langchain_google_vertexai.embeddings.VertexAIEmbeddings¶

class langchain_google_vertexai.embeddings.VertexAIEmbeddings[source]¶

Bases: _VertexAICommon, Embeddings

Google Cloud VertexAI embedding models.

Initialize the sentence_transformer.

param additional_headers: Optional[Dict[str, str]] = None¶

A key-value dictionary representing additional headers for the model call

param api_endpoint: Optional[str] = None (alias 'base_url')¶

Desired API endpoint, e.g., us-central1-aiplatform.googleapis.com

param api_transport: Optional[str] = None¶

The desired API transport method, can be either ‘grpc’ or ‘rest’. Uses the default parameter in vertexai.init if defined.

param client_cert_source: Optional[Callable[[], Tuple[bytes, bytes]]] = None¶

A callback which returns client certificate bytes and private key bytes both

param credentials: Any = None¶

The default custom credentials (google.auth.credentials.Credentials) to use

param full_model_name: Optional[str] = None¶

The full name of the model’s endpoint.

param location: str = 'us-central1'¶

The default location to use when making API calls.

param max_output_tokens: Optional[int] = None (alias 'max_tokens')¶

Token limit determines the maximum amount of text output from one prompt.

param max_retries: int = 6¶

The maximum number of retries to make when generating.

param model_name: str = None (alias 'model')¶

Underlying model name.

param n: int = 1¶

How many completions to generate for each prompt.

param project: Optional[str] = None¶

The default GCP project to use when making Vertex API calls.

param request_parallelism: int = 5¶

The amount of parallelism allowed for requests issued to VertexAI models.

param safety_settings: Optional['SafetySettingsType'] = None¶

The default safety settings to use for all generations.

For example:

from langchain_google_vertexai import HarmBlockThreshold, HarmCategory

safety_settings = {

HarmCategory.HARM_CATEGORY_UNSPECIFIED: HarmBlockThreshold.BLOCK_NONE, HarmCategory.HARM_CATEGORY_DANGEROUS_CONTENT: HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE, HarmCategory.HARM_CATEGORY_HATE_SPEECH: HarmBlockThreshold.BLOCK_ONLY_HIGH, HarmCategory.HARM_CATEGORY_HARASSMENT: HarmBlockThreshold.BLOCK_LOW_AND_ABOVE, HarmCategory.HARM_CATEGORY_SEXUALLY_EXPLICIT: HarmBlockThreshold.BLOCK_NONE,

}

param stop: Optional[List[str]] = None (alias 'stop_sequences')¶

Optional list of stop words to use when generating.

param streaming: bool = False¶

Whether to stream the results or not.

param temperature: Optional[float] = None¶

Sampling temperature, it controls the degree of randomness in token selection.

param top_k: Optional[int] = None¶

How the model selects tokens for output, the next token is selected from

param top_p: Optional[float] = None¶

Tokens are selected from most probable to least until the sum of their

param tuned_model_name: Optional[str] = None¶

The name of a tuned model. If tuned_model_name is passed model_name will be used to determine the model family

async aembed_documents(texts: List[str]) → List[List[float]]¶

Asynchronous Embed search docs.

Parameters

texts (List[str]) – List of text to embed.

Returns

List of embeddings.

Return type

List[List[float]]

async aembed_query(text: str) → List[float]¶

Asynchronous Embed query text.

Parameters

text (str) – Text to embed.

Returns

Embedding.

Return type

List[float]

embed(texts: List[str], batch_size: int = 0, embeddings_task_type: Optional[Literal['RETRIEVAL_QUERY', 'RETRIEVAL_DOCUMENT', 'SEMANTIC_SIMILARITY', 'CLASSIFICATION', 'CLUSTERING', 'QUESTION_ANSWERING', 'FACT_VERIFICATION']] = None, dimensions: Optional[int] = None) → List[List[float]][source]¶

Embed a list of strings.

Parameters

  • texts (List[str]) – List[str] The list of strings to embed.

  • batch_size (int) – [int] The batch size of embeddings to send to the model. If zero, then the largest batch size will be detected dynamically at the first request, starting from 250, down to 5.

  • embeddings_task_type (Optional[Literal['RETRIEVAL_QUERY', 'RETRIEVAL_DOCUMENT', 'SEMANTIC_SIMILARITY', 'CLASSIFICATION', 'CLUSTERING', 'QUESTION_ANSWERING', 'FACT_VERIFICATION']]) –

    [str] optional embeddings task type, one of the following

    RETRIEVAL_QUERY - Text is a query

    in a search/retrieval setting.

    RETRIEVAL_DOCUMENT - Text is a document

    in a search/retrieval setting.

    SEMANTIC_SIMILARITY - Embeddings will be used

    for Semantic Textual Similarity (STS).

    CLASSIFICATION - Embeddings will be used for classification. CLUSTERING - Embeddings will be used for clustering. The following are only supported on preview models: QUESTION_ANSWERING FACT_VERIFICATION

  • dimensions (Optional[int]) – [int] optional. Output embeddings dimensions. Only supported on preview models.

Returns

List of embeddings, one for each text.

Return type

List[List[float]]

embed_documents(texts: List[str], batch_size: int = 0) → List[List[float]][source]¶

Embed a list of documents.

Parameters

  • texts (List[str]) – List[str] The list of texts to embed.

  • batch_size (int) – [int] The batch size of embeddings to send to the model. If zero, then the largest batch size will be detected dynamically at the first request, starting from 250, down to 5.

Returns

List of embeddings, one for each text.

Return type

List[List[float]]

embed_image(image_path: str, contextual_text: Optional[str] = None) → List[float][source]¶

Embed an image.

Parameters

  • image_path (str) – Path to image (local, Google Cloud Storage or web) to generate

  • for. (embeddings) –

  • contextual_text (Optional[str]) – Text to generate embeddings for.

Returns

Embedding for the image.

Return type

List[float]

embed_query(text: str) → List[float][source]¶

Embed a text.

Parameters

text (str) – The text to embed.

Returns

Embedding for the text.

Return type

List[float]

get_num_tokens(text: str) → int¶

Get the number of tokens present in the text.

Useful for checking if an input will fit in a model’s context window.

Parameters

text (str) – The string input to tokenize.

Returns

The integer number of tokens in the text.

Return type

int

property async_prediction_client: PredictionServiceAsyncClient¶

Returns PredictionServiceClient.

property model_type: str¶
property model_version: GoogleEmbeddingModelVersion¶
property prediction_client: PredictionServiceClient¶

Returns PredictionServiceClient.

task_executor: ClassVar[Optional[Executor]] = FieldInfo(exclude=True, extra={})¶

Examples using VertexAIEmbeddings¶