While using LLMGraphTransformer getting AttributeError: 'dict' object has no attribute 'nodes'

[talhaanwarch]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

text = """Sarah is an employee at prismaticAI, a leading technology company based in Westside Valley. She has been working there for the past three years as a software engineer.
Michael is also an employee at prismaticAI, where he works as a data scientist. He joined the company two years ago after completing his graduate studies.
prismaticAI is a well-known technology company that specializes in developing cutting-edge software solutions and artificial intelligence applications. The company has a diverse workforce of talented individuals from various backgrounds.
Both Sarah and Michael are highly skilled professionals who contribute significantly to prismaticAI's success. They work closely with their respective teams to develop innovative products and services that meet the evolving needs of the company's clients."""

with open("sample.txt","w") as f:
    f.write(text)
from langchain.document_loaders import TextLoader
from langchain.text_splitter import CharacterTextSplitter


loader = TextLoader('sample.txt')
documents = loader.load()
text_splitter = CharacterTextSplitter(chunk_size=200, chunk_overlap=20)
texts = text_splitter.split_documents(documents)

from langchain_openai import AzureChatOpenAI
llm = AzureChatOpenAI(
    azure_deployment="gpt-4o",
    api_version="2024-05-01-preview",
    temperature=0,
    openai_api_key="xxxxxxxxxxxxxxxxxxxxxx",
    azure_endpoint="https://xxxxxxx.openai.azure.com/")
    
from langchain_experimental.graph_transformers.llm import LLMGraphTransformer
llm_transformer = LLMGraphTransformer(llm=llm)
graph_documents = llm_transformer.convert_to_graph_documents(texts)

Description

I am trying to use LLMGraphTransformer but getting error AttributeError: 'dict' object has no attribute 'nodes' with gpt-4o. But if i use gpt-4o-miniI did not get any error but no node and relationships are generated as well.

Response in case of gpt-4o-mini

[GraphDocument(nodes=[], relationships=[], source=Document(page_content="Sarah is an employee at prismaticAI, a leading technology company based in Westside Valley. She has been working there for the past three years as a software engineer.\nMichael is also an employee at prismaticAI, where he works as a data scientist. He joined the company two years ago after completing his graduate studies.\nprismaticAI is a well-known technology company that specializes in developing cutting-edge software solutions and artificial intelligence applications. The company has a diverse workforce of talented individuals from various backgrounds.\nBoth Sarah and Michael are highly skilled professionals who contribute significantly to prismaticAI's success. They work closely with their respective teams to develop innovative products and services that meet the evolving needs of the company's clients.", metadata={'source': 'sample.txt'}))]

Error in case of gpt-4o

Error traceback

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Cell In[16], line 4
      2 # Extract Knowledge Graph
      3 llm_transformer = LLMGraphTransformer(llm=llm)
----> 4 graph_documents = llm_transformer.convert_to_graph_documents(texts)

File [~/venv/lib/python3.10/site-packages/langchain_experimental/graph_transformers/llm.py:801](http://localhost:8889/lab/tree/others/knowledgegraph/~/venv/lib/python3.10/site-packages/langchain_experimental/graph_transformers/llm.py#line=800), in LLMGraphTransformer.convert_to_graph_documents(self, documents, config)
    789 def convert_to_graph_documents(
    790     self, documents: Sequence[Document], config: Optional[RunnableConfig] = None
    791 ) -> List[GraphDocument]:
    792     """Convert a sequence of documents into graph documents.
    793 
    794     Args:
   (...)
    799         Sequence[GraphDocument]: The transformed documents as graphs.
    800     """
--> 801     return [self.process_response(document, config) for document in documents]

File [~/venv/lib/python3.10/site-packages/langchain_experimental/graph_transformers/llm.py:801](http://localhost:8889/lab/tree/others/knowledgegraph/~/venv/lib/python3.10/site-packages/langchain_experimental/graph_transformers/llm.py#line=800), in <listcomp>(.0)
    789 def convert_to_graph_documents(
    790     self, documents: Sequence[Document], config: Optional[RunnableConfig] = None
    791 ) -> List[GraphDocument]:
    792     """Convert a sequence of documents into graph documents.
    793 
    794     Args:
   (...)
    799         Sequence[GraphDocument]: The transformed documents as graphs.
    800     """
--> 801     return [self.process_response(document, config) for document in documents]

File [~/venv/lib/python3.10/site-packages/langchain_experimental/graph_transformers/llm.py:742](http://localhost:8889/lab/tree/others/knowledgegraph/~/venv/lib/python3.10/site-packages/langchain_experimental/graph_transformers/llm.py#line=741), in LLMGraphTransformer.process_response(self, document, config)
    740 if self._function_call:
    741     raw_schema = cast(Dict[Any, Any], raw_schema)
--> 742     nodes, relationships = _convert_to_graph_document(raw_schema)
    743 else:
    744     nodes_set = set()

File [~/venv/lib/python3.10/site-packages/langchain_experimental/graph_transformers/llm.py:609](http://localhost:8889/lab/tree/others/knowledgegraph/~/venv/lib/python3.10/site-packages/langchain_experimental/graph_transformers/llm.py#line=608), in _convert_to_graph_document(raw_schema)
    605 else:  # If there are no validation errors use parsed pydantic object
    606     parsed_schema: _Graph = raw_schema["parsed"]
    607     nodes = (
    608         [map_to_base_node(node) for node in parsed_schema.nodes if node.id]
--> 609         if parsed_schema.nodes
    610         else []
    611     )
    613     relationships = (
    614         [
    615             map_to_base_relationship(rel)
   (...)
    620         else []
    621     )
    622 # Title [/](http://localhost:8889/) Capitalize

AttributeError: 'dict' object has no attribute 'nodes'

System Info

System Information

OS: Linux
OS Version: #45~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Wed Sep 11 15:25:05 UTC 2
Python Version: 3.10.12 (main, Sep 11 2024, 15:47:36) [GCC 11.4.0]

Package Information

langchain_core: 0.1.52
langchain: 0.1.14
langchain_community: 0.0.30
langsmith: 0.1.38
langchain_cohere: 0.3.0
langchain_experimental: 0.3.0
langchain_openai: 0.1.1
langchain_text_splitters: 0.0.2

Packages not installed (Not Necessarily a Problem)

The following packages were not found:

langgraph
langserve