Building ML workflows with Amazon SageMaker Pipelines

Model build workflows cover all of the steps performed when developing your model, including data preparation, model training, model tuning, and model deployment. In this case, model deployment can include the tasks necessary to evaluate your model, as well as batch use cases that do not need to be deployed to higher environments. SageMaker Pipelines is a fully managed service that allows you to create automated model build workflows using the SageMaker Python SDK.

SageMaker Pipelines includes built-in step types (https://docs.aws.amazon.com/sagemaker/latest/dg/build-and-manage-steps.html) for executing SageMaker tasks, such as SageMaker Processing for data pre-processing, and SageMaker Training for model training. Pipelines also include steps for controlling how your pipeline works. For example, the pipeline could include conditional steps that could be used to evaluate the output of a previous step to determine whether to...