Is it possible to incrementally add corpus to a retriever?

[ANYMS-A]

Hi,
currently I have a lot of corpus which I dont want to load them all into the memory and use

retriever.save("bm25s_very_big_index", corpus=corpus)

to create the retriever.

Is it possible to incrementally add corpus to a retriever? Thanks!

[xhluca]

This has been discussed here: #20