Jinjing Zhou

@bokesyo Do you have any plan for the multi vector representations? In Colpali report, multi vector showed much better performance than single vector

@bokesyo Also curious are you affiliated to OpenBMB? I saw the embedding model is under a different organization.

Can you try manuually deleta all the files under `pgdata/pgvecto_rs` and reindex all the vector column?

@cho-thinkfree-com Sorry for the problem. We haven't tested such large scale tables with indexes so far. The problem you met might due to insufficient memory. 30M dense vector+30M sparse vector...

Thanks for your interest. If you met any problems, please feel free to raise issues and we would be glad to help. And I think we can make the issue...

It seems weird and probably related the jax's runtime. We may need some time to investigate jax and this issue. Thanks again for your conrtibution!