* CreateAutoMLJobV2 can manage tabular problem types identical to those of
* its previous version CreateAutoMLJob, as well as time-series forecasting,
- * and non-tabular problem types such as image or text classification.
Find guidelines about how to migrate a CreateAutoMLJob to
* CreateAutoMLJobV2 in Migrate a CreateAutoMLJob to CreateAutoMLJobV2.
* CreateAutoMLJobV2 can manage tabular problem types identical to those of
* its previous version CreateAutoMLJob, as well as time-series forecasting,
- * and non-tabular problem types such as image or text classification.
Find guidelines about how to migrate a CreateAutoMLJob to
* CreateAutoMLJobV2 in Migrate a CreateAutoMLJob to CreateAutoMLJobV2.
The type of compression used for an additional data source used in inference or
- * training. Specify None if your additional data source is not compressed.
None if your additional data source is not
+ * compressed.
*/
CompressionType?: CompressionType;
}
@@ -4768,12 +4769,19 @@ export const AutoMLMetricExtendedEnum = {
MAPE: "MAPE",
MASE: "MASE",
MSE: "MSE",
+ PERPLEXITY: "Perplexity",
PRECISION: "Precision",
PRECISION_MACRO: "PrecisionMacro",
R2: "R2",
RECALL: "Recall",
RECALL_MACRO: "RecallMacro",
RMSE: "RMSE",
+ ROUGE1: "Rouge1",
+ ROUGE2: "Rouge2",
+ ROUGEL: "RougeL",
+ ROUGEL_SUM: "RougeLSum",
+ TRAINING_LOSS: "TrainingLoss",
+ VALIDATION_LOSS: "ValidationLoss",
WAPE: "WAPE",
} as const;
@@ -5398,6 +5406,11 @@ export interface AutoMLJobChannel {
* x-application/vnd.amazon+parquet. The default value is
* text/csv;header=present.
*
+ * For text generation (LLMs fine-tuning): text/csv;header=present or
+ * x-application/vnd.amazon+parquet. The default value is
+ * text/csv;header=present.
The maximum number of times a training job is allowed to run.
- *For text and image classification, as well as time-series forecasting problem types, the - * supported value is 1. For tabular problem types, the maximum value is 750.
+ *For text and image classification, time-series forecasting, as well as text generation + * (LLMs fine-tuning) problem types, the supported value is 1. For tabular problem types, the + * maximum value is 750.
*/ MaxCandidates?: number; @@ -5572,7 +5586,7 @@ export interface AutoMLJobConfig { /** * @public - *Specifies a metric to minimize or maximize as the objective of a job.
+ *Specifies a metric to minimize or maximize as the objective of an AutoML job.
*/ export interface AutoMLJobObjective { /** @@ -5581,33 +5595,94 @@ export interface AutoMLJobObjective { * learning system. During training, the model's parameters are updated iteratively to * optimize its performance based on the feedback provided by the objective metric when * evaluating the model on the validation dataset. - *For the list of all available metrics supported by Autopilot, see Autopilot - * metrics.
- *If you do not specify a metric explicitly, the default behavior is to automatically - * use:
+ *The list of available metrics supported by Autopilot and the default metric applied when you + * do not specify a metric name explicitly depend on the problem type.
*For tabular problem types:
*Regression: MSE.
List of available metrics:
+ * Regression: InferenceLatency, MAE,
+ * MSE, R2, RMSE
+ *
Binary classification: Accuracy, AUC,
+ * BalancedAccuracy, F1,
+ * InferenceLatency, LogLoss,
+ * Precision, Recall
+ *
Multiclass classification: Accuracy,
+ * BalancedAccuracy, F1macro,
+ * InferenceLatency, LogLoss,
+ * PrecisionMacro, RecallMacro
+ *
For a description of each metric, see Autopilot metrics for classification and regression.
*Binary classification: F1.
Default objective metrics:
+ *Regression: MSE.
Binary classification: F1.
Multiclass classification: Accuracy.
For image or text classification problem types:
+ *Multiclass classification: Accuracy.
List of available metrics: Accuracy
+ *
For a description of each metric, see Autopilot metrics for text and image classification.
+ *Default objective metrics: Accuracy
+ *
For image or text classification problem types: Accuracy
- *
For time-series forecasting problem types:
+ *List of available metrics: RMSE, wQL,
+ * Average wQL, MASE, MAPE,
+ * WAPE
+ *
For a description of each metric, see Autopilot metrics for + * time-series forecasting.
+ *Default objective metrics: AverageWeightedQuantileLoss
+ *
For time-series forecasting problem types:
- * AverageWeightedQuantileLoss
- *
For text generation problem types (LLMs fine-tuning):
+ * Fine-tuning language models in Autopilot does not
+ * require setting the AutoMLJobObjective field. Autopilot fine-tunes LLMs
+ * without requiring multiple candidates to be trained and evaluated.
+ * Instead, using your dataset, Autopilot directly fine-tunes your target model to enhance a
+ * default objective metric, the cross-entropy loss. After fine-tuning a language model,
+ * you can evaluate the quality of its generated text using different metrics.
+ * For a list of the available metrics, see Metrics for
+ * fine-tuning LLMs in Autopilot.
Stores the configuration information for the image classification problem of an AutoML - * job V2.
+ *The collection of settings used by an AutoML job V2 for the image classification problem + * type.
*/ export interface ImageClassificationJobConfig { /** @@ -5835,8 +5910,7 @@ export type ProblemType = (typeof ProblemType)[keyof typeof ProblemType]; /** * @public - *The collection of settings used by an AutoML job V2 for the TABULAR problem
- * type.
The collection of settings used by an AutoML job V2 for the tabular problem type.
*/ export interface TabularJobConfig { /** @@ -5948,8 +6022,8 @@ export interface TabularJobConfig { /** * @public - *Stores the configuration information for the text classification problem of an AutoML job - * V2.
+ *The collection of settings used by an AutoML job V2 for the text classification problem + * type.
*/ export interface TextClassificationJobConfig { /** @@ -5974,6 +6048,34 @@ export interface TextClassificationJobConfig { TargetLabelColumn: string | undefined; } +/** + * @public + *The collection of settings used by an AutoML job V2 for the text generation problem + * type.
+ *The text generation models that support fine-tuning in Autopilot are currently accessible + * exclusively in regions supported by Canvas. Refer to the documentation of Canvas for the full list of its supported + * Regions.
+ *How long a job is allowed to run, or how many candidates a job is allowed to + * generate.
+ */ + CompletionCriteria?: AutoMLJobCompletionCriteria; + + /** + * @public + *The name of the base model to fine-tune. Autopilot supports fine-tuning a variety of large
+ * language models. For information on the list of supported models, see Text generation models supporting fine-tuning in Autopilot. If no
+ * BaseModelName is provided, the default model used is Falcon-7B-Instruct.
+ *
Stores the holiday featurization attributes applicable to each item of time-series @@ -6223,6 +6325,7 @@ export type AutoMLProblemTypeConfig = | AutoMLProblemTypeConfig.ImageClassificationJobConfigMember | AutoMLProblemTypeConfig.TabularJobConfigMember | AutoMLProblemTypeConfig.TextClassificationJobConfigMember + | AutoMLProblemTypeConfig.TextGenerationJobConfigMember | AutoMLProblemTypeConfig.TimeSeriesForecastingJobConfigMember | AutoMLProblemTypeConfig.$UnknownMember; @@ -6240,6 +6343,7 @@ export namespace AutoMLProblemTypeConfig { TextClassificationJobConfig?: never; TabularJobConfig?: never; TimeSeriesForecastingJobConfig?: never; + TextGenerationJobConfig?: never; $unknown?: never; } @@ -6253,12 +6357,13 @@ export namespace AutoMLProblemTypeConfig { TextClassificationJobConfig: TextClassificationJobConfig; TabularJobConfig?: never; TimeSeriesForecastingJobConfig?: never; + TextGenerationJobConfig?: never; $unknown?: never; } /** * @public - *
Settings used to configure an AutoML job V2 for a tabular problem type (regression, + *
Settings used to configure an AutoML job V2 for the tabular problem type (regression, * classification).
