你当前正在访问 Microsoft Azure Global Edition 技术文档网站。 如果需要访问由世纪互联运营的 Microsoft Azure 中国技术文档网站,请访问 https://docs.azure.cn

Azure.ResourceManager.MachineLearning.Models 命名空间

AccessKeyAuthTypeWorkspaceConnectionProperties

AccessKeyAuthTypeWorkspaceConnectionProperties。

AmlCompute

Azure 机器学习计算。

AmlComputeNodeInformation

与 AmlCompute 相关的计算节点信息。

AmlComputeProperties

AML 计算属性。

AmlComputeScaleSettings

AML 计算的缩放设置。

AmlToken

AML 令牌标识配置。

AmlTokenComputeIdentity

AML 令牌计算标识定义。

ApiKeyAuthWorkspaceConnectionProperties

此连接类型涵盖泛型 ApiKey 身份验证连接类别,例如:AzureOpenAI: Category:= AzureOpenAI AuthType:= ApiKey (类型鉴别器) 凭据:= {ApiKey} as Microsoft.MachineLearning.AccountRP.Contracts.WorkspaceConnection.ApiKey Target:= {ApiBase}

CognitiveService: Category:= CognitiveService AuthType:= ApiKey (类型鉴别器) 凭据:= {SubscriptionKey} as Microsoft.MachineLearning.AccountRP.Contracts.WorkspaceConnection.ApiKey Target:= ServiceRegion={serviceRegion}

CognitiveSearch: Category:= CognitiveSearch AuthType:= ApiKey (作为类型鉴别器) 凭据:= {Key} as Microsoft.MachineLearning.AccountRP.Contracts.WorkspaceConnection.ApiKey Target:= {Endpoint}

对 ApiType、ApiVersion、Kind 和其他元数据字段使用元数据属性包

ArmMachineLearningModelFactory

模型的模型工厂。

AutoDeleteSetting

AutoDeleteSetting。

AutoForecastHorizon

预测范围由系统自动确定。

AutoMLJob

AutoMLJob 类。 使用此类执行 AutoML 任务,如分类/回归等。有关支持的所有任务,请参阅 TaskType 枚举。

AutoMLVertical

AutoML 垂直类。 AutoML 垂直的基类 - TableVertical/ImageVertical/NLPVertical 请注意 AutoMLVertical 是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 ClassificationTask、、MachineLearningForecastingImageClassificationImageClassificationMultilabelImageInstanceSegmentationImageObjectDetectionAutoMLVerticalRegression、、 TextClassificationTextClassificationMultilabel 和 。TextNer

AutoMLVerticalRegression

AutoML 表垂直中的回归任务。

AutoNCrossValidations

自动确定 N 交叉验证。

AutoSeasonality

AutoSeasonality。

AutoTargetLags

AutoTargetLags。

AutoTargetRollingWindowSize

自动确定目标延迟滚动窗口。

AzMonMonitoringAlertNotificationSettings

AzMonMonitoringAlertNotificationSettings。

AzureDevOpsWebhook

特定于 Azure DevOps 的 Webhook 详细信息。

AzureMLBatchInferencingServer

Azure ML 批处理推理服务器配置。

AzureMLOnlineInferencingServer

Azure ML 联机推理配置。

BanditPolicy

根据 slack 条件以及用于评估的频率和延迟间隔,定义提前终止策略。

BaseEnvironmentSource

BaseEnvironmentSource。 请注意 BaseEnvironmentSource ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 BaseEnvironmentType

BaseEnvironmentType

基本环境类型。

BatchDeploymentConfiguration

与不同部署类型相关的属性。 请注意 BatchDeploymentConfiguration ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 BatchPipelineComponentDeploymentConfiguration

BatchPipelineComponentDeploymentConfiguration

Batch 管道组件部署的属性。

BayesianSamplingAlgorithm

定义一个采样算法,该算法基于以前的值生成值。

BlobReferenceForConsumptionDto

BlobReferenceForConsumptionDto。

CategoricalDataDriftMetricThreshold

CategoricalDataDriftMetricThreshold。

CategoricalDataQualityMetricThreshold

CategoricalDataQualityMetricThreshold。

CategoricalPredictionDriftMetricThreshold

CategoricalPredictionDriftMetricThreshold。

ClassificationModelPerformanceMetricThreshold

ClassificationModelPerformanceMetricThreshold。

ClassificationTask

AutoML 表垂直中的分类任务。

ClassificationTrainingSettings

分类 培训相关配置。

CocoExportSummary

CocoExportSummary。

ColumnTransformer

列转换器参数。

ComputeStartStopCronSchedule

ComputeStartStop 计划类型的工作流触发器 cron。

ComputeStartStopRecurrenceSchedule

ComputeStartStop 计划类型的工作流触发器重复周期。

ContainerEndpoint

ContainerEndpoint。

CreateMonitorAction

CreateMonitorAction。

CronTrigger

CronTrigger。

CsvExportSummary

CsvExportSummary。

CustomForecastHorizon

所需的最大预测范围,以时序频率为单位。

CustomInferencingServer

自定义推理服务器配置。

CustomKeysWorkspaceConnectionProperties

Category:= CustomKeys AuthType:= CustomKeys (作为类型鉴别器) 凭据:= {CustomKeys} as Microsoft.MachineLearning.AccountRP.Contracts.WorkspaceConnection.CustomKeys Target:= {any value} 使用 ApiVersion 和其他元数据字段的元数据属性包

CustomMetricThreshold

CustomMetricThreshold。

CustomMonitoringSignal

CustomMonitoringSignal。

CustomNCrossValidations

N 交叉验证由用户指定。

CustomSeasonality

CustomSeasonality。

CustomService

指定自定义服务配置。

CustomTargetLags

CustomTargetLags。

CustomTargetRollingWindowSize

CustomTargetRollingWindowSize。

DatabaseSource

DatabaseSource。

DataCollectionConfiguration

DataCollectionConfiguration。

DataCollector

DataCollector。

DataDriftMetricThresholdBase

DataDriftMetricThresholdBase。 请注意 DataDriftMetricThresholdBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 CategoricalDataDriftMetricThresholdNumericalDataDriftMetricThreshold

DataDriftMonitoringSignal

DataDriftMonitoringSignal。

DataImport

DataImport。

DataImportSource

DataImportSource。 请注意 DataImportSource ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 DatabaseSourceFileSystemSource

DataQualityMetricThresholdBase

DataQualityMetricThresholdBase。 请注意 DataQualityMetricThresholdBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 CategoricalDataQualityMetricThresholdNumericalDataQualityMetricThreshold

DataQualityMonitoringSignal

DataQualityMonitoringSignal。

DatasetExportSummary

DatasetExportSummary。

DockerSetting

DockerSetting。

EmailMonitoringAlertNotificationSettings

EmailMonitoringAlertNotificationSettings。

EncryptionKeyVaultUpdateProperties

EncryptionKeyVaultUpdateProperties。

EncryptionUpdateProperties

EncryptionUpdateProperties。

EnvironmentVariable

EnvironmentVariable。

ExportSummary

ExportSummary。 请注意 ExportSummary , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 CsvExportSummaryCocoExportSummaryDatasetExportSummary

