Josh Phillips
Josh Phillips
I was seeing the same issue with all sorts of different doc lengths and after coming across this post I updated everything to the below versions and now it is...
> I updated everything to the below versions and now it is working python=3.11.6 supabase-1.2.0 langchain=0.0.314 interestingly enough came across one just now that failed with the same error again....
The one that is failing is massive. I run into rate limiting from openAI on the one that fails vs the ones that don't. I am taking a massive json...
I have also followed these steps and it works on others tables setup the same way.
@danielchalef any ideas regarding what we are seeing?
> So my workaround for now is to ship only 100 chunks and 100 vectors in each call to `add_vectors`. put up a pr! I don't know how to use...
> I no longer use supbase and unfortunately, don't have time to dig in to optimize this. The VectorStore can likely be simplified and improved significantly by refactoring to use...
settings chunk size smaller worked on my large dataset. working on a PR
PR is up! Made it so chunk_size is configurable in the SupabaseVectoreStore call! Hopefully gets in soon so folks can tinker to get there databases back up and running! >...