Chapter 7: Practical Exposure to Using SHAP in ML
In the previous chapter, we discussed SHapley Additive exPlanation (SHAP), which is one of the most popular model explainability frameworks. We also covered a practical example of using SHAP for explaining regression models. However, SHAP can explain other types of models trained on different types of datasets. In the previous chapter, you did receive a brief conceptual understanding of the different types of explainers available in SHAP for explaining models trained on different types of datasets. But in this chapter, you will get the practical exposure needed to apply the various types of explainers available in SHAP.
More specifically, you learn how to apply TreeExplainers for explaining tree ensemble models trained on structured tabular data. You will also learn how to apply DeepExplainer and GradientExplainer SHAP with deep learning models trained on image data. As you learned in the previous chapter, the KernelExplainer in...