Allow query embeddings to be passed in max_marginal_relevance_search

[rishabh208gupta]

As part of elastic/elasticsearch-py#2620 , elasticsearch store mmr search allows query or query_embeddings optionally. It would be nice to have it in

langchain-elastic/libs/elasticsearch/langchain_elasticsearch/vectorstores.py

Line 882 in 19cb4ab

def max_marginal_relevance_search(

as well.

P.S. Can I work on it?

[pquentin]

Thanks for your interest @rishabh208gupta! Sure, you can work on this 👍

[rishabh208gupta]

Hi @pquentin, is there any document I can follow to run the tests on local?

[pquentin]

I never worked on this repo before, and it does not look like running tests is documented. You can try reproducing the steps in https://github.com/langchain-ai/langchain-elastic/blob/main/.github/workflows/_test.yml

[rishabh208gupta]

Okay thanks :)

[miguelgrinberg]

Information for contributors isn't currently available, this is something that we plan to improve. But in the meantime, you can run the test suite locally from the lib/elasticsearch directory:

poetry install --with lint,test
make lint
make test
make integration_tests

[rishabh208gupta]

I realise that to make this change, we would need to change the method definition in https://github.com/langchain-ai/langchain/blob/b28bc252c4e7058951c9d3be8ee81595029eab83/libs/core/langchain_core/vectorstores/base.py#L679C9-L679C38

[miguelgrinberg]

@rishabh208gupta as long as it changes in a backwards compatible manner, it is not a problem.