Amazon SageMaker Service Update: This change allows customers to enable data capturing while running a batch transform job, and configure monitoring schedule to monitoring the captured data.

Author: AWS

.changes/next-release/feature-AmazonSageMakerService-4386c3f.json

@@ -0,0 +1,6 @@
+{
+    "type": "feature",
+    "category": "Amazon SageMaker Service",
+    "contributor": "",
+    "description": "This change allows customers to enable data capturing while running a batch transform job, and configure monitoring schedule to monitoring the captured data."
+}

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

@@ -4662,6 +4662,25 @@
         "AWS/Textract/AnalyzeDocument/Forms/V1"
       ]
     },
+    "BatchDataCaptureConfig":{
+      "type":"structure",
+      "required":["DestinationS3Uri"],
+      "members":{
+        "DestinationS3Uri":{
+          "shape":"S3Uri",
+          "documentation":"<p>The Amazon S3 location being used to capture the data.</p>"
+        },
+        "KmsKeyId":{
+          "shape":"KmsKeyId",
+          "documentation":"<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 batch transform job.</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>"
+        },
+        "GenerateInferenceId":{
+          "shape":"Boolean",
+          "documentation":"<p>Flag that indicates whether to append inference id to the output.</p>"
+        }
+      },
+      "documentation":"<p>Configuration to control how SageMaker captures inference data for batch transform jobs.</p>"
+    },
     "BatchDescribeModelPackageError":{
       "type":"structure",
       "required":[
@@ -4757,6 +4776,61 @@
         "SingleRecord"
       ]
     },
+    "BatchTransformInput":{
+      "type":"structure",
+      "required":[
+        "DataCapturedDestinationS3Uri",
+        "DatasetFormat",
+        "LocalPath"
+      ],
+      "members":{
+        "DataCapturedDestinationS3Uri":{
+          "shape":"DestinationS3Uri",
+          "documentation":"<p>The Amazon S3 location being used to capture the data.</p>"
+        },
+        "DatasetFormat":{
+          "shape":"MonitoringDatasetFormat",
+          "documentation":"<p>The dataset format for your batch transform job.</p>"
+        },
+        "LocalPath":{
+          "shape":"ProcessingLocalPath",
+          "documentation":"<p>Path to the filesystem where the batch transform data is available to the container.</p>"
+        },
+        "S3InputMode":{
+          "shape":"ProcessingS3InputMode",
+          "documentation":"<p>Whether the <code>Pipe</code> or <code>File</code> is used as the input mode for transferring data for the monitoring job. <code>Pipe</code> mode is recommended for large datasets. <code>File</code> mode is useful for small files that fit in memory. Defaults to <code>File</code>.</p>"
+        },
+        "S3DataDistributionType":{
+          "shape":"ProcessingS3DataDistributionType",
+          "documentation":"<p>Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Defaults to <code>FullyReplicated</code> </p>"
+        },
+        "FeaturesAttribute":{
+          "shape":"String",
+          "documentation":"<p>The attributes of the input data that are the input features.</p>"
+        },
+        "InferenceAttribute":{
+          "shape":"String",
+          "documentation":"<p>The attribute of the input data that represents the ground truth label.</p>"
+        },
+        "ProbabilityAttribute":{
+          "shape":"String",
+          "documentation":"<p>In a classification problem, the attribute that represents the class probability.</p>"
+        },
+        "ProbabilityThresholdAttribute":{
+          "shape":"ProbabilityThresholdAttribute",
+          "documentation":"<p>The threshold for the class probability to be evaluated as a positive result.</p>"
+        },
+        "StartTimeOffset":{
+          "shape":"MonitoringTimeOffsetString",
+          "documentation":"<p>If specified, monitoring jobs substract this time from the start time. For information about using offsets for scheduling monitoring jobs, see <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/model-monitor-model-quality-schedule.html\">Schedule Model Quality Monitoring Jobs</a>.</p>"
+        },
+        "EndTimeOffset":{
+          "shape":"MonitoringTimeOffsetString",
+          "documentation":"<p>If specified, monitoring jobs substract this time from the end time. For information about using offsets for scheduling monitoring jobs, see <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/model-monitor-model-quality-schedule.html\">Schedule Model Quality Monitoring Jobs</a>.</p>"
+        }
+      },
+      "documentation":"<p>Input object for the batch transform job.</p>"
+    },
     "Bias":{
       "type":"structure",
       "members":{
@@ -8060,6 +8134,10 @@
           "shape":"TransformOutput",
           "documentation":"<p>Describes the results of the transform job.</p>"
         },
+        "DataCaptureConfig":{
+          "shape":"BatchDataCaptureConfig",
+          "documentation":"<p>Configuration to control how SageMaker captures inference data.</p>"
+        },
         "TransformResources":{
           "shape":"TransformResources",
           "documentation":"<p>Describes the resources, including ML instance types and ML instance count, to use for the transform job.</p>"
@@ -8529,9 +8607,12 @@
     },
     "DataQualityJobInput":{
       "type":"structure",
