Find answers from the community

Updated 2 years ago

I would like to use llama index between

I would like to use llama index between 2 documents (which are 2 csv files) i used the llama hub CSV Loader and worked great it loaded it into 2 documents. But im kinda stuck on how do i do I write my build my query engines and indexing them and then later try to give out an sql statement for my response using these 2 tables.

6 comments

LLogan M

Are the CSVs complex and numerical? Or is it like a simple list of QA pairs?

NN01R

It’s a complex dataset from kaggle filled with the list of all the restaurants in Uber eats in one table and their menu with their ratings in another

NN01R

Here's the dataset https://www.kaggle.com/datasets/ahmedshahriarsakib/uber-eats-usa-restaurants-menus

LLogan M

Cool! So to answer questions like this, tbh I would convert it into a sqlite database with two tables, and use the SQL index to answer questions across both tables

tbh it's a little annoying to convert the CSV files as a user for this. We could probably provide some automatic conversion for this at some point _:)

NN01R

Ahhhhh

NN01R

ok cool i think i have a rough idea on whats needed to be written

Add a reply