invoke_endpoint_with_response_stream async ¶

@asynccontextmanager
async def invoke_endpoint_with_response_stream(
    self,
    endpoint_name: "capo_sagemaker_runtime.types.endpoint_name.EndpointName",
    *,
    config_overrides: Optional[AsyncSageMakerRuntimeClientConfig] = None,
    body: Optional["capo_sagemaker_runtime.types.body_blob.BodyBlob"] = None,
    content_type: Optional["capo_sagemaker_runtime.types.header.Header"] = None,
    accept: Optional["capo_sagemaker_runtime.types.header.Header"] = None,
    custom_attributes: Optional[
        "capo_sagemaker_runtime.types.custom_attributes_header.CustomAttributesHeader"
    ] = None,
    target_variant: Optional[
        "capo_sagemaker_runtime.types.target_variant_header.TargetVariantHeader"
    ] = None,
    target_container_hostname: Optional[
        "capo_sagemaker_runtime.types.target_container_hostname_header.TargetContainerHostnameHeader"
    ] = None,
    inference_id: Optional[
        "capo_sagemaker_runtime.types.inference_id.InferenceId"
    ] = None,
    inference_component_name: Optional[
        "capo_sagemaker_runtime.types.inference_component_header.InferenceComponentHeader"
    ] = None,
    session_id: Optional[
        "capo_sagemaker_runtime.types.session_id_header.SessionIdHeader"
    ] = None,
    prefix_aware_id: Optional[
        "capo_sagemaker_runtime.types.prefix_aware_id_header.PrefixAwareIdHeader"
    ] = None,
) -> "AsyncGenerator[capo_sagemaker_runtime.types.invoke_endpoint_with_response_stream_output.InvokeEndpointWithResponseStreamOutput]":
    """<p>Invokes a model at the specified endpoint to return the inference response as a stream. The inference stream provides the response payload incrementally as a series of parts. Before you can get an inference stream, you must have access to a model that's deployed using Amazon SageMaker AI hosting services, and the container for that model must support inference streaming.</p> <p>For more information that can help you use this API, see the following sections in the <i>Amazon SageMaker AI Developer Guide</i>:</p> <ul> <li> <p>For information about how to add streaming support to a model, see <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms-inference-code.html#your-algorithms-inference-code-how-containe-serves-requests">How Containers Serve Requests</a>.</p> </li> <li> <p>For information about how to process the streaming response, see <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/realtime-endpoints-test-endpoints.html">Invoke real-time endpoints</a>.</p> </li> </ul> <p>Before you can use this operation, your IAM permissions must allow the <code>sagemaker:InvokeEndpoint</code> action. For more information about Amazon SageMaker AI actions for IAM policies, see <a href="https://docs.aws.amazon.com/service-authorization/latest/reference/list_amazonsagemaker.html">Actions, resources, and condition keys for Amazon SageMaker AI</a> in the <i>IAM Service Authorization Reference</i>.</p> <p>Amazon SageMaker AI strips all POST headers except those supported by the API. Amazon SageMaker AI might add additional headers. You should not rely on the behavior of headers outside those enumerated in the request syntax. </p> <p>Calls to <code>InvokeEndpointWithResponseStream</code> are authenticated by using Amazon Web Services Signature Version 4. For information, see <a href="https://docs.aws.amazon.com/AmazonS3/latest/API/sig-v4-authenticating-requests.html">Authenticating Requests (Amazon Web Services Signature Version 4)</a> in the <i>Amazon S3 API Reference</i>.</p>

    Args:
        endpoint_name: <p>The name of the endpoint that you specified when you created the endpoint using the <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/API_CreateEndpoint.html">CreateEndpoint</a> API.</p>
        body: <p>Provides input data, in the format specified in the <code>ContentType</code> request header. Amazon SageMaker AI passes all of the data in the body to the model. </p> <p>For information about the format of the request body, see <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/cdf-inference.html">Common Data Formats-Inference</a>.</p>
        content_type: <p>The MIME type of the input data in the request body.</p>
        accept: <p>The desired MIME type of the inference response from the model container.</p>
        custom_attributes: <p>Provides additional information about a request for an inference submitted to a model hosted at an Amazon SageMaker AI endpoint. The information is an opaque value that is forwarded verbatim. You could use this value, for example, to provide an ID that you can use to track a request or to provide other metadata that a service endpoint was programmed to process. The value must consist of no more than 1024 visible US-ASCII characters as specified in <a href="https://datatracker.ietf.org/doc/html/rfc7230#section-3.2.6">Section 3.3.6. Field Value Components</a> of the Hypertext Transfer Protocol (HTTP/1.1). </p> <p>The code in your model is responsible for setting or updating any custom attributes in the response. If your code does not set this value in the response, an empty value is returned. For example, if a custom attribute represents the trace ID, your model can prepend the custom attribute with <code>Trace ID:</code> in your post-processing function. </p> <p>This feature is currently supported in the Amazon Web Services SDKs but not in the Amazon SageMaker AI Python SDK. </p>
        target_variant: <p>Specify the production variant to send the inference request to when invoking an endpoint that is running two or more variants. Note that this parameter overrides the default behavior for the endpoint, which is to distribute the invocation traffic based on the variant weights.</p> <p>For information about how to use variant targeting to perform a/b testing, see <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/model-ab-testing.html">Test models in production</a> </p>
        target_container_hostname: <p>If the endpoint hosts multiple containers and is configured to use direct invocation, this parameter specifies the host name of the container to invoke.</p>
        inference_id: <p>An identifier that you assign to your request.</p>
        inference_component_name: <p>If the endpoint hosts one or more inference components, this parameter specifies the name of inference component to invoke for a streaming response.</p>
        session_id: <p>The ID of a stateful session to handle your request.</p> <p>You can't create a stateful session by using the <code>InvokeEndpointWithResponseStream</code> action. Instead, you can create one by using the <code> <a>InvokeEndpoint</a> </code> action. In your request, you specify <code>NEW_SESSION</code> for the <code>SessionId</code> request parameter. The response to that request provides the session ID for the <code>NewSessionId</code> response parameter.</p>
        prefix_aware_id: <p>An optional, stable identifier that serves as a routing hint for prefix-aware routing. The service routes requests with the same prefix and the same identifier to the same instance. If requests from different applications might have the same prompt prefix, set a different identifier for each application to differentiate their routing decisions.</p> <p>Applies only to endpoints configured with a <code>RoutingStrategy</code> of <code>PREFIX_AWARE</code>.</p>