FeatureAttributionDriftMonitoringSignal

FeatureAttributionDriftMonitoringSignal。

FeatureAttributionMetricThreshold

FeatureAttributionMetricThreshold。

FeatureSetVersionBackfillContent

请求有效负载,用于为给定功能集版本创建回填请求。

FeatureStoreSettings

FeatureStoreSettings。

FeatureSubset

FeatureSubset。

FeatureWindow

指定功能窗口。

FileSystemSource

FileSystemSource。

FixedInputData

修复了输入数据定义。

ForecastHorizon

所需的最大预测范围,以时序频率为单位。 请注意 ForecastHorizon , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 AutoForecastHorizonCustomForecastHorizon

ForecastingSeasonality

预测季节性。 请注意 ForecastingSeasonality , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 AutoSeasonalityCustomSeasonality

ForecastingSettings

预测特定参数。

ForecastingTrainingSettings

与预测训练相关的配置。

FqdnOutboundRule

机器学习工作区的托管网络的 FQDN 出站规则。

GenerationSafetyQualityMetricThreshold

生成安全质量指标阈值定义。

GenerationSafetyQualityMonitoringSignal

生成安全质量监视信号定义。

GenerationTokenStatisticsMetricThreshold

生成令牌统计信息指标阈值定义。

GenerationTokenStatisticsSignal

生成令牌统计信息信号定义。

GridSamplingAlgorithm

定义一个采样算法,该算法会详尽地生成空间中的每个值组合。

HdfsDatastore

HdfsDatastore。

IdleShutdownSetting

在用户定义的非活动时段后停止计算实例。

ImageClassification

图像分类。 当仅使用一组类中的单个标签对图像进行分类时,将使用多类图像分类-例如,每个图像被分类为“猫”、“狗”或“鸭子”的图像。

ImageClassificationMultilabel

图像分类多标签。 当图像可能具有一组标签中的一个或多个标签时,可以使用多标签图像分类-例如,图像可以同时标记“cat”和“dog”。

ImageInstanceSegmentation

图像实例分段。 实例分段用于在像素级别标识图像中的对象,在图像中的每个对象周围绘制多边形。

ImageLimitSettings

AutoML 作业的限制设置。

ImageMetadata

返回有关此计算实例的操作系统映像的元数据。

ImageModelDistributionSettings

用于扫描模型设置值的分布表达式。 <示例> 一些示例包括:

ModelName = "choice('seresnext', 'resnest50')";
LearningRate = "uniform(0.001, 0.01)";
LayersToFreeze = "choice(0, 2)";
```&lt;/example&gt;
All distributions can be specified as distribution_name(min, max) or choice(val1, val2, ..., valn)
where distribution name can be: uniform, quniform, loguniform, etc
For more details on how to compose distribution expressions please check the documentation:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-tune-hyperparameters
For more information on the available settings please visit the official documentation:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-auto-train-image-models.
ImageModelDistributionSettingsClassification

用于扫描模型设置值的分布表达式。 <示例> 一些示例包括:

ModelName = "choice('seresnext', 'resnest50')";
LearningRate = "uniform(0.001, 0.01)";
LayersToFreeze = "choice(0, 2)";
```&lt;/example&gt;
For more details on how to compose distribution expressions please check the documentation:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-tune-hyperparameters
For more information on the available settings please visit the official documentation:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-auto-train-image-models.
ImageModelDistributionSettingsObjectDetection

用于扫描模型设置值的分布表达式。 <示例> 一些示例包括:

