service-2.json

{
  "version":"2.0",
  "metadata":{
    "apiVersion":"2017-05-13",
    "endpointPrefix":"runtime.sagemaker",
    "jsonVersion":"1.1",
    "protocol":"rest-json",
    "protocols":["rest-json"],
    "serviceFullName":"Amazon SageMaker Runtime",
    "serviceId":"SageMaker Runtime",
    "signatureVersion":"v4",
    "signingName":"sagemaker",
    "uid":"runtime.sagemaker-2017-05-13",
    "auth":["aws.auth#sigv4"]
  },
  "operations":{
    "InvokeEndpoint":{
      "name":"InvokeEndpoint",
      "http":{
        "method":"POST",
        "requestUri":"/endpoints/{EndpointName}/invocations"
      },
      "input":{"shape":"InvokeEndpointInput"},
      "output":{"shape":"InvokeEndpointOutput"},
      "errors":[
        {"shape":"InternalFailure"},
        {"shape":"ServiceUnavailable"},
        {"shape":"ValidationError"},
        {"shape":"ModelError"},
        {"shape":"InternalDependencyException"},
        {"shape":"ModelNotReadyException"}
      ],
      "documentation":"<p>After you deploy a model into production using Amazon SageMaker AI hosting services, your client applications use this API to get inferences from the model hosted at the specified endpoint. </p> <p>For an overview of Amazon SageMaker AI, see <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/how-it-works.html\">How It Works</a>. </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>InvokeEndpoint</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> <p>A customer's model containers must respond to requests within 60 seconds. The model itself can have a maximum processing time of 60 seconds before responding to invocations. If your model is going to take 50-60 seconds of processing time, the SDK socket timeout should be set to be 70 seconds.</p> <note> <p>Endpoints are scoped to an individual account, and are not public. The URL does not contain the account ID, but Amazon SageMaker AI determines the account ID from the authentication token that is supplied by the caller.</p> </note>"
    },
    "InvokeEndpointAsync":{
      "name":"InvokeEndpointAsync",
      "http":{
        "method":"POST",
        "requestUri":"/endpoints/{EndpointName}/async-invocations",
        "responseCode":202
      },
      "input":{"shape":"InvokeEndpointAsyncInput"},
      "output":{"shape":"InvokeEndpointAsyncOutput"},
      "errors":[
        {"shape":"InternalFailure"},
        {"shape":"ServiceUnavailable"},
        {"shape":"ValidationError"}
      ],
      "documentation":"<p>After you deploy a model into production using Amazon SageMaker AI hosting services, your client applications use this API to get inferences from the model hosted at the specified endpoint in an asynchronous manner.</p> <p>Inference requests sent to this API are enqueued for asynchronous processing. The processing of the inference request may or may not complete before you receive a response from this API. The response from this API will not contain the result of the inference request but contain information about where you can locate it.</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>InvokeEndpointAsync</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>"
    },
    "InvokeEndpointWithResponseStream":{
      "name":"InvokeEndpointWithResponseStream",
      "http":{
        "method":"POST",
        "requestUri":"/endpoints/{EndpointName}/invocations-response-stream"
      },
      "input":{"shape":"InvokeEndpointWithResponseStreamInput"},
      "output":{"shape":"InvokeEndpointWithResponseStreamOutput"},
      "errors":[
        {"shape":"InternalFailure"},
        {"shape":"ServiceUnavailable"},
        {"shape":"ValidationError"},
        {"shape":"ModelError"},
        {"shape":"ModelStreamError"},
        {"shape":"InternalStreamFailure"}
      ],
      "documentation":"<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>"
    }
  },
  "shapes":{
    "BodyBlob":{
      "type":"blob",
      "max":6291456,
      "sensitive":true
    },
    "CustomAttributesHeader":{
      "type":"string",
      "max":1024,
      "pattern":"\\p{ASCII}*",
      "sensitive":true
    },
    "EnableExplanationsHeader":{
      "type":"string",
      "max":64,
      "min":1,
      "pattern":".*"
    },
    "EndpointName":{
      "type":"string",
      "max":63,
      "pattern":"^[a-zA-Z0-9](-*[a-zA-Z0-9])*"
    },
    "ErrorCode":{"type":"string"},
    "FilenameHeader":{
      "type":"string",
      "max":32,
      "min":1,
      "pattern":"^(?!.*\\..*\\.)[a-zA-Z0-9][a-zA-Z0-9-_\\.]*$"
    },
    "Header":{
      "type":"string",
      "max":1024,
      "pattern":"\\p{ASCII}*"
    },
    "InferenceComponentHeader":{
      "type":"string",
      "max":63,
      "pattern":"^[a-zA-Z0-9]([\\-a-zA-Z0-9]*[a-zA-Z0-9])?$"
    },
    "InferenceId":{
      "type":"string",
      "max":64,
      "min":1,
      "pattern":"\\A\\S[\\p{Print}]*\\z"
    },
    "InputLocationHeader":{
      "type":"string",
      "max":1024,
      "min":1,
      "pattern":"^(https|s3)://([^/]+)/?(.*)$"
    },
    "InternalDependencyException":{
      "type":"structure",
      "members":{
        "Message":{"shape":"Message"}
      },