*/ export interface TabularJobConfigMember { @@ -6266,12 +6371,13 @@ export namespace AutoMLProblemTypeConfig { TextClassificationJobConfig?: never; TabularJobConfig: TabularJobConfig; TimeSeriesForecastingJobConfig?: never; + TextGenerationJobConfig?: never; $unknown?: never; } /** * @public - *Settings used to configure an AutoML job V2 for a time-series forecasting problem + *
Settings used to configure an AutoML job V2 for the time-series forecasting problem * type.
*/ export interface TimeSeriesForecastingJobConfigMember { @@ -6279,6 +6385,26 @@ export namespace AutoMLProblemTypeConfig { TextClassificationJobConfig?: never; TabularJobConfig?: never; TimeSeriesForecastingJobConfig: TimeSeriesForecastingJobConfig; + TextGenerationJobConfig?: never; + $unknown?: never; + } + + /** + * @public + *Settings used to configure an AutoML job V2 for the text generation (LLMs fine-tuning) + * problem type.
+ *The text generation models that support fine-tuning in Autopilot are currently accessible + * exclusively in regions supported by Canvas. Refer to the documentation of Canvas for the full list of its supported + * Regions.
+ *The resolved attributes specific to the TABULAR problem type.
The resolved attributes specific to the tabular problem type.
*/ export interface TabularResolvedAttributes { /** @@ -6347,10 +6478,23 @@ export interface TabularResolvedAttributes { /** * @public - *The resolved attributes specific to the problem type of an AutoML job V2.
+ *The resolved attributes specific to the text generation problem type.
+ */ +export interface TextGenerationResolvedAttributes { + /** + * @public + *The name of the base model to fine-tune.
+ */ + BaseModelName?: string; +} + +/** + * @public + *Stores resolved attributes specific to the problem type of an AutoML job V2.
*/ export type AutoMLProblemTypeResolvedAttributes = | AutoMLProblemTypeResolvedAttributes.TabularResolvedAttributesMember + | AutoMLProblemTypeResolvedAttributes.TextGenerationResolvedAttributesMember | AutoMLProblemTypeResolvedAttributes.$UnknownMember; /** @@ -6359,10 +6503,21 @@ export type AutoMLProblemTypeResolvedAttributes = export namespace AutoMLProblemTypeResolvedAttributes { /** * @public - *Defines the resolved attributes for the TABULAR problem type.
The resolved attributes for the tabular problem type.
*/ export interface TabularResolvedAttributesMember { TabularResolvedAttributes: TabularResolvedAttributes; + TextGenerationResolvedAttributes?: never; + $unknown?: never; + } + + /** + * @public + *The resolved attributes for the text generation problem type.
+ */ + export interface TextGenerationResolvedAttributesMember { + TabularResolvedAttributes?: never; + TextGenerationResolvedAttributes: TextGenerationResolvedAttributes; $unknown?: never; } @@ -6371,17 +6526,21 @@ export namespace AutoMLProblemTypeResolvedAttributes { */ export interface $UnknownMember { TabularResolvedAttributes?: never; + TextGenerationResolvedAttributes?: never; $unknown: [string, any]; } export interface VisitorSpecifies a metric to minimize or maximize as the objective of a job.
+ *Specifies a metric to minimize or maximize as the objective of an AutoML job.
*/ AutoMLJobObjective?: AutoMLJobObjective; @@ -9361,19 +9520,16 @@ export type HyperParameterTuningJobObjectiveType = /** * @public - *Defines the objective metric for a hyperparameter tuning job.
- * Hyperparameter
- * tuning uses the value of this metric to evaluate the training jobs it launches, and
- * returns the training job that results in either the highest or lowest value for this
- * metric, depending on the value you specify for the Type
- * parameter.
Defines the objective metric for a hyperparameter tuning job. Hyperparameter tuning
+ * uses the value of this metric to evaluate the training jobs it launches, and returns the
+ * training job that results in either the highest or lowest value for this metric,
+ * depending on the value you specify for the Type parameter. If you want to
+ * define a custom objective metric, see Define metrics and environment variables.
Whether to - * minimize - * or maximize the objective metric.
+ *Whether to minimize or maximize the objective metric.
*/ Type: HyperParameterTuningJobObjectiveType | undefined; @@ -9922,6 +10078,9 @@ export interface CreateAutoMLJobV2Request { *For time-series forecasting: S3Prefix.
For text generation (LLMs fine-tuning): S3Prefix.
For tabular problem types, you must either provide both the
- * AutoMLJobObjective and indicate the type of supervised learning problem
- * in AutoMLProblemTypeConfig (TabularJobConfig.ProblemType), or
- * none at all.
For tabular problem types: You must either provide both the
+ * AutoMLJobObjective and indicate the type of supervised learning
+ * problem in AutoMLProblemTypeConfig
+ * (TabularJobConfig.ProblemType), or none at all.
For text generation problem types (LLMs fine-tuning):
+ * Fine-tuning language models in Autopilot does not
+ * require setting the AutoMLJobObjective field. Autopilot fine-tunes LLMs
+ * without requiring multiple candidates to be trained and evaluated.
+ * Instead, using your dataset, Autopilot directly fine-tunes your target model to enhance a
+ * default objective metric, the cross-entropy loss. After fine-tuning a language model,
+ * you can evaluate the quality of its generated text using different metrics.
+ * For a list of the available metrics, see Metrics for
+ * fine-tuning LLMs in Autopilot.
A time limit for how long the monitoring job is allowed to run before stopping.
- */ -export interface MonitoringStoppingCondition { - /** - * @public - *The maximum runtime allowed in seconds.
- *The MaxRuntimeInSeconds cannot exceed the frequency of the job. For data
- * quality and model explainability, this can be up to 3600 seconds for an hourly schedule.
- * For model bias and model quality hourly schedules, this can be up to 1800
- * seconds.
The name for the monitoring job definition.
- */ - JobDefinitionName: string | undefined; - - /** - * @public - *Configures the constraints and baselines for the monitoring job.
- */ - DataQualityBaselineConfig?: DataQualityBaselineConfig; - - /** - * @public - *Specifies the container that runs the monitoring job.
- */ - DataQualityAppSpecification: DataQualityAppSpecification | undefined; - - /** - * @public - *A list of inputs for the monitoring job. Currently endpoints are supported as monitoring - * inputs.
- */ - DataQualityJobInput: DataQualityJobInput | undefined; - - /** - * @public - *The output configuration for monitoring jobs.
- */ - DataQualityJobOutputConfig: MonitoringOutputConfig | undefined; - - /** - * @public - *Identifies the resources to deploy for a monitoring job.
- */ - JobResources: MonitoringResources | undefined; - - /** - * @public - *Specifies networking configuration for the monitoring job.
- */ - NetworkConfig?: MonitoringNetworkConfig; - - /** - * @public - *The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker can - * assume to perform tasks on your behalf.
- */ - RoleArn: string | undefined; - - /** - * @public - *A time limit for how long the monitoring job is allowed to run before stopping.
- */ - StoppingCondition?: MonitoringStoppingCondition; - - /** - * @public - *(Optional) An array of key-value pairs. For more information, see - * - * Using Cost Allocation Tags in the Amazon Web Services Billing and Cost Management User Guide.