-      "required":["EndpointInput"],
       "members":{
-        "EndpointInput":{"shape":"EndpointInput"}
+        "EndpointInput":{"shape":"EndpointInput"},
+        "BatchTransformInput":{
+          "shape":"BatchTransformInput",
+          "documentation":"<p>Input object for the batch transform job.</p>"
+        }
       },
       "documentation":"<p>The input for the data quality monitoring job. Currently endpoints are supported for input.</p>"
     },
@@ -12458,6 +12539,10 @@
           "shape":"TransformOutput",
           "documentation":"<p>Identifies the Amazon S3 location where you want Amazon SageMaker to save the results from the transform job.</p>"
         },
+        "DataCaptureConfig":{
+          "shape":"BatchDataCaptureConfig",
+          "documentation":"<p>Configuration to control how SageMaker captures inference data.</p>"
+        },
         "TransformResources":{
           "shape":"TransformResources",
           "documentation":"<p>Describes the resources, including ML instance types and ML instance count, to use for the transform job.</p>"
@@ -20328,12 +20413,13 @@
     },
     "ModelBiasJobInput":{
       "type":"structure",
-      "required":[
-        "EndpointInput",
-        "GroundTruthS3Input"
-      ],
+      "required":["GroundTruthS3Input"],
       "members":{
         "EndpointInput":{"shape":"EndpointInput"},
+        "BatchTransformInput":{
+          "shape":"BatchTransformInput",
+          "documentation":"<p>Input object for the batch transform job.</p>"
+        },
         "GroundTruthS3Input":{
           "shape":"MonitoringGroundTruthS3Input",
           "documentation":"<p>Location of ground truth labels to use in model bias job.</p>"
@@ -20459,9 +20545,12 @@
     },
     "ModelExplainabilityJobInput":{
       "type":"structure",
-      "required":["EndpointInput"],
       "members":{
-        "EndpointInput":{"shape":"EndpointInput"}
+        "EndpointInput":{"shape":"EndpointInput"},
+        "BatchTransformInput":{
+          "shape":"BatchTransformInput",
+          "documentation":"<p>Input object for the batch transform job.</p>"
+        }
       },
       "documentation":"<p>Inputs for the model explainability job.</p>"
     },
@@ -21105,12 +21194,13 @@
     },
     "ModelQualityJobInput":{
       "type":"structure",
-      "required":[
-        "EndpointInput",
-        "GroundTruthS3Input"
-      ],
+      "required":["GroundTruthS3Input"],
       "members":{
         "EndpointInput":{"shape":"EndpointInput"},
+        "BatchTransformInput":{
+          "shape":"BatchTransformInput",
+          "documentation":"<p>Input object for the batch transform job.</p>"
+        },
         "GroundTruthS3Input":{
           "shape":"MonitoringGroundTruthS3Input",
           "documentation":"<p>The ground truth label provided for the model.</p>"
@@ -21250,6 +21340,34 @@
       "max":50,
       "min":1
     },
+    "MonitoringCsvDatasetFormat":{
+      "type":"structure",
+      "members":{
+        "Header":{
+          "shape":"Boolean",
+          "documentation":"<p>Indicates if the CSV data has a header.</p>"
+        }
+      },
+      "documentation":"<p>Represents the CSV dataset format used when running a monitoring job.</p>"
+    },
+    "MonitoringDatasetFormat":{
+      "type":"structure",
+      "members":{
+        "Csv":{
+          "shape":"MonitoringCsvDatasetFormat",
+          "documentation":"<p>The CSV dataset used in the monitoring job.</p>"
+        },
+        "Json":{
+          "shape":"MonitoringJsonDatasetFormat",
+          "documentation":"<p>The JSON dataset used in the monitoring job</p>"
+        },
+        "Parquet":{
+          "shape":"MonitoringParquetDatasetFormat",
+          "documentation":"<p>The Parquet dataset used in the monitoring job</p>"
+        }
+      },
+      "documentation":"<p>Represents the dataset format used when running a monitoring job.</p>"
+    },
     "MonitoringEnvironmentMap":{
       "type":"map",
       "key":{"shape":"ProcessingEnvironmentKey"},
@@ -21333,11 +21451,14 @@
     },
     "MonitoringInput":{
       "type":"structure",
-      "required":["EndpointInput"],
       "members":{
         "EndpointInput":{
           "shape":"EndpointInput",
           "documentation":"<p>The endpoint for a monitoring job.</p>"
+        },
+        "BatchTransformInput":{
+          "shape":"BatchTransformInput",
+          "documentation":"<p>Input object for the batch transform job.</p>"
         }
       },
       "documentation":"<p>The inputs for a monitoring job.</p>"
@@ -21447,6 +21568,16 @@
       "type":"list",
       "member":{"shape":"MonitoringJobDefinitionSummary"}
     },
+    "MonitoringJsonDatasetFormat":{
+      "type":"structure",
+      "members":{
+        "Line":{
+          "shape":"Boolean",
+          "documentation":"<p>Indicates if the file should be read as a json object per line. </p>"
+        }
+      },
+      "documentation":"<p>Represents the JSON dataset format used when running a monitoring job.</p>"
+    },
     "MonitoringMaxRuntimeInSeconds":{
       "type":"integer",
       "max":86400,
@@ -21499,6 +21630,12 @@
       "max":1,
       "min":1
     },
+    "MonitoringParquetDatasetFormat":{
+      "type":"structure",
+      "members":{
+      },
+      "documentation":"<p>Represents the Parquet dataset format used when running a monitoring job.</p>"
+    },
     "MonitoringProblemType":{
       "type":"string",
       "enum":[