RAG

[bruinon]

Folks,

I am testing AutoGen for Retrieval Augmented Code Generation and Question Answering by following the notebook posted here, https://github.com/microsoft/FLAML/blob/main/notebook/autogen_agentchat_RetrieveChat.ipynb

I have a few follow-up questions,

1. How do i specify which embedding function, e.g., openai.Embedding or HuggingFaceEmbedding, can be used in RetrieveUserProxyAgent?
2. how do i configure the CharacterTextSplitter/RecursiveCharacterTextSplitter in RetrieveUserProxyAgent ?
3. I encountered this error when i was going through the example notebook.

   
   File ~/venv/GPT_venv/lib/python3.10/site-packages/autogen/retrieve_utils.py:220, in create_vector_db_from_dir(dir_path, max_tokens, client, db_path, collection_name, get_or_create, chunk_mode, must_break_at_empty_line, embedding_model)
   212     for i in range(0, len(chunks), 40000):
   213         collection.upsert(
   214             documents=chunks[
   215                 i : i + 40000
   216             ],  # we handle tokenization, embedding, and indexing automatically. You can skip that and add your own embeddings as well
   217             ids=[f"doc_{i}" for i in range(i, i + 40000)],  # unique for each doc
   218         )
   219     collection.upsert(
   --> 220         documents=chunks[i : len(chunks)],
   221         ids=[f"doc_{i}" for i in range(i, len(chunks))],  # unique for each doc
   222     )
   223 except ValueError as e:
   224     logger.warning(f"{e}")

UnboundLocalError: local variable 'i' referenced before assignment

[sonichi]

@thinkall

[thinkall]

Hi @bruinon , which version of autogen have you installed? Could you upgrade to the latest version and try this notebook https://github.com/microsoft/autogen/blob/main/notebook/agentchat_RetrieveChat.ipynb?

[thinkall]

For your question 1 and 2, we haven't implemented them yet. Would you like to raise a PR for that?

[bruinon]

UnboundLocalError: local variable 'i' referenced before assignment

Thanks @thinkall for the tips, upgrading from 0.1.3 to 0.1.5 resolved the problem;

Installing collected packages: pyautogen
  Attempting uninstall: pyautogen
    Found existing installation: pyautogen 0.1.3
    Uninstalling pyautogen-0.1.3:
      Successfully uninstalled pyautogen-0.1.3
Successfully installed pyautogen-0.1.5

Do you have insights on the previous two questions? Also in AutoGen framework, is there a way for me to update the metadata for each embedding chunk before inserting into Chroma? this feature is important to identify chunks associated with outdated documents, so that they can be replaced with new embedding chunks.

[thinkall]

UnboundLocalError: local variable 'i' referenced before assignment

Thanks @thinkall for the tips, upgrading from 0.1.3 to 0.1.5 resolved the problem;

Installing collected packages: pyautogen
  Attempting uninstall: pyautogen
    Found existing installation: pyautogen 0.1.3
    Uninstalling pyautogen-0.1.3:
      Successfully uninstalled pyautogen-0.1.3
Successfully installed pyautogen-0.1.5

Do you have insights on the previous two questions? Also in AutoGen framework, is there a way for me to update the metadata for each embedding chunk before inserting into Chroma? this feature is important to identify chunks associated with outdated documents, so that they can be replaced with new embedding chunks.

Adding docs with given doc ids is also doable, but sorry we haven't implemented that yet.

[bruinon]

For your question 1 and 2, we haven't implemented them yet. Would you like to raise a PR for that?

Yes, I would like to raise a PR for that! We would avoid the problems of mixing multiple LLM frameworks in a single application

[thinkall]

Thank you very much @bruinon. Let me know if you need any assistance on the PR.

[FacundoMartinezCampos]

UnboundLocalError: local variable 'i' referenced before assignment

Thanks @thinkall for the tips, upgrading from 0.1.3 to 0.1.5 resolved the problem;

Installing collected packages: pyautogen
  Attempting uninstall: pyautogen
    Found existing installation: pyautogen 0.1.3
    Uninstalling pyautogen-0.1.3:
      Successfully uninstalled pyautogen-0.1.3
Successfully installed pyautogen-0.1.5

Do you have insights on the previous two questions? Also in AutoGen framework, is there a way for me to update the metadata for each embedding chunk before inserting into Chroma? this feature is important to identify chunks associated with outdated documents, so that they can be replaced with new embedding chunks.

Hello, I am having the same issue, I need add metadata in each embedding in order to filter them, were you able to do it?