LangChain-Kùzu Integration

The brand new langchain-kuzu integration package deal is now obtainable on PyPI! This package deal unites the highly effective capabilities of LangChain with the cutting-edge expertise of Kùzu. graph databasepermitting seamless transformation of unstructured textual content into structured graphics. Whether or not you are a knowledge scientist, developer, or AI fanatic, this integration simplifies complicated duties like entity extraction, graph creation, and pure language consulting. Let’s discover what makes this package deal a game-changer in your information workflows.

Studying targets

• Perceive the capabilities of the LangChain-Kùzu integration to remodel unstructured textual content into structured graph databases.
• Discover ways to outline graph schemas, together with nodes and relationships, tailor-made to your particular information wants.
• Grasp the method of making, updating, and querying charts utilizing Kùzu and LangChain’s LLM-based instruments.
• Discover pure language queries of graph databases with LangChain’s GraphQACain to achieve intuitive insights into your information.
• Uncover superior options equivalent to dynamic schema updates, customized LLM pairing, and versatile information import choices into Kùzu.

This text was printed as a part of the Knowledge Science Blogathon.

Fast Kuzu Set up

To get began, merely set up the package deal on google collaboration:

pip set up -U langchain-kuzu langchain-openai langchain-experimental

This set up consists of dependencies for LangChain and Kùzu, together with help for LLM equivalent to OpenAI’s GPT fashions. For those who favor different LLM suppliers, you possibly can set up their respective LangChain-compatible Python packages.

Why select LangChain-Kùzu in your initiatives?

For those who work with unstructured textual content information and wish to create graph-based representations, this package deal is designed for you.

Key options embody:

• Customizable schemes– Outline and extract particular entities and relationships from textual content information effortlessly.
• Textual content to Graphic Transformation: Harness the facility of LLMs to construction significant graphics from plain textual content.
• Question in pure language: Question graphs intuitively utilizing pure language, powered by LangChain’s GraphQACain.
• Seamless integration: Shortly join LangChain’s LLM capabilities with Kùzu for a unified workflow.

Let us take a look at a sensible instance to see this integration in motion.

Create a chart from unstructured textual content

First create a Kùzu database in your native machine and connect with it:

import kuzu

db = kuzu.Database("test_db")
conn = kuzu.Connection(db)

Getting began with LangChain-Kùzu

The mixing of Kùzu with LangChain makes it handy to create and replace graphs from unstructured textual content, and in addition question graphs by means of a Text2Cypher pipeline that makes use of the facility of LangChain’s LLM chains. To begin, we create a KuzuGraph object that makes use of the database object we created earlier together with the KuzuGraph constructor.

from langchain_kuzu.graphs.kuzu_graph import KuzuGraph
graph = KuzuGraph(db, allow_dangerous_requests=True)

Lets say that we wish to remodel the next textual content right into a graphic:

• “Tim Prepare dinner is the CEO of Apple. “Apple is headquartered in California.”

textual content = "Tim Prepare dinner is the CEO of Apple. Apple has its headquarters in California."

Step 1: outline the graphic scheme

First, outline the kinds of entities (nodes) and relationships you wish to embody.

# Outline schema
allowed_nodes = ("Individual", "Firm", "Location")
allowed_relationships = (
    ("Individual", "IS_CEO_OF", "Firm"),
    ("Firm", "HAS_HEADQUARTERS_IN", "Location"),
)

Step 2: Remodel textual content into graphic paperwork

Use the LLMGraphTransformer class to course of textual content in structured graphic paperwork:

from langchain_core.paperwork import Doc
from langchain_experimental.graph_transformers import LLMGraphTransformer
from langchain_openai import ChatOpenAI

# Outline the LLMGraphTransformer
llm_transformer = LLMGraphTransformer(
    llm=ChatOpenAI(mannequin="gpt-4o-mini", temperature=0, api_key='OPENAI_API_KEY'),  # noqa: F821
    allowed_nodes=allowed_nodes,
    allowed_relationships=allowed_relationships,
)

paperwork = (Doc(page_content=textual content))
graph_documents = llm_transformer.convert_to_graph_documents(paperwork)

Step 3: Add graphic paperwork to Kùzu

Add graphic paperwork to Kùzu for additional use:

from langchain_kuzu.graphs.kuzu_graph import KuzuGraph

graph = KuzuGraph(db)
graph.add_graph_documents(graph_documents, include_source=True, allow_dangerous_requests= True)

graph_documents(:2)

Notice: Within the KuzuGraph methodology, set the ‘allow_dangerous_requests’ parameter to True if you happen to get an error.