      "documentation":"<p>Your request caused an exception with an internal dependency. Contact customer support. </p>",
      "error":{"httpStatusCode":530},
      "exception":true,
      "fault":true,
      "synthetic":true
    },
    "InternalFailure":{
      "type":"structure",
      "members":{
        "Message":{"shape":"Message"}
      },
      "documentation":"<p> An internal failure occurred. </p>",
      "error":{"httpStatusCode":500},
      "exception":true,
      "fault":true,
      "synthetic":true
    },
    "InternalStreamFailure":{
      "type":"structure",
      "members":{
        "Message":{"shape":"Message"}
      },
      "documentation":"<p>The stream processing failed because of an unknown error, exception or failure. Try your request again.</p>",
      "exception":true,
      "fault":true,
      "synthetic":true
    },
    "InvocationTimeoutSecondsHeader":{
      "type":"integer",
      "max":3600,
      "min":1
    },
    "InvokeEndpointAsyncInput":{
      "type":"structure",
      "required":[
        "EndpointName",
        "InputLocation"
      ],
      "members":{
        "EndpointName":{
          "shape":"EndpointName",
          "documentation":"<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>",
          "location":"uri",
          "locationName":"EndpointName"
        },
        "ContentType":{
          "shape":"Header",
          "documentation":"<p>The MIME type of the input data in the request body.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Content-Type"
        },
        "Accept":{
          "shape":"Header",
          "documentation":"<p>The desired MIME type of the inference response from the model container.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Accept"
        },
        "CustomAttributes":{
          "shape":"CustomAttributesHeader",
          "documentation":"<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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Custom-Attributes"
        },
        "InferenceId":{
          "shape":"InferenceId",
          "documentation":"<p>The identifier for the inference request. Amazon SageMaker AI will generate an identifier for you if none is specified. </p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Inference-Id"
        },
        "InputLocation":{
          "shape":"InputLocationHeader",
          "documentation":"<p>The Amazon S3 URI where the inference request payload is stored.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-InputLocation"
        },
        "S3OutputPathExtension":{
          "shape":"S3OutputPathExtensionHeader",
          "documentation":"<p>The path extension that is appended to the Amazon S3 output path where the inference response payload is stored.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-S3OutputPathExtension"
        },
        "Filename":{
          "shape":"FilenameHeader",
          "documentation":"<p>The filename for the inference response payload stored in Amazon S3. If not specified, Amazon SageMaker AI generates a filename based on the inference ID.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Filename"
        },
        "RequestTTLSeconds":{
          "shape":"RequestTTLSecondsHeader",
          "documentation":"<p>Maximum age in seconds a request can be in the queue before it is marked as expired. The default is 6 hours, or 21,600 seconds.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-RequestTTLSeconds"
        },
        "InvocationTimeoutSeconds":{
          "shape":"InvocationTimeoutSecondsHeader",
          "documentation":"<p>Maximum amount of time in seconds a request can be processed before it is marked as expired. The default is 15 minutes, or 900 seconds.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-InvocationTimeoutSeconds"
        }
      }
    },
    "InvokeEndpointAsyncOutput":{
      "type":"structure",
      "members":{
        "InferenceId":{
          "shape":"Header",
          "documentation":"<p>Identifier for an inference request. This will be the same as the <code>InferenceId</code> specified in the input. Amazon SageMaker AI will generate an identifier for you if you do not specify one.</p>"
        },
        "OutputLocation":{
          "shape":"Header",
          "documentation":"<p>The Amazon S3 URI where the inference response payload is stored.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-OutputLocation"
        },
        "FailureLocation":{
          "shape":"Header",
          "documentation":"<p>The Amazon S3 URI where the inference failure response payload is stored.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-FailureLocation"
        }
      }
    },
    "InvokeEndpointInput":{
      "type":"structure",
      "required":[
        "EndpointName",
        "Body"
      ],
      "members":{
        "EndpointName":{
          "shape":"EndpointName",
          "documentation":"<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>",
          "location":"uri",
          "locationName":"EndpointName"
        },
        "Body":{
          "shape":"BodyBlob",
          "documentation":"<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>"
