Amazon SageMaker
Developer Guide

CreateTransformJob

Starts a transform job. A transform job uses a trained model to get inferences on a dataset and saves these results to an Amazon S3 location that you specify.

To perform batch transformations, you create a transform job and use the data that you have readily available.

In the request body, you provide the following:

  • TransformJobName - Identifies the transform job. The name must be unique within an AWS Region in an AWS account.

  • ModelName - Identifies the model to use. ModelName must be the name of an existing Amazon SageMaker model in the same AWS Region and AWS account. For information on creating a model, see CreateModel.

  • TransformInput - Describes the dataset to be transformed and the Amazon S3 location where it is stored.

  • TransformOutput - Identifies the Amazon S3 location where you want Amazon SageMaker to save the results from the transform job.

  • TransformResources - Identifies the ML compute instances for the transform job.

For more information about how batch transformation works Amazon SageMaker, see How It Works.

Request Syntax

{ "BatchStrategy": "string", "Environment": { "string" : "string" }, "MaxConcurrentTransforms": number, "MaxPayloadInMB": number, "ModelName": "string", "Tags": [ { "Key": "string", "Value": "string" } ], "TransformInput": { "CompressionType": "string", "ContentType": "string", "DataSource": { "S3DataSource": { "S3DataType": "string", "S3Uri": "string" } }, "SplitType": "string" }, "TransformJobName": "string", "TransformOutput": { "Accept": "string", "AssembleWith": "string", "KmsKeyId": "string", "S3OutputPath": "string" }, "TransformResources": { "InstanceCount": number, "InstanceType": "string", "VolumeKmsKeyId": "string" } }

Request Parameters

For information about the parameters that are common to all actions, see Common Parameters.

The request accepts the following data in JSON format.

BatchStrategy

Determines the number of records included in a single mini-batch. SingleRecord means only one record is used per mini-batch. MultiRecord means a mini-batch is set to contain as many records that can fit within the MaxPayloadInMB limit.

Batch transform will automatically split your input data into whatever payload size is specified if you set SplitType to Line and BatchStrategy to MultiRecord. There's no need to split the dataset into smaller files or to use larger payload sizes unless the records in your dataset are very large.

Type: String

Valid Values: MultiRecord | SingleRecord

Required: No

Environment

The environment variables to set in the Docker container. We support up to 16 key and values entries in the map.

Type: String to string map

Key Length Constraints: Maximum length of 1024.

Key Pattern: [a-zA-Z_][a-zA-Z0-9_]*

Value Length Constraints: Maximum length of 10240.

Required: No

MaxConcurrentTransforms

The maximum number of parallel requests that can be sent to each instance in a transform job. This is good for algorithms that implement multiple workers on larger instances . The default value is 1. To allow Amazon SageMaker to determine the appropriate number for MaxConcurrentTransforms, set the value to 0.

Type: Integer

Valid Range: Minimum value of 0.

Required: No

MaxPayloadInMB

The maximum payload size allowed, in MB. A payload is the data portion of a record (without metadata). The value in MaxPayloadInMB must be greater or equal to the size of a single record. You can approximate the size of a record by dividing the size of your dataset by the number of records. Then multiply this value by the number of records you want in a mini-batch. It is recommended to enter a value slightly larger than this to ensure the records fit within the maximum payload size. The default value is 6 MB. For an unlimited payload size, set the value to 0.

Type: Integer

Valid Range: Minimum value of 0.

Required: No

ModelName

The name of the model that you want to use for the transform job. ModelName must be the name of an existing Amazon SageMaker model within an AWS Region in an AWS account.

Type: String

Length Constraints: Maximum length of 63.

Pattern: ^[a-zA-Z0-9](-*[a-zA-Z0-9])*

Required: Yes

Tags

An array of key-value pairs. Adding tags is optional. For more information, see Using Cost Allocation Tags in the AWS Billing and Cost Management User Guide.

Type: Array of Tag objects

Array Members: Minimum number of 0 items. Maximum number of 50 items.

Required: No

TransformInput

Describes the input source and the way the transform job consumes it.

Type: TransformInput object

Required: Yes

TransformJobName

The name of the transform job. The name must be unique within an AWS Region in an AWS account.

Type: String

Length Constraints: Minimum length of 1. Maximum length of 63.

Pattern: ^[a-zA-Z0-9](-*[a-zA-Z0-9])*

Required: Yes

TransformOutput

Describes the results of the transform job.

Type: TransformOutput object

Required: Yes

TransformResources

Describes the resources, including ML instance types and ML instance count, to use for the transform job.

Type: TransformResources object

Required: Yes

Response Syntax

{ "TransformJobArn": "string" }

Response Elements

If the action is successful, the service sends back an HTTP 200 response.

The following data is returned in JSON format by the service.

TransformJobArn

The Amazon Resource Name (ARN) of the transform job.

Type: String

Length Constraints: Maximum length of 256.

Pattern: arn:aws[a-z\-]*:sagemaker:[a-z0-9\-]*:[0-9]{12}:transform-job/.*

Errors

For information about the errors that are common to all actions, see Common Errors.

ResourceInUse

Resource being accessed is in use.

HTTP Status Code: 400

ResourceLimitExceeded

You have exceeded an Amazon SageMaker resource limit. For example, you might have too many training jobs created.

HTTP Status Code: 400

See Also

For more information about using this API in one of the language-specific AWS SDKs, see the following: