The Amazon Resource Name (ARN) of the AssociationInfo source.
The Amazon Resource Name (ARN) of the AssociationInfo destination.
The data type used to describe the relationship between different sources.
" + }, + "AssociationInfoList":{ + "type":"list", + "member":{"shape":"AssociationInfo"}, + "max":10, + "min":0 + }, "AssociationSummaries":{ "type":"list", "member":{"shape":"AssociationSummary"} @@ -6993,6 +7017,24 @@ "type":"string", "min":1 }, + "BaseModel":{ + "type":"structure", + "members":{ + "HubContentName":{ + "shape":"HubContentName", + "documentation":"The hub content name of the base model.
" + }, + "HubContentVersion":{ + "shape":"HubContentVersion", + "documentation":"The hub content version of the base model.
" + }, + "RecipeName":{ + "shape":"RecipeName", + "documentation":"The recipe name of the base model.
" + } + }, + "documentation":"Identifies the foundation model that was used as the starting point for model customization.
" + }, "BaseModelName":{ "type":"string", "max":256, @@ -7296,6 +7338,10 @@ "ModelApprovalStatus":{ "shape":"ModelApprovalStatus", "documentation":"The approval status of the model.
" + }, + "ModelPackageRegistrationType":{ + "shape":"ModelPackageRegistrationType", + "documentation":"The package registration type of the model package summary.
" } }, "documentation":"Provides summary information about the model package.
" @@ -7602,6 +7648,46 @@ }, "documentation":"Input object for the batch transform job.
" }, + "BedrockCustomModelDeploymentMetadata":{ + "type":"structure", + "members":{ + "Arn":{ + "shape":"String1024", + "documentation":"The Amazon Resource Name (ARN) of the metadata for the Amazon Bedrock custom model deployment.
" + } + }, + "documentation":"The metadata of the Amazon Bedrock custom model deployment.
" + }, + "BedrockCustomModelMetadata":{ + "type":"structure", + "members":{ + "Arn":{ + "shape":"String1024", + "documentation":"The Amazon Resource Name (ARN) of the Amazon Bedrock custom model metadata.
" + } + }, + "documentation":"The metadata of the Amazon Bedrock custom model.
" + }, + "BedrockModelImportMetadata":{ + "type":"structure", + "members":{ + "Arn":{ + "shape":"String1024", + "documentation":"The Amazon Resource Name (ARN) of the Amazon Bedrock model import metadata.
" + } + }, + "documentation":"The metadata of the Amazon Bedrock model import.
" + }, + "BedrockProvisionedModelThroughputMetadata":{ + "type":"structure", + "members":{ + "Arn":{ + "shape":"String1024", + "documentation":"The Amazon Resource Name (ARN) of the Amazon Bedrock provisioned model throughput metadata.
" + } + }, + "documentation":"The metadata of the Amazon Bedrock provisioned model throughput.
" + }, "BestObjectiveNotImproving":{ "type":"structure", "members":{ @@ -7635,6 +7721,11 @@ "box":true, "min":1 }, + "BillableTokenCount":{ + "type":"long", + "box":true, + "min":0 + }, "BlockedReason":{ "type":"string", "max":1024, @@ -9724,6 +9815,7 @@ }, "ClusterOrchestrator":{ "type":"structure", + "required":["Eks"], "members":{ "Eks":{ "shape":"ClusterOrchestratorEksConfig", @@ -12894,6 +12986,10 @@ "shape":"EntityDescription", "documentation":"A description of the model package.
" }, + "ModelPackageRegistrationType":{ + "shape":"ModelPackageRegistrationType", + "documentation":"The package registration type of the model package input.
" + }, "InferenceSpecification":{ "shape":"InferenceSpecification", "documentation":"Specifies details about inference jobs that you can run with models based on this model package, including the following information:
The Amazon ECR paths of containers that contain the inference code and model artifacts.
The instance types that the model package supports for transform jobs and real-time endpoints used for inference.
The input and output content formats that the model package supports for inference.
Contains information about attribute-based access control (ABAC) for the training job.
" + }, + "ServerlessJobConfig":{ + "shape":"ServerlessJobConfig", + "documentation":"The configuration for serverless training jobs.
" + }, + "MlflowConfig":{ + "shape":"MlflowConfig", + "documentation":"The MLflow configuration using SageMaker managed MLflow.
" + }, + "ModelPackageConfig":{ + "shape":"ModelPackageConfig", + "documentation":"The configuration for the model package.
" } } }, @@ -14331,6 +14439,15 @@ "min":1, "pattern":"([\\p{L}\\p{Z}\\p{N}_.:\\/=+\\-@]*)${1,256}" }, + "CustomizationTechnique":{ + "type":"string", + "enum":[ + "SFT", + "DPO", + "RLVR", + "RLAIF" + ] + }, "CustomizedMetricSpecification":{ "type":"structure", "members":{ @@ -14541,6 +14658,10 @@ "FileSystemDataSource":{ "shape":"FileSystemDataSource", "documentation":"The file system that is associated with a channel.
" + }, + "DatasetSource":{ + "shape":"DatasetSource", + "documentation":"The dataset resource that's associated with a channel.
" } }, "documentation":"Describes the location of the channel data.
" @@ -14578,6 +14699,17 @@ }, "documentation":"Configuration for Dataset Definition inputs. The Dataset Definition input must specify exactly one of either AthenaDatasetDefinition or RedshiftDatasetDefinition types.
The Amazon Resource Name (ARN) of the dataset resource.
" + } + }, + "documentation":"Specifies a dataset source for a channel.
" + }, "DebugHookConfig":{ "type":"structure", "required":["S3OutputPath"], @@ -19040,6 +19172,10 @@ "shape":"ModelPackageVersion", "documentation":"The version of the model package.
" }, + "ModelPackageRegistrationType":{ + "shape":"ModelPackageRegistrationType", + "documentation":"The package registration type of the model package output.
" + }, "ModelPackageArn":{ "shape":"ModelPackageArn", "documentation":"The Amazon Resource Name (ARN) of the model package.
" @@ -19688,6 +19824,10 @@ "PipelineVersionId":{ "shape":"PipelineVersionId", "documentation":"The ID of the pipeline version.
" + }, + "MLflowConfig":{ + "shape":"MLflowConfiguration", + "documentation":"The MLflow configuration of the pipeline execution.
" } } }, @@ -20192,7 +20332,7 @@ }, "SecondaryStatus":{ "shape":"SecondaryStatus", - "documentation":" Provides detailed information about the state of the training job. For detailed information on the secondary status of the training job, see StatusMessage under SecondaryStatusTransition.
SageMaker provides primary statuses and secondary statuses that apply to each of them:
Starting - Starting the training job.
Downloading - An optional stage for algorithms that support File training input mode. It indicates that data is being downloaded to the ML storage volumes.
Training - Training is in progress.
Interrupted - The job stopped because the managed spot training instances were interrupted.
Uploading - Training is complete and the model artifacts are being uploaded to the S3 location.
Completed - The training job has completed.
Failed - The training job has failed. The reason for the failure is returned in the FailureReason field of DescribeTrainingJobResponse.
MaxRuntimeExceeded - The job stopped because it exceeded the maximum allowed runtime.
MaxWaitTimeExceeded - The job stopped because it exceeded the maximum allowed wait time.
Stopped - The training job has stopped.
Stopping - Stopping the training job.
Valid values for SecondaryStatus are subject to change.
We no longer support the following secondary statuses:
LaunchingMLInstances
PreparingTraining
DownloadingTrainingImage
Provides detailed information about the state of the training job. For detailed information on the secondary status of the training job, see StatusMessage under SecondaryStatusTransition.
SageMaker provides primary statuses and secondary statuses that apply to each of them:
Starting - Starting the training job.
Pending - The training job is waiting for compute capacity or compute resource provision.
Downloading - An optional stage for algorithms that support File training input mode. It indicates that data is being downloaded to the ML storage volumes.
Training - Training is in progress.
Interrupted - The job stopped because the managed spot training instances were interrupted.
Uploading - Training is complete and the model artifacts are being uploaded to the S3 location.
Completed - The training job has completed.
Failed - The training job has failed. The reason for the failure is returned in the FailureReason field of DescribeTrainingJobResponse.
MaxRuntimeExceeded - The job stopped because it exceeded the maximum allowed runtime.
MaxWaitTimeExceeded - The job stopped because it exceeded the maximum allowed wait time.
Stopped - The training job has stopped.
Stopping - Stopping the training job.
Valid values for SecondaryStatus are subject to change.
We no longer support the following secondary statuses:
LaunchingMLInstances
PreparingTraining
DownloadingTrainingImage
The billable time in seconds. Billable time refers to the absolute wall-clock time.
Multiply BillableTimeInSeconds by the number of instances (InstanceCount) in your training cluster to get the total compute time SageMaker bills you if you run distributed training. The formula is as follows: BillableTimeInSeconds * InstanceCount .
You can calculate the savings from using managed spot training using the formula (1 - BillableTimeInSeconds / TrainingTimeInSeconds) * 100. For example, if BillableTimeInSeconds is 100 and TrainingTimeInSeconds is 500, the savings is 80%.
The billable token count for eligible serverless training jobs.
" + }, "DebugHookConfig":{"shape":"DebugHookConfig"}, "ExperimentConfig":{"shape":"ExperimentConfig"}, "DebugRuleConfigurations":{ @@ -20321,6 +20465,30 @@ "InfraCheckConfig":{ "shape":"InfraCheckConfig", "documentation":"Contains information about the infrastructure health check configuration for the training job.
" + }, + "ServerlessJobConfig":{ + "shape":"ServerlessJobConfig", + "documentation":"The configuration for serverless training jobs.
" + }, + "MlflowConfig":{ + "shape":"MlflowConfig", + "documentation":"The MLflow configuration using SageMaker managed MLflow.
" + }, + "ModelPackageConfig":{ + "shape":"ModelPackageConfig", + "documentation":"The configuration for the model package.
" + }, + "MlflowDetails":{ + "shape":"MlflowDetails", + "documentation":"The MLflow details of this job.
" + }, + "ProgressInfo":{ + "shape":"TrainingProgressInfo", + "documentation":"The Serverless training job progress information.
" + }, + "OutputModelPackageArn":{ + "shape":"ModelPackageArn", + "documentation":"The Amazon Resource Name (ARN) of the output model package containing model weights or checkpoints.
" } } }, @@ -22558,6 +22726,18 @@ }, "documentation":"This is an error field object that contains the error code and the reason for an operation failure.
" }, + "EvaluationType":{ + "type":"string", + "enum":[ + "LLMAJEvaluation", + "CustomScorerEvaluation", + "BenchmarkEvaluation" + ] + }, + "EvaluatorArn":{ + "type":"string", + "pattern":".*" + }, "EventDetails":{ "type":"structure", "members":{ @@ -24099,6 +24279,12 @@ "min":5, "pattern":"\\d{1,4}.\\d{1,4}.\\d{1,4}" }, + "HubDataSetArn":{ + "type":"string", + "max":2048, + "min":0, + "pattern":"(arn:[a-z0-9-\\.]{1,63}:sagemaker:\\w+(?:-\\w+)+:(\\d{12}|aws):hub-content\\/)[a-zA-Z0-9](-*[a-zA-Z0-9]){0,62}\\/DataSet\\/[a-zA-Z0-9](-*[a-zA-Z0-9]){0,63}(\\/\\d{1,4}.\\d{1,4}.\\d{1,4})?" + }, "HubDescription":{ "type":"string", "max":1023, @@ -25636,6 +25822,16 @@ }, "documentation":"The deployment configuration for an endpoint that hosts inference components. The configuration includes the desired deployment strategy and rollback settings.
" }, + "InferenceComponentMetadata":{ + "type":"structure", + "members":{ + "Arn":{ + "shape":"String2048", + "documentation":"The Amazon Resource Name (ARN) of the inference component metadata.
" + } + }, + "documentation":"The metadata of the inference component.
" + }, "InferenceComponentName":{ "type":"string", "max":63, @@ -27352,6 +27548,28 @@ }, "documentation":"Lists a summary of the properties of a lineage group. A lineage group provides a group of shareable lineage entity resources.
" }, + "LineageMetadata":{ + "type":"structure", + "members":{ + "ActionArns":{ + "shape":"MapString2048", + "documentation":"The Amazon Resource Name (ARN) of the lineage metadata action.
" + }, + "ArtifactArns":{ + "shape":"MapString2048", + "documentation":"The Amazon Resource Name (ARN) of the lineage metadata artifact.
" + }, + "ContextArns":{ + "shape":"MapString2048", + "documentation":"The Amazon Resource Name (ARN) of the lineage metadata context.
" + }, + "Associations":{ + "shape":"AssociationInfoList", + "documentation":"The lineage metadata associations.
" + } + }, + "documentation":"The metadata that tracks relationships between ML artifacts, actions, and contexts.
" + }, "LineageType":{ "type":"string", "enum":[ @@ -31807,6 +32025,26 @@ "min":1, "pattern":"[a-zA-Z]+ ?\\d+\\.\\d+(\\.\\d+)?" }, + "MLflowArn":{ + "type":"string", + "max":2048, + "min":0, + "pattern":"arn:aws[a-z\\-]*:sagemaker:[a-z0-9\\-]*:[0-9]{12}:mlflow-[a-zA-Z-]*/.*" + }, + "MLflowConfiguration":{ + "type":"structure", + "members":{ + "MlflowResourceArn":{ + "shape":"MLflowArn", + "documentation":"The Amazon Resource Name (ARN) of MLflow configuration resource.
" + }, + "MlflowExperimentName":{ + "shape":"MlflowExperimentEntityName", + "documentation":"The name of the MLflow configuration.
" + } + }, + "documentation":"The MLflow configuration.
" + }, "MaintenanceStatus":{ "type":"string", "enum":[ @@ -31838,6 +32076,13 @@ "DISABLED" ] }, + "MapString2048":{ + "type":"map", + "key":{"shape":"String2048"}, + "value":{"shape":"String2048"}, + "max":5, + "min":0 + }, "MaxAutoMLJobRuntimeInSeconds":{ "type":"integer", "box":true, @@ -32167,6 +32412,13 @@ "min":0, "pattern":"1|2" }, + "MlFlowResourceArn":{ + "type":"string", + "documentation":"MlflowDetails relevant fields
", + "max":2048, + "min":0, + "pattern":"arn:aws[a-z\\-]*:sagemaker:[a-z0-9\\-]*:[0-9]{12}:mlflow-[a-zA-Z-]*/.*" + }, "MlReservationArn":{ "type":"string", "max":258, @@ -32268,6 +32520,70 @@ "max":2048, "min":0 }, + "MlflowConfig":{ + "type":"structure", + "required":["MlflowResourceArn"], + "members":{ + "MlflowResourceArn":{ + "shape":"MlFlowResourceArn", + "documentation":"The Amazon Resource Name (ARN) of the MLflow resource.
" + }, + "MlflowExperimentName":{ + "shape":"MlflowExperimentName", + "documentation":"The MLflow experiment name used for this job.
" + }, + "MlflowRunName":{ + "shape":"MlflowRunName", + "documentation":"The MLflow run name used for this job.
" + } + }, + "documentation":"The MLflow configuration using SageMaker managed MLflow.
" + }, + "MlflowDetails":{ + "type":"structure", + "members":{ + "MlflowExperimentId":{ + "shape":"MlflowExperimentId", + "documentation":"The MLflow experiment ID used for this job.
" + }, + "MlflowRunId":{ + "shape":"MlflowRunId", + "documentation":"The MLflow run ID used for this job.
" + } + }, + "documentation":"The MLflow details of this job.
" + }, + "MlflowExperimentEntityName":{ + "type":"string", + "max":256, + "min":1, + "pattern":".*" + }, + "MlflowExperimentId":{ + "type":"string", + "max":256, + "min":1, + "pattern":".*" + }, + "MlflowExperimentName":{ + "type":"string", + "documentation":"MlflowConfig relevant fields
", + "max":256, + "min":1, + "pattern":".*" + }, + "MlflowRunId":{ + "type":"string", + "max":256, + "min":1, + "pattern":".*" + }, + "MlflowRunName":{ + "type":"string", + "max":256, + "min":1, + "pattern":".*" + }, "MlflowVersion":{ "type":"string", "max":16, @@ -33245,6 +33561,10 @@ "shape":"ModelPackageVersion", "documentation":"The version number of a versioned model.
" }, + "ModelPackageRegistrationType":{ + "shape":"ModelPackageRegistrationType", + "documentation":"The package registration type of the model package.
" + }, "ModelPackageArn":{ "shape":"ModelPackageArn", "documentation":"The Amazon Resource Name (ARN) of the model package.
" @@ -33367,6 +33687,21 @@ "max":100, "min":1 }, + "ModelPackageConfig":{ + "type":"structure", + "required":["ModelPackageGroupArn"], + "members":{ + "ModelPackageGroupArn":{ + "shape":"ModelPackageGroupArn", + "documentation":"The Amazon Resource Name (ARN) of the model package group of output model package.
" + }, + "SourceModelPackageArn":{ + "shape":"ModelPackageArn", + "documentation":"The Amazon Resource Name (ARN) of the source model package used for continued fine-tuning and custom model evaluation.
" + } + }, + "documentation":"The configuration for the Model package.
" + }, "ModelPackageContainerDefinition":{ "type":"structure", "members":{ @@ -33421,6 +33756,15 @@ "ModelDataETag":{ "shape":"String", "documentation":"The ETag associated with Model Data URL.
" + }, + "IsCheckpoint":{ + "shape":"Boolean", + "documentation":"The checkpoint of the model package.
", + "box":true + }, + "BaseModel":{ + "shape":"BaseModel", + "documentation":"The base model of the package.
" } }, "documentation":"Describes the Docker container for the model package.
" @@ -33542,6 +33886,13 @@ }, "documentation":"The model card associated with the model package. Since ModelPackageModelCard is tied to a model package, it is a specific usage of a model card and its schema is simplified compared to the schema of ModelCard. The ModelPackageModelCard schema does not include model_package_details, and model_overview is composed of the model_creator and model_artifact properties. For more information about the model package model card schema, see Model package model card schema. For more information about the model card associated with the model package, see View the Details of a Model Version.
The approval status of the model. This can be one of the following values.
APPROVED - The model is approved
REJECTED - The model is rejected.
PENDING_MANUAL_APPROVAL - The model is waiting for manual approval.
The package registration type of the model package summary.
" + } }, "documentation":"Provides summary information about a model package.
" }, @@ -36109,6 +36464,10 @@ "fiddler" ] }, + "Peft":{ + "type":"string", + "enum":["LORA"] + }, "PendingDeploymentSummary":{ "type":"structure", "required":["EndpointConfigName"], @@ -36533,6 +36892,30 @@ "EndpointConfig":{ "shape":"EndpointConfigStepMetadata", "documentation":"The endpoint configuration used to create an endpoint during this step execution.
" + }, + "BedrockCustomModel":{ + "shape":"BedrockCustomModelMetadata", + "documentation":"The metadata of the Amazon Bedrock custom model used in the pipeline execution step.
" + }, + "BedrockCustomModelDeployment":{ + "shape":"BedrockCustomModelDeploymentMetadata", + "documentation":"The metadata of the Amazon Bedrock custom model deployment used in pipeline execution step.
" + }, + "BedrockProvisionedModelThroughput":{ + "shape":"BedrockProvisionedModelThroughputMetadata", + "documentation":"The metadata of the Amazon Bedrock provisioned model throughput used in the pipeline execution step.
" + }, + "BedrockModelImport":{ + "shape":"BedrockModelImportMetadata", + "documentation":"The metadata of Amazon Bedrock model import used in pipeline execution step.
" + }, + "InferenceComponent":{ + "shape":"InferenceComponentMetadata", + "documentation":"The metadata of the inference component used in pipeline execution step.
" + }, + "Lineage":{ + "shape":"LineageMetadata", + "documentation":"The metadata of the lineage used in pipeline execution step.
" } }, "documentation":"Metadata for a step execution.
" @@ -38719,6 +39102,11 @@ "type":"list", "member":{"shape":"ProductionVariantInstanceType"} }, + "RecipeName":{ + "type":"string", + "max":255, + "min":0 + }, "RecommendationFailureReason":{"type":"string"}, "RecommendationJobArn":{ "type":"string", @@ -39627,7 +40015,7 @@ }, "VolumeSizeInGB":{ "shape":"OptionalVolumeSizeInGB", - "documentation":"The size of the ML storage volume that you want to provision.
ML storage volumes store model artifacts and incremental states. Training algorithms might also use the ML storage volume for scratch space. If you want to store the training data in the ML storage volume, choose File as the TrainingInputMode in the algorithm specification.
When using an ML instance with NVMe SSD volumes, SageMaker doesn't provision Amazon EBS General Purpose SSD (gp2) storage. Available storage is fixed to the NVMe-type instance's storage capacity. SageMaker configures storage paths for training datasets, checkpoints, model artifacts, and outputs to use the entire capacity of the instance storage. For example, ML instance families with the NVMe-type instance storage include ml.p4d, ml.g4dn, and ml.g5.
When using an ML instance with the EBS-only storage option and without instance storage, you must define the size of EBS volume through VolumeSizeInGB in the ResourceConfig API. For example, ML instance families that use EBS volumes include ml.c5 and ml.p2.
To look up instance types and their instance storage types and volumes, see Amazon EC2 Instance Types.
To find the default local paths defined by the SageMaker training platform, see Amazon SageMaker Training Storage Folders for Training Datasets, Checkpoints, Model Artifacts, and Outputs.
", + "documentation":"The size of the ML storage volume that you want to provision.
SageMaker automatically selects the volume size for serverless training jobs. You cannot customize this setting.
ML storage volumes store model artifacts and incremental states. Training algorithms might also use the ML storage volume for scratch space. If you want to store the training data in the ML storage volume, choose File as the TrainingInputMode in the algorithm specification.
When using an ML instance with NVMe SSD volumes, SageMaker doesn't provision Amazon EBS General Purpose SSD (gp2) storage. Available storage is fixed to the NVMe-type instance's storage capacity. SageMaker configures storage paths for training datasets, checkpoints, model artifacts, and outputs to use the entire capacity of the instance storage. For example, ML instance families with the NVMe-type instance storage include ml.p4d, ml.g4dn, and ml.g5.
When using an ML instance with the EBS-only storage option and without instance storage, you must define the size of EBS volume through VolumeSizeInGB in the ResourceConfig API. For example, ML instance families that use EBS volumes include ml.c5 and ml.p2.
To look up instance types and their instance storage types and volumes, see Amazon EC2 Instance Types.
To find the default local paths defined by the SageMaker training platform, see Amazon SageMaker Training Storage Folders for Training Datasets, Checkpoints, Model Artifacts, and Outputs.
", "box":true }, "VolumeKmsKeyId":{ @@ -40716,6 +41104,59 @@ } } }, + "ServerlessJobBaseModelArn":{ + "type":"string", + "documentation":"ServerlessJobConfig relevant fields
", + "max":2048, + "min":1, + "pattern":"(arn:[a-z0-9-\\.]{1,63}:sagemaker:\\w+(?:-\\w+)+:(\\d{12}|aws):hub-content\\/)[a-zA-Z0-9](-*[a-zA-Z0-9]){0,62}\\/Model\\/[a-zA-Z0-9](-*[a-zA-Z0-9]){0,63}(\\/\\d{1,4}.\\d{1,4}.\\d{1,4})?" + }, + "ServerlessJobConfig":{ + "type":"structure", + "required":[ + "BaseModelArn", + "JobType" + ], + "members":{ + "BaseModelArn":{ + "shape":"ServerlessJobBaseModelArn", + "documentation":"The base model Amazon Resource Name (ARN) in SageMaker Public Hub. SageMaker always selects the latest version of the provided model.
" + }, + "AcceptEula":{ + "shape":"AcceptEula", + "documentation":" Specifies agreement to the model end-user license agreement (EULA). The AcceptEula value must be explicitly defined as True in order to accept the EULA that this model requires. You are responsible for reviewing and complying with any applicable license terms and making sure they are acceptable for your use case before downloading or using a model. For more information, see End-user license agreements section for more details on accepting the EULA.
The serverless training job type.
" + }, + "CustomizationTechnique":{ + "shape":"CustomizationTechnique", + "documentation":"The model customization technique.
" + }, + "Peft":{ + "shape":"Peft", + "documentation":"The parameter-efficient fine-tuning configuration.
" + }, + "EvaluationType":{ + "shape":"EvaluationType", + "documentation":" The evaluation job type. Required when serverless job type is Evaluation.
The evaluator Amazon Resource Name (ARN) used as reward function or reward prompt.
" + } + }, + "documentation":"The configuration for the serverless training job.
" + }, + "ServerlessJobType":{ + "type":"string", + "enum":[ + "FineTuning", + "Evaluation" + ] + }, "ServerlessMaxConcurrency":{ "type":"integer", "box":true, @@ -41497,6 +41938,10 @@ "PipelineVersionId":{ "shape":"PipelineVersionId", "documentation":"The ID of the pipeline version to start execution from.
" + }, + "MlflowExperimentName":{ + "shape":"MlflowExperimentEntityName", + "documentation":"The MLflow experiment name of the start execution.
" } } }, @@ -42644,6 +43089,12 @@ "box":true, "min":0 }, + "TotalStepCountPerEpoch":{ + "type":"long", + "documentation":"TrainingProgressInfo relevant fields
", + "box":true, + "min":0 + }, "TrackingServerArn":{ "type":"string", "max":2048, @@ -42846,6 +43297,16 @@ "min":0, "pattern":"[\\S\\s]*" }, + "TrainingEpochCount":{ + "type":"long", + "box":true, + "min":0 + }, + "TrainingEpochIndex":{ + "type":"long", + "box":true, + "min":0 + }, "TrainingImageConfig":{ "type":"structure", "required":["TrainingRepositoryAccessMode"], @@ -43151,6 +43612,14 @@ "shape":"DebugRuleEvaluationStatuses", "documentation":"Information about the evaluation status of the rules for the training job.
" }, + "OutputModelPackageArn":{ + "shape":"ModelPackageArn", + "documentation":"The output model package Amazon Resource Name (ARN) that contains model weights or checkpoint.
" + }, + "ModelPackageConfig":{ + "shape":"ModelPackageConfig", + "documentation":"The model package configuration.
" + }, "ProfilerConfig":{"shape":"ProfilerConfig"}, "Environment":{ "shape":"TrainingEnvironmentMap", @@ -43565,6 +44034,28 @@ }, "documentation":"Details of the training plan.
For more information about how to reserve GPU capacity for your SageMaker HyperPod clusters using Amazon SageMaker Training Plan, see CreateTrainingPlan .
The total step count per epoch.
" + }, + "CurrentStep":{ + "shape":"TrainingStepIndex", + "documentation":"The current step number.
" + }, + "CurrentEpoch":{ + "shape":"TrainingEpochIndex", + "documentation":"The current epoch number.
" + }, + "MaxEpoch":{ + "shape":"TrainingEpochCount", + "documentation":"The maximum number of epochs for this job.
" + } + }, + "documentation":"The serverless training job progress information.
" + }, "TrainingRepositoryAccessMode":{ "type":"string", "enum":[ @@ -43637,6 +44128,11 @@ }, "documentation":"Defines how the algorithm is used for a training job.
" }, + "TrainingStepIndex":{ + "type":"long", + "box":true, + "min":0 + }, "TrainingTimeInSeconds":{ "type":"integer", "box":true, @@ -45909,6 +46405,10 @@ "shape":"ModelApprovalStatus", "documentation":"The approval status of the model.
" }, + "ModelPackageRegistrationType":{ + "shape":"ModelPackageRegistrationType", + "documentation":"The package registration type of the model package input.
" + }, "ApprovalDescription":{ "shape":"ApprovalDescription", "documentation":"A description for the approval status of the model.
" diff --git a/src/sagemaker_core/main/code_injection/shape_dag.py b/src/sagemaker_core/main/code_injection/shape_dag.py index 014a3e9..235ac0d 100644 --- a/src/sagemaker_core/main/code_injection/shape_dag.py +++ b/src/sagemaker_core/main/code_injection/shape_dag.py @@ -372,6 +372,18 @@ ], "type": "structure", }, + "AssociationInfo": { + "members": [ + {"name": "SourceArn", "shape": "String2048", "type": "string"}, + {"name": "DestinationArn", "shape": "String2048", "type": "string"}, + ], + "type": "structure", + }, + "AssociationInfoList": { + "member_shape": "AssociationInfo", + "member_type": "structure", + "type": "list", + }, "AssociationSummaries": { "member_shape": "AssociationSummary", "member_type": "structure", @@ -828,6 +840,14 @@ ], "type": "structure", }, + "BaseModel": { + "members": [ + {"name": "HubContentName", "shape": "HubContentName", "type": "string"}, + {"name": "HubContentVersion", "shape": "HubContentVersion", "type": "string"}, + {"name": "RecipeName", "shape": "RecipeName", "type": "string"}, + ], + "type": "structure", + }, "BatchAddClusterNodesError": { "members": [ {"name": "InstanceGroupName", "shape": "InstanceGroupName", "type": "string"}, @@ -965,6 +985,11 @@ }, {"name": "ModelPackageStatus", "shape": "ModelPackageStatus", "type": "string"}, {"name": "ModelApprovalStatus", "shape": "ModelApprovalStatus", "type": "string"}, + { + "name": "ModelPackageRegistrationType", + "shape": "ModelPackageRegistrationType", + "type": "string", + }, ], "type": "structure", }, @@ -1228,6 +1253,22 @@ ], "type": "structure", }, + "BedrockCustomModelDeploymentMetadata": { + "members": [{"name": "Arn", "shape": "String1024", "type": "string"}], + "type": "structure", + }, + "BedrockCustomModelMetadata": { + "members": [{"name": "Arn", "shape": "String1024", "type": "string"}], + "type": "structure", + }, + "BedrockModelImportMetadata": { + "members": [{"name": "Arn", "shape": "String1024", "type": "string"}], + "type": "structure", + }, + "BedrockProvisionedModelThroughputMetadata": { + "members": [{"name": "Arn", "shape": "String1024", "type": "string"}], + "type": "structure", + }, "BestObjectiveNotImproving": { "members": [ { @@ -3275,6 +3316,11 @@ {"name": "ModelPackageName", "shape": "EntityName", "type": "string"}, {"name": "ModelPackageGroupName", "shape": "ArnOrName", "type": "string"}, {"name": "ModelPackageDescription", "shape": "EntityDescription", "type": "string"}, + { + "name": "ModelPackageRegistrationType", + "shape": "ModelPackageRegistrationType", + "type": "string", + }, { "name": "InferenceSpecification", "shape": "InferenceSpecification", @@ -3769,6 +3815,9 @@ "shape": "SessionChainingConfig", "type": "structure", }, + {"name": "ServerlessJobConfig", "shape": "ServerlessJobConfig", "type": "structure"}, + {"name": "MlflowConfig", "shape": "MlflowConfig", "type": "structure"}, + {"name": "ModelPackageConfig", "shape": "ModelPackageConfig", "type": "structure"}, ], "type": "structure", }, @@ -4079,6 +4128,7 @@ "members": [ {"name": "S3DataSource", "shape": "S3DataSource", "type": "structure"}, {"name": "FileSystemDataSource", "shape": "FileSystemDataSource", "type": "structure"}, + {"name": "DatasetSource", "shape": "DatasetSource", "type": "structure"}, ], "type": "structure", }, @@ -4100,6 +4150,10 @@ ], "type": "structure", }, + "DatasetSource": { + "members": [{"name": "DatasetArn", "shape": "HubDataSetArn", "type": "string"}], + "type": "structure", + }, "DebugHookConfig": { "members": [ {"name": "LocalPath", "shape": "DirectoryPath", "type": "string"}, @@ -6084,6 +6138,11 @@ {"name": "ModelPackageName", "shape": "EntityName", "type": "string"}, {"name": "ModelPackageGroupName", "shape": "EntityName", "type": "string"}, {"name": "ModelPackageVersion", "shape": "ModelPackageVersion", "type": "integer"}, + { + "name": "ModelPackageRegistrationType", + "shape": "ModelPackageRegistrationType", + "type": "string", + }, {"name": "ModelPackageArn", "shape": "ModelPackageArn", "type": "string"}, {"name": "ModelPackageDescription", "shape": "EntityDescription", "type": "string"}, {"name": "CreationTime", "shape": "CreationTime", "type": "timestamp"}, @@ -6417,6 +6476,7 @@ "type": "structure", }, {"name": "PipelineVersionId", "shape": "PipelineVersionId", "type": "long"}, + {"name": "MLflowConfig", "shape": "MLflowConfiguration", "type": "structure"}, ], "type": "structure", }, @@ -6678,6 +6738,7 @@ {"name": "CheckpointConfig", "shape": "CheckpointConfig", "type": "structure"}, {"name": "TrainingTimeInSeconds", "shape": "TrainingTimeInSeconds", "type": "integer"}, {"name": "BillableTimeInSeconds", "shape": "BillableTimeInSeconds", "type": "integer"}, + {"name": "BillableTokenCount", "shape": "BillableTokenCount", "type": "long"}, {"name": "DebugHookConfig", "shape": "DebugHookConfig", "type": "structure"}, {"name": "ExperimentConfig", "shape": "ExperimentConfig", "type": "structure"}, {"name": "DebugRuleConfigurations", "shape": "DebugRuleConfigurations", "type": "list"}, @@ -6707,6 +6768,12 @@ {"name": "RetryStrategy", "shape": "RetryStrategy", "type": "structure"}, {"name": "RemoteDebugConfig", "shape": "RemoteDebugConfig", "type": "structure"}, {"name": "InfraCheckConfig", "shape": "InfraCheckConfig", "type": "structure"}, + {"name": "ServerlessJobConfig", "shape": "ServerlessJobConfig", "type": "structure"}, + {"name": "MlflowConfig", "shape": "MlflowConfig", "type": "structure"}, + {"name": "ModelPackageConfig", "shape": "ModelPackageConfig", "type": "structure"}, + {"name": "MlflowDetails", "shape": "MlflowDetails", "type": "structure"}, + {"name": "ProgressInfo", "shape": "TrainingProgressInfo", "type": "structure"}, + {"name": "OutputModelPackageArn", "shape": "ModelPackageArn", "type": "string"}, ], "type": "structure", }, @@ -8751,6 +8818,10 @@ ], "type": "structure", }, + "InferenceComponentMetadata": { + "members": [{"name": "Arn", "shape": "String2048", "type": "string"}], + "type": "structure", + }, "InferenceComponentRollingUpdatePolicy": { "members": [ { @@ -9467,6 +9538,15 @@ ], "type": "structure", }, + "LineageMetadata": { + "members": [ + {"name": "ActionArns", "shape": "MapString2048", "type": "map"}, + {"name": "ArtifactArns", "shape": "MapString2048", "type": "map"}, + {"name": "ContextArns", "shape": "MapString2048", "type": "map"}, + {"name": "Associations", "shape": "AssociationInfoList", "type": "list"}, + ], + "type": "structure", + }, "ListActionsRequest": { "members": [ {"name": "SourceUri", "shape": "SourceUri", "type": "string"}, @@ -11346,6 +11426,24 @@ ], "type": "structure", }, + "MLflowConfiguration": { + "members": [ + {"name": "MlflowResourceArn", "shape": "MLflowArn", "type": "string"}, + { + "name": "MlflowExperimentName", + "shape": "MlflowExperimentEntityName", + "type": "string", + }, + ], + "type": "structure", + }, + "MapString2048": { + "key_shape": "String2048", + "key_type": "string", + "type": "map", + "value_shape": "String2048", + "value_type": "string", + }, "MemberDefinition": { "members": [ { @@ -11471,6 +11569,21 @@ ], "type": "structure", }, + "MlflowConfig": { + "members": [ + {"name": "MlflowResourceArn", "shape": "MlFlowResourceArn", "type": "string"}, + {"name": "MlflowExperimentName", "shape": "MlflowExperimentName", "type": "string"}, + {"name": "MlflowRunName", "shape": "MlflowRunName", "type": "string"}, + ], + "type": "structure", + }, + "MlflowDetails": { + "members": [ + {"name": "MlflowExperimentId", "shape": "MlflowExperimentId", "type": "string"}, + {"name": "MlflowRunId", "shape": "MlflowRunId", "type": "string"}, + ], + "type": "structure", + }, "Model": { "members": [ {"name": "ModelName", "shape": "ModelName", "type": "string"}, @@ -11884,6 +11997,11 @@ {"name": "ModelPackageName", "shape": "EntityName", "type": "string"}, {"name": "ModelPackageGroupName", "shape": "EntityName", "type": "string"}, {"name": "ModelPackageVersion", "shape": "ModelPackageVersion", "type": "integer"}, + { + "name": "ModelPackageRegistrationType", + "shape": "ModelPackageRegistrationType", + "type": "string", + }, {"name": "ModelPackageArn", "shape": "ModelPackageArn", "type": "string"}, {"name": "ModelPackageDescription", "shape": "EntityDescription", "type": "string"}, {"name": "CreationTime", "shape": "CreationTime", "type": "timestamp"}, @@ -11940,6 +12058,13 @@ "member_type": "string", "type": "list", }, + "ModelPackageConfig": { + "members": [ + {"name": "ModelPackageGroupArn", "shape": "ModelPackageGroupArn", "type": "string"}, + {"name": "SourceModelPackageArn", "shape": "ModelPackageArn", "type": "string"}, + ], + "type": "structure", + }, "ModelPackageContainerDefinition": { "members": [ {"name": "ContainerHostname", "shape": "ContainerHostname", "type": "string"}, @@ -11959,6 +12084,8 @@ "type": "structure", }, {"name": "ModelDataETag", "shape": "String", "type": "string"}, + {"name": "IsCheckpoint", "shape": "Boolean", "type": "boolean"}, + {"name": "BaseModel", "shape": "BaseModel", "type": "structure"}, ], "type": "structure", }, @@ -12059,6 +12186,11 @@ {"name": "ModelPackageStatus", "shape": "ModelPackageStatus", "type": "string"}, {"name": "ModelApprovalStatus", "shape": "ModelApprovalStatus", "type": "string"}, {"name": "ModelLifeCycle", "shape": "ModelLifeCycle", "type": "structure"}, + { + "name": "ModelPackageRegistrationType", + "shape": "ModelPackageRegistrationType", + "type": "string", + }, ], "type": "structure", }, @@ -13283,6 +13415,32 @@ {"name": "AutoMLJob", "shape": "AutoMLJobStepMetadata", "type": "structure"}, {"name": "Endpoint", "shape": "EndpointStepMetadata", "type": "structure"}, {"name": "EndpointConfig", "shape": "EndpointConfigStepMetadata", "type": "structure"}, + { + "name": "BedrockCustomModel", + "shape": "BedrockCustomModelMetadata", + "type": "structure", + }, + { + "name": "BedrockCustomModelDeployment", + "shape": "BedrockCustomModelDeploymentMetadata", + "type": "structure", + }, + { + "name": "BedrockProvisionedModelThroughput", + "shape": "BedrockProvisionedModelThroughputMetadata", + "type": "structure", + }, + { + "name": "BedrockModelImport", + "shape": "BedrockModelImportMetadata", + "type": "structure", + }, + { + "name": "InferenceComponent", + "shape": "InferenceComponentMetadata", + "type": "structure", + }, + {"name": "Lineage", "shape": "LineageMetadata", "type": "structure"}, ], "type": "structure", }, @@ -14863,6 +15021,18 @@ ], "type": "structure", }, + "ServerlessJobConfig": { + "members": [ + {"name": "BaseModelArn", "shape": "ServerlessJobBaseModelArn", "type": "string"}, + {"name": "AcceptEula", "shape": "AcceptEula", "type": "boolean"}, + {"name": "JobType", "shape": "ServerlessJobType", "type": "string"}, + {"name": "CustomizationTechnique", "shape": "CustomizationTechnique", "type": "string"}, + {"name": "Peft", "shape": "Peft", "type": "string"}, + {"name": "EvaluationType", "shape": "EvaluationType", "type": "string"}, + {"name": "EvaluatorArn", "shape": "EvaluatorArn", "type": "string"}, + ], + "type": "structure", + }, "ServiceCatalogProvisionedProductDetails": { "members": [ {"name": "ProvisionedProductId", "shape": "ServiceCatalogEntityId", "type": "string"}, @@ -15142,6 +15312,11 @@ "type": "structure", }, {"name": "PipelineVersionId", "shape": "PipelineVersionId", "type": "long"}, + { + "name": "MlflowExperimentName", + "shape": "MlflowExperimentEntityName", + "type": "string", + }, ], "type": "structure", }, @@ -15679,6 +15854,8 @@ "shape": "DebugRuleEvaluationStatuses", "type": "list", }, + {"name": "OutputModelPackageArn", "shape": "ModelPackageArn", "type": "string"}, + {"name": "ModelPackageConfig", "shape": "ModelPackageConfig", "type": "structure"}, {"name": "ProfilerConfig", "shape": "ProfilerConfig", "type": "structure"}, {"name": "Environment", "shape": "TrainingEnvironmentMap", "type": "map"}, {"name": "RetryStrategy", "shape": "RetryStrategy", "type": "structure"}, @@ -15804,6 +15981,15 @@ ], "type": "structure", }, + "TrainingProgressInfo": { + "members": [ + {"name": "TotalStepCountPerEpoch", "shape": "TotalStepCountPerEpoch", "type": "long"}, + {"name": "CurrentStep", "shape": "TrainingStepIndex", "type": "long"}, + {"name": "CurrentEpoch", "shape": "TrainingEpochIndex", "type": "long"}, + {"name": "MaxEpoch", "shape": "TrainingEpochCount", "type": "long"}, + ], + "type": "structure", + }, "TrainingRepositoryAuthConfig": { "members": [ { @@ -16791,6 +16977,11 @@ "members": [ {"name": "ModelPackageArn", "shape": "ModelPackageArn", "type": "string"}, {"name": "ModelApprovalStatus", "shape": "ModelApprovalStatus", "type": "string"}, + { + "name": "ModelPackageRegistrationType", + "shape": "ModelPackageRegistrationType", + "type": "string", + }, {"name": "ApprovalDescription", "shape": "ApprovalDescription", "type": "string"}, {"name": "CustomerMetadataProperties", "shape": "CustomerMetadataMap", "type": "map"}, { diff --git a/src/sagemaker_core/main/resources.py b/src/sagemaker_core/main/resources.py index b071e49..84a64a4 100644 --- a/src/sagemaker_core/main/resources.py +++ b/src/sagemaker_core/main/resources.py @@ -17582,7 +17582,21 @@ def get_name(self) -> str: logger.error("Name attribute not found for object mlflow_app") return None + def populate_inputs_decorator(create_func): + @functools.wraps(create_func) + def wrapper(*args, **kwargs): + config_schema_for_resource = {"role_arn": {"type": "string"}} + return create_func( + *args, + **Base.get_updated_kwargs_with_configured_attributes( + config_schema_for_resource, "MlflowApp", **kwargs + ), + ) + + return wrapper + @classmethod + @populate_inputs_decorator @Base.add_validate_call def create( cls, @@ -17750,6 +17764,7 @@ def refresh( transform(response, "DescribeMlflowAppResponse", self) return self + @populate_inputs_decorator @Base.add_validate_call def update( self, @@ -20615,6 +20630,7 @@ class ModelPackage(Base): model_package_status_details: Details about the current status of the model package. model_package_group_name: If the model is a versioned model, the name of the model group that the versioned model belongs to. model_package_version: The version of the model package. + model_package_registration_type: The package registration type of the model package output. model_package_description: A brief summary of the model package. inference_specification: Details about inference jobs that you can run with models based on this model package. source_algorithm_specification: Details about the algorithm that was used to create the model package. @@ -20644,6 +20660,7 @@ class ModelPackage(Base): model_package_name: str model_package_group_name: Optional[str] = Unassigned() model_package_version: Optional[int] = Unassigned() + model_package_registration_type: Optional[str] = Unassigned() model_package_arn: Optional[str] = Unassigned() model_package_description: Optional[str] = Unassigned() creation_time: Optional[datetime.datetime] = Unassigned() @@ -20749,6 +20766,7 @@ def create( model_package_name: Optional[str] = Unassigned(), model_package_group_name: Optional[Union[str, object]] = Unassigned(), model_package_description: Optional[str] = Unassigned(), + model_package_registration_type: Optional[str] = Unassigned(), inference_specification: Optional[shapes.InferenceSpecification] = Unassigned(), validation_specification: Optional[ shapes.ModelPackageValidationSpecification @@ -20785,6 +20803,7 @@ def create( model_package_name: The name of the model package. The name must have 1 to 63 characters. Valid characters are a-z, A-Z, 0-9, and - (hyphen). This parameter is required for unversioned models. It is not applicable to versioned models. model_package_group_name: The name or Amazon Resource Name (ARN) of the model package group that this model version belongs to. This parameter is required for versioned models, and does not apply to unversioned models. model_package_description: A description of the model package. + model_package_registration_type: The package registration type of the model package input. inference_specification: Specifies details about inference jobs that you can run with models based on this model package, including the following information: The Amazon ECR paths of containers that contain the inference code and model artifacts. The instance types that the model package supports for transform jobs and real-time endpoints used for inference. The input and output content formats that the model package supports for inference. validation_specification: Specifies configurations for one or more transform jobs that SageMaker runs to test the model package. source_algorithm_specification: Details about the algorithm that was used to create the model package. @@ -20837,6 +20856,7 @@ def create( "ModelPackageName": model_package_name, "ModelPackageGroupName": model_package_group_name, "ModelPackageDescription": model_package_description, + "ModelPackageRegistrationType": model_package_registration_type, "InferenceSpecification": inference_specification, "ValidationSpecification": validation_specification, "SourceAlgorithmSpecification": source_algorithm_specification, @@ -20967,6 +20987,7 @@ def refresh( def update( self, model_approval_status: Optional[str] = Unassigned(), + model_package_registration_type: Optional[str] = Unassigned(), approval_description: Optional[str] = Unassigned(), customer_metadata_properties: Optional[Dict[str, str]] = Unassigned(), customer_metadata_properties_to_remove: Optional[List[str]] = Unassigned(), @@ -21009,6 +21030,7 @@ def update( operation_input_args = { "ModelPackageArn": self.model_package_arn, "ModelApprovalStatus": model_approval_status, + "ModelPackageRegistrationType": model_package_registration_type, "ApprovalDescription": approval_description, "CustomerMetadataProperties": customer_metadata_properties, "CustomerMetadataPropertiesToRemove": customer_metadata_properties_to_remove, @@ -25678,6 +25700,7 @@ class PipelineExecution(Base): parallelism_configuration: The parallelism configuration applied to the pipeline. selective_execution_config: The selective execution configuration applied to the pipeline run. pipeline_version_id: The ID of the pipeline version. + m_lflow_config: The MLflow configuration of the pipeline execution. """ @@ -25695,6 +25718,7 @@ class PipelineExecution(Base): parallelism_configuration: Optional[shapes.ParallelismConfiguration] = Unassigned() selective_execution_config: Optional[shapes.SelectiveExecutionConfig] = Unassigned() pipeline_version_id: Optional[int] = Unassigned() + m_lflow_config: Optional[shapes.MLflowConfiguration] = Unassigned() def get_name(self) -> str: attributes = vars(self) @@ -28977,7 +29001,7 @@ class TrainingJob(Base): training_job_arn: The Amazon Resource Name (ARN) of the training job. model_artifacts: Information about the Amazon S3 location that is configured for storing model artifacts. training_job_status: The status of the training job. SageMaker provides the following training job statuses: InProgress - The training is in progress. Completed - The training job has completed. Failed - The training job has failed. To see the reason for the failure, see the FailureReason field in the response to a DescribeTrainingJobResponse call. Stopping - The training job is stopping. Stopped - The training job has stopped. For more detailed information, see SecondaryStatus. - secondary_status: Provides