OCIGenAIEmbeddings#

class langchain_community.embeddings.oci_generative_ai.OCIGenAIEmbeddings[source]#

Bases: BaseModel, Embeddings

OCI embedding models.

To authenticate, the OCI client uses the methods described in https://docs.oracle.com/en-us/iaas/Content/API/Concepts/sdk_authentication_methods.htm

The authentifcation method is passed through auth_type and should be one of: API_KEY (default), SECURITY_TOKEN, INSTANCE_PRINCIPLE, RESOURCE_PRINCIPLE

Make sure you have the required policies (profile/roles) to access the OCI Generative AI service. If a specific config profile is used, you must pass the name of the profile (~/.oci/config) through auth_profile.

To use, you must provide the compartment id along with the endpoint url, and model id as named parameters to the constructor.

Example

from langchain.embeddings import OCIGenAIEmbeddings

embeddings = OCIGenAIEmbeddings(
    model_id="MY_EMBEDDING_MODEL",
    service_endpoint="https://inference.generativeai.us-chicago-1.oci.oraclecloud.com",
    compartment_id="MY_OCID"
)

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

param auth_profile: str | None = 'DEFAULT'#: The name of the profile in ~/.oci/config If not specified , DEFAULT will be used

param auth_type: str | None = 'API_KEY'#

Authentication type, could be

API_KEY, SECURITY_TOKEN, INSTANCE_PRINCIPLE, RESOURCE_PRINCIPLE

If not specified, API_KEY will be used

param batch_size: int = 96#: Batch size of OCI GenAI embedding requests. OCI GenAI may handle up to 96 texts per request

param compartment_id: str | None = None#: OCID of compartment

param model_id: str | None = None#: Id of the model to call, e.g., cohere.embed-english-light-v2.0

param model_kwargs: Dict | None = None#: Keyword arguments to pass to the model

param service_endpoint: str | None = None#: service endpoint url

param truncate: str | None = 'END'#: Truncate embeddings that are too long from start or end (“NONE”|”START”|”END”)

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_documents(texts: List[str]) → List[List[float]][source]#

Call out to OCIGenAI’s embedding endpoint.

Parameters:: texts (List[str]) – The list of texts to embed.
Returns:: List of embeddings, one for each text.
Return type:: List[List[float]]

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

Call out to OCIGenAI’s embedding endpoint.

Parameters:: text (str) – The text to embed.
Returns:: Embeddings for the text.
Return type:: List[float]

classmethod validate_environment(values: Dict) → Dict[source]#

Validate that OCI config and python package exists in environment.

Parameters:: values (Dict)
Return type:: Dict

Examples using OCIGenAIEmbeddings