ModelName = "choice('seresnext', 'resnest50')";
LearningRate = "uniform(0.001, 0.01)";
LayersToFreeze = "choice(0, 2)";
```&lt;/example&gt;
For more details on how to compose distribution expressions please check the documentation:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-tune-hyperparameters
For more information on the available settings please visit the official documentation:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-auto-train-image-models.
ImageModelSettings

用于训练模型的设置。 有关可用设置的详细信息,请访问官方文档: https://docs.microsoft.com/en-us/azure/machine-learning/how-to-auto-train-image-models

ImageModelSettingsClassification

用于训练模型的设置。 有关可用设置的详细信息,请访问官方文档: https://docs.microsoft.com/en-us/azure/machine-learning/how-to-auto-train-image-models

ImageModelSettingsObjectDetection

用于训练模型的设置。 有关可用设置的详细信息,请访问官方文档: https://docs.microsoft.com/en-us/azure/machine-learning/how-to-auto-train-image-models

ImageObjectDetection

图像对象检测。 对象检测用于识别图像中的对象,并使用边界框定位每个对象,例如定位图像中的所有狗和猫,并在每个对象周围绘制边界框。

ImageSetting

ImageSetting。

ImageSweepSettings

模型扫描和超参数扫描相关设置。

ImportDataAction

ImportDataAction。

IndexColumn

表示索引列的 Dto 对象。

InferencingServer

InferencingServer。 请注意 InferencingServer , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 AzureMLBatchInferencingServerAzureMLOnlineInferencingServerCustomInferencingServerTritonInferencingServer

IntellectualProperty

资源的知识产权详细信息。

JobAllNodes

所有节点意味着服务将在作业的所有节点上运行。

JobNodes

抽象节点定义 请注意 JobNodes 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 JobAllNodes

JobQueueSettings

JobQueueSettings。

JobStatusMessage

与项目关联的活动消息。

KerberosKeytabCredentials

KerberosKeytabCredentials。

KerberosKeytabSecrets

KerberosKeytabSecrets。

KerberosPasswordCredentials

KerberosPasswordCredentials。

KerberosPasswordSecrets

KerberosPasswordSecrets。

LabelCategory

标签类别定义。

LabelClass

标签类定义。

LabelingDataConfiguration

标记数据配置定义。

LabelingJobImageProperties

图像数据的标记作业的属性。

LabelingJobMediaProperties

标记作业的属性请注意 LabelingJobMediaProperties 是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 LabelingJobImagePropertiesLabelingJobTextProperties

LabelingJobProperties

标记作业定义。

LabelingJobTextProperties

文本数据的标记作业的属性。

LakeHouseArtifact

LakeHouseArtifact。

MachineLearningAccountKeyDatastoreCredentials

帐户密钥数据存储凭据配置。

MachineLearningAccountKeyDatastoreSecrets

数据存储帐户密钥机密。

MachineLearningAksCompute

基于 AKS 的机器学习计算。

MachineLearningAksComputeProperties

AKS 属性。

MachineLearningAksComputeSecrets

与基于 AKS 的机器学习计算相关的机密。

MachineLearningAksNetworkingConfiguration

AKS 网络高级配置。

MachineLearningAllFeatures

MachineLearningAllFeatures。

MachineLearningAssetBase

MachineLearningAssetBase。

MachineLearningAssetContainer

MachineLearningAssetContainer。

MachineLearningAssetReferenceBase

资产引用的基定义。 请注意 MachineLearningAssetReferenceBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningDataPathAssetReferenceMachineLearningOutputPathAssetReferenceMachineLearningIdAssetReference

MachineLearningAssistConfiguration

标记 MLAssist 配置定义 请注意 MachineLearningAssistConfiguration 是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 MLAssistConfigurationDisabledMachineLearningAssistEnabledConfiguration

MachineLearningAssistEnabledConfiguration

启用 MLAssist 时标记 MLAssist 配置定义。

MachineLearningAutoPauseProperties

自动暂停属性。

MachineLearningAutoScaleProperties

自动缩放属性。

MachineLearningAzureBlobDatastore

Azure Blob 数据存储配置。

MachineLearningAzureDataLakeGen1Datastore

Azure Data Lake Gen1 数据存储配置。

MachineLearningAzureDataLakeGen2Datastore

Azure Data Lake Gen2 数据存储配置。

MachineLearningAzureFileDatastore

Azure 文件数据存储配置。

MachineLearningBatchDeploymentPatch

严格用于更新请求。

MachineLearningBatchDeploymentProperties

每个部署的批量推理设置。

MachineLearningBatchEndpointProperties

Batch 终结点配置。

MachineLearningBatchRetrySettings

批量推理操作的重试设置。

MachineLearningBuildContext

Docker 生成上下文的配置设置。

MachineLearningCertificateDatastoreCredentials

证书数据存储凭据配置。

MachineLearningCertificateDatastoreSecrets

数据存储证书机密。

MachineLearningCodeConfiguration

评分代码资产的配置。

MachineLearningCodeContainerProperties

代码资产版本的容器。

MachineLearningCodeVersionProperties

代码资产版本详细信息。

MachineLearningCommandJob

命令作业定义。

MachineLearningCommandJobLimits

命令作业限制类。

MachineLearningComponentContainerProperties

组件容器定义。 <请参阅 href=“https://docs.microsoft.com/en-us/azure/machine-learning/reference-yaml-component-command" />

MachineLearningComponentVersionProperties

组件版本的定义:定义跨组件类型的资源。

MachineLearningComputeInstance

Azure 机器学习计算实例。

MachineLearningComputeInstanceApplication

定义 Aml 实例应用程序及其连接终结点 URI。

MachineLearningComputeInstanceAssignedUser

可分配给计算实例的用户。

MachineLearningComputeInstanceConnectivityEndpoints

定义 ComputeInstance 的所有连接终结点和属性。

MachineLearningComputeInstanceContainer

定义 Aml 实例容器。

MachineLearningComputeInstanceCreatedBy

介绍有关创建此 ComputeInstance 的用户的信息。

MachineLearningComputeInstanceDataDisk

定义 Aml 实例数据磁盘。

MachineLearningComputeInstanceDataMount

定义 Aml 实例 DataMount。

MachineLearningComputeInstanceEnvironmentInfo

环境信息。

MachineLearningComputeInstanceLastOperation

ComputeInstance 上的最后一个操作。

MachineLearningComputeInstanceProperties

计算实例属性。

MachineLearningComputeInstanceSshSettings

指定 SSH 访问的策略和设置。

MachineLearningComputePatch

AmlCompute 更新参数。

MachineLearningComputeProperties

机器学习计算对象。 请注意 MachineLearningComputeProperties ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningAksCompute、、AmlComputeMachineLearningComputeInstanceMachineLearningDataFactoryComputeMachineLearningDataLakeAnalyticsMachineLearningDatabricksComputeMachineLearningHDInsightCompute、、 MachineLearningKubernetesComputeMachineLearningSynapseSpark 和 。MachineLearningVirtualMachineCompute

MachineLearningComputeSecrets

与机器学习计算相关的机密。 对于每种计算类型,可能有所不同。 请注意 MachineLearningComputeSecrets ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningAksComputeSecretsMachineLearningDatabricksComputeSecretsMachineLearningVirtualMachineSecrets

MachineLearningComputeStartStopSchedule

计算启动停止计划属性。

MachineLearningComputeSystemService

在计算上运行的系统服务。

MachineLearningContainerRegistryCredentials

