tencent cloud

Cloud Infinite

Creating a Dataset

다운로드
포커스 모드
폰트 크기
마지막 업데이트 시간: 2026-09-08 16:24:27
AI 번역

Feature Description

This API creates a Dataset, which is a collection of file metadata used to store and manage metadata indexes.

Authorization Description

When using a sub-account, add the ci:CreateDataset permission to the action of the authorization policy. For all operation APIs supported by Cloud Infinite, see CI action.

Use Limits

Only the Jakarta region is supported. In the request Host, Region can only be set to ap-jakarta.
For more usage limitations, see Usage Limits.

Request

Request sample

POST /dataset HTTP/1.1
Host: <AppId>.ci.ap-jakarta.myqcloud.com
Authorization: Auth String
Content-Length: xxx
Content-Type: application/json
Accept: application/json
Note:
Authorization: A request header that carries authentication information and is used to verify the legitimacy of the request. For details, see Request Signature.

Request header

This API only uses common request headers. For details, see Common Request Headers.

Request body

The implementation of this request operation requires the following request body.
{
"DatasetName": "test",
"Description": "test",
"TemplateId": "Official:COSBasicMeta"
}
The specific data content is as follows:
Parameter Name
Description
Type
Required
DatasetName
Dataset name, unique within the same account. Naming rules:
1-64 characters in length
Can contain only lowercase letters, digits, and hyphens (-).
Must start with a letter or digit.
String
Yes
Description
Dataset description. The length is 1 to 256 English or Chinese characters. The default value is empty.
String
No
TemplateId
For the search template associated with a dataset, the backend determines which metadata of files to collect based on the search template when creating metadata indexes. Each search template contains several operators, and different operators represent different processing capabilities. For more information, see Search Templates and Operators.
The default value is empty, which means no search template is associated and no metadata is collected.
String
No
Version
Dataset version. basic or standard. The default value is basic.
String
No
Volume
Estimated data volume of the current dataset. MetaInsight adjusts the data distribution policy based on this parameter to ensure query performance. When Version is basic, the value is 50w. When Version is standard, the default value is 500w, and you can set it to a value from 1 to 10000 in the unit of w. If you set it to 0, the default value is used.
Integer
No
TrainingMode
Source mode of training data. The default value is 0, indicating that training data comes from a specified dataset. When the value is 1, training data comes from files in a bucket directory of cos. It takes effect only when Version is standard.
Integer
No
TrainingDataset
Dataset name of the training data. It takes effect only when TrainingMode is 0.
String
No
TrainingURI
Resource path of the training data. It takes effect only when TrainingMode is 1.
String
No
SceneType
It takes effect when TemplateId=Official:ImageSearch and indicates the scene type. Optional values:
general (default value): general scenario
E-commercial: e-commerce scenario
iDrive: travel scenario
String
No

Response

Response Headers

This API only returns common response headers. For details, see Common Response Headers.

Response Body

{
"Dataset": {
"BindCount": 0,
"CreateTime": "2023-12-25 15:16:20.692674978 +0800 CST",
"DatasetName": "test",
"Description": "test",
"FileCount": 0,
"TemplateId": "Official:COSBasicMeta",
"TotalFileSize": 0,
"UpdateTime": "2023-12-25 15:16:20.692675128 +0800 CST"
},
"RequestId": "NWFjMzQ0MDZfOTBmYTUwXzZkZV8z****"
}
The specific data content of the response body is as follows:
Parameter Name
Description
Type
Dataset
Dataset information
Container
RequestId
Request ID.
String
Dataset
node content:
Parameter Name
Description
Type
BindCount
Number of COS buckets currently bound to the dataset
Long
CreateTime
Timestamp of the data set creation time in RFC3339Nano format
String
DatasetName
Data set name
String
Description
Dataset description
String
FileCount
Number of files in the dataset for which metadata indexes have been created
Long
Region
Region where the dataset resides
String
SceneType
Scene type. Supports general, E-commercial, and iDrive. The default value is general.
String
State
Training status
Unstart: Not started
Running: Training in progress
Success: Training succeeded
Failed: Training failed
String
TaskProgress
Percentage value of training progress. The value ranges from 0 to 100.
Int
TemplateId
Search template
String
TotalFileSize
Total size of files in the dataset for which metadata indexes have been created, in bytes
Long
UpdateTime
The timestamp of the dataset modification time, in RFC3339Nano format.
After a dataset is created, if it has not been updated, the timestamp of the dataset modification time is the same as the timestamp of the dataset creation time.
String
Version
Dataset version
String


도움말 및 지원

문제 해결에 도움이 되었나요?

피드백