The old function got replaced and might get removed in future versions. To ensure that the store is compatible with Neo4j 5.LATEST, this commit changes the store function for embeddings to use the 5.13+ function. The correct baseline database version is already mentioned in the README.main...
The Chroma constructor in the vectorstore section uses the document function when it should be the query function for embeddings. As a result, if the documents parameter is blank when using Chroma, Langchain will error out with a ValidationError. Please change line 95 to be embed_query instead...
In the end, both are just HTTP endpoints, as are all the other API-based endpoints. If I can merge the PRs, then we can use the OpenAI EF. Sorry, something went wrong. Copy link technovangelistMar 23, 2024 @tazarovare you saying Yes, you are just using huggingface under the hood,...