import os
import weaviate
client = weaviate.Client("http://localhost:8080"
)
vector_store = WeaviateVectorStore(
weaviate_client=client, index_name="Test1", text_key="text"
)
retriever = VectorStoreIndex.from_vector_store(vector_store=vector_store, embed_model='local').as_retriever(
similarity_top_k=2
) 100 kwargs.pop("storage_context", None)
101 storage_context = StorageContext.from_defaults(vector_store=vector_store)
--> 103 return cls(
104 nodes=[],
105 embed_model=embed_model,
...
115 )
117 if LCEmbeddings is not None and isinstance(embed_model, LCEmbeddings):
118 try:
ImportError: `llama-index-embeddings-huggingface` package not found, please run `pip install llama-index-embeddings-huggingface`ValueError related to the metadata field in a Weaviate vector store query. The extracts do not cover troubleshooting or fixing errors related to querying vector stores, especially with the metadata field requirements in Weaviate.metadata field of type Test1_metadata_object. This typically means that when querying a GraphQL API (which Weaviate uses for queries), you must specify which fields of the metadata object you want to retrieve in your query. metadata field.metadata field. Please note, this is a general example and not extracted from the provided documentation:{
Get {
Test1(where: {operator: Equal, path: ["page_number"], valueInt: 1}) {
text
metadata {
page_number
# Include other metadata fields you need here
}
}
}
}Test1 and you're interested in entries where the page_number is 1. It specifies that you want to retrieve the text field and a sub-selection of the metadata field, including page_number and potentially other metadata fields.metadata.metadata object containing nested properties. To query such a structure, especially when dealing with an error that requires specifying a sub-selection for the metadata field, you need to adjust your GraphQL query accordingly.{
Get {
Test1(where: {path: ["metadata", "page_number"], operator: Equal, valueString: "1"}) {
text
metadata {
embed_text
chunk_id
previous_chunk_id
next_chunk_id
page_number
file_name
file_id
}
}
}
}Test1 where the page_number within the metadata object equals "1". It specifies that you want to retrieve the text field of the document as well as various fields within the metadata object.where clause in the query uses valueString to match the page_number since your schema indicates that page_number is of data type TEXT. Adjust the valueString to valueInt or another appropriate type if the actual data type of your page_number or any other field differs.where filter in your GraphQL query, specifically related to the page_number argument. The error suggests that the query is missing an argument after page_number, which likely means there's a formatting issue with how the filter is specified.where filter in your GraphQL query is correctly formatted according to Weaviate's requirements for filtering on nested properties within an OBJECT type. where filter is properly formatted. This example assumes you're trying to filter based on the page_number within the metadata object:{
Get {
Test1(where: {path: ["metadata", "page_number"], operator: Equal, valueString: "1"}) {
text
metadata {
embed_text
chunk_id
previous_chunk_id
next_chunk_id
page_number
file_name
file_id
}
}
}
}path is correctly specified as an array, including the object name (metadata) and the property name (page_number).Equal in this case) for your filter condition.valueString in this example) that matches the data type of the field you're filtering on. Since page_number is of type TEXT in your schema, valueString is used.where filters, especially for nested properties. The documentation will provide insights into the correct syntax and available operators for filtering your queries effectively.