From vectors to graphs

Retrieval Augmented Generation (RAG) with LangChain

Meri Nova

Machine Learning Engineer

Vector RAG limitations

The typical vector store RAG workflow.

A bubble containing "themes and relationships".

Retrieval Augmented Generation (RAG) with LangChain

Vector RAG limitations

The typical vector store RAG workflow.

A bubble containing "scalability".

Retrieval Augmented Generation (RAG) with LangChain

Vector RAG limitations

The typical vector store RAG workflow.

A bubble containing "integrating diverse data".

Retrieval Augmented Generation (RAG) with LangChain

Graph databases

A graph showing connections between people, places, and interests.

Retrieval Augmented Generation (RAG) with LangChain

Graph databases - nodes

A graph showing connections between people, places, and interests.

Retrieval Augmented Generation (RAG) with LangChain

Graph databases - edges

A graph showing connections between people, places, and interests.

Retrieval Augmented Generation (RAG) with LangChain

Neo4j graph databases

 

The Neo4j logo.

The LangChain logo.

Retrieval Augmented Generation (RAG) with LangChain

From graphics to graphs...

A graph showing connections between people, places, and interests.

Retrieval Augmented Generation (RAG) with LangChain

From graphics to graphs...

barcelona_graph.jpg

Retrieval Augmented Generation (RAG) with LangChain

From graphics to graphs...

barcelona_graph.jpg

Retrieval Augmented Generation (RAG) with LangChain

Loading and chunking Wikipedia pages

from langchain_community.document_loaders import WikipediaLoader
from langchain_text_splitters import TokenTextSplitter

raw_documents = WikipediaLoader(query="large language model").load()

text_splitter = TokenTextSplitter(chunk_size=100, chunk_overlap=20) 
documents = text_splitter.split_documents(raw_documents[:3])  

print(documents[0])

page_content='A large language model (LLM) is a computational model capable of...' 
metadata={'title': 'Large language model',
          'summary': "A large language model (LLM) is...",
          'source': 'https://en.wikipedia.org/wiki/Large_language_model'}
Retrieval Augmented Generation (RAG) with LangChain

From text to graphs!

from langchain_openai import ChatOpenAI

from langchain_experimental.graph_transformers import LLMGraphTransformer


llm = ChatOpenAI(api_key="...", temperature=0, model_name="gpt-4o-mini")

llm_transformer = LLMGraphTransformer(llm=llm)


graph_documents = llm_transformer.convert_to_graph_documents(documents)
print(graph_documents)
Retrieval Augmented Generation (RAG) with LangChain

From text to graphs!

[GraphDocument(
    nodes=[
        Node(id='Llm', type='Computational model'),
        Node(id='Language Generation', type='Concept'),
        Node(id='Natural Language Processing Tasks', type='Concept'),
        Node(id='Llama Family', type='Computational model'),
        Node(id='Ibm', type='Organization'),
        ..., Node(id='Bert', type='Computational model')],
    relationships=[
        Relationship(source=Node(id='Llm', type='Computational model'),
                     target=Node(id='Language Generation', type='Concept'),
                     type='CAPABLE_OF'),
        ...])]
Retrieval Augmented Generation (RAG) with LangChain

Let's practice!

Retrieval Augmented Generation (RAG) with LangChain

Preparing Video For Download...