    Raises:
        capo_sagemaker_runtime.errors.internal_failure.InternalFailure: <p> An internal failure occurred. </p>
        capo_sagemaker_runtime.errors.internal_stream_failure.InternalStreamFailure: <p>The stream processing failed because of an unknown error, exception or failure. Try your request again.</p>
        capo_sagemaker_runtime.errors.model_error.ModelError: <p> Model (owned by the customer in the container) returned 4xx or 5xx error code. </p>
        capo_sagemaker_runtime.errors.model_stream_error.ModelStreamError: <p> An error occurred while streaming the response body. This error can have the following error codes:</p> <dl> <dt>ModelInvocationTimeExceeded</dt> <dd> <p>The model failed to finish sending the response within the timeout period allowed by Amazon SageMaker AI.</p> </dd> <dt>StreamBroken</dt> <dd> <p>The Transmission Control Protocol (TCP) connection between the client and the model was reset or closed.</p> </dd> </dl>
        capo_sagemaker_runtime.errors.service_unavailable.ServiceUnavailable: <p> The service is unavailable. Try your call again. </p>
        capo_sagemaker_runtime.errors.validation_error.ValidationError: <p> Inspect your request and try again. </p>
        capo_sagemaker_runtime.errors.UnknownServiceError: The service returned an error code this client does not model.
    """

    async def _handler(
        req: "AsyncOperationRequest[capo_sagemaker_runtime.types.invoke_endpoint_with_response_stream_input.InvokeEndpointWithResponseStreamInput]",
    ) -> AsyncOperationResponse[
        "capo_sagemaker_runtime.types.invoke_endpoint_with_response_stream_output.InvokeEndpointWithResponseStreamOutput"
    ]:
        import capo_sagemaker_runtime._operations.amazon_sage_maker_runtime.invoke_endpoint_with_response_stream

        (
            output,
            http_response,
        ) = await capo_sagemaker_runtime._operations.amazon_sage_maker_runtime.invoke_endpoint_with_response_stream.async_invoke_endpoint_with_response_stream(
            req.options, req.input
        )
        return AsyncOperationResponse(output=output, response=http_response)

    def _redact_input(
        input_: "capo_sagemaker_runtime.types.invoke_endpoint_with_response_stream_input.InvokeEndpointWithResponseStreamInput",
    ) -> dict[str, Any]:
        import capo_sagemaker_runtime.types.invoke_endpoint_with_response_stream_input

        return capo_sagemaker_runtime.types.invoke_endpoint_with_response_stream_input.redact_sensitive(
            input_
        )

    def _redact_output(
        output: "capo_sagemaker_runtime.types.invoke_endpoint_with_response_stream_output.InvokeEndpointWithResponseStreamOutput",
    ) -> dict[str, Any]:
        import capo_sagemaker_runtime.types.invoke_endpoint_with_response_stream_output

        return capo_sagemaker_runtime.types.invoke_endpoint_with_response_stream_output.redact_sensitive(
            output
        )

    interceptors_, options_ = self.operation_options(
        config_overrides,
        OperationLogging(
            "InvokeEndpointWithResponseStream",
            redact_input=_redact_input,
            redact_output=_redact_output,
        ),
    )
    input_: capo_sagemaker_runtime.types.invoke_endpoint_with_response_stream_input.InvokeEndpointWithResponseStreamInput = {
        "endpoint_name": endpoint_name
    }
    if body is not None:
        input_["body"] = body
    if content_type is not None:
        input_["content_type"] = content_type
    if accept is not None:
        input_["accept"] = accept
    if custom_attributes is not None:
        input_["custom_attributes"] = custom_attributes
    if target_variant is not None:
        input_["target_variant"] = target_variant
    if target_container_hostname is not None:
        input_["target_container_hostname"] = target_container_hostname
    if inference_id is not None:
        input_["inference_id"] = inference_id
    if inference_component_name is not None:
        input_["inference_component_name"] = inference_component_name
    if session_id is not None:
        input_["session_id"] = session_id
    if prefix_aware_id is not None:
        input_["prefix_aware_id"] = prefix_aware_id

    response = await aexecute_pipeline(
        AsyncOperationRequest(input=input_, options=options_),
        handler=_handler,
        interceptors=list(interceptors_),
    )
    try:
        yield response.output
    finally:
        await response.response.aclose()