Open source frameworks such as Scikit-Learn and TensorFlow have made it simple to write machine learning and deep learning code. They've become immensely popular in the developer community and for good reason. However, managing training and deployment infrastructure still requires a lot of effort and skills that data scientists and machine learning engineers typically do not possess. SageMaker simplifies the whole process. You can go quickly from experimentation to production, without ever worrying about infrastructure.
In this chapter, you learned about the different frameworks available in SageMaker for machine learning and deep learning, as well as how to customize their containers. You also learned how to use Script Mode and Local Mode for fast iteration until you're ready to deploy in production. Finally, you ran several examples, including one that combines Apache Spark and SageMaker.
In the next chapter, you will learn how to use your own custom code...