Could you please advise on the process of querying Chroma DB to retrieve the names of documents for which embeddings are available, leveraging the Lamma index?

Hi, I want to use chroma db instead of in memeory (InMemoryDocumentStore) for generation of test data using RAGAS

documents = SimpleDirectoryReader(
input_files=[
"../data/Xceed Fraud Detection Reference Manual_01March2023_Track Changes.docx",
"../data/FraudDESK Guide - ACH_v6.pdf",
"../data/FraudDESK Guide_Online Banking_v6.pdf",
"../data/FraudUseCases_13July2023.xlsx"
]
).load_data()

azure_model = LangchainLLMWrapper(AzureChatOpenAI(
model=config.chatgpt_model,
azure_deployment=config.openai_deployment_id,
api_key=config.openai_api_key,
azure_endpoint=config.openai_api_base,
api_version=config.openai_api_version,
)
)

embed_model = LangchainEmbeddingsWrapper(AzureOpenAIEmbeddings(
model=config.embed_model,
azure_deployment=config.embed_model_deployment_id,
api_key=config.openai_api_key,
azure_endpoint=config.openai_api_base,
api_version=config.openai_api_version,
))

generator_llm = azure_model
critic_llm = azure_model

splitter = TokenTextSplitter(chunk_size=1000, chunk_overlap=100)
keyphrase_extractor = KeyphraseExtractor(llm=generator_llm)
docstore = InMemoryDocumentStore(
splitter=splitter,
embeddings=embed_model,
extractor=keyphrase_extractor,
)
from ragas.testset import TestsetGenerator
from ragas.testset.evolutions import simple, reasoning, multi_context

test_generator = TestsetGenerator(
generator_llm=generator_llm,
critic_llm=critic_llm,
embeddings=embed_model,
docstore=docstore,
)

testset = test_generator.generate_with_llamaindex_docs(documents=documents[5:6],
test_size=2, distributions={simple: 0.5, reasoning: 0.25,
multi_context: 0.25})
print(testset)

Please help below error appeared when I was evaluating RAG using LlamaDataset from LlamaHub raise mapped_exc(message) from exc
httpx.LocalProtocolError: Illegal header value b'Bearer '

During handling of the above exception, another exception occurred: raise APIConnectionError(request=request) from err
openai.APIConnectionError: Connection error.

And below error appeared : File "C:\Users\akhandey.NICEDEN\copilot\copilot\document_reader.py", line 45, in run_query
index = VectorStoreIndex.from_documents(nodes, storage_context=storage_context)
File "C:\Users\akhandey.NICEDEN\AppData\Roaming\Python\Python310\site-packages\llama_index\indices\base.py", line 97, in from_documents
docstore.set_document_hash(doc.get_doc_id(), doc.hash)
AttributeError: 'TextNode' object has no attribute 'get_doc_id'. Did you mean: 'ref_doc_id'? Please help

The Llama index library that you are using needs openai>=1.0.1 . And existing copilot code needs openai==0.27.6
This is creating a conflict. Does anyone have solution?

How to use azure openai with llamandex?

I have several chorma DB collections, and I want to route queries to the most relevant collection based on a similarity score. Ideally, the document with the highest similarity score within a collection would be retrieved. Can LlamaIndex leverage its retrieval capabilities to achieve this?

I want to log cosine similarity of RAG response while retrieving from chroma db ? Please refer below code for this: .vector_store = self.setup_chroma(collectionName)
storage_context = StorageContext.from_defaults(vector_store=vector_store)

documents = SimpleDirectoryReader(
input_files = files
).load_data()

node_parser = SentenceSplitter(chunk_size=512, chunk_overlap=50)
nodes = node_parser.get_nodes_from_documents(documents)
index = VectorStoreIndex(nodes, storage_context=storage_context)

query_engine = index.as_query_engine()
query_engine.query(query)

but this work very well :vector_store = self.setup_chroma(collectionName)
storage_context = StorageContext.from_defaults(vector_store=vector_store)

documents = SimpleDirectoryReader(
input_files = files
).load_data()

node_parser = SentenceSplitter(chunk_size=180, chunk_overlap=80)
nodes = node_parser.get_nodes_from_documents(documents)
index = VectorStoreIndex(nodes, storage_context=storage_context)

My Requirement is : When updating RAG, only generate embeddings for newly added documents rather than regenerating embeddings for all documents in chroma db using llamaindex.