MachineLearningContainerRegistryCredentials。

MachineLearningContainerResourceRequirements

联机部署中每个容器实例的资源要求。

MachineLearningContainerResourceSettings

MachineLearningContainerResourceSettings。

MachineLearningCustomModelJobInput

MachineLearningCustomModelJobInput。

MachineLearningCustomModelJobOutput

MachineLearningCustomModelJobOutput。

MachineLearningDatabricksCompute

DataFactory 计算。

MachineLearningDatabricksComputeSecrets

与基于 Databricks 的机器学习计算相关的机密。

MachineLearningDatabricksProperties

Databricks 的属性。

MachineLearningDataContainerProperties

数据资产版本的容器。

MachineLearningDataFactoryCompute

DataFactory 计算。

MachineLearningDataLakeAnalytics

DataLakeAnalytics 计算。

MachineLearningDataPathAssetReference

通过资产在数据存储中的路径引用资产。

MachineLearningDatastoreCollectionGetAllOptions

MachineLearningDatastoreCollectionGetAllOptions。

MachineLearningDatastoreCredentials

数据存储凭据的基本定义。 请注意 MachineLearningDatastoreCredentials , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningAccountKeyDatastoreCredentials、、MachineLearningCertificateDatastoreCredentialsKerberosKeytabCredentialsKerberosPasswordCredentialsMachineLearningNoneDatastoreCredentialsMachineLearningSasDatastoreCredentialsMachineLearningServicePrincipalDatastoreCredentials

MachineLearningDatastoreProperties

数据存储内容配置的基本定义。 请注意 MachineLearningDatastoreProperties , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningAzureBlobDatastore、、MachineLearningAzureDataLakeGen1DatastoreMachineLearningAzureDataLakeGen2DatastoreMachineLearningAzureFileDatastoreHdfsDatastoreOneLakeDatastore

MachineLearningDatastoreSecrets

数据存储机密的基本定义。 请注意 MachineLearningDatastoreSecrets , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningCertificateDatastoreSecrets、、KerberosKeytabSecretsKerberosPasswordSecretsMachineLearningSasDatastoreSecretsMachineLearningServicePrincipalDatastoreSecretsMachineLearningAccountKeyDatastoreSecrets

MachineLearningDataVersionCollectionGetAllOptions

MachineLearningDataVersionCollectionGetAllOptions。

MachineLearningDataVersionProperties

数据版本基定义 请注意 MachineLearningDataVersionProperties 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningTableMachineLearningUriFileDataVersionMachineLearningUriFolderDataVersion

MachineLearningDefaultScaleSettings

MachineLearningDefaultScaleSettings。

MachineLearningDeploymentLogs

MachineLearningDeploymentLogs。

MachineLearningDeploymentLogsContent

MachineLearningDeploymentLogsContent。

MachineLearningDeploymentResourceConfiguration

MachineLearningDeploymentResourceConfiguration。

MachineLearningDiagnoseResult

诊断结果。

MachineLearningDiagnoseResultValue

MachineLearningDiagnoseResultValue。

MachineLearningDistributionConfiguration

作业分发配置的基本定义。 请注意 MachineLearningDistributionConfiguration , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 MpiDistributionConfigurationPyTorchDistributionConfigurationRayDistributionConfigurationTensorFlowDistributionConfiguration

MachineLearningEarlyTerminationPolicy

提前终止策略允许在完成之前取消性能不佳的运行。请注意 MachineLearningEarlyTerminationPolicy 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 BanditPolicyMedianStoppingPolicyTruncationSelectionPolicy

MachineLearningEncryptionKeyVaultProperties

MachineLearningEncryptionKeyVaultProperties。

MachineLearningEncryptionSetting

MachineLearningEncryptionSetting。

MachineLearningEndpointAuthKeys

终结点身份验证的密钥。

MachineLearningEndpointAuthToken

服务令牌。

MachineLearningEndpointDeploymentProperties

终结点部署的基本定义。

MachineLearningEndpointKeyRegenerateContent

MachineLearningEndpointKeyRegenerateContent。

MachineLearningEndpointProperties

推理终结点基定义。

MachineLearningEndpointScheduleAction

MachineLearningEndpointScheduleAction。

MachineLearningEnvironmentContainerProperties

环境规范版本的容器。

MachineLearningEnvironmentVersionProperties

环境版本详细信息。

MachineLearningError

对所有 Azure 资源管理器 API 的常见错误响应,可返回失败操作的错误详细信息。 (这也遵循 OData 错误响应格式.) 。

MachineLearningEstimatedVmPrice

使用特定 OS 类型、层等的 VM 的估计价格信息。

MachineLearningEstimatedVmPrices

使用 VM 的估计价格信息。

MachineLearningFeatureProperties

表示功能的 Dto 对象。

MachineLearningFeatureSetContainerCollectionGetAllOptions

MachineLearningFeatureSetContainerCollectionGetAllOptions。

MachineLearningFeatureSetContainerProperties

表示功能集的 Dto 对象。

MachineLearningFeatureSetJob

表示功能集作业的 Dto 对象。

MachineLearningFeatureSetVersionCollectionGetAllOptions

MachineLearningFeatureSetVersionCollectionGetAllOptions。

MachineLearningFeatureSetVersionProperties

表示功能集版本的 Dto 对象。

MachineLearningFeatureStoreEntityContainerCollectionGetAllOptions

MachineLearningFeatureStoreEntityContainerCollectionGetAllOptions。

MachineLearningFeatureStoreEntityContainerProperties

表示特征实体的 Dto 对象。

MachineLearningFeaturestoreEntityVersionCollectionGetAllOptions

MachineLearningFeaturestoreEntityVersionCollectionGetAllOptions。

MachineLearningFeatureStoreEntityVersionProperties

表示功能实体版本的 Dto 对象。

MachineLearningFeaturizationSettings

特征化配置。

MachineLearningFlavorData

MachineLearningFlavorData。

MachineLearningFlowModelJobInput

MachineLearningFlowModelJobInput。

MachineLearningFlowModelJobOutput

MachineLearningFlowModelJobOutput。

MachineLearningForecasting

AutoML 表垂直中的预测任务。

MachineLearningFqdnEndpoint

MachineLearningFqdnEndpoint。

MachineLearningFqdnEndpointDetail

MachineLearningFqdnEndpointDetail。

MachineLearningFqdnEndpoints

FQDN 终结点结果的属性包。

MachineLearningFqdnEndpointsProperties

MachineLearningFqdnEndpointsProperties。

MachineLearningHDInsightCompute

HDInsight 计算。

MachineLearningHDInsightProperties

HDInsight 计算属性。

MachineLearningIdAssetReference

通过其 ARM 资源 ID 引用资产。

MachineLearningIdentityConfiguration

标识配置的基本定义。 请注意 MachineLearningIdentityConfiguration , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 AmlTokenMachineLearningManagedIdentityMachineLearningUserIdentity

MachineLearningInferenceContainerProperties

MachineLearningInferenceContainerProperties。

MachineLearningInferenceContainerRoute

MachineLearningInferenceContainerRoute。

MachineLearningInstanceTypeSchema

实例类型架构。

