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。
Bicep 资源定义
可以使用目标作部署工作区/标记作业资源类型:
- 资源组 - 请参阅 资源组部署命令
有关每个 API 版本中已更改属性的列表,请参阅 更改日志。
资源格式
若要创建 Microsoft.MachineLearningServices/workspaces/labelingJobs 资源,请将以下 Bicep 添加到模板。
resource symbolicname 'Microsoft.MachineLearningServices/workspaces/labelingJobs@2022-12-01-preview' = {
parent: resourceSymbolicName
name: 'string'
properties: {
componentId: 'string'
computeId: 'string'
dataConfiguration: {
dataId: 'string'
incrementalDataRefresh: 'string'
}
description: 'string'
displayName: 'string'
experimentName: 'string'
identity: {
identityType: 'string'
// For remaining properties, see IdentityConfiguration objects
}
isArchived: bool
jobInstructions: {
uri: 'string'
}
jobType: 'string'
labelCategories: {
{customized property}: {
classes: {
{customized property}: {
displayName: 'string'
subclasses: {
{customized property}: ...
}
}
}
displayName: 'string'
multiSelect: 'string'
}
}
labelingJobMediaProperties: {
mediaType: 'string'
// For remaining properties, see LabelingJobMediaProperties objects
}
mlAssistConfiguration: {
mlAssist: 'string'
// For remaining properties, see MLAssistConfiguration objects
}
properties: {
{customized property}: 'string'
}
services: {
{customized property}: {
endpoint: 'string'
jobServiceType: 'string'
nodes: {
nodesValueType: 'string'
// For remaining properties, see Nodes objects
}
port: int
properties: {
{customized property}: 'string'
}
}
}
tags: {
{customized property}: 'string'
}
}
}
MLAssistConfiguration 对象
设置 mlAssist 属性以指定对象的类型。
对于 禁用,请使用:
{
mlAssist: 'Disabled'
}
对于 “已启用”,请使用:
{
inferencingComputeBinding: 'string'
mlAssist: 'Enabled'
trainingComputeBinding: 'string'
}
节点对象
设置 nodesValueType 属性以指定对象的类型。
对于 “全部”,请使用:
{
nodesValueType: 'All'
}
IdentityConfiguration 对象
设置 identityType 属性以指定对象的类型。
对于 AMLToken,请使用:
{
identityType: 'AMLToken'
}
对于 托管,请使用:
{
clientId: 'string'
identityType: 'Managed'
objectId: 'string'
resourceId: 'string'
}
对于 UserIdentity,请使用:
{
identityType: 'UserIdentity'
}
LabelingJobMediaProperties 对象
设置 mediaType 属性以指定对象的类型。
对于 映像,请使用:
{
annotationType: 'string'
mediaType: 'Image'
}
对于 文本,请使用:
{
annotationType: 'string'
mediaType: 'Text'
}
属性值
Microsoft.MachineLearningServices/workspaces/labelingJobs
名称 | DESCRIPTION | 价值 |
---|---|---|
姓名 | 资源名称 | 字符串 约束: 模式 = ^[a-zA-Z0-9][a-zA-Z0-9\-_]{0,254}$ (必需) |
父级 | 在 Bicep 中,可以为子资源指定父资源。 仅当子资源在父资源外部声明时,才需要添加此属性。 有关详细信息,请参阅 父资源外部的子资源。 |
类型资源的符号名称: 工作区 |
属性 | [必需]实体的其他属性。 | LabelingJobProperties (必需) |
所有节点