- */ - Tags?: Tag[]; -} diff --git a/clients/client-sagemaker/src/models/models_1.ts b/clients/client-sagemaker/src/models/models_1.ts index 638b8475d4828..3bd2bedb200dd 100644 --- a/clients/client-sagemaker/src/models/models_1.ts +++ b/clients/client-sagemaker/src/models/models_1.ts @@ -43,6 +43,9 @@ import { ContentClassifier, ContinuousParameterRange, ConvergenceDetected, + DataQualityAppSpecification, + DataQualityBaselineConfig, + DataQualityJobInput, EndpointInput, HyperParameterScalingType, HyperParameterTuningJobObjective, @@ -56,7 +59,6 @@ import { MonitoringOutputConfig, MonitoringResources, MonitoringStatisticsResource, - MonitoringStoppingCondition, OutputDataConfig, ProcessingInstanceType, ProcessingS3DataDistributionType, @@ -76,6 +78,93 @@ import { VpcConfig, } from "./models_0"; +/** + * @public + *A time limit for how long the monitoring job is allowed to run before stopping.
+ */ +export interface MonitoringStoppingCondition { + /** + * @public + *The maximum runtime allowed in seconds.
+ *The MaxRuntimeInSeconds cannot exceed the frequency of the job. For data
+ * quality and model explainability, this can be up to 3600 seconds for an hourly schedule.
+ * For model bias and model quality hourly schedules, this can be up to 1800
+ * seconds.
The name for the monitoring job definition.
+ */ + JobDefinitionName: string | undefined; + + /** + * @public + *Configures the constraints and baselines for the monitoring job.
+ */ + DataQualityBaselineConfig?: DataQualityBaselineConfig; + + /** + * @public + *Specifies the container that runs the monitoring job.
+ */ + DataQualityAppSpecification: DataQualityAppSpecification | undefined; + + /** + * @public + *A list of inputs for the monitoring job. Currently endpoints are supported as monitoring + * inputs.
+ */ + DataQualityJobInput: DataQualityJobInput | undefined; + + /** + * @public + *The output configuration for monitoring jobs.
+ */ + DataQualityJobOutputConfig: MonitoringOutputConfig | undefined; + + /** + * @public + *Identifies the resources to deploy for a monitoring job.
+ */ + JobResources: MonitoringResources | undefined; + + /** + * @public + *Specifies networking configuration for the monitoring job.
+ */ + NetworkConfig?: MonitoringNetworkConfig; + + /** + * @public + *The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker can + * assume to perform tasks on your behalf.
+ */ + RoleArn: string | undefined; + + /** + * @public + *A time limit for how long the monitoring job is allowed to run before stopping.
+ */ + StoppingCondition?: MonitoringStoppingCondition; + + /** + * @public + *(Optional) An array of key-value pairs. For more information, see + * + * Using Cost Allocation Tags in the Amazon Web Services Billing and Cost Management User Guide.
+ */ + Tags?: Tag[]; +} + /** * @public */ @@ -3588,12 +3677,11 @@ export interface HyperParameterTrainingJobDefinition { /** * @public - *Defines the objective metric for a hyperparameter tuning job.
- * Hyperparameter
- * tuning uses the value of this metric to evaluate the training jobs it launches, and
- * returns the training job that results in either the highest or lowest value for this
- * metric, depending on the value you specify for the Type
- * parameter.
Defines the objective metric for a hyperparameter tuning job. Hyperparameter tuning
+ * uses the value of this metric to evaluate the training jobs it launches, and returns the
+ * training job that results in either the highest or lowest value for this metric,
+ * depending on the value you specify for the Type parameter. If you want to
+ * define a custom objective metric, see Define metrics and environment variables.
The name of the context to delete.
- */ - ContextName: string | undefined; -} - -/** - * @public - */ -export interface DeleteContextResponse { - /** - * @public - *The Amazon Resource Name (ARN) of the context.
- */ - ContextArn?: string; -} - /** * @internal */ diff --git a/clients/client-sagemaker/src/models/models_2.ts b/clients/client-sagemaker/src/models/models_2.ts index 70b99ee7ca84e..79e9447f1c40f 100644 --- a/clients/client-sagemaker/src/models/models_2.ts +++ b/clients/client-sagemaker/src/models/models_2.ts @@ -58,7 +58,6 @@ import { MonitoringNetworkConfig, MonitoringOutputConfig, MonitoringResources, - MonitoringStoppingCondition, NeoVpcConfig, ObjectiveStatus, OutputConfig, @@ -138,6 +137,7 @@ import { ModelQualityBaselineConfig, ModelQualityJobInput, MonitoringScheduleConfig, + MonitoringStoppingCondition, MonitoringType, NetworkConfig, NotebookInstanceAcceleratorType, @@ -177,6 +177,28 @@ import { VendorGuidance, } from "./models_1"; +/** + * @public + */ +export interface DeleteContextRequest { + /** + * @public + *The name of the context to delete.
+ */ + ContextName: string | undefined; +} + +/** + * @public + */ +export interface DeleteContextResponse { + /** + * @public + *The Amazon Resource Name (ARN) of the context.
+ */ + ContextArn?: string; +} + /** * @public */ @@ -1468,7 +1490,7 @@ export interface ModelDeployResult { export interface ResolvedAttributes { /** * @public - *Specifies a metric to minimize or maximize as the objective of a job.
+ *Specifies a metric to minimize or maximize as the objective of an AutoML job.
*/ AutoMLJobObjective?: AutoMLJobObjective; @@ -10747,102 +10769,6 @@ export interface MonitoringSchedule { Tags?: Tag[]; } -/** - * @public - *A hosted endpoint for real-time inference.
- */ -export interface Endpoint { - /** - * @public - *The name of the endpoint.
- */ - EndpointName: string | undefined; - - /** - * @public - *The Amazon Resource Name (ARN) of the endpoint.
- */ - EndpointArn: string | undefined; - - /** - * @public - *The endpoint configuration associated with the endpoint.
- */ - EndpointConfigName: string | undefined; - - /** - * @public - *A list of the production variants hosted on the endpoint. Each production variant is a - * model.
- */ - ProductionVariants?: ProductionVariantSummary[]; - - /** - * @public - *The currently active data capture configuration used by your Endpoint.
- */ - DataCaptureConfig?: DataCaptureConfigSummary; - - /** - * @public - *The status of the endpoint.
- */ - EndpointStatus: EndpointStatus | undefined; - - /** - * @public - *If the endpoint failed, the reason it failed.
- */ - FailureReason?: string; - - /** - * @public - *The time that the endpoint was created.
- */ - CreationTime: Date | undefined; - - /** - * @public - *The last time the endpoint was modified.
- */ - LastModifiedTime: Date | undefined; - - /** - * @public - *A list of monitoring schedules for the endpoint. For information about model - * monitoring, see Amazon SageMaker Model Monitor.
- */ - MonitoringSchedules?: MonitoringSchedule[]; - - /** - * @public - *A list of the tags associated with the endpoint. For more information, see Tagging Amazon Web Services resources in the Amazon Web Services General - * Reference Guide.
- */ - Tags?: Tag[]; - - /** - * @public - *A list of the shadow variants hosted on the endpoint. Each shadow variant is a model - * in shadow mode with production traffic replicated from the production variant.