MachineLearningInstanceTypeSchemaResources

此实例类型的资源请求/限制。

MachineLearningJobCollectionGetAllOptions

MachineLearningJobCollectionGetAllOptions。

MachineLearningJobInput

命令作业定义。 请注意 MachineLearningJobInput , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningCustomModelJobInput、、MachineLearningLiteralJobInputMachineLearningFlowModelJobInputMachineLearningTableJobInputMachineLearningTritonModelJobInputMachineLearningUriFileJobInputMachineLearningUriFolderJobInput

MachineLearningJobLimits

MachineLearningJobLimits。 请注意 MachineLearningJobLimits , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningSweepJobLimitsMachineLearningCommandJobLimits

MachineLearningJobOutput

有关在何处查找作业输出/日志的作业输出定义容器信息。 请注意 MachineLearningJobOutput , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningCustomModelJobOutput、、MachineLearningFlowModelJobOutputMachineLearningTableJobOutputMachineLearningTritonModelJobOutputMachineLearningUriFileJobOutputMachineLearningUriFolderJobOutput

MachineLearningJobPatch

Azure 资源管理器更新请求中严格使用的资源信封。

MachineLearningJobProperties

作业的基本定义。 请注意 MachineLearningJobProperties , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 AutoMLJob、、MachineLearningCommandJobLabelingJobPropertiesMachineLearningPipelineJobSparkJobMachineLearningSweepJob

MachineLearningJobResourceConfiguration

MachineLearningJobResourceConfiguration。

MachineLearningJobScheduleAction

MachineLearningJobScheduleAction。

MachineLearningJobService

作业终结点定义。

MachineLearningKubernetesCompute

基于 Kubernetes 计算的机器学习计算。

MachineLearningKubernetesOnlineDeployment

特定于 KubernetesOnlineDeployment 的属性。

MachineLearningKubernetesProperties

Kubernetes 属性。

MachineLearningLiteralJobInput

文本输入类型。

MachineLearningManagedIdentity

托管标识配置。

MachineLearningManagedIdentityAuthTypeWorkspaceConnection

MachineLearningManagedIdentityAuthTypeWorkspaceConnection。

MachineLearningManagedOnlineDeployment

特定于 ManagedOnlineDeployment 的属性。

MachineLearningModelContainerProperties

MachineLearningModelContainerProperties。

MachineLearningModelVersionCollectionGetAllOptions

MachineLearningModelVersionCollectionGetAllOptions。

MachineLearningModelVersionProperties

模型资产版本详细信息。

MachineLearningNodeStateCounts

amlCompute 上各种计算节点状态的计数。

MachineLearningNoneAuthTypeWorkspaceConnection

MachineLearningNoneAuthTypeWorkspaceConnection。

MachineLearningNoneDatastoreCredentials

数据存储凭据为空/无。

MachineLearningNotebookPreparationError

MachineLearningNotebookPreparationError。

MachineLearningNotebookResourceInfo

MachineLearningNotebookResourceInfo。

MachineLearningObjective

优化目标。

MachineLearningOnlineDeploymentPatch

严格用于更新请求。

MachineLearningOnlineDeploymentProperties

MachineLearningOnlineDeploymentProperties。 请注意 MachineLearningOnlineDeploymentProperties , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningKubernetesOnlineDeploymentMachineLearningManagedOnlineDeployment

MachineLearningOnlineEndpointCollectionGetAllOptions

MachineLearningOnlineEndpointCollectionGetAllOptions。

MachineLearningOnlineEndpointProperties

联机终结点配置。

MachineLearningOnlineRequestSettings

联机部署评分请求配置。

MachineLearningOnlineScaleSettings

联机部署缩放配置。 请注意 MachineLearningOnlineScaleSettings , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningDefaultScaleSettingsMachineLearningTargetUtilizationScaleSettings

MachineLearningOutboundRule

机器学习工作区的托管网络的出站规则。 请注意 MachineLearningOutboundRule , 是基类。 根据方案,可能需要在此处分配基类的派生类,或者需要将此属性强制转换为可能的派生类之一。 可用的派生类包括 FqdnOutboundRulePrivateEndpointOutboundRuleServiceTagOutboundRule

MachineLearningOutputPathAssetReference

通过资产在作业输出中的路径引用资产。

MachineLearningPartialManagedServiceIdentity

托管服务标识 (系统分配的标识和/或用户分配的标识)

MachineLearningPasswordDetail

MachineLearningPasswordDetail。

MachineLearningPatAuthTypeWorkspaceConnection

MachineLearningPatAuthTypeWorkspaceConnection。

MachineLearningPipelineJob

管道作业定义:定义 MFE 属性的泛型。

MachineLearningPrivateEndpoint

专用终结点资源。

MachineLearningPrivateLinkResource

专用链接资源。

MachineLearningPrivateLinkServiceConnectionState

有关服务使用者和提供程序之间的连接状态的信息集合。

MachineLearningProbeSettings

部署容器运行情况/就绪情况探测配置。

MachineLearningQuotaProperties

配额更新或检索的属性。

MachineLearningQuotaUpdateContent

配额更新参数。

MachineLearningRecurrenceSchedule

MachineLearningRecurrenceSchedule。

MachineLearningRecurrenceTrigger

MachineLearningRecurrenceTrigger。

MachineLearningRegistryModelVersionCollectionGetAllOptions

MachineLearningRegistryModelVersionCollectionGetAllOptions。

MachineLearningRegistryPatch

严格用于更新请求。

MachineLearningResourceBase

MachineLearningResourceBase。

MachineLearningResourceConfiguration

MachineLearningResourceConfiguration。

MachineLearningResourceName

资源名称。

MachineLearningResourcePatch

严格用于更新请求。

MachineLearningResourcePatchWithIdentity

严格用于更新请求。

MachineLearningResourceQuota

分配给资源的配额。

MachineLearningSasAuthTypeWorkspaceConnection

MachineLearningSasAuthTypeWorkspaceConnection。

MachineLearningSasDatastoreCredentials

SAS 数据存储凭据配置。

MachineLearningSasDatastoreSecrets

数据存储 SAS 机密。

MachineLearningScheduleAction

MachineLearningScheduleAction。 请注意 MachineLearningScheduleAction ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningJobScheduleActionCreateMonitorActionImportDataActionMachineLearningEndpointScheduleAction

MachineLearningScheduleBase

MachineLearningScheduleBase。

MachineLearningScheduleProperties

计划的基本定义。

MachineLearningScriptReference

脚本参考。

MachineLearningScriptsToExecute

自定义安装脚本。

MachineLearningServicePrincipalDatastoreCredentials

服务主体数据存储凭据配置。

MachineLearningServicePrincipalDatastoreSecrets

数据存储服务主体机密。

MachineLearningSharedPrivateLinkResource

MachineLearningSharedPrivateLinkResource。

MachineLearningSku

表示 SKU 的资源模型定义。

MachineLearningSkuCapacity

SKU 容量信息。

MachineLearningSkuDetail

满足 ARM 协定要求以列出资源的所有可用 SKU。

MachineLearningSkuPatch

通用 SKU 定义。

MachineLearningSkuSetting

SkuSetting 满足在 ARM 协定中剥离 SKU 信息的需求。

MachineLearningSslConfiguration

用于评分的 ssl 配置。

MachineLearningStackEnsembleSettings

高级设置以自定义 StackEnsemble 运行。

MachineLearningSweepJob

扫描作业定义。

MachineLearningSweepJobLimits

