namedVectors with custom embedder?

Support
1

Description

I literally can’t find documentation on how to use (create and perform vector search) namedVectors (like 2 vectors per collection) using my custom vectors. Can anyone point me to the sample script showing how to do that, preferably using V4 syntax?

2

hi @noban !!

Welcome to our community!

Sorry! Missed this message. :thinking:
Just discovered some messages that went under my radar :frowning:

Hope you were able to solve this already :frowning:

Luckly, I have just crafted some sample code that can help you on this topic (or for other in the future):

This example shows how to use a custom embed model (running locally), that is also offered by Hugging face.

Now, if you provide the named vectors, like in here:

            batch.add_object(
                properties={"title": line},
                vector={
                    "title_vector": vector
                }
            )

Weaviate will not reach out to Hugging face to vectorize the object.

While performing searches, you can do the same. For example:

client.connect()
collection = client.collections.get("Test")

query_vector = model.encode(["pet animals"])

objects = collection.query.near_vector(
    near_vector=query_vector[0], 
    return_metadata=wvc.query.MetadataQuery(distance=True),
    include_vector=True,
    target_vector="title_vector"
).objects
for object in objects:
    print("#" * 10)
    print(object.metadata.distance)
    print(object.properties)

Let me know if this helps!

Thanks!