A follow up question, like the tutorial you sent, there's various documents for each city "Boston", "Houston", so on and so forth. What is the best way to handle multi-modal data for each of the cities? For example, if "Boston" had a text document and a CSV file and you wanted to use different indices for each data type
If you want to use different indices for each data type (rather than combining them), I think you just have to get a little more specific with the "summaries" of each index.
Like, what's in the csv that's not in the text document? The summary should be clear
