Building a machine learning project from setup to report
In this section, you will further improve the practical example of diamond cuts described in Chapter 3, Model Evaluation in Comet, and deployed in Chapter 6, Integrating Comet into DevOps. In this chapter, you will focus on the following aspects:
- Reviewing the scenario
- Selecting the best model
- Calculating the SHAP value
- Building the final report
Let’s start with the first step: reviewing the scenario.
Reviewing the scenario
As our use case, we will use the diamonds dataset provided by ggplot2 under the MIT licenses (https://ggplot2.tidyverse.org/reference/diamonds.html) and available on Kaggle as a CSV file (https://www.kaggle.com/shivam2503/diamonds). With respect to the original version, already described in Figure 3.3 in Chapter 3, we use the cleaned version produced in the same chapter and shown in the following figure:
Figure 8.6 – The cleaned version of...