- */ - ShadowProductionVariants?: ProductionVariantSummary[]; -} - -/** - * @public - * @enum - */ -export const EndpointConfigSortKey = { - CreationTime: "CreationTime", - Name: "Name", -} as const; - -/** - * @public - */ -export type EndpointConfigSortKey = (typeof EndpointConfigSortKey)[keyof typeof EndpointConfigSortKey]; - /** * @internal */ diff --git a/clients/client-sagemaker/src/models/models_3.ts b/clients/client-sagemaker/src/models/models_3.ts index 7b98708f2bc58..c34b37ecdf477 100644 --- a/clients/client-sagemaker/src/models/models_3.ts +++ b/clients/client-sagemaker/src/models/models_3.ts @@ -19,8 +19,6 @@ import { AutoMLJobSummary, AutoMLSortBy, AutoMLSortOrder, - BatchDataCaptureConfig, - BatchStrategy, CacheHitResult, CallbackStepMetadata, CandidateSortBy, @@ -38,17 +36,13 @@ import { ModelPackageStatus, OutputParameter, Tag, - TransformInput, - TransformOutput, - TransformResources, UserContext, VpcConfig, } from "./models_0"; import { _InstanceType, - DataProcessing, + DataCaptureConfigSummary, EdgeOutputConfig, - ExperimentConfig, FeatureDefinition, FeatureType, HyperParameterTrainingJobDefinition, @@ -61,7 +55,6 @@ import { LabelingJobInputConfig, ModelCardSecurityConfig, ModelCardStatus, - ModelClientConfig, MonitoringType, OfflineStoreConfig, OnlineStoreConfig, @@ -83,7 +76,6 @@ import { EdgePackagingJobStatus, EdgePackagingJobSummary, EMRStepMetadata, - EndpointConfigSortKey, EndpointOutputConfiguration, EndpointStatus, ExecutionStatus, @@ -110,12 +102,14 @@ import { ModelConfiguration, ModelPackageGroupStatus, MonitoringExecutionSummary, + MonitoringSchedule, NotebookInstanceStatus, ObjectiveStatusCounters, OfflineStoreStatus, OfflineStoreStatusValue, PipelineExecutionStatus, ProcessingJobStatus, + ProductionVariantSummary, ProjectStatus, RecommendationJobStatus, RecommendationMetrics, @@ -134,6 +128,102 @@ import { Workteam, } from "./models_2"; +/** + * @public + *A hosted endpoint for real-time inference.
+ */ +export interface Endpoint { + /** + * @public + *The name of the endpoint.
+ */ + EndpointName: string | undefined; + + /** + * @public + *The Amazon Resource Name (ARN) of the endpoint.
+ */ + EndpointArn: string | undefined; + + /** + * @public + *The endpoint configuration associated with the endpoint.
+ */ + EndpointConfigName: string | undefined; + + /** + * @public + *A list of the production variants hosted on the endpoint. Each production variant is a + * model.
+ */ + ProductionVariants?: ProductionVariantSummary[]; + + /** + * @public + *The currently active data capture configuration used by your Endpoint.
+ */ + DataCaptureConfig?: DataCaptureConfigSummary; + + /** + * @public + *The status of the endpoint.
+ */ + EndpointStatus: EndpointStatus | undefined; + + /** + * @public + *If the endpoint failed, the reason it failed.
+ */ + FailureReason?: string; + + /** + * @public + *The time that the endpoint was created.
+ */ + CreationTime: Date | undefined; + + /** + * @public + *The last time the endpoint was modified.
+ */ + LastModifiedTime: Date | undefined; + + /** + * @public + *A list of monitoring schedules for the endpoint. For information about model + * monitoring, see Amazon SageMaker Model Monitor.
+ */ + MonitoringSchedules?: MonitoringSchedule[]; + + /** + * @public + *A list of the tags associated with the endpoint. For more information, see Tagging Amazon Web Services resources in the Amazon Web Services General + * Reference Guide.
+ */ + Tags?: Tag[]; + + /** + * @public + *A list of the shadow variants hosted on the endpoint. Each shadow variant is a model + * in shadow mode with production traffic replicated from the production variant.
+ */ + ShadowProductionVariants?: ProductionVariantSummary[]; +} + +/** + * @public + * @enum + */ +export const EndpointConfigSortKey = { + CreationTime: "CreationTime", + Name: "Name", +} as const; + +/** + * @public + */ +export type EndpointConfigSortKey = (typeof EndpointConfigSortKey)[keyof typeof EndpointConfigSortKey]; + /** * @public *Provides summary information for an endpoint configuration.
@@ -10354,250 +10444,6 @@ export interface ModelCard { ModelPackageGroupName?: string; } -/** - * @public - *An endpoint that hosts a model displayed in the Amazon SageMaker Model Dashboard.
- */ -export interface ModelDashboardEndpoint { - /** - * @public - *The endpoint name.
- */ - EndpointName: string | undefined; - - /** - * @public - *The Amazon Resource Name (ARN) of the endpoint.
- */ - EndpointArn: string | undefined; - - /** - * @public - *A timestamp that indicates when the endpoint was created.
- */ - CreationTime: Date | undefined; - - /** - * @public - *The last time the endpoint was modified.
- */ - LastModifiedTime: Date | undefined; - - /** - * @public - *The endpoint status.
- */ - EndpointStatus: EndpointStatus | undefined; -} - -/** - * @public - *A batch transform job. For information about SageMaker batch transform, see Use Batch - * Transform.
- */ -export interface TransformJob { - /** - * @public - *The name of the transform job.
- */ - TransformJobName?: string; - - /** - * @public - *The Amazon Resource Name (ARN) of the transform job.
- */ - TransformJobArn?: string; - - /** - * @public - *The status of the transform job.
- *Transform job statuses are:
- *
- * InProgress - The job is in progress.
- * Completed - The job has completed.
- * Failed - The transform job has failed. To see the reason for the failure,
- * see the FailureReason field in the response to a
- * DescribeTransformJob call.
- * Stopping - The transform job is stopping.
- * Stopped - The transform job has stopped.
If the transform job failed, the reason it failed.
- */ - FailureReason?: string; - - /** - * @public - *The name of the model associated with the transform job.
- */ - ModelName?: string; - - /** - * @public - *The maximum number of parallel requests that can be sent to each instance in a transform
- * job. If MaxConcurrentTransforms is set to 0 or left unset, SageMaker checks the
- * optional execution-parameters to determine the settings for your chosen algorithm. If the
- * execution-parameters endpoint is not enabled, the default value is 1. For built-in algorithms,
- * you don't need to set a value for MaxConcurrentTransforms.
Configures the timeout and maximum number of retries for processing a transform job - * invocation.
- */ - ModelClientConfig?: ModelClientConfig; - - /** - * @public - *The maximum allowed size of the payload, in MB. A payload is the data portion of a record
- * (without metadata). The value in MaxPayloadInMB must be greater than, or equal
- * to, the size of a single record. To estimate the size of a record in MB, divide the size of
- * your dataset by the number of records. To ensure that the records fit within the maximum
- * payload size, we recommend using a slightly larger value. The default value is 6 MB. For cases
- * where the payload might be arbitrarily large and is transmitted using HTTP chunked encoding,
- * set the value to 0. This feature works only in supported algorithms. Currently, SageMaker built-in
- * algorithms do not support HTTP chunked encoding.
Specifies the number of records to include in a mini-batch for an HTTP inference request. - * A record is a single unit of input data that inference can be made on. For example, a single - * line in a CSV file is a record.
- */ - BatchStrategy?: BatchStrategy; - - /** - * @public - *The environment variables to set in the Docker container. We support up to 16 key and - * values entries in the map.
- */ - Environment?: RecordDescribes the input source of a transform job and the way the transform job consumes - * it.
- */ - TransformInput?: TransformInput; - - /** - * @public - *Describes the results of a transform job.
- */ - TransformOutput?: TransformOutput; - - /** - * @public - *Describes the resources, including ML instance types and ML instance count, to use for - * transform job.
- */ - TransformResources?: TransformResources; - - /** - * @public - *A timestamp that shows when the transform Job was created.
