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。
Ingestion Jobs - Create
创建具有指定作业 ID 的引入作业。
PUT {endpoint}/openai/ingestion/jobs/{job-id}?api-version=2024-07-01-preview
URI 参数
名称 | 在 | 必需 | 类型 | 说明 |
---|---|---|---|---|
endpoint
|
path | True |
string url |
支持的认知服务终结点(协议和主机名,例如:https://aoairesource.openai.azure.com。将“aoairesource”替换为 Azure OpenAI 帐户名称。 |
job-id
|
path | True |
string |
将创建的作业的 ID。 |
api-version
|
query | True |
string |
请求的 API 版本。 |
请求头
名称 | 必需 | 类型 | 说明 |
---|---|---|---|
mgmt-user-token |
string |
用于访问工作区的令牌(仅适用于用户计算作业)。 |
|
aml-user-token |
string |
用于访问工作区中作业中的资源的令牌(仅适用于用户计算作业)。 |
请求正文
请求正文可以为下列任一内容:
名称 | 说明 |
---|---|
Ingestion |
|
Ingestion |
IngestionJobSystemCompute
名称 | 必需 | 类型 | 说明 |
---|---|---|---|
kind | True |
string:
System |
IngestionJobType |
completionAction |
完成操作。 |
||
dataRefreshIntervalInHours |
integer |
||
datasource | SystemComputeDatasource: |
SystemComputeDatasource |
|
jobId |
string |
||
searchServiceConnection | BaseConnection: |
BaseConnection |
IngestionJobUserCompute
名称 | 必需 | 类型 | 说明 |
---|---|---|---|
kind | True |
string:
User |
IngestionJobType |
workspaceId | True |
string |
|
compute | JobCompute: |
JobCompute |
|
dataRefreshIntervalInHours |
integer |
||
datasource | UserComputeDatasource: |
UserComputeDatasource |
|
jobId |
string |
||
target | TargetIndex: |
TargetIndex |
响应
名称 | 类型 | 说明 |
---|---|---|
200 OK | IngestionJob: |
成功 |
Other Status Codes |
发生错误。 |
安全性
api-key
API 密钥身份验证
类型:
apiKey
在:
header
OAuth2Auth
OAuth2 身份验证
类型:
oauth2
流向:
implicit
授权 URL:
https://login.microsoftonline.com/common/oauth2/v2.0/authorize
作用域
名称 | 说明 |
---|---|
https://cognitiveservices.azure.com/.default |
示例
Create a system-compute ingestion job |
Create a user-compute ingestion job |
Create a system-compute ingestion job
示例请求
PUT {endpoint}/openai/ingestion/jobs/ingestion-job?api-version=2024-07-01-preview
{
"kind": "SystemCompute",
"searchServiceConnection": {
"kind": "EndpointWithManagedIdentity",
"endpoint": "https://aykame-dev-search.search.windows.net"
},
"datasource": {
"kind": "Storage",
"connection": {
"kind": "EndpointWithManagedIdentity",
"endpoint": "https://mystorage.blob.core.windows.net/",
"resourceId": "/subscriptions/1234567-abcd-1234-5678-1234abcd/resourceGroups/my-resource/providers/Microsoft.Storage/storageAccounts/mystorage"
},
"containerName": "container",
"chunking": {
"maxChunkSizeInTokens": 2048
},
"embeddings": [
{
"connection": {
"kind": "RelativeConnection"
},
"deploymentName": "Ada"
}
]
},
"dataRefreshIntervalInHours": 24,
"completionAction": "keepAllAssets"
}
示例响应
operation-location: https://aoairesource.openai.azure.com/openai/ingestion/jobs/ingestion-job/runs/72a2792ef7d24ba7b82c7fe4a37e379f?api-version=2024-07-01-preview
{
"kind": "SystemCompute",
"jobId": "ingestion-job",
"searchServiceConnection": {
"kind": "EndpointWithManagedIdentity",
"endpoint": "https://aykame-dev-search.search.windows.net"
},
"datasource": {
