Join our book community on Discord
https://packt.link/EarlyAccessCommunity
As we reach the end of the cookbook, it’s time to bring the reality of machine learning (ML) in production into the spotlight. Everything we’ve covered up to this point is of little use in the business world if it just sits on your laptop. Models must be deployed in compute environments that allow for scalability and high throughput while still maintaining predictive performance at or above the business rules that govern them. Although we are only devoting a single chapter to this topic, you should keep in mind that production ML deployment, monitoring, benchmarking, and the CI/CD/CT cycle (among other topics) makes up the lion’s share of real-world challenges for utilizing ML in business. Many of the considerations are non-technical as well: how do I determine how well a model needs to perform in order to achieve a given ROI, or, how do I know when I need to retrain my model, or, do I need...