- */ - CreationTime?: Date; - - /** - * @public - *Indicates when the transform job starts on ML instances. You are billed for the time
- * interval between this time and the value of TransformEndTime.
Indicates when the transform job has been completed, or has stopped or failed. You are
- * billed for the time interval between this time and the value of
- * TransformStartTime.
The Amazon Resource Name (ARN) of the labeling job that created the transform job.
- */ - LabelingJobArn?: string; - - /** - * @public - *The Amazon Resource Name (ARN) of the AutoML job that created the transform job.
- */ - AutoMLJobArn?: string; - - /** - * @public - *The data structure used to specify the data to be used for inference in a batch - * transform job and to associate the data that is relevant to the prediction results in - * the output. The input filter provided allows you to exclude input data that is not - * needed for inference in a batch transform job. The output filter provided allows you to - * include input data relevant to interpreting the predictions in the output from the job. - * For more information, see Associate Prediction - * Results with their Corresponding Input Records.
- */ - DataProcessing?: DataProcessing; - - /** - * @public - *Associates a SageMaker job as a trial component with an experiment and trial. Specified when - * you call the following APIs:
- *- * CreateProcessingJob - *
- *- * CreateTrainingJob - *
- *- * CreateTransformJob - *
- *A list of tags associated with the transform job.
- */ - Tags?: Tag[]; - - /** - * @public - *Configuration to control how SageMaker captures inference data for batch transform jobs.
- */ - DataCaptureConfig?: BatchDataCaptureConfig; -} - /** * @internal */ diff --git a/clients/client-sagemaker/src/models/models_4.ts b/clients/client-sagemaker/src/models/models_4.ts index 18b413a4ddaa5..dd9cea4494e09 100644 --- a/clients/client-sagemaker/src/models/models_4.ts +++ b/clients/client-sagemaker/src/models/models_4.ts @@ -7,6 +7,8 @@ import { AlgorithmSpecification, AppSecurityGroupManagement, AppSpecification, + BatchDataCaptureConfig, + BatchStrategy, BooleanOperator, Channel, CheckpointConfig, @@ -20,12 +22,16 @@ import { ResourceConfig, StoppingCondition, Tag, + TransformInput, + TransformOutput, + TransformResources, UserContext, VpcConfig, } from "./models_0"; import { _InstanceType, CrossAccountFilterOption, + DataProcessing, DebugHookConfig, DebugRuleConfiguration, DebugRuleEvaluationStatus, @@ -42,6 +48,7 @@ import { MemberDefinition, ModelCardSecurityConfig, ModelCardStatus, + ModelClientConfig, ModelMetrics, ModelPackageValidationSpecification, ModelVariantConfig, @@ -87,7 +94,7 @@ import { Direction, DomainSettingsForUpdate, Edge, - Endpoint, + EndpointStatus, FeatureParameter, MetricData, ModelArtifacts, @@ -105,6 +112,7 @@ import { SelectiveExecutionConfig, ServiceCatalogProvisionedProductDetails, TrainingJobStatus, + TransformJobStatus, TrialComponentMetricSummary, TrialComponentSource, TrialSource, @@ -112,6 +120,7 @@ import { Workteam, } from "./models_2"; import { + Endpoint, Experiment, FeatureGroup, FeatureMetadata, @@ -123,13 +132,255 @@ import { Model, ModelCard, ModelCardFilterSensitiveLog, - ModelDashboardEndpoint, MonitoringAlertSummary, Parameter, ResourceType, - TransformJob, } from "./models_3"; +/** + * @public + *An endpoint that hosts a model displayed in the Amazon SageMaker Model Dashboard.
+ */ +export interface ModelDashboardEndpoint { + /** + * @public + *The endpoint name.
+ */ + EndpointName: string | undefined; + + /** + * @public + *The Amazon Resource Name (ARN) of the endpoint.
+ */ + EndpointArn: string | undefined; + + /** + * @public + *A timestamp that indicates when the endpoint was created.
+ */ + CreationTime: Date | undefined; + + /** + * @public + *The last time the endpoint was modified.
+ */ + LastModifiedTime: Date | undefined; + + /** + * @public + *The endpoint status.
+ */ + EndpointStatus: EndpointStatus | undefined; +} + +/** + * @public + *A batch transform job. For information about SageMaker batch transform, see Use Batch + * Transform.
+ */ +export interface TransformJob { + /** + * @public + *The name of the transform job.
+ */ + TransformJobName?: string; + + /** + * @public + *The Amazon Resource Name (ARN) of the transform job.
+ */ + TransformJobArn?: string; + + /** + * @public + *The status of the transform job.
+ *Transform job statuses are:
+ *
+ * InProgress - The job is in progress.
+ * Completed - The job has completed.
+ * Failed - The transform job has failed. To see the reason for the failure,
+ * see the FailureReason field in the response to a
+ * DescribeTransformJob call.
+ * Stopping - The transform job is stopping.
+ * Stopped - The transform job has stopped.
If the transform job failed, the reason it failed.
+ */ + FailureReason?: string; + + /** + * @public + *The name of the model associated with the transform job.
+ */ + ModelName?: string; + + /** + * @public + *The maximum number of parallel requests that can be sent to each instance in a transform
+ * job. If MaxConcurrentTransforms is set to 0 or left unset, SageMaker checks the
+ * optional execution-parameters to determine the settings for your chosen algorithm. If the
+ * execution-parameters endpoint is not enabled, the default value is 1. For built-in algorithms,
+ * you don't need to set a value for MaxConcurrentTransforms.
Configures the timeout and maximum number of retries for processing a transform job + * invocation.
+ */ + ModelClientConfig?: ModelClientConfig; + + /** + * @public + *The maximum allowed size of the payload, in MB. A payload is the data portion of a record
+ * (without metadata). The value in MaxPayloadInMB must be greater than, or equal
+ * to, the size of a single record. To estimate the size of a record in MB, divide the size of
+ * your dataset by the number of records. To ensure that the records fit within the maximum
+ * payload size, we recommend using a slightly larger value. The default value is 6 MB. For cases
+ * where the payload might be arbitrarily large and is transmitted using HTTP chunked encoding,
+ * set the value to 0. This feature works only in supported algorithms. Currently, SageMaker built-in
+ * algorithms do not support HTTP chunked encoding.
Specifies the number of records to include in a mini-batch for an HTTP inference request. + * A record is a single unit of input data that inference can be made on. For example, a single + * line in a CSV file is a record.
+ */ + BatchStrategy?: BatchStrategy; + + /** + * @public + *The environment variables to set in the Docker container. We support up to 16 key and + * values entries in the map.
+ */ + Environment?: RecordDescribes the input source of a transform job and the way the transform job consumes + * it.
+ */ + TransformInput?: TransformInput; + + /** + * @public + *Describes the results of a transform job.
+ */ + TransformOutput?: TransformOutput; + + /** + * @public + *Describes the resources, including ML instance types and ML instance count, to use for + * transform job.
+ */ + TransformResources?: TransformResources; + + /** + * @public + *A timestamp that shows when the transform Job was created.
+ */ + CreationTime?: Date; + + /** + * @public + *Indicates when the transform job starts on ML instances. You are billed for the time
+ * interval between this time and the value of TransformEndTime.
Indicates when the transform job has been completed, or has stopped or failed. You are
+ * billed for the time interval between this time and the value of
+ * TransformStartTime.
The Amazon Resource Name (ARN) of the labeling job that created the transform job.
+ */ + LabelingJobArn?: string; + + /** + * @public + *The Amazon Resource Name (ARN) of the AutoML job that created the transform job.