名称 | DESCRIPTION | 价值 |
---|---|---|
nodesValueType | [必需]Nodes 值的类型 | “全部”(必需) |
AmlToken
名称 | DESCRIPTION | 价值 |
---|---|---|
身份类型 | [必需]指定标识框架的类型。 | “AMLToken”(必需) |
身份配置
名称 | DESCRIPTION | 价值 |
---|---|---|
身份类型 | 对于 AmlToken 类型,设置为“AMLToken”。 设置为 “ManagedIdentity”类型。 对于 UserIdentity 类型,设置为“UserIdentity”。 | “AMLToken” “Managed” “UserIdentity”(必需) |
JobBase服务
名称 | DESCRIPTION | 价值 |
---|
工作服务
名称 | DESCRIPTION | 价值 |
---|---|---|
终结点 | 终结点的 URL。 | 字符串 |
jobServiceType | 终结点类型。 | 字符串 |
节点 | 用户希望启动服务的节点。 如果未将 Nodes 设置为 null,则服务将仅在领导节点上启动。 |
节点 |
移植 | 用户设置的终结点的端口。 | 整数 (int) |
属性 | 在终结点上设置的其他属性。 | JobServiceProperties |
JobServiceProperties
名称 | DESCRIPTION | 价值 |
---|
标签类别
名称 | DESCRIPTION | 价值 |
---|---|---|
类 | 此类别中标签类的字典。 | 标签类别类 |
displayName (显示名称) | 标签类别的显示名称。 | 字符串 |
多选 | 指示是否允许在此类别中选择多个类。 | “Disabled” “Enabled” |
标签类别类
名称 | DESCRIPTION | 价值 |
---|
标签类
名称 | DESCRIPTION | 价值 |
---|---|---|
displayName (显示名称) | 标签类的显示名称。 | 字符串 |
子 | 标签类的子类的字典。 | LabelClass子类 |
LabelClass子类
名称 | DESCRIPTION | 价值 |
---|
LabelingDataConfiguration
名称 | DESCRIPTION | 价值 |
---|---|---|
数据 ID | 要执行标记的数据资产的资源 ID。 | 字符串 |
incrementalDataRefresh | 指示是否启用增量数据刷新。 | “Disabled” “Enabled” |
LabelingJobImageProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
annotationType | 图像标记作业的批注类型。 | “BoundingBox” “Classification” “InstanceSegmentation” |
mediaType | [必需]作业的媒体类型。 | “Image”(必需) |
TagJobInstructions
名称 | DESCRIPTION | 价值 |
---|---|---|
URI | 指向页面的链接,其中包含标签器的详细标记说明。 | 字符串 |
LabelingJobLabel类别
名称 | DESCRIPTION | 价值 |
---|
LabelingJobMediaProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
mediaType | 设置为“Image”,类型 为 LabelingJobImageProperties。 对于 LabelingJobTextProperties 类型,设置为“Text”。 | “Image” “Text”(必需) |
LabelingJobProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
组件 ID | 组件资源的 ARM 资源 ID。 | 字符串 |
计算 ID | 计算资源的 ARM 资源 ID。 | 字符串 |
dataConfiguration | 配置作业中使用的数据。 | LabelingDataConfiguration |
描述 | 资产说明文本。 | 字符串 |
displayName (显示名称) | 作业的显示名称。 | 字符串 |
实验名称 | 作业所属的试验的名称。 如果未设置,作业将放置在“默认”试验中。 | 字符串 |
标识 | 标识配置。 如果已设置,则应为 AmlToken、ManagedIdentity、UserIdentity 或 null 之一。 如果为 null,则默认为 AmlToken。 |
身份配置 |
isArchived 已存档 | 资产是否已存档? | 布尔 |
工作说明 | 标记作业的说明。 | TagJobInstructions |
职位类型 | [必需]指定作业的类型。 | “AutoML” “Command” “标记” “Pipeline” “Spark” “扫描”(必需) |
标签类别 | 作业的标签类别。 | LabelingJobLabel类别 |
labelingJobMediaProperties | 作业中的媒体类型特定属性。 | LabelingJobMediaProperties |
mlAssistConfiguration | 在作业中配置 MLAssist 功能。 | MLAssistConfiguration |
属性 | 资产属性字典。 | ResourceBaseProperties |
