Automating ML with CI/CD
If you recall from Chapter 1, Getting Started with Automated Machine Learning on AWS, I highlighted that the typical ML process is manual and iterative. If you compare Figure 1.2, showing a realistic overview of the ML process, with Figure 4.3, showing the CI/CD process, I'm sure you will note that there are significant dissimilarities between the two processes:
Figure 4.4 – A realistic overview of the ML process
However, since the focus of this chapter is to address the limitations of both the typical ML process and the AutoML methodology, specifically when it comes to bridging the gap for model deployment, there are several similarities between these processes. So, if you take a deployment-centric approach (Figure 4.3), as opposed to an experiment-centric approach (Figure 1.2), the procedure for deploying an optimized model into production is exactly the same as the procedure for deploying software code changes into...