+ */ + AutoMLJobArn?: string; + + /** + * @public + *The data structure used to specify the data to be used for inference in a batch + * transform job and to associate the data that is relevant to the prediction results in + * the output. The input filter provided allows you to exclude input data that is not + * needed for inference in a batch transform job. The output filter provided allows you to + * include input data relevant to interpreting the predictions in the output from the job. + * For more information, see Associate Prediction + * Results with their Corresponding Input Records.
+ */ + DataProcessing?: DataProcessing; + + /** + * @public + *Associates a SageMaker job as a trial component with an experiment and trial. Specified when + * you call the following APIs:
+ *+ * CreateProcessingJob + *
+ *+ * CreateTrainingJob + *
+ *+ * CreateTransformJob + *
+ *A list of tags associated with the transform job.
+ */ + Tags?: Tag[]; + + /** + * @public + *Configuration to control how SageMaker captures inference data for batch transform jobs.
+ */ + DataCaptureConfig?: BatchDataCaptureConfig; +} + /** * @public *The model card for a model displayed in the Amazon SageMaker Model Dashboard.
diff --git a/clients/client-sagemaker/src/protocols/Aws_json1_1.ts b/clients/client-sagemaker/src/protocols/Aws_json1_1.ts index 8ccce0045d84c..24356d50d53c2 100644 --- a/clients/client-sagemaker/src/protocols/Aws_json1_1.ts +++ b/clients/client-sagemaker/src/protocols/Aws_json1_1.ts @@ -892,7 +892,6 @@ import { CreateCodeRepositoryInput, CreateCompilationJobRequest, CreateContextRequest, - CreateDataQualityJobDefinitionRequest, DataQualityAppSpecification, DataQualityBaselineConfig, DataQualityJobInput, @@ -938,7 +937,6 @@ import { MonitoringResources, MonitoringS3Output, MonitoringStatisticsResource, - MonitoringStoppingCondition, MultiModelConfig, NeoVpcConfig, OutputConfig, @@ -960,6 +958,7 @@ import { Tag, TargetPlatform, TextClassificationJobConfig, + TextGenerationJobConfig, TimeSeriesConfig, TimeSeriesForecastingJobConfig, TimeSeriesForecastingSettings, @@ -983,6 +982,7 @@ import { WorkspaceSettings, } from "../models/models_0"; import { + CreateDataQualityJobDefinitionRequest, CreateDeviceFleetRequest, CreateDomainRequest, CreateEdgeDeploymentPlanRequest, @@ -1042,7 +1042,6 @@ import { DeleteArtifactRequest, DeleteAssociationRequest, DeleteCodeRepositoryInput, - DeleteContextRequest, DeploymentConfig, DeploymentStage, DeviceSelectionConfig, @@ -1121,6 +1120,7 @@ import { MonitoringInput, MonitoringJobDefinition, MonitoringScheduleConfig, + MonitoringStoppingCondition, NetworkConfig, NotebookInstanceAcceleratorType, NotebookInstanceLifecycleHook, @@ -1193,6 +1193,7 @@ import { WorkforceVpcConfigRequest, } from "../models/models_1"; import { + DeleteContextRequest, DeleteDataQualityJobDefinitionRequest, DeleteDeviceFleetRequest, DeleteDomainRequest, @@ -1363,7 +1364,6 @@ import { EdgeModel, EdgePackagingJobSummary, EnableSagemakerServicecatalogPortfolioInput, - Endpoint, FeatureParameter, FinalHyperParameterTuningJobObjectiveMetric, HyperParameterTrainingJobSummary, @@ -1391,6 +1391,7 @@ import { Workteam, } from "../models/models_2"; import { + Endpoint, EndpointConfigSummary, EndpointSummary, Experiment, @@ -1568,7 +1569,6 @@ import { ModelCardExportJobSummary, ModelCardSummary, ModelCardVersionSummary, - ModelDashboardEndpoint, ModelMetadataFilter, ModelMetadataSearchExpression, ModelPackageGroupSummary, @@ -1594,13 +1594,13 @@ import { StudioLifecycleConfigDetails, SuggestionQuery, TrainingJobSummary, - TransformJob, TransformJobSummary, TrialComponentSummary, TrialSummary, UserProfileDetails, } from "../models/models_3"; import { + ModelDashboardEndpoint, ModelDashboardModel, ModelDashboardModelCard, ModelDashboardMonitoringSchedule, @@ -1649,6 +1649,7 @@ import { StopTrainingJobRequest, StopTransformJobRequest, TrainingJob, + TransformJob, Trial, TrialComponent, TrialComponentSimpleSummary, @@ -21999,6 +22000,8 @@ const se_StopPipelineExecutionRequest = (input: StopPipelineExecutionRequest, co // se_TextClassificationJobConfig omitted. +// se_TextGenerationJobConfig omitted. + // se_TimeSeriesConfig omitted. // se_TimeSeriesForecastingJobConfig omitted. @@ -28045,6 +28048,10 @@ const de_TargetTrackingScalingPolicyConfiguration = ( // de_TextClassificationJobConfig omitted. +// de_TextGenerationJobConfig omitted. + +// de_TextGenerationResolvedAttributes omitted. + // de_TimeSeriesConfig omitted. // de_TimeSeriesForecastingJobConfig omitted. diff --git a/codegen/sdk-codegen/aws-models/sagemaker.json b/codegen/sdk-codegen/aws-models/sagemaker.json index e9c7ed0761575..ce1f6f9ffafbe 100644 --- a/codegen/sdk-codegen/aws-models/sagemaker.json +++ b/codegen/sdk-codegen/aws-models/sagemaker.json @@ -400,7 +400,7 @@ "CompressionType": { "target": "com.amazonaws.sagemaker#CompressionType", "traits": { - "smithy.api#documentation": "The type of compression used for an additional data source used in inference or\n training. Specify None if your additional data source is not compressed.
The type of compression used for an additional data source used in inference or\n training. Specify None if your additional data source is not\n compressed.
The content type of the data from the input source. The following are the allowed\n content types for different problems:
\nFor tabular problem types: text/csv;header=present or\n x-application/vnd.amazon+parquet. The default value is\n text/csv;header=present.
For image classification: image/png, image/jpeg, or\n image/*. The default value is image/*.
For text classification: text/csv;header=present or\n x-application/vnd.amazon+parquet. The default value is\n text/csv;header=present.
For time-series forecasting: text/csv;header=present or\n x-application/vnd.amazon+parquet. The default value is\n text/csv;header=present.
The content type of the data from the input source. The following are the allowed\n content types for different problems:
\nFor tabular problem types: text/csv;header=present or\n x-application/vnd.amazon+parquet. The default value is\n text/csv;header=present.
For image classification: image/png, image/jpeg, or\n image/*. The default value is image/*.
For text classification: text/csv;header=present or\n x-application/vnd.amazon+parquet. The default value is\n text/csv;header=present.
For time-series forecasting: text/csv;header=present or\n x-application/vnd.amazon+parquet. The default value is\n text/csv;header=present.
For text generation (LLMs fine-tuning): text/csv;header=present or\n x-application/vnd.amazon+parquet. The default value is\n text/csv;header=present.
The maximum number of times a training job is allowed to run.
\nFor text and image classification, as well as time-series forecasting problem types, the\n supported value is 1. For tabular problem types, the maximum value is 750.
" + "smithy.api#documentation": "The maximum number of times a training job is allowed to run.
\nFor text and image classification, time-series forecasting, as well as text generation\n (LLMs fine-tuning) problem types, the supported value is 1. For tabular problem types, the\n maximum value is 750.
" } }, "MaxRuntimePerTrainingJobInSeconds": { @@ -2790,13 +2790,13 @@ "target": "com.amazonaws.sagemaker#AutoMLMetricEnum", "traits": { "smithy.api#clientOptional": {}, - "smithy.api#documentation": "The name of the objective metric used to measure the predictive quality of a machine\n learning system. During training, the model's parameters are updated iteratively to\n optimize its performance based on the feedback provided by the objective metric when\n evaluating the model on the validation dataset.
