Find answers from the community

Updated 2 months ago

hey ravitheja 0475 Logan M 8260

hey ,
we have seen the Auto Retriever (https://gpt-index.readthedocs.io/en/latest/examples/vector_stores/pinecone_auto_retriever.html)
I had some questions
1> will this work with a nodes having multiple tags per category
2> will this scale well with our use-case (i.e. a company can have a lot of documents and about 15-20 tags and documents many will have similar tags),
so should you advise using this for better filtration of nodes?
(as for us we will have to summarize the whole document then classify under multiple classes for storing and then passing the multiple classes in MetadataInfo for finding the best tags for the user's query)
S
L
2 comments
please assist here
cc: @Logan M @ravitheja @jerryjliu0
Tbh I can't comment for sure on either point. I would suggest trying it with a small dataset to see what happens though!
Add a reply
Sign up and join the conversation on Discord