We have implemented all sorts of models in Go, including regressions, classifications, clustering, and more. You have also learned about some of the process around developing a machine learning model. Our models have successfully predicted disease progression, flower species, and objects within images. Yet, we are still missing an important piece of the machine learning puzzle: deployment, maintenance, and scaling.
If our models just stay on our laptops, they are not doing any good or creating value within a company. We need to know how to take our machine learning workflows and integrate them into the systems that are already deployed in our organization, and we need to know how to scale, update, and maintain these workflows over time.
The fact that our machine learning workflows are, by their very nature, multi-stage workflows might...
