create_endpoint_config ¶

def create_endpoint_config(
    self,
    *,
    config_overrides: Optional[SageMakerClientConfig] = None,
    endpoint_config_name: Optional[
        "capo_sagemaker.types.endpoint_config_name.EndpointConfigName"
    ] = None,
    production_variants: Optional[
        "capo_sagemaker.types.production_variant_list.ProductionVariantList"
    ] = None,
    data_capture_config: Optional[
        "capo_sagemaker.types.data_capture_config.DataCaptureConfig"
    ] = None,
    tags: Optional["capo_sagemaker.types.tag_list.TagList"] = None,
    kms_key_id: Optional["capo_sagemaker.types.kms_key_id.KmsKeyId"] = None,
    async_inference_config: Optional[
        "capo_sagemaker.types.async_inference_config.AsyncInferenceConfig"
    ] = None,
    explainer_config: Optional[
        "capo_sagemaker.types.explainer_config.ExplainerConfig"
    ] = None,
    shadow_production_variants: Optional[
        "capo_sagemaker.types.production_variant_list.ProductionVariantList"
    ] = None,
    execution_role_arn: Optional["capo_sagemaker.types.role_arn.RoleArn"] = None,
    vpc_config: Optional["capo_sagemaker.types.vpc_config.VpcConfig"] = None,
    enable_network_isolation: Optional[
        "capo_sagemaker.types.boolean.Boolean"
    ] = None,
    metrics_config: Optional[
        "capo_sagemaker.types.metrics_config.MetricsConfig"
    ] = None,
) -> (
    "capo_sagemaker.types.create_endpoint_config_output.CreateEndpointConfigOutput"
):
    """<p>Creates an endpoint configuration that SageMaker hosting services uses to deploy models. In the configuration, you identify one or more models, created using the <code>CreateModel</code> API, to deploy and the resources that you want SageMaker to provision. Then you call the <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateEndpoint.html">CreateEndpoint</a> API.</p> <note> <p> Use this API if you want to use SageMaker hosting services to deploy models into production. </p> </note> <p>In the request, you define a <code>ProductionVariant</code>, for each model that you want to deploy. Each <code>ProductionVariant</code> parameter also describes the resources that you want SageMaker to provision. This includes the number and type of ML compute instances to deploy. </p> <p>If you are hosting multiple models, you also assign a <code>VariantWeight</code> to specify how much traffic you want to allocate to each model. For example, suppose that you want to host two models, A and B, and you assign traffic weight 2 for model A and 1 for model B. SageMaker distributes two-thirds of the traffic to Model A, and one-third to model B. </p> <note> <p>When you call <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateEndpoint.html">CreateEndpoint</a>, a load call is made to DynamoDB to verify that your endpoint configuration exists. When you read data from a DynamoDB table supporting <a href="https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.ReadConsistency.html"> <code>Eventually Consistent Reads</code> </a>, the response might not reflect the results of a recently completed write operation. The response might include some stale data. If the dependent entities are not yet in DynamoDB, this causes a validation error. If you repeat your read request after a short time, the response should return the latest data. So retry logic is recommended to handle these possible issues. We also recommend that customers call <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeEndpointConfig.html">DescribeEndpointConfig</a> before calling <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateEndpoint.html">CreateEndpoint</a> to minimize the potential impact of a DynamoDB eventually consistent read.</p> </note>

