DataWorks task scheduling is widely used in machine learning scenarios. It allows you to run DataWorks tasks periodically to update your model, helping you create a model training pipeline. Alibaba Cloud Machine Learning Platform for AI is integrated with DTplus to provide the DataWorks task scheduling service. DTplus is a set of Apsara Stack system management, operations, and maintenance platforms.
After you have successfully run all nodes in an experiment, you can deploy the experiment to DataWorks and then schedule DataWorks to periodically run the experiment. This topic uses the heart disease prediction case as an example to describe how to schedule DataWorks tasks.
Note: Before you schedule DataWorks tasks, make sure that you have successfully run all the nodes in your experiment, and you have activated DataStudio in DataWorks.
Log on to the Alibaba Cloud Machine Learning Platform for AI console.
Select Deploy > Schedule DataWorks Tasks to go to DataStudio of DataWorks.
In the DataStudio console, Select Create > Machine Learning Platform for AI, and then create a machine learning experiment node.
In the Create Node dialog box, enter the node name, select the target folder, and click Submit.
After the experiment node is created, perform the following steps on the canvas:
Select the experiment from the drop-down list.
Configure task scheduling parameters, including the recurrence, input, and output parameters.
Click Submit. The task will be executed the next day.
Click Administration in the upper-right corner to go to the administration page. You can view the status of the machine learning task and the system log. You can also perform other operations. For example, you can add retroactive data or test the experiment.