class SageMakerCreateTrainingJob (construct)

Language	Type name
.NET	Amazon.CDK.AWS.StepFunctions.Tasks.SageMakerCreateTrainingJob
Go	github.com/aws/aws-cdk-go/awscdk/v2/awsstepfunctionstasks#SageMakerCreateTrainingJob
Java	software.amazon.awscdk.services.stepfunctions.tasks.SageMakerCreateTrainingJob
Python	aws_cdk.aws_stepfunctions_tasks.SageMakerCreateTrainingJob
TypeScript (source)	aws-cdk-lib » aws_stepfunctions_tasks » SageMakerCreateTrainingJob

Implements IConstruct, IDependable, IChainable, INextable, IGrantable, IConnectable

Class representing the SageMaker Create Training Job task.

Example

new tasks.SageMakerCreateTrainingJob(this, 'TrainSagemaker', {
  trainingJobName: sfn.JsonPath.stringAt('$.JobName'),
  algorithmSpecification: {
    algorithmName: 'BlazingText',
    trainingInputMode: tasks.InputMode.FILE,
  },
  inputDataConfig: [{
    channelName: 'train',
    dataSource: {
      s3DataSource: {
        s3DataType: tasks.S3DataType.S3_PREFIX,
        s3Location: tasks.S3Location.fromJsonExpression('$.S3Bucket'),
      },
    },
  }],
  outputDataConfig: {
    s3OutputLocation: tasks.S3Location.fromBucket(s3.Bucket.fromBucketName(this, 'Bucket', 'amzn-s3-demo-bucket'), 'myoutputpath'),
  },
  resourceConfig: {
    instanceCount: 1,
    instanceType: new ec2.InstanceType(sfn.JsonPath.stringAt('$.InstanceType')),
    volumeSize: Size.gibibytes(50),
  }, // optional: default is 1 instance of EC2 `M4.XLarge` with `10GB` volume
  stoppingCondition: {
    maxRuntime: Duration.hours(2),
  }, // optional: default is 1 hour
});

Initializer

new SageMakerCreateTrainingJob(scope: Construct, id: string, props: SageMakerCreateTrainingJobProps)

Parameters

• scope Construct
• id string — Descriptive identifier for this chainable.
• props SageMakerCreateTrainingJobProps

Construct Props

Name	Type	Description
algorithmSpecification	AlgorithmSpecification	Identifies the training algorithm to use.
outputDataConfig	OutputDataConfig	Identifies the Amazon S3 location where you want Amazon SageMaker to save the results of model training.
trainingJobName	string	Training Job Name.
comment?	string	An optional description for this state.
credentials?	Credentials	Credentials for an IAM Role that the State Machine assumes for executing the task.
enableNetworkIsolation?	boolean	Isolates the training container.
environment?	{ [string]: string }	Environment variables to set in the Docker container.
heartbeat?⚠️	Duration	Timeout for the heartbeat.
heartbeatTimeout?	Timeout	Timeout for the heartbeat.
hyperparameters?	{ [string]: any }	Algorithm-specific parameters that influence the quality of the model.
inputDataConfig?	Channel[]	Describes the various datasets (e.g. train, validation, test) and the Amazon S3 location where stored.
inputPath?	string	JSONPath expression to select part of the state to be the input to this state.
integrationPattern?	IntegrationPattern	AWS Step Functions integrates with services directly in the Amazon States Language.
outputPath?	string	JSONPath expression to select select a portion of the state output to pass to the next state.
resourceConfig?	ResourceConfig	Specifies the resources, ML compute instances, and ML storage volumes to deploy for model training.
resultPath?	string	JSONPath expression to indicate where to inject the state's output.
resultSelector?	{ [string]: any }	The JSON that will replace the state's raw result and become the effective result before ResultPath is applied.
role?	IRole	Role for the Training Job.
stateName?	string	Optional name for this state.
stoppingCondition?	StoppingCondition	Sets a time limit for training.
tags?	{ [string]: string }	Tags to be applied to the train job.
taskTimeout?	Timeout	Timeout for the task.
timeout?⚠️	Duration	Timeout for the task.
vpcConfig?	VpcConfig	Specifies the VPC that you want your training job to connect to.

---

algorithmSpecification

Type: AlgorithmSpecification

Identifies the training algorithm to use.

---

outputDataConfig

Type: OutputDataConfig

Identifies the Amazon S3 location where you want Amazon SageMaker to save the results of model training.

---

trainingJobName

Type: string

Training Job Name.

---

comment?

Type: string (optional, default: No comment)

An optional description for this state.

---

credentials?

Type: Credentials (optional, default: None (Task is executed using the State Machine's execution role))

Credentials for an IAM Role that the State Machine assumes for executing the task.

This enables cross-account resource invocations.

See also: https://docs.aws.amazon.com/step-functions/latest/dg/concepts-access-cross-acct-resources.html

---

enableNetworkIsolation?

Type: boolean (optional, default: false)

Isolates the training container.