服务 | JobEndpoints 列表。 对于本地作业,作业终结点的终结点值为 FileStreamObject。 |
JobBase服务 |
标签 | 标记字典。 可以添加、删除和更新标记。 | ResourceBaseTags |
LabelingJobTextProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
annotationType | 文本标记作业的批注类型。 | “Classification” 'NamedEntityRecognition' (命名实体识别) |
mediaType | [必需]作业的媒体类型。 | “Text”(必需) |
ManagedIdentity (托管身份)
名称 | DESCRIPTION | 价值 |
---|---|---|
客户端 ID | 按客户端 ID 指定用户分配的标识。 对于系统分配,请不要设置此字段。 | 字符串 约束: 最小长度 = 36 最大长度 = 36 图案 = ^[0-9a-fA-F]{8}-([0-9a-fA-F]{4}-){3}[0-9a-fA-F]{12}$ |
身份类型 | [必需]指定标识框架的类型。 | “Managed”(必需) |
对象 ID | 按对象 ID 指定用户分配的标识。 对于系统分配,请不要设置此字段。 | 字符串 约束: 最小长度 = 36 最大长度 = 36 图案 = ^[0-9a-fA-F]{8}-([0-9a-fA-F]{4}-){3}[0-9a-fA-F]{12}$ |
资源ID | 按 ARM 资源 ID 指定用户分配的标识。 对于系统分配,请不要设置此字段。 | 字符串 |
MLAssistConfiguration
名称 | DESCRIPTION | 价值 |
---|---|---|
mlAssist | 设置为 MLAssistConfigurationDisabled 类型的“Disabled”。 设置为 MLAssistConfigurationEnabled 类型的“Enabled”。 | “Disabled” “Enabled”(必需) |
MLAssistConfigurationDisabled
名称 | DESCRIPTION | 价值 |
---|---|---|
mlAssist | [必需]指示是否启用 MLAssist 功能。 | “Disabled”(必需) |
MLAssistConfiguration已启用
名称 | DESCRIPTION | 价值 |
---|---|---|
inferencingComputeBinding | [必需]推理中使用的 AML 计算绑定。 | 字符串 约束: 模式 = [a-zA-Z0-9_] (必需) |
mlAssist | [必需]指示是否启用 MLAssist 功能。 | “Enabled”(必需) |
trainingComputeBinding 的 | [必需]训练中使用的 AML 计算绑定。 | 字符串 约束: 模式 = [a-zA-Z0-9_] (必需) |
节点
名称 | DESCRIPTION | 价值 |
---|---|---|
nodesValueType | 对于 AllNodes 类型,设置为“All”。 | “全部”(必需) |
ResourceBaseProperties
名称 | DESCRIPTION | 价值 |
---|
ResourceBaseTags
名称 | DESCRIPTION | 价值 |
---|
用户身份
名称 | DESCRIPTION | 价值 |
---|---|---|
身份类型 | [必需]指定标识框架的类型。 | “UserIdentity”(必需) |
ARM 模板资源定义
可以使用目标作部署工作区/标记作业资源类型:
- 资源组 - 请参阅 资源组部署命令
有关每个 API 版本中已更改属性的列表,请参阅 更改日志。
资源格式
若要创建 Microsoft.MachineLearningServices/workspaces/labelingJobs 资源,请将以下 JSON 添加到模板。
{
"type": "Microsoft.MachineLearningServices/workspaces/labelingJobs",
"apiVersion": "2022-12-01-preview",
"name": "string",
"properties": {
"componentId": "string",
"computeId": "string",
"dataConfiguration": {
"dataId": "string",
"incrementalDataRefresh": "string"
},
"description": "string",
"displayName": "string",
"experimentName": "string",
"identity": {
"identityType": "string"
// For remaining properties, see IdentityConfiguration objects
},
"isArchived": "bool",
"jobInstructions": {
"uri": "string"
},
"jobType": "string",
"labelCategories": {
"{customized property}": {
"classes": {
"{customized property}": {
"displayName": "string",
"subclasses": {
"{customized property}": ...
