API Overview¶
All ModelArts APIs are proprietary.
You can use these APIs to manage datasets, training jobs, models, and services.
Data Management APIs¶
API | Description |
---|---|
Query created datasets by page based on specified conditions. | |
Create a new dataset and determine whether to enable team labeling.
| |
Query details about a dataset, including the dataset name, type, and version name based on the dataset ID. | |
Modify basic information about a dataset, such as the dataset name, description, version, or labels. Modified labels take effect in the entire dataset, including the samples in the dataset. | |
Delete a dataset based on the dataset ID to release resources. | |
Query dataset statistics, such as sample statistics, label statistics, or hard examples based on specified conditions. | |
Query the monitoring data of a dataset within a specified period, such as the number of labeled samples, number of unlabeled samples, and total number of samples at each time point within the period. |
API | Description |
---|---|
Query the versions of a dataset based on the dataset ID to learn about the dataset version evolution. | |
Publish a modified dataset as a new version. The modification includes labeling samples, adding samples, and deleting samples in the dataset. | |
Query details about a specified dataset labeling version, including the name, description, number of files, and storage path based on the dataset ID and version ID. | |
Delete a dataset version based on the dataset ID and version ID. |
API | Description |
---|---|
Query dataset samples by page based on specified conditions. | |
Add samples to a dataset in batches for data labeling. | |
Delete unused samples from a dataset in batches. | |
Query a single sample based on the sample ID, including the sample status and labels. | |
Obtain sample search criteria, such as the label list and attribute key-value pairs of the dataset based on the dataset ID. | |
Query the samples of a team labeling task on the data labeling platform by page based on the dataset ID and team labeling task ID. | |
Query details about a sample in a team labeling task on the data labeling platform based on the dataset ID, team labeling task ID, and sample ID. |
API | Description |
---|---|
Query the labels in a specified dataset version. | |
Create labels during dataset labeling. This function is available only in datasets of the text classification and named entity types. | |
Modify dataset labels in batches. The modification takes effect in the entire dataset, including the samples in the dataset. | |
Delete dataset labels in batches and determine whether to delete the samples with the labels. | |
Modify a label in a dataset based on the label name. The modification takes effect in the entire dataset, including the samples in the dataset. | |
Delete a label in a dataset based on the label name and determine whether to delete the samples with the label. |
API | Description |
---|---|
Label multiple samples in a dataset in batches.
This API uses a new label list to overwrite the original one to update the sample labels. For example, if an empty label list is used to overwrite the original one, all sample labels will be deleted. |
API | Description |
---|---|
Query the team labeling tasks of a dataset based on the dataset ID. | |
Create a team labeling task based on an existing dataset so that multiple members can concurrently label the dataset.
| |
Query details about a team labeling task based on the dataset ID and team labeling task ID, including the task name, data, and team information. | |
Assign and start a team labeling task on the data labeling platform based on the dataset ID and team labeling task ID as the team labeling task manager. | |
Update the description, name, and team information of a team labeling task based on the dataset ID and team labeling task ID. | |
Delete a team labeling task based on the dataset ID and team labeling task ID. | |
Start an acceptance task for a team labeling task based on the dataset ID and team labeling task ID. | |
Query the report and statistics of a team labeling acceptance task based on the dataset ID and team labeling task ID. | |
Determine the acceptance scope for a team labeling task, including all labeled data, and update the sample data accordingly. | |
Query statistics for a team labeling task on the data labeling platform, such as the sample statistics, label statistics, and hard example set based on the dataset ID and team labeling task ID. | |
Obtaining Details About the Progress of a Team Labeling Task Member | Query details about the progress of a team labeling task member in a team labeling task based on the dataset ID and team labeling task ID. |
Obtain all team labeling tasks on the data labeling platform by page as a member in a team labeling task. |
API | Description |
---|---|
During the acceptance of a team labeling task, provide review comments on samples, including the review result and score. | |
