Creating a pipeline

You can create a pipeline by using the GUI pipeline editor. This pipeline automates the notebook workflow that you used earlier to train a model and save it to S3 storage.

Prerequisites

  • You configured a pipeline server as described in Enabling AI pipelines.

  • If you configured the pipeline server after you created your workbench, you stopped and then started your workbench.

Procedure

  1. Open your workbench’s JupyterLab environment. If the launcher is not visible, click + to open it.

    Pipeline buttons

  2. Click Pipeline Editor.

    Pipeline Editor button

    You have created a blank pipeline.

  3. Set the default runtime image for when you run your notebook or Python code.

    1. In the pipeline editor, click Open Panel.

      Open Panel

    2. Select the Pipeline Properties tab.

      Pipeline Properties Tab

    3. In the Pipeline Properties panel, scroll down to Generic Node Defaults and Runtime Image. Set the value to Runtime | Tensorflow | Cuda | Python 3.12.

      Pipeline Runtime Image0

  4. Select File → Save Pipeline.

Verification

  • In the JupyterLab file browser, the pipeline file (for example, untitled.pipeline) appears in your working directory.

Next step

Adding nodes to your pipeline