Interpreting and Evaluating Machine Learning Models
The promise and potential of machine learning systems to create systems that can make decisions without the need for hardcoded rules or heuristics is huge. However, this promise is often far from straightforward to fulfil, and in developing machine learning models or leading teams who develop machine learning models, great care needs to be taken to ensure their accuracy and reliability.
In this chapter, we will explore how to interpret and evaluate different machine learning models.
This is one of, if not the most important skill you can have in your toolkit as a decision-maker working on data science projects.
While it can be convenient to allow data scientists to evaluate their own models and “mark their own homework,” this is a risky decision to make and will, invariably, eventually lead to problems.
This chapter covers the following topics:
- How do I know whether this model will be accurate? ...
