feat: Added LlamaIndex integration with Neo4j

[mahimairaja]

Fixes

This PR resolves #2268

What does this PR do?

This PR adds the notebook to demonstrate the integration of LlamaIndex with Neo4j

Screenshots

Overall Architecture

Building - Knowledge Graph

Limitation:

Building the Knowledge graph for various data might be difficult and each data connectors needs an unique approach.

[srini047]

@mahimairaja Just two quick things:

• Could you add a demo video of running the notebook in the Naas lab environment? The notebook flow is correct and as expected.
• Also make sure to add import requests in the Import libraries section.

[mahimairaja]

@mahimairaja Just two quick things:

• Could you add a demo video of running the notebook in the Naas lab environment? The notebook flow is correct and as expected.
• Also make sure to add import requests in the Import libraries section.

Thank you for the response @srini047, Will make it asap

[srini047]

@mahimairaja Just two quick things:

• Could you add a demo video of running the notebook in the Naas lab environment? The notebook flow is correct and as expected.
• Also make sure to add import requests in the Import libraries section.

Thank you for the response @srini047, Will make it asap

Great, Looking forward to it.

[mahimairaja]

Hi @srini047,

I just tested the notebook with the google colab Free version, it works as expected!

But, I am facing some conflicts while running the same in the naas environment. I am not able to install llama_index. Kindly help me to proceed further. Thanks.

Error:

TypeError: multiple bases have instance lay-out conflict

Click to toggle Full `Error Message`

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-6-b6c670793068> in <module>
      1 try:
----> 2     import llama_index
      3 except ModuleNotFoundError:
      4     get_ipython().system('pip install -q llama-index')
      5     import llama_index

/opt/conda/lib/python3.9/site-packages/llama_index/__init__.py in <module>
     15 
     16 # embeddings
---> 17 from llama_index.embeddings.langchain import LangchainEmbedding
     18 from llama_index.embeddings.openai import OpenAIEmbedding
     19 

/opt/conda/lib/python3.9/site-packages/llama_index/embeddings/__init__.py in <module>
     14 from llama_index.embeddings.huggingface_utils import DEFAULT_HUGGINGFACE_EMBEDDING_MODEL
     15 from llama_index.embeddings.instructor import InstructorEmbedding
---> 16 from llama_index.embeddings.langchain import LangchainEmbedding
     17 from llama_index.embeddings.llm_rails import LLMRailsEmbeddings
     18 from llama_index.embeddings.openai import OpenAIEmbedding

/opt/conda/lib/python3.9/site-packages/llama_index/embeddings/langchain.py in <module>
      3 from typing import List, Optional
      4 
----> 5 from llama_index.bridge.langchain import Embeddings as LCEmbeddings
      6 from llama_index.bridge.pydantic import PrivateAttr
      7 from llama_index.callbacks import CallbackManager

/opt/conda/lib/python3.9/site-packages/llama_index/bridge/langchain.py in <module>
      1 import langchain
----> 2 from langchain.agents import AgentExecutor, AgentType, initialize_agent
      3 
      4 # agents and tools
      5 from langchain.agents.agent_toolkits.base import BaseToolkit

~/.local/lib/python3.9/site-packages/langchain/agents/__init__.py in <module>
     29 
     30 """  # noqa: E501
---> 31 from langchain.agents.agent import (
     32     Agent,
     33     AgentExecutor,

~/.local/lib/python3.9/site-packages/langchain/agents/agent.py in <module>
     21 import yaml
     22 
---> 23 from langchain.agents.agent_iterator import AgentExecutorIterator
     24 from langchain.agents.agent_types import AgentType
     25 from langchain.agents.tools import InvalidTool

~/.local/lib/python3.9/site-packages/langchain/agents/agent_iterator.py in <module>
     19 )
     20 
---> 21 from langchain.callbacks.manager import (
     22     AsyncCallbackManager,
     23     AsyncCallbackManagerForChainRun,

~/.local/lib/python3.9/site-packages/langchain/callbacks/__init__.py in <module>
      8 """
      9 
---> 10 from langchain.callbacks.aim_callback import AimCallbackHandler
     11 from langchain.callbacks.argilla_callback import ArgillaCallbackHandler
     12 from langchain.callbacks.arize_callback import ArizeCallbackHandler

~/.local/lib/python3.9/site-packages/langchain/callbacks/aim_callback.py in <module>
      3 
      4 from langchain.callbacks.base import BaseCallbackHandler
----> 5 from langchain.schema import AgentAction, AgentFinish, LLMResult
      6 
      7 

~/.local/lib/python3.9/site-packages/langchain/schema/__init__.py in <module>
      1 """**Schemas** are the LangChain Base Classes and Interfaces."""
----> 2 from langchain.schema.agent import AgentAction, AgentFinish
      3 from langchain.schema.cache import BaseCache
      4 from langchain.schema.chat_history import BaseChatMessageHistory
      5 from langchain.schema.document import BaseDocumentTransformer, Document

~/.local/lib/python3.9/site-packages/langchain/schema/agent.py in <module>
      4 
      5 from langchain.load.serializable import Serializable
----> 6 from langchain.schema.messages import BaseMessage
      7 
      8 

~/.local/lib/python3.9/site-packages/langchain/schema/messages.py in <module>
    155 
    156 
--> 157 class HumanMessageChunk(HumanMessage, BaseMessageChunk):
    158     """A Human Message chunk."""
    159 

/opt/conda/lib/python3.9/site-packages/pydantic/main.cpython-39-x86_64-linux-gnu.so in pydantic.main.ModelMetaclass.__new__()

/opt/conda/lib/python3.9/abc.py in __new__(mcls, name, bases, namespace, **kwargs)
    104         """
    105         def __new__(mcls, name, bases, namespace, **kwargs):
--> 106             cls = super().__new__(mcls, name, bases, namespace, **kwargs)
    107             _abc_init(cls)
    108             return cls

TypeError: multiple bases have instance lay-out conflict

[srini047]

Hi @srini047,

I just tested the notebook with the google colab Free version, it works as expected!

But, I am facing some conflicts while running the same in the naas environment. I am not able to install llama_index. Kindly help me to proceed further. Thanks.

Error:

TypeError: multiple bases have instance lay-out conflict

Click to toggle Full Error Message

@mahimairaja Just add how to generate the Neo4j graph instance code.

[mahimairaja]

@mahimairaja Just add how to generate the Neo4j graph instance code.

Thanks for letting know, I have added now, please check it

[srini047]

LGTM🚀
Nice work @mahimairaja. @FlorentLvr could have a final look.

[mahimairaja]

Kindly share your valuable comments @FlorentLvr

[github-actions]

The template is now available on the master branch on this link:
https://github.com/jupyter-naas/awesome-notebooks/blob/master/LlamaIndex/LlamaIndex_Integrate_with_Neo4j.ipynb

[github-actions]

Thank you for your contribution @mahimairaja, your PR has been merged into the master branch of awesome-notebook.
Here is the contribution certificate you can share on social media so everybody knows how awesome you are 🤙🌎.
Spread the #opensource love 💚