"kind": "Storage",
"connection": {
"kind": "EndpointWithManagedIdentity",
"endpoint": "https://mystorage.blob.core.windows.net/",
"resourceId": "/subscriptions/1234567-abcd-1234-5678-1234abcd/resourceGroups/my-resource/providers/Microsoft.Storage/storageAccounts/mystorage"
},
"containerName": "container",
"chunking": {
"maxChunkSizeInTokens": 2048
},
"embeddings": [
{
"connection": {
"kind": "RelativeConnection"
},
"deploymentName": "Ada"
}
]
},
"dataRefreshIntervalInHours": 24,
"completionAction": "keepAllAssets"
}
Create a user-compute ingestion job
示例请求
PUT {endpoint}/openai/ingestion/jobs/ingestion-job?api-version=2024-07-01-preview
{
"kind": "UserCompute",
"workspaceId": "/subscriptions/f375b912-331c-4fc5-8e9f-2d7205e3e036/resourceGroups/adrama-copilot-demo/providers/Microsoft.MachineLearningServices/workspaces/adrama-rag-dev",
"compute": {
"kind": "ServerlessCompute"
},
"target": {
"kind": "AzureAISearch",
"connectionId": "/subscriptions/f375b912-331c-4fc5-8e9f-2d7205e3e036/resourceGroups/adrama-copilot-demo/providers/Microsoft.MachineLearningServices/workspaces/adrama-rag-dev/connections/search-connection"
},
"datasource": {
"kind": "Dataset",
"datasetId": "azureml://locations/centraluseuap/workspaces/83317fe6-efa6-4e4a-b020-d0edd11ec382/data/PlainText/versions/1",
"datasetType": "uri_folder"
}
}
示例响应
operation-location: https://aoairesource.openai.azure.com/openai/ingestion/jobs/ingestion-job/runs/72a2792ef7d24ba7b82c7fe4a37e379f?api-version=2024-07-01-preview
{
"kind": "UserCompute",
"jobId": "ingestion-job",
"workspaceId": "/subscriptions/f375b912-331c-4fc5-8e9f-2d7205e3e036/resourceGroups/adrama-copilot-demo/providers/Microsoft.MachineLearningServices/workspaces/adrama-rag-dev",
"compute": {
"kind": "ServerlessCompute"
},
"target": {
"kind": "AzureAISearch",
"connectionId": "/subscriptions/f375b912-331c-4fc5-8e9f-2d7205e3e036/resourceGroups/adrama-copilot-demo/providers/Microsoft.MachineLearningServices/workspaces/adrama-rag-dev/connections/search-connection"
},
"datasource": {
"kind": "Dataset",
"datasetId": "azureml://locations/centraluseuap/workspaces/83317fe6-efa6-4e4a-b020-d0edd11ec382/data/PlainText/versions/1",
"datasetType": "uri_folder"
}
}
定义
名称 | 说明 |
---|---|
Azure |
Azure AI 搜索索引。 |
Chunking |
ChunkingSettings |
Compute |
计算类型。 |
Connection |
连接字符串连接。 |
Connection |
连接类型。 |
Cosmos |
CosmosDB 索引。 |
Crawling |
CrawlingSettings |
Custom |
自定义计算。 |
Deployment |
相对部署连接。 |
Endpoint |
终结点密钥连接。 |
Endpoint |
终结点托管标识连接。 |
Error |
错误 |
Error |
ErrorCode |
Error |
ErrorResponse |
Generic |
ConnectionEmbeddingSettings |
Ingestion |
完成操作。 |
Ingestion |
|
Ingestion |
IngestionJobType |
Ingestion |
|
Inner |
InnerError |
Inner |
InnerErrorCode |
Pinecone |
Pinecone 索引。 |
Serverless |
无服务器计算。 |
System |
数据源类型。 |
System |
SystemComputeStorage |
System |
SystemComputeUrl |
Target |
目标类型。 |
User |
UserComputeStorage |
User |
数据源类型。 |
User |
UserComputeUrl |
Workspace |
AML 工作区连接。 |
Workspace |