扫描作业限制类。

MachineLearningSynapseSpark

SynapseSpark 计算。

MachineLearningSynapseSparkProperties

MachineLearningSynapseSparkProperties。

MachineLearningTable

MLTable 数据定义。

MachineLearningTableJobInput

MachineLearningTableJobInput。

MachineLearningTableJobOutput

MachineLearningTableJobOutput。

MachineLearningTargetUtilizationScaleSettings

MachineLearningTargetUtilizationScaleSettings。

MachineLearningTrainingSettings

与训练相关的配置。

MachineLearningTrialComponent

试用组件定义。

MachineLearningTriggerBase

MachineLearningTriggerBase。 请注意 MachineLearningTriggerBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 CronTriggerMachineLearningRecurrenceTrigger

MachineLearningTritonModelJobInput

MachineLearningTritonModelJobInput。

MachineLearningTritonModelJobOutput

MachineLearningTritonModelJobOutput。

MachineLearningUriFileDataVersion

uri-file 数据版本实体。

MachineLearningUriFileJobInput

MachineLearningUriFileJobInput。

MachineLearningUriFileJobOutput

MachineLearningUriFileJobOutput。

MachineLearningUriFolderDataVersion

uri-folder 数据版本实体。

MachineLearningUriFolderJobInput

MachineLearningUriFolderJobInput。

MachineLearningUriFolderJobOutput

MachineLearningUriFolderJobOutput。

MachineLearningUsage

介绍 AML 资源使用情况。

MachineLearningUsageName

用法名称。

MachineLearningUserAccountCredentials

在计算节点上创建的每个用户帐户的设置。

MachineLearningUserFeature

为工作区启用的功能。

MachineLearningUserIdentity

用户标识配置。

MachineLearningUsernamePasswordAuthTypeWorkspaceConnection

MachineLearningUsernamePasswordAuthTypeWorkspaceConnection。

MachineLearningVirtualMachineCompute

基于 Azure 虚拟机的机器学习计算。

MachineLearningVirtualMachineProperties

MachineLearningVirtualMachineProperties。

MachineLearningVirtualMachineSecrets

与基于 AKS 的机器学习计算相关的机密。

MachineLearningVmSize

描述 VM 大小的属性。

MachineLearningVmSshCredentials

管理员虚拟机的凭据。

MachineLearningWebhook

Webhook 基请注意 MachineLearningWebhook 是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 AzureDevOpsWebhook

MachineLearningWorkspaceConnectionManagedIdentity

MachineLearningWorkspaceConnectionManagedIdentity。

MachineLearningWorkspaceConnectionPatch

将用于更新机器学习工作区连接的属性。

MachineLearningWorkspaceConnectionProperties

MachineLearningWorkspaceConnectionProperties。 请注意 MachineLearningWorkspaceConnectionProperties ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningManagedIdentityAuthTypeWorkspaceConnectionMachineLearningNoneAuthTypeWorkspaceConnectionMachineLearningPatAuthTypeWorkspaceConnectionMachineLearningSasAuthTypeWorkspaceConnectionMachineLearningUsernamePasswordAuthTypeWorkspaceConnection

MachineLearningWorkspaceConnectionUsernamePassword

MachineLearningWorkspaceConnectionUsernamePassword。

MachineLearningWorkspaceDiagnoseContent

用于诊断工作区的参数。

MachineLearningWorkspaceDiagnoseProperties

MachineLearningWorkspaceDiagnoseProperties。

MachineLearningWorkspaceDiagnoseResult

MachineLearningWorkspaceDiagnoseResult。

MachineLearningWorkspaceGetKeysResult

MachineLearningWorkspaceGetKeysResult。

MachineLearningWorkspaceGetNotebookKeysResult

MachineLearningWorkspaceGetNotebookKeysResult。

MachineLearningWorkspaceGetStorageAccountKeysResult

MachineLearningWorkspaceGetStorageAccountKeysResult。

MachineLearningWorkspaceNotebookAccessTokenResult

MachineLearningWorkspaceNotebookAccessTokenResult。

MachineLearningWorkspacePatch

用于更新机器学习工作区的参数。

MachineLearningWorkspaceQuotaUpdate

更新配额响应的属性。

ManagedComputeIdentity

托管计算标识定义。

ManagedNetworkProvisionContent

机器学习工作区托管网络的托管网络预配选项。

ManagedNetworkProvisionStatus

机器学习工作区托管网络的预配状态。

ManagedNetworkSettings

机器学习工作区的托管网络设置。

MaterializationSettings

MaterializationSettings。

MedianStoppingPolicy

基于所有运行的主要指标的移动平均定义提前终止策略。

MLAssistConfigurationDisabled

禁用 MLAssist 时标记 MLAssist 配置定义。

ModelConfiguration

模型配置选项。

ModelPackageContent

模型包操作请求属性。

ModelPackageInput

模型包输入选项。

ModelPackageResult

异步包操作成功完成后返回的包响应。

ModelPerformanceMetricThresholdBase

ModelPerformanceMetricThresholdBase。 请注意 ModelPerformanceMetricThresholdBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 ClassificationModelPerformanceMetricThresholdRegressionModelPerformanceMetricThreshold

ModelPerformanceSignal

模型性能信号定义。

MonitorComputeConfigurationBase

监视计算配置基定义。 请注意 MonitorComputeConfigurationBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 MonitorServerlessSparkCompute

MonitorComputeIdentityBase

监视计算标识基定义。 请注意 MonitorComputeIdentityBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 AmlTokenComputeIdentityManagedComputeIdentity

MonitorDefinition

MonitorDefinition。

MonitoringAlertNotificationSettingsBase

MonitoringAlertNotificationSettingsBase。 请注意 MonitoringAlertNotificationSettingsBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 AzMonMonitoringAlertNotificationSettingsEmailMonitoringAlertNotificationSettings

MonitoringDataSegment

MonitoringDataSegment。

MonitoringFeatureFilterBase

MonitoringFeatureFilterBase。 请注意 MonitoringFeatureFilterBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 MachineLearningAllFeaturesFeatureSubsetTopNFeaturesByAttribution

MonitoringInputDataBase

监视输入数据基定义。 请注意 MonitoringInputDataBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 FixedInputDataStaticInputDataTrailingInputData

MonitoringSignalBase

MonitoringSignalBase。 请注意 MonitoringSignalBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 CustomMonitoringSignal、、DataDriftMonitoringSignalDataQualityMonitoringSignalFeatureAttributionDriftMonitoringSignalGenerationSafetyQualityMonitoringSignalGenerationTokenStatisticsSignalModelPerformanceSignalPredictionDriftMonitoringSignal

MonitoringTarget

监视目标定义。

MonitoringWorkspaceConnection

监视工作区连接定义。

MonitorServerlessSparkCompute

监视无服务器 Spark 计算定义。

MountBindOptions

MountBindOptions。

MpiDistributionConfiguration

MPI 分发配置。

NCrossValidations

N 交叉验证值。 请注意 NCrossValidations ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 AutoNCrossValidationsCustomNCrossValidations

NlpFixedParameters

修复了在 AutoML NLP 训练期间不会被扫过的训练参数。

NlpParameterSubspace

每个参数的字符串化搜索空间。 请参阅以下示例。

NlpSweepSettings

与模型扫描和超参数优化相关的设置。