Manufacturing:

(GraphDocument(nodes=(Node(id='Tim Prepare dinner', kind="Individual", properties={}), 
Node(id='Apple', kind="Firm", properties={}), Node(id='California', 
kind="Location", properties={})), relationships=(Relationship(supply=Node(id='Tim
Prepare dinner', kind="Individual", properties={}), goal=Node(id='Apple', kind="Firm", 
properties={}), kind="IS_CEO_OF", properties={}), 
Relationship(supply=Node(id='Apple', kind="Firm", properties={}), 
goal=Node(id='California', kind="Location", properties={}), 
kind="HAS_HEADQUARTERS_IN", properties={})), supply=Doc(metadata={}, 
page_content="Tim Prepare dinner is the CEO of Apple. Apple has its headquarters in 
California.")))

Consulting the graph

With the KuzuQAChainyou possibly can seek the advice of the graph utilizing pure language:

# Add the graph doc to the graph
graph.add_graph_documents(
    graph_documents,
    include_source=True,
)

from langchain_kuzu.chains.graph_qa.kuzu import KuzuQAChain

# Create the KuzuQAChain with verbosity enabled to see the generated Cypher queries
chain = KuzuQAChain.from_llm(
    llm=ChatOpenAI(mannequin="gpt-4o-mini", temperature=0.3, api_key='OPENAI_API_KEY'),  # noqa: F821
    graph=graph,
    verbose=True,
    allow_dangerous_requests=True,
)

chain.invoke("The place is Apple headquartered?")

Manufacturing:

> Getting into new KuzuQAChain chain...
Generated Cypher:
MATCH (c:Firm {id: 'Apple'})-(:HAS_HEADQUARTERS_IN)->(l:Location) RETURN l
Full Context:
({'l': {'_id': {'offset': 0, 'desk': 1}, '_label': 'Location', 'id': 'California', 'kind': 'entity'}})

> Completed chain.
{'question': 'The place is Apple headquartered?',
 'consequence': 'Apple is headquartered in California.'}

Unlocking superior options

The LangChain-Kùzu integration presents a number of superior options to enhance your workflows:

• Dynamic schema updates– Mechanically updates outlines when the chart is up to date.
• Customized LLM Matching: Use separate LLMs for Cypher technology and response technology to optimize efficiency.
• Full graphic inspection– Simply examine nodes, relationships, and schemas with intuitive instructions.

Kùzu is a high-performance, embeddable graph database constructed for contemporary functions. Highlights embody:

• Encrypted question help: Declaratively question property graphs utilizing Cypher.
• Built-in structure: runs in course of with out the necessity to configure the server.
• Versatile information import– Handles information from varied codecs equivalent to CSV, JSON, and relational databases.

Discover extra within the kùzu documentation.

Getting began with LangChain-Kùzu

To begin your journey:

• Set up the package deal: Begin with pip set up langchain-kuzu.
• Outline your graphic scheme: adapt it to your particular wants.
• Leverage LLMs: Use LangChain instruments to effortlessly create and question charts.

Go to the PyPI web page for extra detailed examples and updates. Do not forget to focus on our repository on GitHub and share your suggestions – your contributions drive our progress!

Conclusion

The langchain-kuzu integration redefines the best way you work together with unstructured information. Whether or not reworking textual content into structured graphs or querying these graphs with pure language, this package deal opens up highly effective potentialities for gaining insights from AI-powered information. Strive it right this moment and uncover a extra intuitive method to work with graph information!

Key takeaways

• The LangChain-Kùzu integration simplifies the transformation of unstructured textual content into structured graph databases effortlessly.
• Outline customizable graphical schemas to extract significant entities and relationships tailor-made to your information.
• Leverage LangChain LLMs for intuitive pure language queries and text-to-graph conversion.
• Kùzu presents dynamic schema updates, seamless integration, and help for Cypher queries for improved workflows.
• This integration permits AI lovers, builders, and information scientists to unlock invaluable insights from graph information.

Incessantly requested questions

The media proven on this article is just not the property of Analytics Vidhya and is used on the writer’s discretion.

Hey! I’m Adarsh, a Enterprise Analytics graduate from ISB, at the moment immersed in analysis and exploring new frontiers. I’m keen about information science, synthetic intelligence, and all of the revolutionary methods they will remodel industries. Whether or not it is constructing fashions, engaged on information pipelines, or diving into machine studying, I really like experimenting with the newest expertise. AI is not only my curiosity, it is the place I see the longer term going, and I am all the time excited to be part of that journey!