When I'm trying to import SImpleDirectoryREader using this line: from llama_index.core import SimpleDirectoryReader, Somehow it's giving me this error: ImportError: llama-index-readers-file package not found. I tried uninstalling but still face the same issue.

Solved by defining embed_model and service context but got a new problem. WHEN I DO pack.run it says: ValueError: Only one free input key is allowed.

I'm an azure open ai user, it seems like now it's fairly hard to add the embedding model to my service context definition. IS there an example that I can follow?

I'm trying to use AzureOpenAI and the latest gpt-4-1106 preview model to try and extract tables from PDF. I was able to set up the AzureOpenAI object and run a complete method for completion. However, when I try to create a VectorstoreIndex, it fails and gives me the error AttributeError: module 'openai' has no attribute 'error'. ANyone running into this issue before?

I'm using MIlvus Vector Store as my backend, wondering if I'm using the vecstoreindex implementation by llama index, how can I better utilize the metadata I fetched from the nodes.

Say I don't have acces to the OPenAI 0613 function calling api, but only have access to the 0315 one. How can I leverage pydantic for more structured outputs?

I'm trying to use the SImpleDIrectoryREader to read ta directory of pptx files, but I was getting the OSError: cannot find loader for this WMF file. WOndering what's the fastest way for me to bypass the error to just skip the file. I think the error happens because I have images in the files that can't be extracted.

I'm using metadata extractor to generate some metadata for documents. However, when I try to load the documents into MIlvus, I was getting the error: TypeError: Object of type set is not JSON serializable. Fix: So I'm using the entity extractor and it's returning a set

When using the SubQuestionQueryEngine, I keep running into the issue of 'KeyError: 'comparison'. I have multiple documents as query engine tool and feed them in the query_engine_tools parameter in constructing the SubQuestionQueryENgine.

If I have the doc_id, is there a way to access the doc text using the doc_id?

Token indices sequence length is longer than the specified maximum sequence length for this model (1215 > 1024). Running this sequence through the model will result in indexing errors. I updaetd the library to latest one and saw this error. WOnder where I could specify the length to be more than 1024.

Say I already have a collection of tree indices. How should I compose a graph over these tree indices to handle users' questions on them. How to generate a effective summary over them?

Will the choice of chunk_size_limit also afffect the query performance for GPTListIndex?

Hi all- I'm trying to compose a simplekeywordtableIndex over a tree index but when I queried the graph, it's giving me the error: integer divison or modulo by zero

When I tried to use the AZureOPenAI llm as input for the openai agent, it SAYS LLM MUST BE A OPENAI INSTANCE. Does it not take a azure instance as the llm?

I have a usecase where a set of documents need to be categorized using one or a few of the tags in a pre-defined list. WIll llama index be a good tool for this use case?

I'm using the PDFReader to read in my document. Is there a way to specify how many pages you want in one chunk? I think for now, the default is to have each page as one document object. I'd like to have 5-10 pages in one document object.

I was using QA summary graph as my query engine. One question that I want to ask my document is a value extraction question, which uses the vecstore index. The response i get is incorrect but when I went back and checked the log. I realized it got it right in the initial attempt but messed up with the refine template. I know this issue is kinda vague but are there ways for us to stop refining after getting it right in the first place.

I didn't see a load_from_disk method now for indices