Machine Learning Engineering and MLOps with Google Cloud
It’s generally estimated that almost 90% of data science projects never make it to production. Data scientists spend a lot of time training and experimenting with models in the lab, but often don’t succeed in bringing those workloads out into the real world. A major reason for this is because, as we have discussed in the previous chapters of this book, there are difficult challenges at every step in the model development lifecycle. Following on from our previous chapter, we will now dive into more detail on deployment concepts and challenges, and describe the importance of Machine Learning Operations (MLOps) in addressing these challenges for large-scale production AI/ML workloads.
Specifically, this chapter will cover the following topics:
- An introduction to MLOps
- Why MLOps is needed for deploying large-scale ML workloads
- MLOps tools
- Implementing MLOps on Google Cloud using Vertex AI Pipelines...
