Vector for Target Multidimensional Target Vectors

[nathanwilk7]

I was debugging the script below from @rlmanrique using this commit bd7db7cb34bf94ece9b1c5f9a56e31afc06e6f0e on Weaviate main and believe that we should still set the target vectors for "multidimensional vectors" in the same way we do for other vectors. I'm not familiar with the Python client code so please correct me if I'm mistaken.

Also, left a code clarification question as a PR comment.

I hit an issue I haven't had time to fix when trying to run tests locally, so I have not run the test suite locally, curious to see the test results are here.

import weaviate
import weaviate.classes.config as wvc


if __name__ == "__main__":

    client = weaviate.connect_to_local()

    # Delete collection
    client.collections.delete("CollectionMultivector")

    vector_index_config = wvc.Configure.VectorIndex.hnsw(
        ef=512,
        multi_vector=wvc.Configure.VectorIndex.MultiVector.multi_vector(),
    )
    vectorizers = [
        wvc.Configure.NamedVectors.none(
            name="colbert",
            vector_index_config=vector_index_config,
        ),
        wvc.Configure.NamedVectors.none(
            name="regular",
            vector_index_config=vector_index_config,
        )
    ]

    collection = client.collections.create(
        name="CollectionMultivector",
        vectorizer_config=vectorizers,
        replication_config=wvc.Configure.replication(factor=1),
    )

    vector = {
        "colbert": weaviate.classes.query.NearVector.multidimensional([[1, 2], [3, 4]])
    }
    response = collection.query.near_vector(
        near_vector=vector, limit=10, target_vector="colbert"
    )

    client.close()

Error from server:

E           	details = "extract target vectors: class CollectionRF1 has multiple vectors, but no target vectors were provided"

[orca-security-eu]

Orca Security Scan Summary

Status	Check	Issues by priority
Passed	Secrets	0   0   0   0	View in Orca

[weaviate-git-bot]

To avoid any confusion in the future about your contribution to Weaviate, we work with a Contributor License Agreement. If you agree, you can simply add a comment to this PR that you agree with the CLA so that we can merge.

beep boop - the Weaviate bot 👋🤖

PS:
Are you already a member of the Weaviate Slack channel?