        },
        "ContentType":{
          "shape":"Header",
          "documentation":"<p>The MIME type of the input data in the request body.</p>",
          "location":"header",
          "locationName":"Content-Type"
        },
        "Accept":{
          "shape":"Header",
          "documentation":"<p>The desired MIME type of the inference response from the model container.</p>",
          "location":"header",
          "locationName":"Accept"
        },
        "CustomAttributes":{
          "shape":"CustomAttributesHeader",
          "documentation":"<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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Custom-Attributes"
        },
        "TargetModel":{
          "shape":"TargetModelHeader",
          "documentation":"<p>The model to request for inference when invoking a multi-model endpoint.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Target-Model"
        },
        "TargetVariant":{
          "shape":"TargetVariantHeader",
          "documentation":"<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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Target-Variant"
        },
        "TargetContainerHostname":{
          "shape":"TargetContainerHostnameHeader",
          "documentation":"<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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Target-Container-Hostname"
        },
        "InferenceId":{
          "shape":"InferenceId",
          "documentation":"<p>If you provide a value, it is added to the captured data when you enable data capture on the endpoint. For information about data capture, see <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/model-monitor-data-capture.html\">Capture Data</a>.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Inference-Id"
        },
        "EnableExplanations":{
          "shape":"EnableExplanationsHeader",
          "documentation":"<p>An optional JMESPath expression used to override the <code>EnableExplanations</code> parameter of the <code>ClarifyExplainerConfig</code> API. See the <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/clarify-online-explainability-create-endpoint.html#clarify-online-explainability-create-endpoint-enable\">EnableExplanations</a> section in the developer guide for more information. </p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Enable-Explanations"
        },
        "InferenceComponentName":{
          "shape":"InferenceComponentHeader",
          "documentation":"<p>If the endpoint hosts one or more inference components, this parameter specifies the name of inference component to invoke.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Inference-Component"
        },
        "SessionId":{
          "shape":"SessionIdOrNewSessionConstantHeader",
          "documentation":"<p>Creates a stateful session or identifies an existing one. You can do one of the following:</p> <ul> <li> <p>Create a stateful session by specifying the value <code>NEW_SESSION</code>.</p> </li> <li> <p>Send your request to an existing stateful session by specifying the ID of that session.</p> </li> </ul> <p>With a stateful session, you can send multiple requests to a stateful model. When you create a session with a stateful model, the model must create the session ID and set the expiration time. The model must also provide that information in the response to your request. You can get the ID and timestamp from the <code>NewSessionId</code> response parameter. For any subsequent request where you specify that session ID, SageMaker AI routes the request to the same instance that supports the session.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Session-Id"
        }
      },
      "payload":"Body"
    },
    "InvokeEndpointOutput":{
      "type":"structure",
      "required":["Body"],
      "members":{
        "Body":{
          "shape":"BodyBlob",
          "documentation":"<p>Includes the inference provided by the model. </p> <p>For information about the format of the response body, see <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/cdf-inference.html\">Common Data Formats-Inference</a>.</p> <p>If the explainer is activated, the body includes the explanations provided by the model. For more information, see the <b>Response section</b> under <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/clarify-online-explainability-invoke-endpoint.html#clarify-online-explainability-response\">Invoke the Endpoint</a> in the Developer Guide.</p>"
        },
        "ContentType":{
          "shape":"Header",
          "documentation":"<p>The MIME type of the inference returned from the model container.</p>",
          "location":"header",
          "locationName":"Content-Type"
        },
        "InvokedProductionVariant":{
          "shape":"Header",
          "documentation":"<p>Identifies the production variant that was invoked.</p>",
          "location":"header",
          "locationName":"x-Amzn-Invoked-Production-Variant"
        },
        "CustomAttributes":{
          "shape":"CustomAttributesHeader",
