Machine Learning Designer generates models and debugs models by using pipelines. You must first create a pipeline and streamline different components in the pipeline based on the logic of modeling. This topic describes how to create a pipeline.

Prerequisites

Machine Learning Platform for AI is activated.

Background information

Machine Learning Designer provides various pipeline templates. You can use these pipeline templates based on your business requirements. You can also create a pipeline template. The following example shows the steps that are required to create a pipeline.

Procedure

  1. Go to the Visualized Modeling (Machine Learning Designer) page.
    1. Log on to the Machine Learning Platform for AI console.
    2. In the left-side navigation pane, click Workspaces. On the Workspace list page, click the name of the workspace that you want to manage.
    3. In the left-side navigation pane, choose Model Training > Visualized Modeling (Designer) to go to the Visualized Modeling (Machine Learning Designer) page.
  2. Create a pipeline.
    1. Choose Create Pipeline > New blank canvas.
    2. In the New blank canvas dialog box, set the parameters and click OK to create a pipeline.
      In this example, enter a name for the pipeline and use the default settings for other parameters.
  3. On the Pipeline list tab, select the pipeline that you created and click Enter pipeline.

What to do next

After you create a pipeline, you need to prepare data and perform data preprocessing. For more information, see Preprocess data.