}
}
},
"displayName": "string",
"multiSelect": "string"
}
},
"labelingJobMediaProperties": {
"mediaType": "string"
// For remaining properties, see LabelingJobMediaProperties objects
},
"mlAssistConfiguration": {
"mlAssist": "string"
// For remaining properties, see MLAssistConfiguration objects
},
"properties": {
"{customized property}": "string"
},
"services": {
"{customized property}": {
"endpoint": "string",
"jobServiceType": "string",
"nodes": {
"nodesValueType": "string"
// For remaining properties, see Nodes objects
},
"port": "int",
"properties": {
"{customized property}": "string"
}
}
},
"tags": {
"{customized property}": "string"
}
}
}
MLAssistConfiguration 对象
设置 mlAssist 属性以指定对象的类型。
对于 禁用,请使用:
{
"mlAssist": "Disabled"
}
对于 “已启用”,请使用:
{
"inferencingComputeBinding": "string",
"mlAssist": "Enabled",
"trainingComputeBinding": "string"
}
节点对象
设置 nodesValueType 属性以指定对象的类型。
对于 “全部”,请使用:
{
"nodesValueType": "All"
}
IdentityConfiguration 对象
设置 identityType 属性以指定对象的类型。
对于 AMLToken,请使用:
{
"identityType": "AMLToken"
}
对于 托管,请使用:
{
"clientId": "string",
"identityType": "Managed",
"objectId": "string",
"resourceId": "string"
}
对于 UserIdentity,请使用:
{
"identityType": "UserIdentity"
}
LabelingJobMediaProperties 对象
设置 mediaType 属性以指定对象的类型。
对于 映像,请使用:
{
"annotationType": "string",
"mediaType": "Image"
}
对于 文本,请使用:
{
"annotationType": "string",
"mediaType": "Text"
}
属性值
Microsoft.MachineLearningServices/workspaces/labelingJobs
名称 | DESCRIPTION | 价值 |
---|---|---|
apiVersion | API 版本 | “2022-12-01-预览” |
姓名 | 资源名称 | 字符串 约束: 模式 = ^[a-zA-Z0-9][a-zA-Z0-9\-_]{0,254}$ (必需) |
属性 | [必需]实体的其他属性。 | LabelingJobProperties (必需) |
类型 | 资源类型 | “Microsoft.MachineLearningServices/workspaces/labelingJobs” |
所有节点
名称 | DESCRIPTION | 价值 |
---|---|---|
nodesValueType | [必需]Nodes 值的类型 | “全部”(必需) |
AmlToken
名称 | DESCRIPTION | 价值 |
---|---|---|
身份类型 | [必需]指定标识框架的类型。 | “AMLToken”(必需) |
身份配置
名称 | DESCRIPTION | 价值 |
---|---|---|
身份类型 | 对于 AmlToken 类型,设置为“AMLToken”。 设置为 “ManagedIdentity”类型。 对于 UserIdentity 类型,设置为“UserIdentity”。 | “AMLToken” “Managed” “UserIdentity”(必需) |
JobBase服务
名称 | DESCRIPTION | 价值 |
---|
工作服务
名称 | DESCRIPTION | 价值 |
---|---|---|
终结点 | 终结点的 URL。 | 字符串 |
jobServiceType | 终结点类型。 | 字符串 |
节点 | 用户希望启动服务的节点。 如果未将 Nodes 设置为 null,则服务将仅在领导节点上启动。 |
节点 |
移植 | 用户设置的终结点的端口。 | 整数 (int) |
属性 | 在终结点上设置的其他属性。 | JobServiceProperties |
JobServiceProperties
名称 | DESCRIPTION | 价值 |
---|
标签类别
名称 | DESCRIPTION | 价值 |
---|---|---|
类 | 此类别中标签类的字典。 | 标签类别类 |
displayName (显示名称) | 标签类别的显示名称。 | 字符串 |
