WeaviateQueryError: Query call with protocol GRPC search failed

[Agam]

Description

from dotenv import load_dotenv

load_dotenv()

client = weaviate.connect_to_weaviate_cloud(
    cluster_url=os.getenv("WEAVIATE_URL"),  # Replace with your Weaviate Cloud URL
    auth_credentials=weaviate.auth.AuthApiKey(
        os.getenv("WEAVIATE_KEY")
    ),  # Replace with your Weaviate Cloud key
    headers={
        "X-OpenAI-Api-key": os.getenv("OPENAI_APIKEY")
    },  # Replace with your OpenAI API key
)


def create_collection(collection_name: str):
    client.collections.delete(collection_name)
    client.collections.create(
        collection_name,
        properties=[
            Property(name="page", data_type=DataType.INT),
            Property(name="text", data_type=DataType.TEXT),
        ],
    )


def store_new_doc(collection_name: str):
    extracted_text = [
        {
            "text": "Some test about dogs.",
            "page": 0,
        },
        {
            "text": "Test some text about cats.",
            "page": 1,
        },
    ]
    print(extracted_text)

    collection = client.collections.get(collection_name)

    with collection.batch.dynamic() as batch:
        for data_row in extracted_text:
            batch.add_object(
                properties=data_row,
            )


def query(collection_name: str, query_text: str):
    collection = client.collections.get(collection_name)

    response = collection.query.near_text(
        query=query_text,
        limit=3,
        return_metadata=MetadataQuery(score=True),
    )

    for o in response.objects:
        print(o.properties)
        print(o.metadata.score)


if __name__ == "__main__":
    create_collection("Test")
    store_new_doc("Test")
    query("Test", "What animal says meow?")

And I get:

    raise WeaviateQueryError(e.details(), "GRPC search")  # pyright: ignore
weaviate.exceptions.WeaviateQueryError: Query call with protocol GRPC search failed with message panic occurred: ValidateParam was called without any known params present.

Server Setup Information

• Weaviate Server Version: Cloud
• Deployment Method:
• Multi Node? Number of Running Nodes: 1
• Client Language and Version: EN, 4.6.4
• Multitenancy?: Nope

Any additional Information

[DudaNogueira]

Hi @Agam !!

Welcome to our community

The issue is that while you have passed the OpenAI api key, you never specified what vectorizer and generative integration you want.

Here, this should be you collection creation code:

from weaviate.classes.config import Configure
client.collections.create(
    collection_name,
    vectorizer_config=Configure.Vectorizer.text2vec_openai(),
    generative_config=Configure.Generative.openai(),
    properties=[
        Property(name="page", data_type=DataType.INT),
        Property(name="text", data_type=DataType.TEXT),
    ],
)

Check here for this step in our Quickstart:

Quickstart Tutorial | Weaviate - Vector Database

Let me know if this helps!

Thanks!

[DudaNogueira]

By the way, This error message doesn’t give much on the actual issue

We are aware

This error message should improve in future versions.

Thanks!

[Agam]

@DudaNogueira

from weaviate.classes.config import Configure

Awesome, yeah the error was kind of cryptic, good to hear you are looking into improving the DX.

Now I’m getting:

    self.__pydantic_validator__.validate_python(data, self_instance=self)
pydantic_core._pydantic_core.ValidationError: 1 validation error for _CollectionConfigCreate
generative_config
  Input should be a valid dictionary or instance of _GenerativeConfigCreate [type=model_type, input_value=<function _Generative.openai at 0x7f47a346ca60>, input_type=function]

I tried following the tutorial, ended with this:

def create_collection(collection_name: str):
    client.collections.delete(collection_name)
    client.collections.create(
        collection_name,
        vectorizer_config=Configure.Vectorizer.text2vec_openai(),
        generative_config=Configure.Generative.openai,
        properties=[
            Property(name="text", data_type=DataType.TEXT),
            Property(name="page", data_type=DataType.INT),
        ],
    )

[Agam]

Ok your example was missing () for the open ai.
Now it works!

[DudaNogueira]

That’s right!!

Glad we figured it out.

If you have any other issues, let us know!