\nFor the list of all available metrics supported by Autopilot, see Autopilot\n metrics.
\nIf you do not specify a metric explicitly, the default behavior is to automatically\n use:
\nFor tabular problem types:
\nRegression: MSE.
Binary classification: F1.
Multiclass classification: Accuracy.
For image or text classification problem types: Accuracy\n
For time-series forecasting problem types:\n AverageWeightedQuantileLoss\n
The name of the objective metric used to measure the predictive quality of a machine\n learning system. During training, the model's parameters are updated iteratively to\n optimize its performance based on the feedback provided by the objective metric when\n evaluating the model on the validation dataset.
\nThe list of available metrics supported by Autopilot and the default metric applied when you\n do not specify a metric name explicitly depend on the problem type.
\nFor tabular problem types:
\nList of available metrics:
\n Regression: InferenceLatency, MAE,\n MSE, R2, RMSE\n
Binary classification: Accuracy, AUC,\n BalancedAccuracy, F1,\n InferenceLatency, LogLoss,\n Precision, Recall\n
Multiclass classification: Accuracy,\n BalancedAccuracy, F1macro,\n InferenceLatency, LogLoss,\n PrecisionMacro, RecallMacro\n
For a description of each metric, see Autopilot metrics for classification and regression.
\nDefault objective metrics:
\nRegression: MSE.
Binary classification: F1.
Multiclass classification: Accuracy.
For image or text classification problem types:
\nList of available metrics: Accuracy\n
For a description of each metric, see Autopilot metrics for text and image classification.
\nDefault objective metrics: Accuracy\n
For time-series forecasting problem types:
\nList of available metrics: RMSE, wQL,\n Average wQL, MASE, MAPE,\n WAPE\n
For a description of each metric, see Autopilot metrics for\n time-series forecasting.
\nDefault objective metrics: AverageWeightedQuantileLoss\n
For text generation problem types (LLMs fine-tuning): \n Fine-tuning language models in Autopilot does not\n require setting the AutoMLJobObjective field. Autopilot fine-tunes LLMs\n without requiring multiple candidates to be trained and evaluated. \n Instead, using your dataset, Autopilot directly fine-tunes your target model to enhance a\n default objective metric, the cross-entropy loss. After fine-tuning a language model,\n you can evaluate the quality of its generated text using different metrics. \n For a list of the available metrics, see Metrics for\n fine-tuning LLMs in Autopilot.
Specifies a metric to minimize or maximize as the objective of a job.
" + "smithy.api#documentation": "Specifies a metric to minimize or maximize as the objective of an AutoML job.
" } }, "com.amazonaws.sagemaker#AutoMLJobObjectiveType": { @@ -3296,6 +3296,48 @@ "traits": { "smithy.api#enumValue": "AverageWeightedQuantileLoss" } + }, + "ROUGE1": { + "target": "smithy.api#Unit", + "traits": { + "smithy.api#enumValue": "Rouge1" + } + }, + "ROUGE2": { + "target": "smithy.api#Unit", + "traits": { + "smithy.api#enumValue": "Rouge2" + } + }, + "ROUGEL": { + "target": "smithy.api#Unit", + "traits": { + "smithy.api#enumValue": "RougeL" + } + }, + "ROUGEL_SUM": { + "target": "smithy.api#Unit", + "traits": { + "smithy.api#enumValue": "RougeLSum" + } + }, + "PERPLEXITY": { + "target": "smithy.api#Unit", + "traits": { + "smithy.api#enumValue": "Perplexity" + } + }, + "VALIDATION_LOSS": { + "target": "smithy.api#Unit", + "traits": { + "smithy.api#enumValue": "ValidationLoss" + } + }, + "TRAINING_LOSS": { + "target": "smithy.api#Unit", + "traits": { + "smithy.api#enumValue": "TrainingLoss" + } } } }, @@ -3398,13 +3440,19 @@ "TabularJobConfig": { "target": "com.amazonaws.sagemaker#TabularJobConfig", "traits": { - "smithy.api#documentation": "Settings used to configure an AutoML job V2 for a tabular problem type (regression,\n classification).
" + "smithy.api#documentation": "Settings used to configure an AutoML job V2 for the tabular problem type (regression,\n classification).
" } }, "TimeSeriesForecastingJobConfig": { "target": "com.amazonaws.sagemaker#TimeSeriesForecastingJobConfig", "traits": { - "smithy.api#documentation": "Settings used to configure an AutoML job V2 for a time-series forecasting problem\n type.
" + "smithy.api#documentation": "Settings used to configure an AutoML job V2 for the time-series forecasting problem\n type.
" + } + }, + "TextGenerationJobConfig": { + "target": "com.amazonaws.sagemaker#TextGenerationJobConfig", + "traits": { + "smithy.api#documentation": "Settings used to configure an AutoML job V2 for the text generation (LLMs fine-tuning)\n problem type.
\nThe text generation models that support fine-tuning in Autopilot are currently accessible\n exclusively in regions supported by Canvas. Refer to the documentation of Canvas for the full list of its supported\n Regions.
\nDefines the resolved attributes for the TABULAR problem type.
The resolved attributes for the tabular problem type.
" + } + }, + "TextGenerationResolvedAttributes": { + "target": "com.amazonaws.sagemaker#TextGenerationResolvedAttributes", + "traits": { + "smithy.api#documentation": "The resolved attributes for the text generation problem type.
" } } }, "traits": { - "smithy.api#documentation": "The resolved attributes specific to the problem type of an AutoML job V2.
" + "smithy.api#documentation": "Stores resolved attributes specific to the problem type of an AutoML job V2.
" } }, "com.amazonaws.sagemaker#AutoMLProcessingUnit": { @@ -3707,6 +3767,16 @@ } } }, + "com.amazonaws.sagemaker#BaseModelName": { + "type": "string", + "traits": { + "smithy.api#length": { + "min": 1, + "max": 256 + }, + "smithy.api#pattern": "^[a-zA-Z0-9](-*[a-zA-Z0-9])*$" + } + }, "com.amazonaws.sagemaker#BatchDataCaptureConfig": { "type": "structure", "members": { @@ -7065,7 +7135,7 @@ } ], "traits": { - "smithy.api#documentation": "Creates an Autopilot job also referred to as Autopilot experiment or AutoML job.
\nWe recommend using the new versions CreateAutoMLJobV2 and DescribeAutoMLJobV2, which offer backward compatibility.
\n\n CreateAutoMLJobV2 can manage tabular problem types identical to those of\n its previous version CreateAutoMLJob, as well as time-series forecasting,\n and non-tabular problem types such as image or text classification.
Find guidelines about how to migrate a CreateAutoMLJob to\n CreateAutoMLJobV2 in Migrate a CreateAutoMLJob to CreateAutoMLJobV2.
You can find the best-performing model after you run an AutoML job by calling DescribeAutoMLJobV2 (recommended) or DescribeAutoMLJob.
" + "smithy.api#documentation": "Creates an Autopilot job also referred to as Autopilot experiment or AutoML job.
\nWe recommend using the new versions CreateAutoMLJobV2 and DescribeAutoMLJobV2, which offer backward compatibility.
\n\n CreateAutoMLJobV2 can manage tabular problem types identical to those of\n its previous version CreateAutoMLJob, as well as time-series forecasting,\n non-tabular problem types such as image or text classification, and text generation\n (LLMs fine-tuning).
Find guidelines about how to migrate a CreateAutoMLJob to\n CreateAutoMLJobV2 in Migrate a CreateAutoMLJob to CreateAutoMLJobV2.
