Exploring LinearExplainer in SHAP
LinearExplainer in SHAP is particularly developed for linear machine learning models. In the previous section, we have seen that although KernelExplainer is model-agnostic, it can be very slow. So, I think that is one of the main motivations behind using LinearExplainer to explain a linear model with independent features and even consider feature correlation. In this section, we will discuss applying the LinearExplainer method in practice. The detailed notebook tutorial is available at https://github.com/PacktPublishing/Applied-Machine-Learning-Explainability-Techniques/blob/main/Chapter07/LinearExplainers.ipynb. We have used the same Red Wine Quality dataset as used for the tutorial discussed in Chapter 6, Model Interpretability Using SHAP. You can refer to the same tutorial to learn more about the dataset as we will only focus on the LinearExplainer application part in this section.
Application of LinearExplainer in SHAP
For this example, we...