WorkspaceConnectionEmbeddingSettings |
AzureAISearchIndex
Azure AI 搜索索引。
名称 | 类型 | 说明 |
---|---|---|
connectionId |
string |
指向 Azure AI 搜索索引的连接 ID。 |
kind |
string:
Azure |
目标类型。 |
ChunkingSettings
ChunkingSettings
名称 | 类型 | 说明 |
---|---|---|
maxChunkSizeInTokens |
integer |
ComputeType
计算类型。
名称 | 类型 | 说明 |
---|---|---|
CustomCompute |
string |
自定义用户计算。 |
ServerlessCompute |
string |
无服务器用户计算。 |
ConnectionStringConnection
连接字符串连接。
名称 | 类型 | 说明 |
---|---|---|
connectionString |
string |
连接字符串 |
kind |
string:
Connection |
连接类型。 |
ConnectionType
连接类型。
名称 | 类型 | 说明 |
---|---|---|
ConnectionString |
string |
连接字符串。 |
EndpointWithKey |
string |
终结点和密钥连接。 |
EndpointWithManagedIdentity |
string |
终结点和托管标识。 |
RelativeConnection |
string |
相对部署 |
WorkspaceConnection |
string |
AML 工作区连接。 |
CosmosDBIndex
CosmosDB 索引。
名称 | 类型 | 说明 |
---|---|---|
collectionName |
string |
cosmos DB 集合的名称。 |
connectionId |
string |
指向 cosmos DB 的连接 ID。 |
databaseName |
string |
cosmos DB 数据库的名称。 |
kind |
string:
CosmosDB |
目标类型。 |
CrawlingSettings
CrawlingSettings
名称 | 类型 | 说明 |
---|---|---|
maxCrawlDepth |
integer |
|
maxCrawlTimeInMins |
integer |
|
maxDownloadTimeInMins |
integer |
|
maxFileSize |
integer |
|
maxFiles |
integer |
|
maxRedirects |
integer |
CustomCompute
自定义计算。
名称 | 类型 | 说明 |
---|---|---|
computeId |
string |
自定义计算的 ID |
kind | string: |
计算类型。 |
DeploymentConnection
相对部署连接。
名称 | 类型 | 说明 |
---|---|---|
kind |
string:
Relative |
连接类型。 |
EndpointKeyConnection
终结点密钥连接。
名称 | 类型 | 说明 |
---|---|---|
endpoint |
string |
端点 |
key |
string |
钥匙 |
kind |
string:
Endpoint |
连接类型。 |
EndpointMIConnection
终结点托管标识连接。
名称 | 类型 | 说明 |
---|---|---|
endpoint |
string |
端点 |
kind |
string:
Endpoint |
连接类型。 |
resourceId |
string |
资源 ID |
Error
错误
名称 | 类型 | 说明 |
---|---|---|
code |
ErrorCode |
|
details |
Error[] |
错误详细信息(如果可用)。 |
innererror |
InnerError |
|
message |
string |
此错误的消息。 |
target |
string |
发生错误的位置(如果可用)。 |
ErrorCode
ErrorCode
名称 | 类型 | 说明 |
---|---|---|
conflict |
string |
请求的操作与当前资源状态冲突。 |
contentFilter |
string |
由于我们的安全系统,映像生成失败。 |
fileImportFailed |
string |
导入文件失败。 |
forbidden |
string |
当前用户/API 密钥禁止此操作。 |
internalFailure |
string |
内部错误。 请重试。 |
invalidPayload |
string |
此操作的请求数据无效。 |
itemDoesAlreadyExist |
string |
该项已存在。 |
jsonlValidationFailed |
string |
jsonl 数据的验证失败。 |
notFound |
string |
找不到资源。 |
quotaExceeded |
string |
超出配额。 |
serviceUnavailable |
string |
该服务当前不可用。 |
tooManyRequests |
string |
请求过多。 请稍后重试。 |
unauthorized |
string |
当前用户/API 密钥未获得操作的授权。 |
unexpectedEntityState |
string |
无法以当前资源的状态执行该操作。 |
ErrorResponse
ErrorResponse
名称 | 类型 | 说明 |
---|---|---|
error |
错误 |
GenericEmbeddingSettings
ConnectionEmbeddingSettings
名称 | 类型 | 说明 |
---|---|---|
connection | BaseConnection: |
BaseConnection |
deploymentName |
string |
|
modelName |
string |
IngestionJobCompletionAction
完成操作。
名称 | 类型 | 说明 |
---|---|---|
cleanUpTempAssets |
string |
清理在引入过程中创建的中间资产。 |
keepAllAssets |
string |
不会清理引入过程中创建的任何中间资产。 |
