memory provider etc. in AI applications.
+
+The easiest way to run these notebooks is in Google Colab by clicking on the [Open In Colab](https://colab.research.google.com/assets/colab-badge.svg) button at the top of the notebook.
+
+Some notebooks also have written articles/tutorials associated with them. To read the article, click the [View Article](https://img.shields.io/badge/View%20Article-blue) at the top of the notebook.