Find answers from the community

Updated 9 months ago

Hello, is there a way to persist the

At a glance

The community member is asking if there is a way to persist the index that the bm25retriever builds to disk. The comments suggest that the index cannot be directly persisted, but the community members recommend saving and loading the nodes and then re-inputting them into the bm25 model. Additionally, it is mentioned that the bm25retriever is built on top of the rank_bm25 library.

aanonymous_lambda

Hello, is there a way to persist the index that the bm25retriever builds to disk? Thank you.

2 comments

LLogan M

nope, you just have to save and load the nodes and throw them back into bm25

LLogan M

under the hood its just using the rank_bm25 library

Add a reply