NlpVerticalLimitSettings

作业执行约束。

NotificationSetting

通知的配置。

NumericalDataDriftMetricThreshold

NumericalDataDriftMetricThreshold。

NumericalDataQualityMetricThreshold

NumericalDataQualityMetricThreshold。

NumericalPredictionDriftMetricThreshold

NumericalPredictionDriftMetricThreshold。

OneLakeArtifact

OneLake 项目 (数据源) 配置。 请注意 OneLakeArtifact ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 LakeHouseArtifact

OneLakeDatastore

OneLake (Trident) 数据存储配置。

OnlineInferenceConfiguration

联机推理配置选项。

PackageInputPathBase

PackageInputPathBase。 请注意 PackageInputPathBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 PackageInputPathIdPackageInputPathVersionPackageInputPathUri

PackageInputPathId

使用资源 ID 指定的包输入路径。

PackageInputPathUri

指定为 URL 的包输入路径。

PackageInputPathVersion

使用名称和版本指定的包输入路径。

PendingUploadCredentialDto

PendingUploadCredentialDto。 请注意 PendingUploadCredentialDto ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 SasCredentialDto

PendingUploadRequestDto

PendingUploadRequestDto。

PendingUploadResponseDto

PendingUploadResponseDto。

PredictionDriftMetricThresholdBase

PredictionDriftMetricThresholdBase。 请注意 PredictionDriftMetricThresholdBase ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 CategoricalPredictionDriftMetricThresholdNumericalPredictionDriftMetricThreshold

PredictionDriftMonitoringSignal

PredictionDriftMonitoringSignal。

PrivateEndpointBase

专用终结点资源。

PrivateEndpointDestination

机器学习工作区托管网络的专用终结点出站规则的专用终结点目标。

PrivateEndpointOutboundRule

机器学习工作区托管网络的专用终结点出站规则。

ProgressMetrics

进度指标定义。

PyTorchDistributionConfiguration

PyTorch 分发配置。

RandomSamplingAlgorithm

定义随机生成值的采样算法。

RayDistributionConfiguration

光线分布配置。

RegistryAcrDetails

要用于注册表的 ACR 帐户的详细信息。

RegistryPrivateEndpoint

链接到此 PE 连接的 PE 网络资源。

RegistryPrivateEndpointConnection

专用终结点连接定义。

RegistryPrivateLinkServiceConnectionState

连接状态。

RegistryRegionArmDetails

注册表所在的每个区域的详细信息。

RegressionModelPerformanceMetricThreshold

RegressionModelPerformanceMetricThreshold。

RegressionTrainingSettings

回归训练相关配置。

SamplingAlgorithm

用于生成超参数值的采样算法以及用于配置算法的属性请注意 SamplingAlgorithm 是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 BayesianSamplingAlgorithmGridSamplingAlgorithmRandomSamplingAlgorithm

SasCredentialDto

SasCredentialDto。

SecretConfiguration

机密配置定义。

ServicePrincipalAuthTypeWorkspaceConnectionProperties

ServicePrincipalAuthTypeWorkspaceConnectionProperties。

ServiceTagDestination

机器学习工作区托管网络的服务标记出站规则的服务标记目标。

ServiceTagOutboundRule

机器学习工作区托管网络的服务标记出站规则。

SparkJob

Spark 作业定义。

SparkJobEntry

Spark 作业入口点定义。 请注意 SparkJobEntry ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 SparkJobPythonEntrySparkJobScalaEntry

SparkJobPythonEntry

SparkJobPythonEntry。

SparkJobScalaEntry

SparkJobScalaEntry。

SparkResourceConfiguration

SparkResourceConfiguration。

StaticInputData

静态输入数据定义。

StorageAccountDetails

要用于注册表的存储帐户的详细信息。

SystemCreatedAcrAccount

SystemCreatedAcrAccount。

SystemCreatedStorageAccount

SystemCreatedStorageAccount。

TableFixedParameters

修复了在 AutoML 表训练期间不会遍历的训练参数。

TableParameterSubspace

TableParameterSubspace。

TableSweepSettings

TableSweepSettings。

TableVerticalFeaturizationSettings

特征化配置。

TableVerticalLimitSettings

作业执行约束。

TargetLags

滞后于目标列的过去周期数。 请注意 TargetLags ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 AutoTargetLagsCustomTargetLags

TargetRollingWindowSize

预测目标滚动窗口大小。 请注意 TargetRollingWindowSize ,是基类。 根据方案,可能需要在此处分配基类的派生类,或者此属性需要强制转换为可能的派生类之一。 可用的派生类包括 AutoTargetRollingWindowSizeCustomTargetRollingWindowSize

TensorFlowDistributionConfiguration

TensorFlow 分发配置。

TextClassification

AutoML NLP 垂直中的文本分类任务。 NLP - 自然语言处理。

TextClassificationMultilabel

AutoML NLP 垂直中的文本分类多标签任务。 NLP - 自然语言处理。

TextNer

AutoML NLP 垂直中的文本 NER 任务。 NER - 命名实体识别。 NLP - 自然语言处理。

TopNFeaturesByAttribution

TopNFeaturesByAttribution。

TrailingInputData

尾随输入数据定义。

TritonInferencingServer

Triton 推理服务器配置。

TruncationSelectionPolicy

定义一个提前终止策略,该策略在每个评估间隔内取消给定百分比的运行。

VolumeDefinition

VolumeDefinition。

WorkspaceConnectionAccessKey

WorkspaceConnectionAccessKey。

WorkspaceConnectionServicePrincipal

WorkspaceConnectionServicePrincipal。

WorkspaceHubConfig

WorkspaceHub 的配置对象。

结构

AutoDeleteCondition

AutoDeleteCondition。

AutoMLVerticalRegressionModel

AutoML 支持的所有回归模型的枚举。

AutoMLVerticalRegressionPrimaryMetric

回归任务的主要指标。

AutoRebuildSetting

派生映像的“自动重新生成”设置。

BlockedTransformer

AutoML 支持的所有分类模型的枚举。

CategoricalDataDriftMetric

CategoricalDataDriftMetric。

CategoricalDataQualityMetric

CategoricalDataQualityMetric。

CategoricalPredictionDriftMetric

CategoricalPredictionDriftMetric。

ClassificationModel

AutoML 支持的所有分类模型的枚举。

ClassificationModelPerformanceMetric

ClassificationModelPerformanceMetric。

ClassificationMultilabelPrimaryMetric

分类多标签任务的主要指标。

ClassificationPrimaryMetric

分类任务的主要指标。

ContainerCommunicationProtocol

通过此终结点进行通信的协议。

DataCollectionMode

DataCollectionMode。

EmailNotificationEnableType

用于确定电子邮件通知类型的枚举。

EnvironmentVariableType

环境变量的类型。 可能的值为:local - 局部变量。

FeatureAttributionMetric

FeatureAttributionMetric。

FeatureDataType

FeatureDataType。

FeatureStoreJobType

FeatureStoreJobType。

ForecastingModel

AutoML 支持的所有预测模型的枚举。

ForecastingPrimaryMetric

预测任务的主要指标。

GenerationSafetyQualityMetric

生成安全质量指标枚举。

GenerationTokenStatisticsMetric

生成令牌统计信息指标枚举。

ImageAnnotationType

图像数据的批注类型。

ImageType

图像的类型。 可能的值为:docker - 用于 docker 映像。 azureml - 适用于 AzureML 映像。

IncrementalDataRefresh

是否启用 IncrementalDataRefresh。

InstanceSegmentationPrimaryMetric

InstanceSegmentation 任务的主要指标。

IntellectualProtectionLevel

与知识产权关联的保护级别。

IsolationMode

机器学习工作区的托管网络的隔离模式。

JobInputType

用于确定作业输入类型的枚举。

JobProvisioningState

用于确定作业预配状态的枚举。

JobStatusMessageLevel

JobStatusMessageLevel。

JobTier

用于确定作业层的枚举。

LabelCategoryMultiSelect

是否启用了 multiSelect。

LearningRateScheduler

学习速率计划程序枚举。

LogTrainingMetric

LogTrainingMetric。

LogValidationLoss

LogValidationLoss。

MachineLearningAllocationState

计算的分配状态。 可能的值为:稳定 - 指示计算未调整大小。 正在进行的计算节点数没有变化。 计算在创建时进入此状态,并且未对计算执行任何操作来更改计算节点数。 调整大小 - 指示计算正在调整大小;也就是说,正在计算中添加或删除计算节点。