detailed information about the state of the training job. For detailed information on the secondary status of the training job, see StatusMessage under SecondaryStatusTransition. SageMaker provides primary statuses and secondary statuses that apply to each of them: InProgress Starting - Starting the training job. Downloading - An optional stage for algorithms that support File training input mode. It indicates that data is being downloaded to the ML storage volumes. Training - Training is in progress. Interrupted - The job stopped because the managed spot training instances were interrupted. Uploading - Training is complete and the model artifacts are being uploaded to the S3 location. Completed Completed - The training job has completed. Failed Failed - The training job has failed. The reason for the failure is returned in the FailureReason field of DescribeTrainingJobResponse. Stopped MaxRuntimeExceeded - The job stopped because it exceeded the maximum allowed runtime. MaxWaitTimeExceeded - The job stopped because it exceeded the maximum allowed wait time. Stopped - The training job has stopped. Stopping Stopping - Stopping the training job. Valid values for SecondaryStatus are subject to change. We no longer support the following secondary statuses: LaunchingMLInstances PreparingTraining DownloadingTrainingImage + secondary_status: Provides detailed information about the state of the training job. For detailed information on the secondary status of the training job, see StatusMessage under SecondaryStatusTransition. SageMaker provides primary statuses and secondary statuses that apply to each of them: InProgress Starting - Starting the training job. Pending - The training job is waiting for compute capacity or compute resource provision. Downloading - An optional stage for algorithms that support File training input mode. It indicates that data is being downloaded to the ML storage volumes. Training - Training is in progress. Interrupted - The job stopped because the managed spot training instances were interrupted. Uploading - Training is complete and the model artifacts are being uploaded to the S3 location. Completed Completed - The training job has completed. Failed Failed - The training job has failed. The reason for the failure is returned in the FailureReason field of DescribeTrainingJobResponse. Stopped MaxRuntimeExceeded - The job stopped because it exceeded the maximum allowed runtime. MaxWaitTimeExceeded - The job stopped because it exceeded the maximum allowed wait time. Stopped - The training job has stopped. Stopping Stopping - Stopping the training job. Valid values for SecondaryStatus are subject to change. We no longer support the following secondary statuses: LaunchingMLInstances PreparingTraining DownloadingTrainingImage stopping_condition: Specifies a limit to how long a model training job can run. It also specifies how long a managed Spot training job has to complete. When the job reaches the time limit, SageMaker ends the training job. Use this API to cap model training costs. To stop a job, SageMaker sends the algorithm the SIGTERM signal, which delays job termination for 120 seconds. Algorithms can use this 120-second window to save the model artifacts, so the results of training are not lost. creation_time: A timestamp that indicates when the training job was created. tuning_job_arn: The Amazon Resource Name (ARN) of the associated hyperparameter tuning job if the training job was launched by a hyperparameter tuning job. @@ -29003,6 +29027,7 @@ class TrainingJob(Base): checkpoint_config: training_time_in_seconds: The training time in seconds. billable_time_in_seconds: The billable time in seconds. Billable time refers to the absolute wall-clock time. Multiply BillableTimeInSeconds by the number of instances (InstanceCount) in your training cluster to get the total compute time SageMaker bills you if you run distributed training. The formula is as follows: BillableTimeInSeconds \* InstanceCount . You can calculate the savings from using managed spot training using the formula (1 - BillableTimeInSeconds / TrainingTimeInSeconds) \* 100. For example, if BillableTimeInSeconds is 100 and TrainingTimeInSeconds is 500, the savings is 80%. + billable_token_count: The billable token count for eligible serverless training jobs. debug_hook_config: experiment_config: debug_rule_configurations: Configuration information for Amazon SageMaker Debugger rules for debugging output tensors. @@ -29016,6 +29041,12 @@ class TrainingJob(Base): retry_strategy: The number of times to retry the job when the job fails due to an InternalServerError. remote_debug_config: Configuration for remote debugging. To learn more about the remote debugging functionality of SageMaker, see Access a training container through Amazon Web Services Systems Manager (SSM) for remote debugging. infra_check_config: Contains information about the infrastructure health check configuration for the training job. + serverless_job_config: The configuration for serverless training jobs. + mlflow_config: The MLflow configuration using SageMaker managed MLflow. + model_package_config: The configuration for the model package. + mlflow_details: The MLflow details of this job. + progress_info: The Serverless training job progress information. + output_model_package_arn: The Amazon Resource Name (ARN) of the output model package containing model weights or checkpoints. """ @@ -29049,6 +29080,7 @@ class TrainingJob(Base): checkpoint_config: Optional[shapes.CheckpointConfig] = Unassigned() training_time_in_seconds: Optional[int] = Unassigned() billable_time_in_seconds: Optional[int] = Unassigned() + billable_token_count: Optional[int] = Unassigned() debug_hook_config: Optional[shapes.DebugHookConfig] = Unassigned() experiment_config: Optional[shapes.ExperimentConfig] = Unassigned() debug_rule_configurations: Optional[List[shapes.DebugRuleConfiguration]] = Unassigned() @@ -29064,6 +29096,12 @@ class TrainingJob(Base): retry_strategy: Optional[shapes.RetryStrategy] = Unassigned() remote_debug_config: Optional[shapes.RemoteDebugConfig] = Unassigned() infra_check_config: Optional[shapes.InfraCheckConfig] = Unassigned() + serverless_job_config: Optional[shapes.ServerlessJobConfig] = Unassigned() + mlflow_config: Optional[shapes.MlflowConfig] = Unassigned() + model_package_config: Optional[shapes.ModelPackageConfig] = Unassigned() + mlflow_details: Optional[shapes.MlflowDetails] = Unassigned() + progress_info: Optional[shapes.TrainingProgressInfo] = Unassigned() + output_model_package_arn: Optional[str] = Unassigned() def get_name(self) -> str: attributes = vars(self) @@ -29142,6 +29180,9 @@ def create( remote_debug_config: Optional[shapes.RemoteDebugConfig] = Unassigned(), infra_check_config: Optional[shapes.InfraCheckConfig] = Unassigned(), session_chaining_config: Optional[shapes.SessionChainingConfig] = Unassigned(), + serverless_job_config: Optional[shapes.ServerlessJobConfig] = Unassigned(), + mlflow_config: Optional[shapes.MlflowConfig] = Unassigned(), + model_package_config: Optional[shapes.ModelPackageConfig] = Unassigned(), session: Optional[Session] = None, region: Optional[str] = None, ) -> Optional["TrainingJob"]: @@ -29174,6 +29215,9 @@ def create( remote_debug_config: Configuration for remote debugging. To learn more about the remote debugging functionality of SageMaker, see Access a training container through Amazon Web Services Systems Manager (SSM) for remote debugging. infra_check_config: Contains information about the infrastructure health check configuration for the training job. session_chaining_config: Contains information about attribute-based access control (ABAC) for the training job. + serverless_job_config: The configuration for serverless training jobs. + mlflow_config: The MLflow configuration using SageMaker managed MLflow. + model_package_config: The configuration for the model package. session: Boto3 session. region: Region name. @@ -29229,6 +29273,9 @@ def create( "RemoteDebugConfig": remote_debug_config, "InfraCheckConfig": infra_check_config, "SessionChainingConfig": session_chaining_config, + "ServerlessJobConfig": serverless_job_config, + "MlflowConfig": mlflow_config, + "ModelPackageConfig": model_package_config, } operation_input_args = Base.populate_chained_attributes( diff --git a/src/sagemaker_core/main/shapes.py b/src/sagemaker_core/main/shapes.py index 6dcc41d..fc3f9f4 100644 --- a/src/sagemaker_core/main/shapes.py +++ b/src/sagemaker_core/main/shapes.py @@ -650,6 +650,23 @@ class AdditionalS3DataSource(Base): e_tag: Optional[str] = Unassigned() +class BaseModel(Base): + """ + BaseModel + Identifies the foundation model that was used as the starting point for model customization. + + Attributes + ---------------------- + hub_content_name: The hub content name of the base model. + hub_content_version: The hub content version of the base model. + recipe_name: The recipe name of the base model. + """ + + hub_content_name: Optional[Union[str, object]] = Unassigned() + hub_content_version: Optional[str] = Unassigned() + recipe_name: Optional[str] = Unassigned() + + class ModelPackageContainerDefinition(Base): """ ModelPackageContainerDefinition @@ -670,6 +687,8 @@ class ModelPackageContainerDefinition(Base): nearest_model_name: The name of a pre-trained machine learning benchmarked by Amazon SageMaker Inference Recommender model that matches your model. You can find a list of benchmarked models by calling ListModelMetadata. additional_s3_data_source: The additional data source that is used during inference in the Docker container for your model package. model_data_e_tag: The ETag associated with Model Data URL. + is_checkpoint: The checkpoint of the model package. + base_model: The base model of the package. """ container_hostname: Optional[str] = Unassigned() @@ -685,6 +704,8 @@ class ModelPackageContainerDefinition(Base): nearest_model_name: Optional[str] = Unassigned() additional_s3_data_source: Optional[AdditionalS3DataSource] = Unassigned() model_data_e_tag: Optional[str] = Unassigned() + is_checkpoint: Optional[bool] = Unassigned() + base_model: Optional[BaseModel] = Unassigned() class AdditionalInferenceSpecificationDefinition(Base): @@ -948,6 +969,19 @@ class FileSystemDataSource(Base): directory_path: str +class DatasetSource(Base): + """ + DatasetSource + Specifies a dataset source for a channel. + + Attributes + ---------------------- + dataset_arn: The Amazon Resource Name (ARN) of the dataset resource. + """ + + dataset_arn: str + + class DataSource(Base): """ DataSource @@ -957,10 +991,12 @@ class DataSource(Base): ---------------------- s3_data_source: The S3 location of the data source that is associated with a channel. file_system_data_source: The file system that is associated with a channel. + dataset_source: The dataset resource that's associated with a channel. """ s3_data_source: Optional[S3DataSource] = Unassigned() file_system_data_source: Optional[FileSystemDataSource] = Unassigned() + dataset_source: Optional[DatasetSource] = Unassigned() class ShuffleConfig(Base): @@ -1074,7 +1110,7 @@ class ResourceConfig(Base): ---------------------- instance_type: The ML compute instance type. instance_count: The number of ML compute instances to use. For distributed training, provide a value greater than 1. - volume_size_in_gb: The size of the ML storage volume that you want to provision. ML storage volumes store model artifacts and incremental states. Training algorithms might also use the ML storage volume for scratch space. If you want to store the training data in the ML storage volume, choose File as the TrainingInputMode in the algorithm specification. When using an ML instance with NVMe SSD volumes, SageMaker doesn't provision Amazon EBS General Purpose SSD (gp2) storage. Available storage is fixed to the NVMe-type instance's storage capacity. SageMaker configures storage paths for training datasets, checkpoints, model artifacts, and outputs to use the entire capacity of the instance storage. For example, ML instance families with the NVMe-type instance storage include ml.p4d, ml.g4dn, and ml.g5. When using an ML instance with the EBS-only storage option and without instance storage, you must define the size of EBS volume through VolumeSizeInGB in the ResourceConfig API. For example, ML instance families that use EBS volumes include ml.c5 and ml.p2. To look up instance types and their instance storage types and volumes, see Amazon EC2 Instance Types. To find the default local paths defined by the SageMaker training platform, see Amazon SageMaker Training Storage Folders for Training Datasets, Checkpoints, Model Artifacts, and Outputs. + volume_size_in_gb: The size of the ML storage volume that you want to provision. SageMaker automatically selects the volume size for serverless training jobs. You cannot customize this setting. ML storage volumes store model artifacts and incremental states. Training algorithms might also use the ML storage volume for scratch space. If you want to store the training data in the ML storage volume, choose File as the TrainingInputMode in the algorithm specification. When using an ML instance with NVMe SSD volumes, SageMaker doesn't provision Amazon EBS General Purpose SSD (gp2) storage. Available storage is fixed to the NVMe-type instance's storage capacity. SageMaker configures storage paths for training datasets, checkpoints, model artifacts, and outputs to use the entire capacity of the instance storage. For example, ML instance families with the NVMe-type instance storage include ml.p4d, ml.g4dn, and ml.g5. When using an ML instance with the EBS-only storage option and without instance storage, you must define the size of EBS volume through VolumeSizeInGB in the ResourceConfig API. For example, ML instance families that use EBS volumes include ml.c5 and ml.p2. To look up instance types and their instance storage types and volumes, see Amazon EC2 Instance Types. To find the default local paths defined by the SageMaker training platform, see Amazon SageMaker Training Storage Folders for Training Datasets, Checkpoints, Model Artifacts, and Outputs. volume_kms_key_id: The Amazon Web Services KMS key that SageMaker uses to encrypt data on the storage volume attached to the ML compute instance(s) that run the training job. Certain Nitro-based instances include local storage, dependent on the instance type. Local storage volumes are encrypted using a hardware module on the instance. You can't request a VolumeKmsKeyId when using an instance type with local storage. For a list of instance types that support local instance storage, see Instance Store Volumes. For more information about local instance storage encryption, see SSD Instance Store Volumes. The VolumeKmsKeyId can be in any of the following formats: // KMS Key ID "1234abcd-12ab-34cd-56ef-1234567890ab" // Amazon Resource Name (ARN) of a KMS Key "arn:aws:kms:us-west-2:111122223333:key/1234abcd-12ab-34cd-56ef-1234567890ab" keep_alive_period_in_seconds: The duration of time in seconds to retain configured resources in a warm pool for subsequent training jobs. instance_groups: The configuration of a heterogeneous cluster in JSON format. @@ -1571,6 +1607,21 @@ class ArtifactSummary(Base): last_modified_time: Optional[datetime.datetime] = Unassigned() +class AssociationInfo(Base): + """ + AssociationInfo + The data type used to describe the relationship between different sources. + + Attributes + ---------------------- + source_arn: The Amazon Resource Name (ARN) of the AssociationInfo source. + destination_arn: The Amazon Resource Name (ARN) of the AssociationInfo destination. + """ + + source_arn: str + destination_arn: str + + class IamIdentity(Base): """ IamIdentity @@ -2646,6 +2697,7 @@ class BatchDescribeModelPackageSummary(Base): inference_specification model_package_status: The status of the mortgage package. model_approval_status: The approval status of the model. + model_package_registration_type: The package registration type of the model package summary. """ model_package_group_name: Union[str, object] @@ -2656,6 +2708,7 @@ class BatchDescribeModelPackageSummary(Base): model_package_version: Optional[int] = Unassigned() model_package_description: Optional[str] = Unassigned() model_approval_status: Optional[str] = Unassigned() + model_package_registration_type: Optional[str] = Unassigned() class BatchDescribeModelPackageOutput(Base): @@ -2830,6 +2883,58 @@ class BatchTransformInput(Base): exclude_features_attribute: Optional[str] = Unassigned() +class BedrockCustomModelDeploymentMetadata(Base): + """ + BedrockCustomModelDeploymentMetadata + The metadata of the Amazon Bedrock custom model deployment. + + Attributes + ---------------------- + arn: The Amazon Resource Name (ARN) of the metadata for the Amazon Bedrock custom model deployment. + """ + + arn: Optional[str] = Unassigned() + + +class BedrockCustomModelMetadata(Base): + """ + BedrockCustomModelMetadata + The metadata of the Amazon Bedrock custom model. + + Attributes + ---------------------- + arn: The Amazon Resource Name (ARN) of the Amazon Bedrock custom model metadata. + """ + + arn: Optional[str] = Unassigned() + + +class BedrockModelImportMetadata(Base): + """ + BedrockModelImportMetadata + The metadata of the Amazon Bedrock model import. + + Attributes + ---------------------- + arn: The Amazon Resource Name (ARN) of the Amazon Bedrock model import metadata. + """ + + arn: Optional[str] = Unassigned() + + +class BedrockProvisionedModelThroughputMetadata(Base): + """ + BedrockProvisionedModelThroughputMetadata + The metadata of the Amazon Bedrock provisioned model throughput. + + Attributes + ---------------------- + arn: The Amazon Resource Name (ARN) of the Amazon Bedrock provisioned model throughput metadata. + """ + + arn: Optional[str] = Unassigned() + + class BestObjectiveNotImproving(Base): """ BestObjectiveNotImproving @@ -4149,7 +4254,7 @@ class ClusterOrchestrator(Base): eks: The Amazon EKS cluster used as the orchestrator for the SageMaker HyperPod cluster. """ - eks: Optional[ClusterOrchestratorEksConfig] = Unassigned() + eks: ClusterOrchestratorEksConfig class FSxLustreConfig(Base): @@ -8727,6 +8832,63 @@ class SessionChainingConfig(Base): enable_session_tag_chaining: Optional[bool] = Unassigned() +class ServerlessJobConfig(Base): + """ + ServerlessJobConfig + The configuration for the serverless training job. + + Attributes + ---------------------- + base_model_arn: The base model Amazon Resource Name (ARN) in SageMaker Public Hub. SageMaker always selects the latest version of the provided model. + accept_eula: Specifies agreement to the model end-user license agreement (EULA). The AcceptEula value must be explicitly defined as True in order to accept the EULA that this model requires. You are responsible for reviewing and complying with any applicable license terms and making sure they are acceptable for your use case before downloading or using a model. For more information, see End-user license agreements section for more details on accepting the EULA. + job_type: The serverless training job type. + customization_technique: The model customization technique. + peft: The parameter-efficient fine-tuning configuration. + evaluation_type: The