Review the team labeling task on the data labeling platform based on the dataset ID and team labeling task ID, determine the review result, and provide review comments as the team labeling task manager. | |
Update sample labels on the data labeling platform in batches, including adding, modifying, and deleting the sample labels. Ensure that only the labels in the dataset can be added or modified. |
API | Description |
---|---|
Query all labeling teams by page. | |
Add a labeling team. | |
Query details about a labeling team, including the team name, description, and total number of members based on the team ID. | |
Update the name and description of a labeling team based on the team ID. | |
Delete a labeling team based on the team ID. |
API | Description |
---|---|
Send an email to members in a labeling team to notify them of starting the team labeling task after the task is created. | |
Query all labeling team members by page based on specified conditions. | |
Obtain members in a labeling team by page based on the team ID. | |
Add new members to a labeling team. | |
Delete multiple members from a labeling team in batches. | |
Query details about a member in a labeling team, including the member description, email address, and role based on the team ID and member ID. | |
Update the description and role of a member in a labeling team based on the team ID and member ID. | |
Delete a member from a labeling team based on the team ID and member ID. |
API | Description |
---|---|
Query historical tasks imported to a dataset by page based on the dataset ID. | |
Create a dataset import task to import labels and data (such as manifest files and OBS data) from a storage system to the dataset. | |
Query details about a dataset import task based on the dataset ID and task ID to learn about the data source, import mode, and task status. |
API | Description |
---|---|
Query historical tasks exported from a dataset by page based on the dataset ID. | |
Export certain data as a new dataset or to OBS. | |
Query details about a dataset export task based on the dataset ID and task ID to learn about the export type, task status, and number of samples. |
API | Description |
---|---|
Synchronize data and labels from the dataset input path to the dataset. | |
Query the status of a data source synchronization task based on the dataset ID. |
API | Description |
---|---|
Query the to-be-confirmed auto labeling samples in a dataset by page based on the dataset ID. | |
Query information of a single auto labeling sample based on the dataset ID and sample ID, such as the sample labels, hard example details, and sample type. | |
Query all auto labeling tasks by page based on the dataset ID. | |
Start an auto labeling task for unlabeled data to quickly label the data. After the auto labeling task is complete, call the API for updating sample labels in batches to check the labeling result. | |
Obtain details about an auto labeling task based on the dataset ID and task ID to learn about the task configuration, name, and status. | |
Stop an ongoing auto labeling task based on the dataset ID and task ID. |
API | Description |
---|---|
Query all data processing tasks by page. Data processing extracts valuable data from a large amount of disordered, difficult-to-understand data. | |
Create a data processing task. Data processing involves data validation, data cleansing, data selection, and data augmentation. | |
Query the algorithms available for a data processing type. | |
Query details about a data processing task, including the task description, status, and version based on the task ID. | |
Update the description of a data processing task based on the task ID. | |
Delete a data processing task based on the task ID. | |
Query the versions of a data processing task by page based on the task ID. | |
Publish a data processing task as a new version after the task parameters are modified. | |
Query details about a version of a data processing task, including the version description, task status, and algorithm template based on the task ID and version ID. | |
Delete a version of a data processing task based on the task ID and version ID. | |
Query results of a data processing task by page based on the task ID and version ID. | |
Stop an ongoing data processing task based on the task ID and version ID. |
DevEnviron APIs¶
Type | API | Description |
---|---|---|
Managing DevEnviron instances | Query the list of notebook instances meeting search criteria. | |
Create a notebook instance based on parameters such as the instance flavor, AI engine, and storage. | ||
Query details about a notebook instance. | ||
Update a notebook instance. | ||
Delete the container and all storage resources of a notebook instance. | ||
Save the running instance as a container image. In the saved image, the installed dependency package (pip package) is retained. In VS Code remote development, the plug-ins installed on the server are retained. | ||
Obtain available flavors. | ||
Obtain the flavors available for a notebook instance. | ||
