From the course: Data Science Foundations: Data Assessment for Predictive Modeling
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
How to prepare for eventual deployment
From the course: Data Science Foundations: Data Assessment for Predictive Modeling
How to prepare for eventual deployment
- [Instructor] The final phase of CRISP-DM is deployment. And we want to be ready, because if we don't plan for it now, it either won't happen efficiently or it won't happen at all. We want to be especially attentive to discoveries and decisions that we're making during data understanding that will impact deployment because we probably won't be around. Sure, we might be around at first, but what about a year from now or five years from now? Each of the tasks that we've discussed will have at least one aspect that will have a direct impact on deployment. Let's discuss some of the most important. Those same data sources needed for integration are going to be needed in production, as well, and the work that you did in describing the data could be used to flag out of range values, thereby removing that row of data during scoring. Some of the inputs are pulled in directly, of course, just like you did during data loading,…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.