多选 | 指示是否允许在此类别中选择多个类。 | “Disabled” “Enabled” |
标签类别类
名称 | DESCRIPTION | 价值 |
---|
标签类
名称 | DESCRIPTION | 价值 |
---|---|---|
displayName (显示名称) | 标签类的显示名称。 | 字符串 |
子 | 标签类的子类的字典。 | LabelClass子类 |
LabelClass子类
名称 | DESCRIPTION | 价值 |
---|
LabelingDataConfiguration
名称 | DESCRIPTION | 价值 |
---|---|---|
数据 ID | 要执行标记的数据资产的资源 ID。 | 字符串 |
incrementalDataRefresh | 指示是否启用增量数据刷新。 | “Disabled” “Enabled” |
LabelingJobImageProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
annotationType | 图像标记作业的批注类型。 | “BoundingBox” “Classification” “InstanceSegmentation” |
mediaType | [必需]作业的媒体类型。 | “Image”(必需) |
TagJobInstructions
名称 | DESCRIPTION | 价值 |
---|---|---|
URI | 指向页面的链接,其中包含标签器的详细标记说明。 | 字符串 |
LabelingJobLabel类别
名称 | DESCRIPTION | 价值 |
---|
LabelingJobMediaProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
mediaType | 设置为“Image”,类型 为 LabelingJobImageProperties。 对于 LabelingJobTextProperties 类型,设置为“Text”。 | “Image” “Text”(必需) |
LabelingJobProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
组件 ID | 组件资源的 ARM 资源 ID。 | 字符串 |
计算 ID | 计算资源的 ARM 资源 ID。 | 字符串 |
dataConfiguration | 配置作业中使用的数据。 | LabelingDataConfiguration |
描述 | 资产说明文本。 | 字符串 |
displayName (显示名称) | 作业的显示名称。 | 字符串 |
实验名称 | 作业所属的试验的名称。 如果未设置,作业将放置在“默认”试验中。 | 字符串 |
标识 | 标识配置。 如果已设置,则应为 AmlToken、ManagedIdentity、UserIdentity 或 null 之一。 如果为 null,则默认为 AmlToken。 |
身份配置 |
isArchived 已存档 | 资产是否已存档? | 布尔 |
工作说明 | 标记作业的说明。 | TagJobInstructions |
职位类型 | [必需]指定作业的类型。 | “AutoML” “Command” “标记” “Pipeline” “Spark” “扫描”(必需) |
标签类别 | 作业的标签类别。 | LabelingJobLabel类别 |
labelingJobMediaProperties | 作业中的媒体类型特定属性。 | LabelingJobMediaProperties |
mlAssistConfiguration | 在作业中配置 MLAssist 功能。 | MLAssistConfiguration |
属性 | 资产属性字典。 | ResourceBaseProperties |
服务 | JobEndpoints 列表。 对于本地作业,作业终结点的终结点值为 FileStreamObject。 |
JobBase服务 |
标签 | 标记字典。 可以添加、删除和更新标记。 | ResourceBaseTags |
LabelingJobTextProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
annotationType | 文本标记作业的批注类型。 | “Classification” 'NamedEntityRecognition' (命名实体识别) |
mediaType | [必需]作业的媒体类型。 | “Text”(必需) |
ManagedIdentity (托管身份)
名称 | DESCRIPTION | 价值 |
---|---|---|
客户端 ID | 按客户端 ID 指定用户分配的标识。 对于系统分配,请不要设置此字段。 | 字符串 约束: 最小长度 = 36 最大长度 = 36 图案 = ^[0-9a-fA-F]{8}-([0-9a-fA-F]{4}-){3}[0-9a-fA-F]{12}$ |
身份类型 | [必需]指定标识框架的类型。 | “Managed”(必需) |
对象 ID | 按对象 ID 指定用户分配的标识。 对于系统分配,请不要设置此字段。 | 字符串 约束: 最小长度 = 36 最大长度 = 36 图案 = ^[0-9a-fA-F]{8}-([0-9a-fA-F]{4}-){3}[0-9a-fA-F]{12}$ |
资源ID | 按 ARM 资源 ID 指定用户分配的标识。 对于系统分配,请不要设置此字段。 | 字符串 |
MLAssistConfiguration
名称 | DESCRIPTION | 价值 |
---|---|---|