No inbound or outbound network calls can be made to or from the training container.

---

environment?

Type: { [string]: string } (optional, default: No environment variables)

Environment variables to set in the Docker container.

---

heartbeat?⚠️

⚠️ Deprecated: use heartbeatTimeout

Type: Duration (optional, default: None)

Timeout for the heartbeat.

---

heartbeatTimeout?

Type: Timeout (optional, default: None)

Timeout for the heartbeat.

[disable-awslint:duration-prop-type] is needed because all props interface in aws-stepfunctions-tasks extend this interface

---

hyperparameters?

Type: { [string]: any } (optional, default: No hyperparameters)

Algorithm-specific parameters that influence the quality of the model.

Set hyperparameters before you start the learning process. For a list of hyperparameters provided by Amazon SageMaker

See also: https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html

---

inputDataConfig?

Type: Channel[] (optional, default: No inputDataConfig)

Describes the various datasets (e.g. train, validation, test) and the Amazon S3 location where stored.

---

inputPath?

Type: string (optional, default: The entire task input (JSON path '$'))

JSONPath expression to select part of the state to be the input to this state.

May also be the special value JsonPath.DISCARD, which will cause the effective input to be the empty object {}.

---

integrationPattern?

Type: IntegrationPattern (optional, default: IntegrationPattern.REQUEST_RESPONSE for most tasks. IntegrationPattern.RUN_JOB for the following exceptions: BatchSubmitJob, EmrAddStep, EmrCreateCluster, EmrTerminationCluster, and EmrContainersStartJobRun.)

AWS Step Functions integrates with services directly in the Amazon States Language.

You can control these AWS services using service integration patterns.

Depending on the AWS Service, the Service Integration Pattern availability will vary.

See also: https://docs.aws.amazon.com/step-functions/latest/dg/connect-supported-services.html

---

outputPath?

Type: string (optional, default: The entire JSON node determined by the state input, the task result, and resultPath is passed to the next state (JSON path '$'))

JSONPath expression to select select a portion of the state output to pass to the next state.

May also be the special value JsonPath.DISCARD, which will cause the effective output to be the empty object {}.

---

resourceConfig?

Type: ResourceConfig (optional, default: 1 instance of EC2 M4.XLarge with 10GB volume)

Specifies the resources, ML compute instances, and ML storage volumes to deploy for model training.

---

resultPath?

Type: string (optional, default: Replaces the entire input with the result (JSON path '$'))

JSONPath expression to indicate where to inject the state's output.

May also be the special value JsonPath.DISCARD, which will cause the state's input to become its output.

---

resultSelector?

Type: { [string]: any } (optional, default: None)

The JSON that will replace the state's raw result and become the effective result before ResultPath is applied.

You can use ResultSelector to create a payload with values that are static or selected from the state's raw result.

See also: https://docs.aws.amazon.com/step-functions/latest/dg/input-output-inputpath-params.html#input-output-resultselector

---

role?

Type: IRole (optional, default: a role will be created.)

Role for the Training Job.

The role must be granted all necessary permissions for the SageMaker training job to be able to operate.

See https://docs.aws.amazon.com/fr_fr/sagemaker/latest/dg/sagemaker-roles.html#sagemaker-roles-createtrainingjob-perms

---

stateName?

Type: string (optional, default: The construct ID will be used as state name)

Optional name for this state.

---

stoppingCondition?

Type: StoppingCondition (optional, default: max runtime of 1 hour)

Sets a time limit for training.

---

tags?

Type: { [string]: string } (optional, default: No tags)

Tags to be applied to the train job.

---

taskTimeout?

Type: Timeout (optional, default: None)

Timeout for the task.

[disable-awslint:duration-prop-type] is needed because all props interface in aws-stepfunctions-tasks extend this interface

---

timeout?⚠️

⚠️ Deprecated: use taskTimeout

Type: Duration (optional, default: None)

Timeout for the task.

---

vpcConfig?

Type: VpcConfig (optional, default: No VPC)

Specifies the VPC that you want your training job to connect to.

Properties

Name	Type	Description
connections	Connections	Allows specify security group connections for instances of this fleet.
endStates	INextable[]	Continuable states of this Chainable.
grantPrincipal	IPrincipal	The principal to grant permissions to.
id	string	Descriptive identifier for this chainable.
node	Node	The tree node.
role	IRole	The execution role for the Sagemaker training job.
startState	State	First state of this Chainable.
stateId	string	Tokenized string that evaluates to the state's ID.
taskMetrics?	TaskMetricsConfig
taskPolicies?	PolicyStatement[]

---

connections

Type: Connections

Allows specify security group connections for instances of this fleet.

---

endStates

Type: INextable[]

Continuable states of this Chainable.

---

grantPrincipal

Type: IPrincipal

The principal to grant permissions to.

---

id

Type: string

Descriptive identifier for this chainable.

---

node

Type: Node

The tree node.

---

role

Type: IRole

The execution role for the Sagemaker training job.

Only available after task has been added to a state machine.

