Find answers from the community

Updated 3 months ago

Hello we re trying to use our support

Hello, we're trying to use our support chat history (from our database) to feed the llama_index, right now I've exported all the data to a JSON file and put it on the "data" folder (millions of lines). But I was wondering how will the AI know what is the customer question vs what is the chat attendant answer? Is there a way to send like an index for each phrase (QUESTION:, ANSWER:)?

5 comments

LLogan M

you may need to write your own data loader and add that manually.

It's pretty easy actually! Just load the JSON in python, iterate over it, and manually create Document objects

Full guide on documents here: https://gpt-index.readthedocs.io/en/latest/core_modules/data_modules/documents_and_nodes/usage_documents.html

MMarco Castellari

Thank you Logan, I'll take a look!

MMarco Castellari

Hey Logan, sorry for being annoying, but can u give me one more help?

MMarco Castellari

Like on this sample, how would I fill in a way that AI will know what is a "question" and what is the "answer" for this question?

Attachment

Captura_de_Tela_2023-07-19_as_19.36.44.png

LLogan M

I would make each document a labeled question/answer pair

So you can set the text to be exactly that

Plain Text

Document(text="Question: question text?/n/nAnswer: answer text")

Add a reply