          "documentation":"<p>Provides additional information in the response about the inference returned by 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 return an ID received in the <code>CustomAttributes</code> header of a request or other metadata that a service endpoint was programmed to produce. The value must consist of no more than 1024 visible US-ASCII characters as specified in <a href=\"https://tools.ietf.org/html/rfc7230#section-3.2.6\">Section 3.3.6. Field Value Components</a> of the Hypertext Transfer Protocol (HTTP/1.1). If the customer wants the custom attribute returned, the model must set the custom attribute to be included on the way back. </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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Custom-Attributes"
        },
        "NewSessionId":{
          "shape":"NewSessionResponseHeader",
          "documentation":"<p>If you created a stateful session with your request, the ID and expiration time that the model assigns to that session.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-New-Session-Id"
        },
        "ClosedSessionId":{
          "shape":"SessionIdHeader",
          "documentation":"<p>If you closed a stateful session with your request, the ID of that session.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Closed-Session-Id"
        }
      },
      "payload":"Body"
    },
    "InvokeEndpointWithResponseStreamInput":{
      "type":"structure",
      "required":[
        "EndpointName",
        "Body"
      ],
      "members":{
        "EndpointName":{
          "shape":"EndpointName",
          "documentation":"<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>",
          "location":"uri",
          "locationName":"EndpointName"
        },
        "Body":{
          "shape":"BodyBlob",
          "documentation":"<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>"
        },
        "ContentType":{
          "shape":"Header",
          "documentation":"<p>The MIME type of the input data in the request body.</p>",
          "location":"header",
          "locationName":"Content-Type"
        },
        "Accept":{
          "shape":"Header",
          "documentation":"<p>The desired MIME type of the inference response from the model container.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Accept"
        },
        "CustomAttributes":{
          "shape":"CustomAttributesHeader",
          "documentation":"<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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Custom-Attributes"
        },
        "TargetVariant":{
          "shape":"TargetVariantHeader",
          "documentation":"<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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Target-Variant"
        },
        "TargetContainerHostname":{
          "shape":"TargetContainerHostnameHeader",
          "documentation":"<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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Target-Container-Hostname"
        },
        "InferenceId":{
          "shape":"InferenceId",
          "documentation":"<p>An identifier that you assign to your request.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Inference-Id"
        },
        "InferenceComponentName":{
          "shape":"InferenceComponentHeader",
          "documentation":"<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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Inference-Component"
        },
        "SessionId":{
          "shape":"SessionIdHeader",
          "documentation":"<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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Session-Id"
        }
      },
      "payload":"Body"
    },
    "InvokeEndpointWithResponseStreamOutput":{
      "type":"structure",
      "required":["Body"],
      "members":{
        "Body":{"shape":"ResponseStream"},
        "ContentType":{
          "shape":"Header",
          "documentation":"<p>The MIME type of the inference returned from the model container.</p>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Content-Type"
        },
        "InvokedProductionVariant":{
          "shape":"Header",
          "documentation":"<p>Identifies the production variant that was invoked.</p>",
          "location":"header",
          "locationName":"x-Amzn-Invoked-Production-Variant"
        },
        "CustomAttributes":{
          "shape":"CustomAttributesHeader",
          "documentation":"<p>Provides additional information in the response about the inference returned by 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 return an ID received in the <code>CustomAttributes</code> header of a request or other metadata that a service endpoint was programmed to produce. The value must consist of no more than 1024 visible US-ASCII characters as specified in <a href=\"https://tools.ietf.org/html/rfc7230#section-3.2.6\">Section 3.3.6. Field Value Components</a> of the Hypertext Transfer Protocol (HTTP/1.1). If the customer wants the custom attribute returned, the model must set the custom attribute to be included on the way back. </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>",
