Define the Pipeline Function

With the pipeline steps tagged, you are now ready to decorate the pipeline function. Kale can infer the pipeline structure by examining the order of the function calls inside the function.

2. Add @pipeline decorator to the main entry point ml_pipeline function.

Adding the @pipeline decorator above the main entry point function indicates that this function defines the overall Kubeflow pipeline. Add the following above the ml_pipeline function to indicate that this code block is the main pipeline entry point, as well as the name of the pipeline and the name of the experiment.

@pipeline(name="binary-classification", experiment="kale-tutorial")

Now that you have tagged the ml_pipeline function you are ready to run the pipeline via the command line.