I am using LangChain's Neo4j Vector Index, and its Neo4jVector.from_existing_index function.
This is an example I found for the structure:
movie_plot_vector = Neo4jVector.from_existing_index(
embedding_provider,
url="url",
username="neo4j",
password="pasword",
index_name="customerIndex",
embedding_node_property="embedding_customer_name",
text_node_property="customer_name",
)
plot_retriever = RetrievalQA.from_llm(
llm=llm,
retriever=movie_plot_vector.as_retriever(),
verbose=True,
return_source_documents=True
)
I want to know if there is a way to search on multiple indexes, what I would imagine this would look like:
movie_plot_vector = Neo4jVector.from_existing_index(
embedding_provider,
url="url",
username="neo4j",
password="pasword",
index_name=["customerIndex","problemIndex"]
embedding_node_property=["embedding_customer_name","embedding_problem"]
text_node_property=["customer_name","problem_description"]
)
plot_retriever = RetrievalQA.from_llm(
llm=llm,
retriever=movie_plot_vector.as_retriever(),
verbose=True,
return_source_documents=True
)
Or is there another function from Neo4jVector which does what I need? Can anyone guide me to find a solution please?