    Args:
        endpoint_config_name: <p>The name of the endpoint configuration. You specify this name in a <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateEndpoint.html">CreateEndpoint</a> request. </p>
        production_variants: <p>An array of <code>ProductionVariant</code> objects, one for each model that you want to host at this endpoint.</p>
        tags: <p>An array of key-value pairs. You can use tags to categorize your Amazon Web Services resources in different ways, for example, by purpose, owner, or environment. For more information, see <a href="https://docs.aws.amazon.com/general/latest/gr/aws_tagging.html">Tagging Amazon Web Services Resources</a>.</p>
        kms_key_id: <p>The Amazon Resource Name (ARN) of a Amazon Web Services Key Management Service key that SageMaker uses to encrypt data on the storage volume attached to the ML compute instance that hosts the endpoint.</p> <p>The KmsKeyId can be any of the following formats: </p> <ul> <li> <p>Key ID: <code>1234abcd-12ab-34cd-56ef-1234567890ab</code> </p> </li> <li> <p>Key ARN: <code>arn:aws:kms:us-west-2:111122223333:key/1234abcd-12ab-34cd-56ef-1234567890ab</code> </p> </li> <li> <p>Alias name: <code>alias/ExampleAlias</code> </p> </li> <li> <p>Alias name ARN: <code>arn:aws:kms:us-west-2:111122223333:alias/ExampleAlias</code> </p> </li> </ul> <p>The KMS key policy must grant permission to the IAM role that you specify in your <code>CreateEndpoint</code>, <code>UpdateEndpoint</code> requests. For more information, refer to the Amazon Web Services Key Management Service section<a href="https://docs.aws.amazon.com/kms/latest/developerguide/key-policies.html"> Using Key Policies in Amazon Web Services KMS </a> </p> <note> <p>Certain Nitro-based instances include local storage, dependent on the instance type. Local storage volumes are encrypted using a hardware module on the instance. If any of the models that you specify in the <code>ProductionVariants</code> parameter use nitro-based instances with local storage, the <code>KmsKeyId</code> parameter does not encrypt instance local storage.</p> <p>For a list of instance types that support local instance storage, see <a href="https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/InstanceStorage.html#instance-store-volumes">Instance Store Volumes</a>.</p> <p>For more information about local instance storage encryption, see <a href="https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ssd-instance-store.html">SSD Instance Store Volumes</a>.</p> </note>
        async_inference_config: <p>Specifies configuration for how an endpoint performs asynchronous inference. This is a required field in order for your Endpoint to be invoked using <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_runtime_InvokeEndpointAsync.html">InvokeEndpointAsync</a>.</p>
        explainer_config: <p>A member of <code>CreateEndpointConfig</code> that enables explainers.</p>
        shadow_production_variants: <p>An array of <code>ProductionVariant</code> objects, one for each model that you want to host at this endpoint in shadow mode with production traffic replicated from the model specified on <code>ProductionVariants</code>. If you use this field, you can only specify one variant for <code>ProductionVariants</code> and one variant for <code>ShadowProductionVariants</code>.</p>
        execution_role_arn: <p>The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker AI can assume to perform actions on your behalf. For more information, see <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-roles.html">SageMaker AI Roles</a>. </p> <note> <p>To be able to pass this role to Amazon SageMaker AI, the caller of this action must have the <code>iam:PassRole</code> permission.</p> </note>
        enable_network_isolation: <p>Sets whether all model containers deployed to the endpoint are isolated. If they are, no inbound or outbound network calls can be made to or from the model containers.</p>
        metrics_config: <p>The configuration parameters for utilization metrics.</p>

    Raises:
        capo_sagemaker.errors.resource_limit_exceeded.ResourceLimitExceeded: <p> You have exceeded an SageMaker resource limit. For example, you might have too many training jobs created. </p>
        capo_sagemaker.errors.UnknownServiceError: The service returned an error code this client does not model.
    """

    def _handler(
        req: "OperationRequest[capo_sagemaker.types.create_endpoint_config_input.CreateEndpointConfigInput]",
    ) -> OperationResponse[
        "capo_sagemaker.types.create_endpoint_config_output.CreateEndpointConfigOutput"
    ]:
        import capo_sagemaker._operations.sage_maker.create_endpoint_config

        output, http_response = (
            capo_sagemaker._operations.sage_maker.create_endpoint_config.create_endpoint_config(
                req.options, req.input
            )
        )
        return OperationResponse(output=output, response=http_response)

    interceptors_, options_ = self.operation_options(
        config_overrides, OperationLogging("CreateEndpointConfig")
    )
    input_: capo_sagemaker.types.create_endpoint_config_input.CreateEndpointConfigInput = {}
    if endpoint_config_name is not None:
        input_["endpoint_config_name"] = endpoint_config_name
    if production_variants is not None:
        input_["production_variants"] = production_variants
    if data_capture_config is not None:
        input_["data_capture_config"] = data_capture_config
    if tags is not None:
        input_["tags"] = tags
    if kms_key_id is not None:
        input_["kms_key_id"] = kms_key_id
    if async_inference_config is not None:
        input_["async_inference_config"] = async_inference_config
    if explainer_config is not None:
        input_["explainer_config"] = explainer_config
    if shadow_production_variants is not None:
        input_["shadow_production_variants"] = shadow_production_variants
    if execution_role_arn is not None:
        input_["execution_role_arn"] = execution_role_arn
    if vpc_config is not None:
        input_["vpc_config"] = vpc_config
    if enable_network_isolation is not None:
        input_["enable_network_isolation"] = enable_network_isolation
    if metrics_config is not None:
        input_["metrics_config"] = metrics_config

    response = execute_pipeline(
        OperationRequest(input=input_, options=options_),
        handler=_handler,
        interceptors=list(interceptors_),
    )
    response.response.close()
    return response.output