You can find the best-performing model after you run an AutoML job by calling DescribeAutoMLJobV2 (recommended) or DescribeAutoMLJob.
" } }, "com.amazonaws.sagemaker#CreateAutoMLJobRequest": { @@ -7177,7 +7247,7 @@ } ], "traits": { - "smithy.api#documentation": "Creates an Autopilot job also referred to as Autopilot experiment or AutoML job V2.
\n\n CreateAutoMLJobV2 and DescribeAutoMLJobV2 are new versions of CreateAutoMLJob\n and DescribeAutoMLJob which offer backward compatibility.
\n\n CreateAutoMLJobV2 can manage tabular problem types identical to those of\n its previous version CreateAutoMLJob, as well as time-series forecasting,\n and non-tabular problem types such as image or text classification.
Find guidelines about how to migrate a CreateAutoMLJob to\n CreateAutoMLJobV2 in Migrate a CreateAutoMLJob to CreateAutoMLJobV2.
For the list of available problem types supported by CreateAutoMLJobV2, see\n AutoMLProblemTypeConfig.
You can find the best-performing model after you run an AutoML job V2 by calling DescribeAutoMLJobV2.
" + "smithy.api#documentation": "Creates an Autopilot job also referred to as Autopilot experiment or AutoML job V2.
\n\n CreateAutoMLJobV2 and DescribeAutoMLJobV2 are new versions of CreateAutoMLJob\n and DescribeAutoMLJob which offer backward compatibility.
\n\n CreateAutoMLJobV2 can manage tabular problem types identical to those of\n its previous version CreateAutoMLJob, as well as time-series forecasting,\n non-tabular problem types such as image or text classification, and text generation\n (LLMs fine-tuning).
Find guidelines about how to migrate a CreateAutoMLJob to\n CreateAutoMLJobV2 in Migrate a CreateAutoMLJob to CreateAutoMLJobV2.
For the list of available problem types supported by CreateAutoMLJobV2, see\n AutoMLProblemTypeConfig.
You can find the best-performing model after you run an AutoML job V2 by calling DescribeAutoMLJobV2.
" } }, "com.amazonaws.sagemaker#CreateAutoMLJobV2Request": { @@ -7195,7 +7265,7 @@ "target": "com.amazonaws.sagemaker#AutoMLJobInputDataConfig", "traits": { "smithy.api#clientOptional": {}, - "smithy.api#documentation": "An array of channel objects describing the input data and their location. Each channel\n is a named input source. Similar to the InputDataConfig attribute in the CreateAutoMLJob input parameters.\n The supported formats depend on the problem type:
For tabular problem types: S3Prefix,\n ManifestFile.
For image classification: S3Prefix, ManifestFile,\n AugmentedManifestFile.
For text classification: S3Prefix.
For time-series forecasting: S3Prefix.
An array of channel objects describing the input data and their location. Each channel\n is a named input source. Similar to the InputDataConfig attribute in the CreateAutoMLJob input parameters.\n The supported formats depend on the problem type:
For tabular problem types: S3Prefix,\n ManifestFile.
For image classification: S3Prefix, ManifestFile,\n AugmentedManifestFile.
For text classification: S3Prefix.
For time-series forecasting: S3Prefix.
For text generation (LLMs fine-tuning): S3Prefix.
Specifies a metric to minimize or maximize as the objective of a job. If not specified,\n the default objective metric depends on the problem type. For the list of default values\n per problem type, see AutoMLJobObjective.
\nFor tabular problem types, you must either provide both the\n AutoMLJobObjective and indicate the type of supervised learning problem\n in AutoMLProblemTypeConfig (TabularJobConfig.ProblemType), or\n none at all.
Specifies a metric to minimize or maximize as the objective of a job. If not specified,\n the default objective metric depends on the problem type. For the list of default values\n per problem type, see AutoMLJobObjective.
\nFor tabular problem types: You must either provide both the\n AutoMLJobObjective and indicate the type of supervised learning\n problem in AutoMLProblemTypeConfig\n (TabularJobConfig.ProblemType), or none at all.
For text generation problem types (LLMs fine-tuning): \n Fine-tuning language models in Autopilot does not\n require setting the AutoMLJobObjective field. Autopilot fine-tunes LLMs\n without requiring multiple candidates to be trained and evaluated. \n Instead, using your dataset, Autopilot directly fine-tunes your target model to enhance a\n default objective metric, the cross-entropy loss. After fine-tuning a language model,\n you can evaluate the quality of its generated text using different metrics. \n For a list of the available metrics, see Metrics for\n fine-tuning LLMs in Autopilot.
Whether to\n minimize\n or maximize the objective metric.
", + "smithy.api#documentation": "Whether to minimize or maximize the objective metric.
", "smithy.api#required": {} } }, @@ -27714,7 +27784,7 @@ } }, "traits": { - "smithy.api#documentation": "Defines the objective metric for a hyperparameter tuning job.\n Hyperparameter\n tuning uses the value of this metric to evaluate the training jobs it launches, and\n returns the training job that results in either the highest or lowest value for this\n metric, depending on the value you specify for the Type\n parameter.
Defines the objective metric for a hyperparameter tuning job. Hyperparameter tuning\n uses the value of this metric to evaluate the training jobs it launches, and returns the\n training job that results in either the highest or lowest value for this metric,\n depending on the value you specify for the Type parameter. If you want to\n define a custom objective metric, see Define metrics and environment variables.
Stores the configuration information for the image classification problem of an AutoML\n job V2.
" + "smithy.api#documentation": "The collection of settings used by an AutoML job V2 for the image classification problem\n type.
" } }, "com.amazonaws.sagemaker#ImageConfig": { @@ -55028,7 +55098,7 @@ } }, "traits": { - "smithy.api#documentation": "The collection of settings used by an AutoML job V2 for the TABULAR problem\n type.
The collection of settings used by an AutoML job V2 for the tabular problem type.
" } }, "com.amazonaws.sagemaker#TabularResolvedAttributes": { @@ -55042,7 +55112,7 @@ } }, "traits": { - "smithy.api#documentation": "The resolved attributes specific to the TABULAR problem type.
The resolved attributes specific to the tabular problem type.
" } }, "com.amazonaws.sagemaker#Tag": { @@ -55658,7 +55728,38 @@ } }, "traits": { - "smithy.api#documentation": "Stores the configuration information for the text classification problem of an AutoML job\n V2.
" + "smithy.api#documentation": "The collection of settings used by an AutoML job V2 for the text classification problem\n type.
" + } + }, + "com.amazonaws.sagemaker#TextGenerationJobConfig": { + "type": "structure", + "members": { + "CompletionCriteria": { + "target": "com.amazonaws.sagemaker#AutoMLJobCompletionCriteria" + }, + "BaseModelName": { + "target": "com.amazonaws.sagemaker#BaseModelName", + "traits": { + "smithy.api#documentation": "The name of the base model to fine-tune. Autopilot supports fine-tuning a variety of large\n language models. For information on the list of supported models, see Text generation models supporting fine-tuning in Autopilot. If no\n BaseModelName is provided, the default model used is Falcon-7B-Instruct.\n
The collection of settings used by an AutoML job V2 for the text generation problem\n type.
\nThe text generation models that support fine-tuning in Autopilot are currently accessible\n exclusively in regions supported by Canvas. Refer to the documentation of Canvas for the full list of its supported\n Regions.
\nThe name of the base model to fine-tune.
" + } + } + }, + "traits": { + "smithy.api#documentation": "The resolved attributes specific to the text generation problem type.
" } }, "com.amazonaws.sagemaker#ThingName": { @@ -61159,7 +61260,7 @@ "type": "float", "traits": { "smithy.api#range": { - "min": 0 + "min": 0.0 } } },