Skip to main content

​
GoogleEncoder Objects

class GoogleEncoder(DenseEncoder)
GoogleEncoder class for generating embeddings using Google’s AI Platform. The GoogleEncoder class is a subclass of DenseEncoder and utilizes the TextEmbeddingModel from the Google AI Platform to generate embeddings for given documents. It requires a Google Cloud project ID and supports customization of the pre-trained model, score threshold, location, and API endpoint. Example usage: 
from semantic_router.encoders.google_encoder import GoogleEncoder

encoder = GoogleEncoder(project_id="your-project-id")
embeddings = encoder(["document1", "document2"])
Attributes:
  • client - An instance of the TextEmbeddingModel client.
  • type - The type of the encoder, which is “google”.

​
__init__

def __init__(name: Optional[str] = None,
             score_threshold: float = 0.75,
             project_id: Optional[str] = None,
             location: Optional[str] = None,
             api_endpoint: Optional[str] = None)
Initializes the GoogleEncoder. Arguments:
  • model_name (str): The name of the pre-trained model to use for embedding. If not provided, the default model specified in EncoderDefault will be used.
  • score_threshold (float): The threshold for similarity scores.
  • project_id (str): The Google Cloud project ID. If not provided, it will be retrieved from the GOOGLE_PROJECT_ID environment variable.
  • location (str): The location of the AI Platform resources. If not provided, it will be retrieved from the GOOGLE_LOCATION environment variable, defaulting to “us-central1”.
  • api_endpoint (str): The API endpoint for the AI Platform. If not provided, it will be retrieved from the GOOGLE_API_ENDPOINT environment variable.
Raises:
  • str0: If the Google Project ID is not provided or if the AI Platform client fails to initialize.

​
__call__

def __call__(docs: List[str]) -> List[List[float]]
Generates embeddings for the given documents. Arguments:
  • docs (List[str]): A list of strings representing the documents to embed.
Raises:
  • ValueError: If the Google AI Platform client is not initialized or if the API call fails.
Returns: List[List[float]]: A list of lists, where each inner list contains the embedding values for a document.