mlAssist | 设置为 MLAssistConfigurationDisabled 类型的“Disabled”。 设置为 MLAssistConfigurationEnabled 类型的“Enabled”。 | “Disabled” “Enabled”(必需) |
MLAssistConfigurationDisabled
名称 | DESCRIPTION | 价值 |
---|---|---|
mlAssist | [必需]指示是否启用 MLAssist 功能。 | “Disabled”(必需) |
MLAssistConfiguration已启用
名称 | DESCRIPTION | 价值 |
---|---|---|
inferencingComputeBinding | [必需]推理中使用的 AML 计算绑定。 | 字符串 约束: 模式 = [a-zA-Z0-9_] (必需) |
mlAssist | [必需]指示是否启用 MLAssist 功能。 | “Enabled”(必需) |
trainingComputeBinding 的 | [必需]训练中使用的 AML 计算绑定。 | 字符串 约束: 模式 = [a-zA-Z0-9_] (必需) |
节点
名称 | DESCRIPTION | 价值 |
---|---|---|
nodesValueType | 对于 AllNodes 类型,设置为“All”。 | “全部”(必需) |
ResourceBaseProperties
名称 | DESCRIPTION | 价值 |
---|
ResourceBaseTags
名称 | DESCRIPTION | 价值 |
---|
用户身份
名称 | DESCRIPTION | 价值 |
---|---|---|
身份类型 | [必需]指定标识框架的类型。 | “UserIdentity”(必需) |
用法示例
Terraform (AzAPI 提供程序)资源定义
可以使用目标作部署工作区/标记作业资源类型:
- 资源组
有关每个 API 版本中已更改属性的列表,请参阅 更改日志。
资源格式
若要创建 Microsoft.MachineLearningServices/workspaces/labelingJobs 资源,请将以下 Terraform 添加到模板。
resource "azapi_resource" "symbolicname" {
type = "Microsoft.MachineLearningServices/workspaces/labelingJobs@2022-12-01-preview"
name = "string"
parent_id = "string"
body = {
properties = {
componentId = "string"
computeId = "string"
dataConfiguration = {
dataId = "string"
incrementalDataRefresh = "string"
}
description = "string"
displayName = "string"
experimentName = "string"
identity = {
identityType = "string"
// For remaining properties, see IdentityConfiguration objects
}
isArchived = bool
jobInstructions = {
uri = "string"
}
jobType = "string"
labelCategories = {
{customized property} = {
classes = {
{customized property} = {
displayName = "string"
subclasses = {
{customized property} = ...
}
}
}
displayName = "string"
multiSelect = "string"
}
}
labelingJobMediaProperties = {
mediaType = "string"
// For remaining properties, see LabelingJobMediaProperties objects
}
mlAssistConfiguration = {
mlAssist = "string"
// For remaining properties, see MLAssistConfiguration objects
}
properties = {
{customized property} = "string"
}
services = {
{customized property} = {
endpoint = "string"
jobServiceType = "string"
nodes = {
nodesValueType = "string"
// For remaining properties, see Nodes objects
}
port = int
properties = {
{customized property} = "string"
}
}
}
tags = {
{customized property} = "string"
}
}
}
}
MLAssistConfiguration 对象
设置 mlAssist 属性以指定对象的类型。
对于 禁用,请使用:
{
mlAssist = "Disabled"
}
对于 “已启用”,请使用:
{
inferencingComputeBinding = "string"
mlAssist = "Enabled"
