Index a document set for retrieval

Embeds every chunk with a small encoder (by default BAAI/bge-small-en-v1.5, about 130 MB, downloaded once) and stores the vectors next to the chunks. dragon_documents() does this for you unless index = FALSE; call this after editing files, or to switch encoders. Nothing is re-embedded when the chunks have not changed.

Usage

dragon_index(docs, model = "BAAI/bge-small-en-v1.5", force = FALSE)

Arguments

docs

A dragon_documents object or its directory.

model

Hugging Face id of a sentence-embedding model.

force

Re-embed even if the index already matches the chunks.

Value

The dragon_documents object, reloaded, invisibly.