Amazon SageMaker Service Update: SageMaker Inference Recommender now supports a new API ListInferenceRecommendationJobSteps to return the details of all the benchmark we create for an inference recommendation job.

Author: AWS

.changes/next-release/feature-AmazonSageMakerService-1ebdae5.json

@@ -0,0 +1,6 @@
+{
+    "type": "feature",
+    "category": "Amazon SageMaker Service",
+    "contributor": "",
+    "description": "SageMaker Inference Recommender now supports a new API ListInferenceRecommendationJobSteps to return the details of all the benchmark we create for an inference recommendation job."
+}

services/sagemaker/src/main/resources/codegen-resources/paginators-1.json

@@ -156,6 +156,12 @@
       "limit_key": "MaxResults",
       "result_key": "Images"
     },
+    "ListInferenceRecommendationsJobSteps": {
+      "input_token": "NextToken",
+      "output_token": "NextToken",
+      "limit_key": "MaxResults",
+      "result_key": "Steps"
+    },
     "ListInferenceRecommendationsJobs": {
       "input_token": "NextToken",
       "output_token": "NextToken",

services/sagemaker/src/main/resources/codegen-resources/service-2.json

@@ -2170,6 +2170,16 @@
       "output":{"shape":"ListImagesResponse"},
       "documentation":"<p>Lists the images in your account and their properties. The list can be filtered by creation time or modified time, and whether the image name contains a specified string.</p>"
     },
+    "ListInferenceRecommendationsJobSteps":{
+      "name":"ListInferenceRecommendationsJobSteps",
+      "http":{
+        "method":"POST",
+        "requestUri":"/"
+      },
+      "input":{"shape":"ListInferenceRecommendationsJobStepsRequest"},
+      "output":{"shape":"ListInferenceRecommendationsJobStepsResponse"},
+      "documentation":"<p>Returns a list of the subtasks for an Inference Recommender job.</p> <p>The supported subtasks are benchmarks, which evaluate the performance of your model on different instance types.</p>"
+    },
     "ListInferenceRecommendationsJobs":{
       "name":"ListInferenceRecommendationsJobs",
       "http":{
@@ -4540,7 +4550,8 @@
         "PrecisionMacro",
         "Recall",
         "RecallMacro",
-        "LogLoss"
+        "LogLoss",
+        "InferenceLatency"
       ]
     },
     "AutoMLMode":{
@@ -16258,6 +16269,37 @@
       },
       "documentation":"<p>A structure that contains a list of recommendation jobs.</p>"
     },
+    "InferenceRecommendationsJobStep":{
+      "type":"structure",
+      "required":[
+        "StepType",
+        "JobName",
+        "Status"
+      ],
+      "members":{
+        "StepType":{
+          "shape":"RecommendationStepType",
+          "documentation":"<p>The type of the subtask.</p> <p> <code>BENCHMARK</code>: Evaluate the performance of your model on different instance types.</p>"
+        },
+        "JobName":{
+          "shape":"RecommendationJobName",
+          "documentation":"<p>The name of the Inference Recommender job.</p>"
+        },
+        "Status":{
+          "shape":"RecommendationJobStatus",
+          "documentation":"<p>The current status of the benchmark.</p>"
+        },
+        "InferenceBenchmark":{
+          "shape":"RecommendationJobInferenceBenchmark",
+          "documentation":"<p>The details for a specific benchmark.</p>"
+        }
+      },
+      "documentation":"<p>A returned array object for the <code>Steps</code> response field in the <a href=\"https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_InferenceRecommendationsJobStep.html\">ListInferenceRecommendationsJobSteps</a> API command.</p>"
+    },
+    "InferenceRecommendationsJobSteps":{
+      "type":"list",
+      "member":{"shape":"InferenceRecommendationsJobStep"}
+    },
     "InferenceRecommendationsJobs":{
       "type":"list",
       "member":{"shape":"InferenceRecommendationsJob"}
@@ -18470,6 +18512,45 @@
         }
       }
     },
+    "ListInferenceRecommendationsJobStepsRequest":{
+      "type":"structure",
+      "required":["JobName"],
+      "members":{
+        "JobName":{
+          "shape":"RecommendationJobName",
+          "documentation":"<p>The name for the Inference Recommender job.</p>"
+        },
+        "Status":{
+          "shape":"RecommendationJobStatus",
+          "documentation":"<p>A filter to return benchmarks of a specified status. If this field is left empty, then all benchmarks are returned.</p>"
+        },
+        "StepType":{
+          "shape":"RecommendationStepType",
+          "documentation":"<p>A filter to return details about the specified type of subtask.</p> <p> <code>BENCHMARK</code>: Evaluate the performance of your model on different instance types.</p>"
+        },
+        "MaxResults":{
+          "shape":"MaxResults",
+          "documentation":"<p>The maximum number of results to return.</p>"
+        },
+        "NextToken":{
+          "shape":"NextToken",
+          "documentation":"<p>A token that you can specify to return more results from the list. Specify this field if you have a token that was returned from a previous request.</p>"
+        }
+      }
+    },
+    "ListInferenceRecommendationsJobStepsResponse":{
+      "type":"structure",
+      "members":{
+        "Steps":{
+          "shape":"InferenceRecommendationsJobSteps",
+          "documentation":"<p>A list of all subtask details in Inference Recommender.</p>"
+        },
+        "NextToken":{
+          "shape":"NextToken",
+          "documentation":"<p>A token that you can specify in your next request to return more results from the list.</p>"
+        }
+      }
+    },
     "ListInferenceRecommendationsJobsRequest":{
       "type":"structure",
       "members":{
@@ -24597,6 +24678,7 @@
       "type":"list",
       "member":{"shape":"ProductionVariantInstanceType"}
     },
+    "RecommendationFailureReason":{"type":"string"},
     "RecommendationJobArn":{
       "type":"string",
       "max":256,
@@ -24650,6 +24732,20 @@
       "type":"string",
       "max":128
     },
+    "RecommendationJobInferenceBenchmark":{
+      "type":"structure",
+      "required":["ModelConfiguration"],
+      "members":{
+        "Metrics":{"shape":"RecommendationMetrics"},
+        "EndpointConfiguration":{"shape":"EndpointOutputConfiguration"},
+        "ModelConfiguration":{"shape":"ModelConfiguration"},
+        "FailureReason":{
+          "shape":"RecommendationFailureReason",
+          "documentation":"<p>The reason why a benchmark failed.</p>"
+        }
+      },
+      "documentation":"<p>The details for a specific benchmark from an Inference Recommender job.</p>"
+    },
     "RecommendationJobInputConfig":{
       "type":"structure",
       "required":["ModelPackageVersionArn"],
@@ -24805,6 +24901,10 @@
       },
       "documentation":"<p>The metrics of recommendations.</p>"
     },
+    "RecommendationStepType":{
+      "type":"string",
+      "enum":["BENCHMARK"]
+    },
     "RecordWrapper":{
       "type":"string",
       "enum":[