trainingComputeBinding = "string"
}
节点对象
设置 nodesValueType 属性以指定对象的类型。
对于 “全部”,请使用:
{
nodesValueType = "All"
}
IdentityConfiguration 对象
设置 identityType 属性以指定对象的类型。
对于 AMLToken,请使用:
{
identityType = "AMLToken"
}
对于 托管,请使用:
{
clientId = "string"
identityType = "Managed"
objectId = "string"
resourceId = "string"
}
对于 UserIdentity,请使用:
{
identityType = "UserIdentity"
}
LabelingJobMediaProperties 对象
设置 mediaType 属性以指定对象的类型。
对于 映像,请使用:
{
annotationType = "string"
mediaType = "Image"
}
对于 文本,请使用:
{
annotationType = "string"
mediaType = "Text"
}
属性值
Microsoft.MachineLearningServices/workspaces/labelingJobs
名称 | DESCRIPTION | 价值 |
---|---|---|
姓名 | 资源名称 | 字符串 约束: 模式 = ^[a-zA-Z0-9][a-zA-Z0-9\-_]{0,254}$ (必需) |
parent_id | 此资源的父资源的 ID。 | 类型资源的 ID: 工作区 |
属性 | [必需]实体的其他属性。 | LabelingJobProperties (必需) |
类型 | 资源类型 | “Microsoft.MachineLearningServices/workspaces/labelingJobs@2022-12-01-preview” |
所有节点
名称 | DESCRIPTION | 价值 |
---|---|---|
nodesValueType | [必需]Nodes 值的类型 | “全部”(必需) |
AmlToken
名称 | DESCRIPTION | 价值 |
---|---|---|
身份类型 | [必需]指定标识框架的类型。 | “AMLToken”(必需) |
身份配置
名称 | DESCRIPTION | 价值 |
---|---|---|
身份类型 | 对于 AmlToken 类型,设置为“AMLToken”。 设置为 “ManagedIdentity”类型。 对于 UserIdentity 类型,设置为“UserIdentity”。 | “AMLToken” “Managed” “UserIdentity”(必需) |
JobBase服务
名称 | DESCRIPTION | 价值 |
---|
工作服务
名称 | DESCRIPTION | 价值 |
---|---|---|
终结点 | 终结点的 URL。 | 字符串 |
jobServiceType | 终结点类型。 | 字符串 |
节点 | 用户希望启动服务的节点。 如果未将 Nodes 设置为 null,则服务将仅在领导节点上启动。 |
节点 |
移植 | 用户设置的终结点的端口。 | 整数 (int) |
属性 | 在终结点上设置的其他属性。 | JobServiceProperties |
JobServiceProperties
名称 | DESCRIPTION | 价值 |
---|
标签类别
名称 | DESCRIPTION | 价值 |
---|---|---|
类 | 此类别中标签类的字典。 | 标签类别类 |
displayName (显示名称) | 标签类别的显示名称。 | 字符串 |
多选 | 指示是否允许在此类别中选择多个类。 | “Disabled” “Enabled” |
标签类别类
名称 | DESCRIPTION | 价值 |
---|
标签类
名称 | DESCRIPTION | 价值 |
---|---|---|
displayName (显示名称) | 标签类的显示名称。 | 字符串 |
子 | 标签类的子类的字典。 | LabelClass子类 |
LabelClass子类
名称 | DESCRIPTION | 价值 |
---|
LabelingDataConfiguration
名称 | DESCRIPTION | 价值 |
---|---|---|
数据 ID | 要执行标记的数据资产的资源 ID。 | 字符串 |
incrementalDataRefresh | 指示是否启用增量数据刷新。 | “Disabled” “Enabled” |
LabelingJobImageProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
annotationType | 图像标记作业的批注类型。 | “BoundingBox” “Classification” “InstanceSegmentation” |
mediaType | [必需]作业的媒体类型。 | “Image”(必需) |
TagJobInstructions
名称 | DESCRIPTION | 价值 |
---|---|---|
URI | 指向页面的链接,其中包含标签器的详细标记说明。 | 字符串 |
LabelingJobLabel类别
名称 | DESCRIPTION | 价值 |
---|
LabelingJobMediaProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
