Exploring the ML life cycle with Feast
In this section, let's discuss what ML model development looks like when you are using a feature store. We went through the ML life cycle in Chapter 1, An Overview of the Machine Learning Life Cycle. This makes it easy to understand how it changes with a feature store and enables us to skip through a few steps that will be redundant.
Problem statement (plan and create)
The problem statement remains the same as it was in Chapter 1, An Overview of the Machine Learning Life Cycle. Let's assume that you own a retail business and would like to improve the customer experience. First and foremost, you want to find your customer segments and customer lifetime value.
Data (preparation and cleaning)
Unlike in Chapter 1, An Overview of the Machine Learning Life Cycle, before exploring the data and figuring out the access and more, here the starting point for model building is...