Hello @All.
How can I get the exact answer from the document?
After a period of use, I found that many questions the chatbot cannot answer even though the data contains information and there are some cases where the chatbot cannot answer similar questions.
How to optimize the chatbot.

Hi all.

Can llamaindex store OpenAI's vector-storage instead of the current VectorStoreIndex?

Thank you!

Hi @All

I have 2 documents as 2 guide files on different topics in one index. When querying a guide, both documents contain that guide, and when the source returns, it also provides nodes from those 2 documents. How can I select the node that best matches the answer?

Hi @all.

I cannot delete a doc in the index using the doc_id.

doc_id = f"{id}"
index.delete(doc_id)

Save the index after the update

index.storage_context.persist(persist_dir=f"{Constants.index_path}/{folder}")

I got the error:
NotImplementedError: Delete not yet implemented for Faiss index.

Hi @all.
I am using CondensePlusContextChatEngine for chatting, but I am facing an issue where the response time is too long, averaging over 12 seconds to return an answer. How can I optimize this response time?

Hi all.

I am using CondensePlusContextChatEngine. How can I generate questions from my question?

Thanks!

Hi all

I have an index for querying, and after the query is done, I want to retrieve a few queries related to the previous query. How can I do that?.

I using RetrieverQueryEngine

Hi
I'm using PrevNextNodePostprocessor with RetrieverQueryEngine. When querying, I receive the error message ValueError: doc_id not found. Please tell me what this error is caused by and how to solve it?

Settings.chunk_size = 512
docstore = SimpleDocumentStore()
documents = SimpleDirectoryReader(f"{Constants.docs_path}").load_data()
nodes = Settings.node_parser.get_nodes_from_documents(documents)
docstore.add_documents(nodes)

node_postprocessor = PrevNextNodePostprocessor(docstore=docstore, num_nodes=4)
query_engine = RetrieverQueryEngine.from_args(retriever, streaming=True, text_qa_template=prompt_tmpl, service_context=service_context, node_postprocessors=[node_postprocessor])

Hi all.
I am currently using QueryFusionRetriever for querying, but I am facing an issue where it cannot fetch additional related nodes to answer a question; it only relies on the information from one node that it finds. For example, if I have a question whose answer lies across three consecutive nodes with relationships, the results only provide data from the first node. How can I retrieve the information from all three nodes? Here is my code:

retriever = QueryFusionRetriever(
indexes,
similarity_top_k=6,
num_queries=3,
mode="simple", #reciprocal_rerank
use_async=True,
verbose=True,
)

Thank you!

Hello. I am using OpenAIAgent along with QueryEngineTool and FunctionTool, but I am facing an issue where, despite training on specific data, it still returns results from outside the trained dataset. I would like to know if there are any options to limit responses to only the data I have trained on?
Thanks all.

Hi all.
I have a question. why using service_context in CondensePlusContextChatEngine to serve for token is not working. When I run it always get 0 value
response = chat_engine.chat(input_text) print(str(token_counter.total_llm_token_count))

Hi all. Can we currently use the gpt-4o-mini model?

Hi everyone.

I'm encountering an issue with querying when training a docx file containing a description of the content I need to search for. However, when I query, I can't retrieve the desired content. Instead, the result I receive is from a different part within the file.

Please let me know how to handle this.
Thank you all very much.

Hello everybody.
How do I know if my docx file is up to standard or not? What criteria should be considered for making a standard file in terms of the data inside and during training? I'm asking because my data contains content, but when I query, I cannot retrieve that content. It only indicates that the content is not provided. Additionally, I want to ask how to query for more detailed data rather than just summarizing the main points of the content.