encode_queries()
This operation takes in a list of query strings and encodes each query into a vector embedding.
Request syntax
encode_queries(
queries: List[str],
) -> List[np.array]
PARAMETERS:
queries (List[str])
A list of string values, where each string represents a query that will be passed to the embedding model for encoding. The model will generate an embedding vector for each string in the list.
RETURN TYPE:
List[np.array]
RETURNS:
A list where each element is a NumPy array.
Exceptions:
ImportError
This exception will be raised when the Voyage module is not installed.
Examples
from pymilvus.model.dense import VoyageEmbeddingFunction
voyage_ef = VoyageEmbeddingFunction(
model_name="voyage-lite-02-instruct", # Defaults to `voyage-2`
api_key='YOUR_API_KEY' # Replace with your own Voyage API key
)
queries = ["When was artificial intelligence founded",
"Where was Alan Turing born?"]
query_embeddings = voyage_ef.encode_queries(queries)
print("Embeddings:", query_embeddings)
print("Dim", voyage_ef.dim, query_embeddings[0].shape)
# Embeddings: [array([ 0.01733501, -0.0230672 , -0.05208827, ..., -0.00957995,
# 0.04493361, 0.01485138]), array([ 0.05937521, -0.00729363, -0.02184347, ..., -0.02107683,
# 0.05706626, 0.0263358 ])]
# Dim 1024 (1024,)