bclavie / RAGatouille

Easily use and train state of the art late-interaction retrieval methods (ColBERT) in any RAG pipeline. Designed for modularity and ease-of-use, backed by research.
Apache License 2.0
2.45k stars 173 forks source link

Index metadata does not change after add_to_index or delete_from_index #192

Open newmanar opened 2 months ago

newmanar commented 2 months ago

After I built an index, I thought I was being clever by reading in docid_metadata_map.json to be able to review metadata and select documents I want to work with. However, when adding (add_to_index) or removing (delete_from_index) from the index, it appears that while the operations seem successful, the docid_metadata_map.json and pid_docid_map.json files are not updated.

Is this expected behavior? I'd think we'd want to have those files updated to allow for effective metadata management.

Thanks for any insights.