---

startState

Type: State

First state of this Chainable.

---

stateId

Type: string

Tokenized string that evaluates to the state's ID.

---

taskMetrics?

Type: TaskMetricsConfig (optional)

---

taskPolicies?

Type: PolicyStatement[] (optional)

Methods

Name	Description
addCatch(handler, props?)	Add a recovery handler for this state.
addPrefix(x)	Add a prefix to the stateId of this state.
addRetry(props?)	Add retry configuration for this state.
addSecurityGroup(securityGroup)	Add the security group to all instances via the launch configuration security groups array.
bindToGraph(graph)	Register this state as part of the given graph.
metric(metricName, props?)	Return the given named metric for this Task.
metricFailed(props?)	Metric for the number of times this activity fails.
metricHeartbeatTimedOut(props?)	Metric for the number of times the heartbeat times out for this activity.
metricRunTime(props?)	The interval, in milliseconds, between the time the Task starts and the time it closes.
metricScheduleTime(props?)	The interval, in milliseconds, for which the activity stays in the schedule state.
metricScheduled(props?)	Metric for the number of times this activity is scheduled.
metricStarted(props?)	Metric for the number of times this activity is started.
metricSucceeded(props?)	Metric for the number of times this activity succeeds.
metricTime(props?)	The interval, in milliseconds, between the time the activity is scheduled and the time it closes.
metricTimedOut(props?)	Metric for the number of times this activity times out.
next(next)	Continue normal execution with the given state.
toStateJson()	Return the Amazon States Language object for this state.
toString()	Returns a string representation of this construct.

---

addCatch(handler, props?)

public addCatch(handler: IChainable, props?: CatchProps): TaskStateBase

Parameters

• handler IChainable
• props CatchProps

Returns

• TaskStateBase

Add a recovery handler for this state.

When a particular error occurs, execution will continue at the error handler instead of failing the state machine execution.

---

addPrefix(x)

public addPrefix(x: string): void

Parameters

• x string

Add a prefix to the stateId of this state.

---

addRetry(props?)

public addRetry(props?: RetryProps): TaskStateBase

Parameters

• props RetryProps

Returns

• TaskStateBase

Add retry configuration for this state.

This controls if and how the execution will be retried if a particular error occurs.

---

addSecurityGroup(securityGroup)

public addSecurityGroup(securityGroup: ISecurityGroup): void

Parameters

• securityGroup ISecurityGroup — : The security group to add.

Add the security group to all instances via the launch configuration security groups array.

---

bindToGraph(graph)

public bindToGraph(graph: StateGraph): void

Parameters

• graph StateGraph

Register this state as part of the given graph.

Don't call this. It will be called automatically when you work with states normally.

---

metric(metricName, props?)

public metric(metricName: string, props?: MetricOptions): Metric

Parameters

• metricName string
• props MetricOptions

Returns

• Metric

Return the given named metric for this Task.

---

metricFailed(props?)

public metricFailed(props?: MetricOptions): Metric

Parameters

• props MetricOptions

Returns

• Metric

Metric for the number of times this activity fails.

---

metricHeartbeatTimedOut(props?)

public metricHeartbeatTimedOut(props?: MetricOptions): Metric

Parameters

• props MetricOptions

Returns

• Metric

Metric for the number of times the heartbeat times out for this activity.

---

metricRunTime(props?)

public metricRunTime(props?: MetricOptions): Metric

Parameters

• props MetricOptions

Returns

• Metric

The interval, in milliseconds, between the time the Task starts and the time it closes.

---

metricScheduleTime(props?)

public metricScheduleTime(props?: MetricOptions): Metric

Parameters

• props MetricOptions

Returns

• Metric

The interval, in milliseconds, for which the activity stays in the schedule state.

---

metricScheduled(props?)

public metricScheduled(props?: MetricOptions): Metric

Parameters

• props MetricOptions

Returns

• Metric

Metric for the number of times this activity is scheduled.

---

metricStarted(props?)

public metricStarted(props?: MetricOptions): Metric

Parameters

• props MetricOptions

Returns

• Metric

Metric for the number of times this activity is started.

---

metricSucceeded(props?)

public metricSucceeded(props?: MetricOptions): Metric

Parameters

• props MetricOptions

Returns

• Metric

Metric for the number of times this activity succeeds.

---

metricTime(props?)

public metricTime(props?: MetricOptions): Metric

Parameters

• props MetricOptions

Returns

• Metric

The interval, in milliseconds, between the time the activity is scheduled and the time it closes.

---

metricTimedOut(props?)

public metricTimedOut(props?: MetricOptions): Metric

Parameters

• props MetricOptions

Returns

• Metric

Metric for the number of times this activity times out.

---

next(next)

public next(next: IChainable): Chain

Parameters

• next IChainable

Returns

• Chain

Continue normal execution with the given state.

---

toStateJson()

public toStateJson(): json

Returns

• json

Return the Amazon States Language object for this state.

---

toString()

public toString(): string

Returns

• string

Returns a string representation of this construct.