Platform for AI (PAI) provides the Pipeline Tasks tab that allows you to manage pipeline tasks in a centralized and visualized manner. This topic describes how to manage pipeline tasks.
Background information
The Pipeline Tasks tab displays the following types of pipeline tasks:
Pipeline tasks manually executed in Machine Learning Designer. For more information, see Model training.
Pipeline tasks periodically scheduled in DataWorks. For more information, see Use DataWorks tasks to schedule pipelines in Machine Learning Designer.
Procedure
Go to the Distributed Training Jobs page
Log on to the PAI console.
In the left-side navigation pane, click Workspaces. On the Workspaces page, click the name of the workspace that you want to manage.
In the left-side navigation pane, choose to go to the Distributed Training Jobs page.
On the Pipeline Tasks tab, you can view the details of a task, filter tasks, and stop or start a task.
View the details of a task
Find the task that you want to manage and click the task name. The Pipeline Details page of the task appears You can click the Details and Task Result tabs to view details.
On the Details tab, the nodes that were run in the task are displayed. You can click a node on the canvas to view the details of the node on the right. The details include the Operation Info, Job Logs, and Output Result information.
You can also click Share in the upper-right corner to share the pipeline task details page with other users.
On the Task Result tab, move the pointer over Result in the Actions column to view the output URLs of all nodes in the current pipeline task.
Filter tasks to find the desired task
On the Pipeline Tasks page, click the filter button next to Status or Task Source column to filter the pipeline that you want to manage.
Stop or start a task
Click Stop in the Actions column to stop a running pipeline task. Click Clone in the Actions column to restart the pipeline task.