evaluation job type. Required when serverless job type is Evaluation. + evaluator_arn: The evaluator Amazon Resource Name (ARN) used as reward function or reward prompt. + """ + + base_model_arn: str + job_type: str + accept_eula: Optional[bool] = Unassigned() + customization_technique: Optional[str] = Unassigned() + peft: Optional[str] = Unassigned() + evaluation_type: Optional[str] = Unassigned() + evaluator_arn: Optional[str] = Unassigned() + + +class MlflowConfig(Base): + """ + MlflowConfig + The MLflow configuration using SageMaker managed MLflow. + + Attributes + ---------------------- + mlflow_resource_arn: The Amazon Resource Name (ARN) of the MLflow resource. + mlflow_experiment_name: The MLflow experiment name used for this job. + mlflow_run_name: The MLflow run name used for this job. + """ + + mlflow_resource_arn: str + mlflow_experiment_name: Optional[str] = Unassigned() + mlflow_run_name: Optional[str] = Unassigned() + + +class ModelPackageConfig(Base): + """ + ModelPackageConfig + The configuration for the Model package. + + Attributes + ---------------------- + model_package_group_arn: The Amazon Resource Name (ARN) of the model package group of output model package. + source_model_package_arn: The Amazon Resource Name (ARN) of the source model package used for continued fine-tuning and custom model evaluation. + """ + + model_package_group_arn: str + source_model_package_arn: Optional[str] = Unassigned() + + class ModelClientConfig(Base): """ ModelClientConfig @@ -10085,6 +10247,21 @@ class SelectiveExecutionConfig(Base): source_pipeline_execution_arn: Optional[str] = Unassigned() +class MLflowConfiguration(Base): + """ + MLflowConfiguration + The MLflow configuration. + + Attributes + ---------------------- + mlflow_resource_arn: The Amazon Resource Name (ARN) of MLflow configuration resource. + mlflow_experiment_name: The name of the MLflow configuration. + """ + + mlflow_resource_arn: Optional[str] = Unassigned() + mlflow_experiment_name: Optional[str] = Unassigned() + + class ServiceCatalogProvisionedProductDetails(Base): """ ServiceCatalogProvisionedProductDetails @@ -10229,6 +10406,40 @@ class ProfilerRuleEvaluationStatus(Base): last_modified_time: Optional[datetime.datetime] = Unassigned() +class MlflowDetails(Base): + """ + MlflowDetails + The MLflow details of this job. + + Attributes + ---------------------- + mlflow_experiment_id: The MLflow experiment ID used for this job. + mlflow_run_id: The MLflow run ID used for this job. + """ + + mlflow_experiment_id: Optional[str] = Unassigned() + mlflow_run_id: Optional[str] = Unassigned() + + +class TrainingProgressInfo(Base): + """ + TrainingProgressInfo + The serverless training job progress information. + + Attributes + ---------------------- + total_step_count_per_epoch: The total step count per epoch. + current_step: The current step number. + current_epoch: The current epoch number. + max_epoch: The maximum number of epochs for this job. + """ + + total_step_count_per_epoch: Optional[int] = Unassigned() + current_step: Optional[int] = Unassigned() + current_epoch: Optional[int] = Unassigned() + max_epoch: Optional[int] = Unassigned() + + class ReservedCapacitySummary(Base): """ ReservedCapacitySummary @@ -11527,6 +11738,19 @@ class ImageVersion(Base): failure_reason: Optional[str] = Unassigned() +class InferenceComponentMetadata(Base): + """ + InferenceComponentMetadata + The metadata of the inference component. + + Attributes + ---------------------- + arn: The Amazon Resource Name (ARN) of the inference component metadata. + """ + + arn: Optional[str] = Unassigned() + + class InferenceComponentSummary(Base): """ InferenceComponentSummary @@ -11777,6 +12001,25 @@ class LineageGroupSummary(Base): last_modified_time: Optional[datetime.datetime] = Unassigned() +class LineageMetadata(Base): + """ + LineageMetadata + The metadata that tracks relationships between ML artifacts, actions, and contexts. + + Attributes + ---------------------- + action_arns: The Amazon Resource Name (ARN) of the lineage metadata action. + artifact_arns: The Amazon Resource Name (ARN) of the lineage metadata artifact. + context_arns: The Amazon Resource Name (ARN) of the lineage metadata context. + associations: The lineage metadata associations. + """ + + action_arns: Optional[Dict[str, str]] = Unassigned() + artifact_arns: Optional[Dict[str, str]] = Unassigned() + context_arns: Optional[Dict[str, str]] = Unassigned() + associations: Optional[List[AssociationInfo]] = Unassigned() + + class MonitoringJobDefinitionSummary(Base): """ MonitoringJobDefinitionSummary @@ -11999,6 +12242,7 @@ class ModelPackageSummary(Base): model_package_status: The overall status of the model package. model_approval_status: The approval status of the model. This can be one of the following values. APPROVED - The model is approved REJECTED - The model is rejected. PENDING_MANUAL_APPROVAL - The model is waiting for manual approval. model_life_cycle + model_package_registration_type: The package registration type of the model package summary. """ model_package_arn: str @@ -12010,6 +12254,7 @@ class ModelPackageSummary(Base): model_package_description: Optional[str] = Unassigned() model_approval_status: Optional[str] = Unassigned() model_life_cycle: Optional[ModelLifeCycle] = Unassigned() + model_package_registration_type: Optional[str] = Unassigned() class ModelSummary(Base): @@ -12360,6 +12605,12 @@ class PipelineExecutionStepMetadata(Base): auto_ml_job: The Amazon Resource Name (ARN) of the AutoML job that was run by this step. endpoint: The endpoint that was invoked during this step execution. endpoint_config: The endpoint configuration used to create an endpoint during this step execution. + bedrock_custom_model: The metadata of the Amazon Bedrock custom model used in the pipeline execution step. + bedrock_custom_model_deployment: The metadata of the Amazon Bedrock custom model deployment used in pipeline execution step. + bedrock_provisioned_model_throughput: The metadata of the Amazon Bedrock provisioned model throughput used in the pipeline execution step. + bedrock_model_import: The metadata of Amazon Bedrock model import used in pipeline execution step. + inference_component: The metadata of the inference component used in pipeline execution step. + lineage: The metadata of the lineage used in pipeline execution step. """ training_job: Optional[TrainingJobStepMetadata] = Unassigned() @@ -12378,6 +12629,14 @@ class PipelineExecutionStepMetadata(Base): auto_ml_job: Optional[AutoMLJobStepMetadata] = Unassigned() endpoint: Optional[EndpointStepMetadata] = Unassigned() endpoint_config: Optional[EndpointConfigStepMetadata] = Unassigned() + bedrock_custom_model: Optional[BedrockCustomModelMetadata] = Unassigned() + bedrock_custom_model_deployment: Optional[BedrockCustomModelDeploymentMetadata] = Unassigned() + bedrock_provisioned_model_throughput: Optional[BedrockProvisionedModelThroughputMetadata] = ( + Unassigned() + ) + bedrock_model_import: Optional[BedrockModelImportMetadata] = Unassigned() + inference_component: Optional[InferenceComponentMetadata] = Unassigned() + lineage: Optional[LineageMetadata] = Unassigned() class SelectiveExecutionResult(Base): @@ -13147,6 +13406,7 @@ class ModelPackage(Base): model_package_name: The name of the model package. The name can be as follows: For a versioned model, the name is automatically generated by SageMaker Model Registry and follows the format 'ModelPackageGroupName/ModelPackageVersion'. For an unversioned model, you must provide the name. model_package_group_name: The model group to which the model belongs. model_package_version: The version number of a versioned model. + model_package_registration_type: The package registration type of the model package. model_package_arn: The Amazon Resource Name (ARN) of the model package. model_package_description: The description of the model package. creation_time: The time that the model package was created. @@ -13180,6 +13440,7 @@ class ModelPackage(Base): model_package_name: Optional[Union[str, object]] = Unassigned() model_package_group_name: Optional[Union[str, object]] = Unassigned() model_package_version: Optional[int] = Unassigned() + model_package_registration_type: Optional[str] = Unassigned() model_package_arn: Optional[str] = Unassigned() model_package_description: Optional[str] = Unassigned() creation_time: Optional[datetime.datetime] = Unassigned() @@ -13719,6 +13980,8 @@ class TrainingJob(Base): debug_rule_configurations: Information about the debug rule configuration. tensor_board_output_config debug_rule_evaluation_statuses: Information about the evaluation status of the rules for the training job. + output_model_package_arn: The output model package Amazon Resource Name (ARN) that contains model weights or checkpoint. + model_package_config: The model package configuration. profiler_config environment: The environment variables to set in the Docker container. retry_strategy: The number of times to retry the job when the job fails due to an InternalServerError. @@ -13759,6 +14022,8 @@ class TrainingJob(Base): debug_rule_configurations: Optional[List[DebugRuleConfiguration]] = Unassigned() tensor_board_output_config: Optional[TensorBoardOutputConfig] = Unassigned() debug_rule_evaluation_statuses: Optional[List[DebugRuleEvaluationStatus]] = Unassigned() + output_model_package_arn: Optional[str] = Unassigned() + model_package_config: Optional[ModelPackageConfig] = Unassigned() profiler_config: Optional[ProfilerConfig] = Unassigned() environment: Optional[Dict[str, str]] = Unassigned() retry_strategy: Optional[RetryStrategy] = Unassigned()