Obtaining the Available Duration of a Running Notebook Instance | Obtain the available duration of a running notebook instance. | |
Prolong the available duration of a running notebook instance. | ||
Start a notebook instance. | ||
Stop a notebook instance. | ||
Dynamically mounting OBS | Obtain the notebook instances with OBS storage mounted. | |
Dynamically mount OBS to a notebook instance in the running state. | ||
Obtaining Details About a Notebook Instance with OBS Storage Mounted | Obtain details about a notebook instance with OBS storage mounted. | |
Dynamically unmount OBS from a notebook instance. | ||
Image management | Obtain all images by page based on specified conditions. | |
Register a custom image with ModelArts image management. | ||
Obtain the overview of user image information. Image names are used for aggregation. | ||
Obtain details about an image. | ||
Delete an image. For a private image, you can also delete its SWR image using parameters. |
API | Description |
---|---|
Create a development environment instance for code development. | |
Obtain the development environment instances that meet the search criteria. | |
Query details about a notebook instance. | |
Modifying the Description of a Development Environment Instance | Modify the description of a development environment instance. |
Delete a development environment instance. | |
Start or stop a development environment instance. |
Training Management APIs¶
API | Description |
---|---|
Create an algorithm. | |
Obtain algorithms. | |
Obtain a specified algorithm based on the algorithm ID. | |
Update an algorithm. | |
Delete an algorithm. |
API | Description |
---|---|
Create a training job. | |
Obtain details about a training job. | |
Modify the description of a training job. | |
Delete a training job. | |
Terminate a training job. Only jobs in the creating, awaiting, or running state can be terminated. | |
Obtaining the Logs of a Specified Task in a Training Job (Preview) | Obtain the logs of a specified task in a training job (preview). |
Obtaining the Logs of a Specified Task in a Training Job (OBS Link) | Obtain the logs of a specified task in a training job (OBS link). You can view all logs or download the logs. |
Obtaining the Runtime Metrics of a Specified Task in a Training Job | Obtain the runtime metrics of a specified task in a training job. |
Obtain the created training jobs by search criteria. |
API | Description |
---|---|
Obtaining the General Specifications Supported by a Training Job | Obtain the general specifications supported by a training job. |
Obtaining the Preset AI Frameworks Supported by a Training Job | Obtain the preset AI frameworks supported by a training job. |
Model Management APIs¶
API | Description |
---|---|
Import a model. | |
Obtain the models that meet the search criteria. | |
View details about a model based on the model ID. | |
Delete a specified model based on the model ID. All versions of the model can be deleted in cascading mode. |
Service Management APIs¶
API | Description |
---|---|
Deploy a model service. | |
Obtain model services. | |
Obtain details about a model service based on the service ID. | |
Update a model service. | |
Obtain service monitoring information. | |
Obtain the update logs of a real-time service. | |
Obtain service event logs, including service operation records, key actions during deployment, and deployment failure causes. | |
Delete a model service. | |
Obtain supported service deployment specifications. | |
Obtain dedicated resource pools. |
Resource Management APIs¶
API | Description |
---|---|
Obtain the configuration parameters of the ModelArts OS service, such as the CIDR block and user resource quota. |
API | Description |
---|---|
Obtain the quotas of some ModelArts OS resources, such as the quotas for resource pool quotas and networks. |
API | Description |
---|---|
Obtain details of a specified plug-in template. |
API | Description |
---|---|
Obtain the nodes in a resource pool. | |
Delete nodes from a specific resource pool in batches. At least one node must be reserved in the resource pool. |
API | Description |
---|---|
Obtain events. |
API | Description |
---|---|
Create network resources. | |
Obtain a list of network resources. | |
Obtain details about a specified network resource. | |
Delete a specified network resource. | |
Update a specified network resource. |
API | Description |
---|---|
Obtain the real-time usage of all resource pools in the current project. |
API | Description |
---|---|
Create a resource pool. | |
Obtain resource pools. | |
Obtain details about a specified resource pool. | |
Delete a specified resource pool. | |
Update a specified resource pool. | |
Obtain the monitoring information of a resource pool. | |
Obtain the statistics of a resource pool. |
API | Description |
---|---|
Obtain resource flavors. |
API | Description |
---|---|
Obtain dedicated resource pool jobs. | |
Obtain statistics for dedicated resource pool jobs. |