mediaType | 设置为“Image”,类型 为 LabelingJobImageProperties。 对于 LabelingJobTextProperties 类型,设置为“Text”。 | “Image” “Text”(必需) |
LabelingJobProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
组件 ID | 组件资源的 ARM 资源 ID。 | 字符串 |
计算 ID | 计算资源的 ARM 资源 ID。 | 字符串 |
dataConfiguration | 配置作业中使用的数据。 | LabelingDataConfiguration |
描述 | 资产说明文本。 | 字符串 |
displayName (显示名称) | 作业的显示名称。 | 字符串 |
实验名称 | 作业所属的试验的名称。 如果未设置,作业将放置在“默认”试验中。 | 字符串 |
标识 | 标识配置。 如果已设置,则应为 AmlToken、ManagedIdentity、UserIdentity 或 null 之一。 如果为 null,则默认为 AmlToken。 |
身份配置 |
isArchived 已存档 | 资产是否已存档? | 布尔 |
工作说明 | 标记作业的说明。 | TagJobInstructions |
职位类型 | [必需]指定作业的类型。 | “AutoML” “Command” “标记” “Pipeline” “Spark” “扫描”(必需) |
标签类别 | 作业的标签类别。 | LabelingJobLabel类别 |
labelingJobMediaProperties | 作业中的媒体类型特定属性。 | LabelingJobMediaProperties |
mlAssistConfiguration | 在作业中配置 MLAssist 功能。 | MLAssistConfiguration |
属性 | 资产属性字典。 | ResourceBaseProperties |
服务 | JobEndpoints 列表。 对于本地作业,作业终结点的终结点值为 FileStreamObject。 |
JobBase服务 |
标签 | 标记字典。 可以添加、删除和更新标记。 | ResourceBaseTags |
LabelingJobTextProperties
名称 | DESCRIPTION | 价值 |
---|---|---|
annotationType | 文本标记作业的批注类型。 | “Classification” 'NamedEntityRecognition' (命名实体识别) |
mediaType | [必需]作业的媒体类型。 | “Text”(必需) |
ManagedIdentity (托管身份)
名称 | DESCRIPTION | 价值 |
---|---|---|
客户端 ID | 按客户端 ID 指定用户分配的标识。 对于系统分配,请不要设置此字段。 | 字符串 约束: 最小长度 = 36 最大长度 = 36 图案 = ^[0-9a-fA-F]{8}-([0-9a-fA-F]{4}-){3}[0-9a-fA-F]{12}$ |
身份类型 | [必需]指定标识框架的类型。 | “Managed”(必需) |
对象 ID | 按对象 ID 指定用户分配的标识。 对于系统分配,请不要设置此字段。 | 字符串 约束: 最小长度 = 36 最大长度 = 36 图案 = ^[0-9a-fA-F]{8}-([0-9a-fA-F]{4}-){3}[0-9a-fA-F]{12}$ |
资源ID | 按 ARM 资源 ID 指定用户分配的标识。 对于系统分配,请不要设置此字段。 | 字符串 |
MLAssistConfiguration
名称 | DESCRIPTION | 价值 |
---|---|---|
mlAssist | 设置为 MLAssistConfigurationDisabled 类型的“Disabled”。 设置为 MLAssistConfigurationEnabled 类型的“Enabled”。 | “Disabled” “Enabled”(必需) |
MLAssistConfigurationDisabled
名称 | DESCRIPTION | 价值 |
---|---|---|
mlAssist | [必需]指示是否启用 MLAssist 功能。 | “Disabled”(必需) |
MLAssistConfiguration已启用
名称 | DESCRIPTION | 价值 |
---|---|---|
inferencingComputeBinding | [必需]推理中使用的 AML 计算绑定。 | 字符串 约束: 模式 = [a-zA-Z0-9_] (必需) |
mlAssist | [必需]指示是否启用 MLAssist 功能。 | “Enabled”(必需) |
trainingComputeBinding 的 | [必需]训练中使用的 AML 计算绑定。 | 字符串 约束: 模式 = [a-zA-Z0-9_] (必需) |
节点
名称 | DESCRIPTION | 价值 |
---|---|---|
nodesValueType | 对于 AllNodes 类型,设置为“All”。 | “全部”(必需) |
ResourceBaseProperties
名称 | DESCRIPTION | 价值 |
---|
ResourceBaseTags
名称 | DESCRIPTION | 价值 |
---|
用户身份
名称 | DESCRIPTION | 价值 |
---|---|---|
身份类型 | [必需]指定标识框架的类型。 | “UserIdentity”(必需) |