You're reading from Interpretable Machine Learning with Python Learn to build interpretable high-performance models with hands-on real-world examples

Product type Paperback

Published in Mar 2021

Publisher Packt

ISBN-13 9781800203907

Length 736 pages

Edition 1st Edition

Languages

Python

Tools

TensorFlow

Concepts

Machine Learning

Author (1):

Serg Masís

View More author details

Table of Contents (19) Chapters

Preface

1. Section 1: Introduction to Machine Learning Interpretation

2. Chapter 1: Interpretation, Interpretability, and Explainability; and Why Does It All Matter? FREE CHAPTER

3. Chapter 2: Key Concepts of Interpretability

4. Chapter 3: Interpretation Challenges

5. Section 2: Mastering Interpretation Methods

6. Chapter 4: Fundamentals of Feature Importance and Impact

7. Chapter 5: Global Model-Agnostic Interpretation Methods

8. Chapter 6: Local Model-Agnostic Interpretation Methods

9. Chapter 7: Anchor and Counterfactual Explanations

10. Chapter 8: Visualizing Convolutional Neural Networks

11. Chapter 9: Interpretation Methods for Multivariate Forecasting and Sensitivity Analysis

12. Section 3:Tuning for Interpretability

13. Chapter 10: Feature Selection and Engineering for Interpretability

14. Chapter 11: Bias Mitigation and Causal Inference Methods

15. Chapter 12: Monotonic Constraints and Model Tuning for Interpretability

16. Chapter 13: Adversarial Robustness

17. Chapter 14: What's Next for Machine Learning Interpretability?

18. Other Books You May Enjoy

Leveraging SHAP's KernelExplainer for local interpretations with SHAP values

For this section, and for subsequent use, we will train a Support Vector Classifier (SVC) model first.

Training a C-SVC model

SVM is a family of model classes that operate in high-dimensional space to find an optimal hyperplane, where they attempt to separate the classes with the maximum margin between them. Support vectors are the points closest to the decision boundary (the dividing hyperplane) that would change it if were removed. To find the best hyperplane, they use a cost function called hinge loss and a computationally cheap method to operate in high-dimensional space, called the kernel trick, and even though a hyperplane suggests linear separability, it's not always limited to a linear kernel.

The scikit-learn implementation we will use is called C-SVC. SVC uses an L2 regularization parameter called C and, by default, uses a kernel called the radial basis function (RBF), which is...