Logan comments

Results 685 comments of


                                            Logan

[Question]: Is it possible integrate langchain ConversationBufferMemory ?

Yea go for it

Updating an index based on changes

This should be working now, assuming you set an explicit/consistent doc_id for each document. More details here: https://gpt-index.readthedocs.io/en/latest/how_to/index/document_management.html https://github.com/jerryjliu/llama_index/tree/main/docs/examples/discover_llamaindex/document_management/ https://www.youtube.com/watch?v=j6dJcODLd_c

[Bug]: `NodeParser` Previous and Next Node Relationships Cross Document Boundaries

Yea good catch. Feel free to propose a PR :)

[Bug]: Error on building and retrieving data using query with AI Search + llama-index

I dont have access to azure to test this. But I know the azure team is working on updating this integration, hopefully soon :)

[Documentation]: Add "Retrieving Sources"-Notebook /Guide

Good suggestion! as a quick note, I will point out that the response object always has the source nodes used to create the response `print(response.source_nodes)` -- each source node has...

[Question]: Why is this llama-index query to Ollama (llama3) always timing out?

Set the request timeout larger, I usually do something like `request_timeout=3000.0`

[Question]: OpenAIAgent.from_tools not working with streamlit, works fine outside it.

Can you give more details? Is it writing the same sql query? Is it returning a response? It could just be variation when you have different sql data or different...

[Question]: OpenAIAgent.from_tools not working with streamlit, works fine outside it.

Seems like maybe a difference in installed dependencies? Thats my best guess

improve agent ux with chatgpt

Sounds good! I know ChatGPT has proven to be a little... obtuse to work with lol

how to use a model loaded from HuggingFace transformers?

Yea, I second this. Is there no way to use LangChain with a model that is loaded locally?