          "location":"header",
          "locationName":"X-Amzn-SageMaker-Custom-Attributes"
        }
      },
      "payload":"Body"
    },
    "LogStreamArn":{"type":"string"},
    "Message":{
      "type":"string",
      "max":2048
    },
    "ModelError":{
      "type":"structure",
      "members":{
        "Message":{"shape":"Message"},
        "OriginalStatusCode":{
          "shape":"StatusCode",
          "documentation":"<p> Original status code. </p>"
        },
        "OriginalMessage":{
          "shape":"Message",
          "documentation":"<p> Original message. </p>"
        },
        "LogStreamArn":{
          "shape":"LogStreamArn",
          "documentation":"<p> The Amazon Resource Name (ARN) of the log stream. </p>"
        }
      },
      "documentation":"<p> Model (owned by the customer in the container) returned 4xx or 5xx error code. </p>",
      "error":{"httpStatusCode":424},
      "exception":true
    },
    "ModelNotReadyException":{
      "type":"structure",
      "members":{
        "Message":{"shape":"Message"}
      },
      "documentation":"<p>Either a serverless endpoint variant's resources are still being provisioned, or a multi-model endpoint is still downloading or loading the target model. Wait and try your request again.</p>",
      "error":{"httpStatusCode":429},
      "exception":true,
      "synthetic":true
    },
    "ModelStreamError":{
      "type":"structure",
      "members":{
        "Message":{"shape":"Message"},
        "ErrorCode":{
          "shape":"ErrorCode",
          "documentation":"<p>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>"
        }
      },
      "documentation":"<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>",
      "exception":true,
      "synthetic":true
    },
    "NewSessionResponseHeader":{
      "type":"string",
      "max":256,
      "pattern":"^[a-zA-Z0-9](-*[a-zA-Z0-9])*;\\sExpires=[0-9]{4}-[0-9]{2}-[0-9]{2}T[0-9]{2}:[0-9]{2}:[0-9]{2}Z$"
    },
    "PartBlob":{
      "type":"blob",
      "sensitive":true
    },
    "PayloadPart":{
      "type":"structure",
      "members":{
        "Bytes":{
          "shape":"PartBlob",
          "documentation":"<p>A blob that contains part of the response for your streaming inference request.</p>",
          "eventpayload":true
        }
      },
      "documentation":"<p>A wrapper for pieces of the payload that's returned in response to a streaming inference request. A streaming inference response consists of one or more payload parts. </p>",
      "event":true
    },
    "RequestTTLSecondsHeader":{
      "type":"integer",
      "max":21600,
      "min":60
    },
    "ResponseStream":{
      "type":"structure",
      "members":{
        "PayloadPart":{
          "shape":"PayloadPart",
          "documentation":"<p>A wrapper for pieces of the payload that's returned in response to a streaming inference request. A streaming inference response consists of one or more payload parts. </p>"
        },
        "ModelStreamError":{
          "shape":"ModelStreamError",
          "documentation":"<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>"
        },
        "InternalStreamFailure":{
          "shape":"InternalStreamFailure",
          "documentation":"<p>The stream processing failed because of an unknown error, exception or failure. Try your request again.</p>"
        }
      },
      "documentation":"<p>A stream of payload parts. Each part contains a portion of the response for a streaming inference request.</p>",
      "eventstream":true
    },
    "S3OutputPathExtensionHeader":{
      "type":"string",
      "max":512,
      "min":1,
      "pattern":"^(?!s3:|https:)[a-zA-Z0-9!_.*'()/-]+$"
    },
    "ServiceUnavailable":{
      "type":"structure",
      "members":{
        "Message":{"shape":"Message"}
      },
      "documentation":"<p> The service is unavailable. Try your call again. </p>",
      "error":{"httpStatusCode":503},
      "exception":true,
      "fault":true,
      "synthetic":true
    },
    "SessionIdHeader":{
      "type":"string",
      "max":256,
      "pattern":"^[a-zA-Z0-9](-*[a-zA-Z0-9])*$"
    },
    "SessionIdOrNewSessionConstantHeader":{
      "type":"string",
      "max":256,
      "pattern":"^(NEW_SESSION)$|^[a-zA-Z0-9](-*[a-zA-Z0-9])*$"
    },
    "StatusCode":{"type":"integer"},
    "TargetContainerHostnameHeader":{
      "type":"string",
      "max":63,
      "pattern":"^[a-zA-Z0-9](-*[a-zA-Z0-9])*"
    },
    "TargetModelHeader":{
      "type":"string",
      "max":1024,
      "min":1,
      "pattern":"\\A\\S[\\p{Print}]*\\z"
    },
    "TargetVariantHeader":{
      "type":"string",
      "max":63,
      "pattern":"^[a-zA-Z0-9](-*[a-zA-Z0-9])*"
    },
    "ValidationError":{
      "type":"structure",
      "members":{
        "Message":{"shape":"Message"}
      },
      "documentation":"<p> Inspect your request and try again. </p>",
      "error":{"httpStatusCode":400},
      "exception":true,
      "synthetic":true
    }
  },
  "documentation":"<p> The Amazon SageMaker AI runtime API. </p>"
}