IngestionJobSystemCompute
名称 | 类型 | 说明 |
---|---|---|
completionAction |
完成操作。 |
|
dataRefreshIntervalInHours |
integer |
|
datasource | SystemComputeDatasource: |
SystemComputeDatasource |
jobId |
string |
|
kind |
string:
System |
IngestionJobType |
searchServiceConnection | BaseConnection: |
BaseConnection |
IngestionJobType
IngestionJobType
名称 | 类型 | 说明 |
---|---|---|
SystemCompute |
string |
在服务拥有的资源上运行的作业。 |
UserCompute |
string |
在用户拥有的工作区上运行的作业。 |
IngestionJobUserCompute
名称 | 类型 | 说明 |
---|---|---|
compute | JobCompute: |
JobCompute |
dataRefreshIntervalInHours |
integer |
|
datasource | UserComputeDatasource: |
UserComputeDatasource |
jobId |
string |
|
kind |
string:
User |
IngestionJobType |
target | TargetIndex: |
TargetIndex |
workspaceId |
string |
InnerError
InnerError
名称 | 类型 | 说明 |
---|---|---|
code |
InnerErrorCode |
|
innererror |
InnerError |
InnerErrorCode
InnerErrorCode
名称 | 类型 | 说明 |
---|---|---|
invalidPayload |
string |
此操作的请求数据无效。 |
PineconeIndex
Pinecone 索引。
名称 | 类型 | 说明 |
---|---|---|
connectionId |
string |
指向 pinecone 的连接 ID。 |
kind |
string:
Pinecone |
目标类型。 |
ServerlessCompute
无服务器计算。
名称 | 类型 | 说明 |
---|---|---|
instanceCount |
integer |
要运行作业的实例计数。 |
kind | string: |
计算类型。 |
sku |
string |
SKU 级别 |
SystemComputeDatasourceType
数据源类型。
名称 | 类型 | 说明 |
---|---|---|
Storage |
string |
Azure 存储帐户。 |
Urls |
string |
URL。 |
SystemComputeStorage
SystemComputeStorage
名称 | 类型 | 说明 |
---|---|---|
chunking |
ChunkingSettings |
|
connection | BaseConnection: |
BaseConnection |
containerName |
string |
容器名称 |
embeddings |
ConnectionEmbeddingSettings |
|
kind |
string:
Storage |
数据源类型。 |
SystemComputeUrl
SystemComputeUrl
名称 | 类型 | 说明 |
---|---|---|
chunking |
ChunkingSettings |
|
connection | BaseConnection: |
BaseConnection |
containerName |
string |
容器名称 |
crawling |
CrawlingSettings |
|
embeddings |
ConnectionEmbeddingSettings |
|
kind |
string:
Urls |
数据源类型。 |
urls |
string[] |
TargetType
目标类型。
名称 | 类型 | 说明 |
---|---|---|
AzureAISearch |
string |
Azure AI 搜索索引。 |
CosmosDB |
string |
CosmosDB 索引。 |
Pinecone |
string |
Pinecone 索引。 |
UserComputeDataset
UserComputeStorage
名称 | 类型 | 说明 |
---|---|---|
chunking |
ChunkingSettings |
|
datasetId |
string |
|
datasetType |
string |
|
embeddings |
WorkspaceConnectionEmbeddingSettings |
|
kind |
string:
Dataset |
数据源类型。 |
UserComputeDatasourceType
数据源类型。
名称 | 类型 | 说明 |
---|---|---|
Dataset |
string |
工作区数据集。 |
Urls |
string |
URL。 |
UserComputeUrl
UserComputeUrl
名称 | 类型 | 说明 |
---|---|---|
chunking |
ChunkingSettings |
|
crawling |
CrawlingSettings |
|
embeddings |
WorkspaceConnectionEmbeddingSettings |
|
kind |
string:
Urls |
数据源类型。 |
urls |
string[] |
WorkspaceConnection
AML 工作区连接。
名称 | 类型 | 说明 |
---|---|---|
connectionId |
string |
ConnectionId |
kind | string: |
连接类型。 |
WorkspaceConnectionEmbeddingSettings
WorkspaceConnectionEmbeddingSettings
名称 | 类型 | 说明 |
---|---|---|
connectionId |
string |
|
deploymentName |
string |
|
modelName |
string |