MachineLearningApplicationSharingPolicy

用于在父工作区的用户之间共享此计算实例上的应用程序的策略。 如果为“个人”,则只有创建者可以访问此计算实例上的应用程序。 共享时,任何工作区用户都可以访问此实例上的应用程序,具体取决于他/她分配的角色。

MachineLearningBatchLoggingLevel

记录批处理推理的详细程度。 增加日志记录的详细顺序为:警告、信息和调试。 默认值为“信息”。

MachineLearningBatchOutputAction

用于确定批处理推理如何处理输出的枚举。

MachineLearningBillingCurrency

三个字母代码,指定 VM 价格的货币。 示例:USD。

MachineLearningCachingType

数据磁盘的缓存类型。

MachineLearningClusterPurpose

群集的预期用法。

MachineLearningComputeInstanceAuthorizationType

计算实例授权类型。 可用值为个人 (默认) 。

MachineLearningComputeInstanceAutosave

自动保存设置。

MachineLearningComputeInstanceState

ComputeInstance 的当前状态。

MachineLearningComputePowerAction

[必需]计算能力操作。

MachineLearningComputeProvisioningStatus

计划的当前部署状态。

MachineLearningConnectionCategory

连接的类别。

MachineLearningContainerType

要从中检索日志的容器的类型。

MachineLearningDataType

用于确定数据类型的枚举。

MachineLearningDayOfWeek

工作日的枚举。

MachineLearningDeploymentProvisioningState

DeploymentProvisioningState 的可能值。

MachineLearningDiagnoseResultLevel

工作区设置错误的级别。

MachineLearningEgressPublicNetworkAccessType

用于确定对部署的出口是启用还是禁用 PublicNetworkAccess 的枚举。

MachineLearningEncryptionStatus

指示是否为工作区启用加密。

MachineLearningEndpointAuthMode

用于确定终结点身份验证模式的枚举。

MachineLearningEndpointComputeType

用于确定终结点计算类型的枚举。

MachineLearningEndpointProvisioningState

终结点预配的状态。

MachineLearningEnvironmentType

环境类型由 Azure ML 服务创建或策展。

MachineLearningFeatureLag

用于为数字特征生成滞后的标志。

MachineLearningFeaturizationMode

特征化模式 - 确定数据特征化模式。

MachineLearningFlowAutoLogger

指示是否为笔记本启用 mlflow 自动记录器。

MachineLearningFlowAutoLoggerState

用于确定 mlflow 自动记录器状态的枚举。

MachineLearningGoal

定义超参数优化支持的指标目标。

MachineLearningInputDeliveryMode

用于确定输入数据传递模式的枚举。

MachineLearningJobStatus

作业的状态。

MachineLearningKeyType

MachineLearningKeyType。

MachineLearningListViewType

MachineLearningListViewType。

MachineLearningLoadBalancerType

负载均衡器类型。

MachineLearningLogVerbosity

用于设置日志详细性的枚举。

MachineLearningModelSize

图像模型大小。

MachineLearningMountAction

装载操作。

MachineLearningMountState

装载状态。

MachineLearningNetwork

此容器的网络。

MachineLearningNodeState

计算节点的状态。 值是空闲、正在运行、正在准备、不可用、离开和抢占。

MachineLearningOperatingSystemType

操作系统的类型。

MachineLearningOperationName

最后一个操作的名称。

MachineLearningOperationStatus

操作状态。

MachineLearningOperationTrigger

操作的触发器。

MachineLearningOrderString

MachineLearningOrderString。

MachineLearningOSType

计算 OS 类型。

MachineLearningOutputDeliveryMode

输出数据传递模式枚举。

MachineLearningPrivateEndpointConnectionProvisioningState

当前预配状态。

MachineLearningPrivateEndpointServiceConnectionStatus

服务使用者与服务提供商的连接状态。

MachineLearningProvisioningState

群集的预配状态。 有效值为 Unknown、Updating、Provisioning、Succeeded 和 Failed。

MachineLearningPublicNetworkAccess

是否允许来自公用网络的请求。

MachineLearningPublicNetworkAccessType

用于确定 PublicNetworkAccess 是已启用还是已禁用的枚举。

MachineLearningQuotaUnit

描述配额度量单位的枚举。

MachineLearningRecurrenceFrequency

用于描述定期计划频率的枚举。

MachineLearningRemoteLoginPortPublicAccess

公共 SSH 端口的状态。 可能的值为:Disabled - 指示群集的所有节点上的公共 ssh 端口已关闭。 已启用 - 指示群集的所有节点上的公用 ssh 端口已打开。 NotSpecified - 指示如果定义了 VNet,则群集的所有节点上的公共 ssh 端口已关闭,否则将打开所有公共节点。 它只能在创建群集时处于默认状态,在创建后,它将启用或禁用。

MachineLearningScheduleListViewType

MachineLearningScheduleListViewType。

MachineLearningScheduleProvisioningState

计划的当前部署状态。

MachineLearningScheduleProvisioningStatus

MachineLearningScheduleProvisioningStatus。

MachineLearningScheduleStatus

计划是启用还是禁用?。

MachineLearningServiceDataAccessAuthIdentity

MachineLearningServiceDataAccessAuthIdentity。

MachineLearningShortSeriesHandlingConfiguration

定义 AutoML 应如何处理短时序的参数。

MachineLearningSkuScaleType

计算 SKU 的节点缩放设置。

MachineLearningSourceType

数据源类型。

MachineLearningSshPublicAccess

公共 SSH 端口的状态。 可能的值为:Disabled - 指示在此实例上关闭公共 SSH 端口。 已启用 - 指示公共 ssh 端口已打开,并且根据 VNet/子网策略(如果适用)进行访问。

MachineLearningSslConfigStatus

启用或禁用用于评分的 ssl。

MachineLearningStackMetaLearnerType

元学习器是在单个异类模型的输出上训练的模型。 如果) 启用了交叉验证,则默认元学习者是用于分类任务的 LogisticRegression (或 LogisticRegressionCV;如果) 启用了交叉验证,则为 ElasticNet 对回归/预测 (任务启用 ElasticNet 或 ElasticNetCV。 此参数可以是以下字符串之一:LogisticRegression、LogisticRegressionCV、LightGBMClassifier、ElasticNet、ElasticNetCV、LightGBMRegressor 或 LinearRegression

MachineLearningStorageAccountType

此存储帐户的类型。

MachineLearningTriggerType

MachineLearningTriggerType。

MachineLearningUnderlyingResourceAction

MachineLearningUnderlyingResourceAction。

MachineLearningUnitOfMeasure

指定 VM 价格的时间度量单位。 示例:OneHour。

MachineLearningUsageUnit

描述使用情况度量单位的枚举。

MachineLearningUseStl

配置时序目标列的 STL 分解。

MachineLearningVmPriceOSType

VM 使用的操作系统类型。

MachineLearningVmPriority

虚拟机优先级。

MachineLearningVmTier

VM 的类型。

MachineLearningWorkspaceQuotaStatus

更新工作区配额的状态。

ManagedNetworkStatus

机器学习工作区的托管网络的状态。

MaterializationStoreType

MaterializationStoreType。

ModelTaskType

模型任务类型枚举。

MonitoringFeatureDataType

MonitoringFeatureDataType。

MonitoringModelType

MonitoringModelType。

MonitoringNotificationMode

MonitoringNotificationMode。

NetworkingRuleAction

网络规则的操作枚举。

NlpLearningRateScheduler

与 HF 支持的学习速率计划程序一致的枚举。

NumericalDataDriftMetric

NumericalDataDriftMetric。

NumericalDataQualityMetric

NumericalDataQualityMetric。

NumericalPredictionDriftMetric

NumericalPredictionDriftMetric。

ObjectDetectionPrimaryMetric

Image ObjectDetection 任务的主要指标。

OutboundRuleCategory

机器学习工作区的托管网络出站规则的类别。

OutboundRuleStatus

机器学习工作区的托管网络出站规则的类型。

PackageBuildState

包响应中返回的包生成状态。

PackageInputDeliveryMode

模型或输入的装载类型。

PackageInputType

输入的类型。

PendingUploadType

要用于挂起上传位置的存储类型。

RandomSamplingAlgorithmRule

随机算法的特定类型。

RegistryAssetProvisioningState

预配注册表资产的状态。

RegressionModelPerformanceMetric

RegressionModelPerformanceMetric。

RollingRateType

RollingRateType。

SamplingAlgorithmType

SamplingAlgorithmType。

StochasticOptimizer

图像模型的随机优化器。

TargetAggregationFunction

目标聚合函数。

TextAnnotationType

文本数据的批注类型。

TrainingMode

训练模式指示是否使用分布式训练。

ValidationMetricType

用于图像任务中的验证指标的指标计算方法。

VolumeDefinitionType

卷定义的类型。 可能的值:bind、volume、tmpfs、npipe。

枚举

MachineLearningSkuTier

如果服务具有多个层,但 PUT 上不需要此字段,则资源提供程序必须实现此字段。