pulumi-aws-native-kotlin/com.pulumi.awsnative.sagemaker.kotlin.outputs/ModelExplainabilityJobDefinitionBatchTransformInput

ModelExplainabilityJobDefinitionBatchTransformInput

data class ModelExplainabilityJobDefinitionBatchTransformInput(val dataCapturedDestinationS3Uri: String, val datasetFormat: ModelExplainabilityJobDefinitionDatasetFormat, val featuresAttribute: String? = null, val inferenceAttribute: String? = null, val localPath: String, val probabilityAttribute: String? = null, val s3DataDistributionType: ModelExplainabilityJobDefinitionBatchTransformInputS3DataDistributionType? = null, val s3InputMode: ModelExplainabilityJobDefinitionBatchTransformInputS3InputMode? = null)

The batch transform input for a monitoring job.

Constructors

ModelExplainabilityJobDefinitionBatchTransformInput

fun ModelExplainabilityJobDefinitionBatchTransformInput(dataCapturedDestinationS3Uri: String, datasetFormat: ModelExplainabilityJobDefinitionDatasetFormat, featuresAttribute: String? = null, inferenceAttribute: String? = null, localPath: String, probabilityAttribute: String? = null, s3DataDistributionType: ModelExplainabilityJobDefinitionBatchTransformInputS3DataDistributionType? = null, s3InputMode: ModelExplainabilityJobDefinitionBatchTransformInputS3InputMode? = null)

Types

object Companion

Properties

dataCapturedDestinationS3Uri

val dataCapturedDestinationS3Uri: String

A URI that identifies the Amazon S3 storage location where Batch Transform Job captures data.

val datasetFormat: ModelExplainabilityJobDefinitionDatasetFormat

featuresAttribute

val featuresAttribute: String? = null

JSONpath to locate features in JSONlines dataset

inferenceAttribute

val inferenceAttribute: String? = null

Index or JSONpath to locate predicted label(s)

val localPath: String

Path to the filesystem where the endpoint data is available to the container.

probabilityAttribute

val probabilityAttribute: String? = null

Index or JSONpath to locate probabilities

s3DataDistributionType

val s3DataDistributionType: ModelExplainabilityJobDefinitionBatchTransformInputS3DataDistributionType? = null

Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Defauts to FullyReplicated

val s3InputMode: ModelExplainabilityJobDefinitionBatchTransformInputS3InputMode? = null

Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Pipe mode is recommended for large datasets. File mode is useful for small files that fit in memory. Defaults to File.