A Gentle Introduction to Predictive Modelling
Introduction
1. Python Basics
Introduction to python
Working with DataFrames
2. Linear Models and Overfitting
Linear Models: a Brief Theory
Linear Model in action
The first Villain: overfitting
3. Unbiased predictive performance estimates
Training and Test sets
Cross-validation
4. Fighting Overfitting - The Advent of Machine Learning
Feature Reduction and Regularization: a brief theory
Regularized Models in Action
The Second Villain: Feature Leakage
5. Hyperparameter Optimization and Nested Cross-Validation
Cross-validation Revisited
Nested Cross-validation
6. Generalizability, Validity, Fairness
Generalizability
Validity and specificity
Fairness
7. Model Explanation
Explainability of “Classical” Models
Explainability vs. Black Box Models
8. Complex Models: from Ensembles to Deep Learning
9. Scientific Examples
References
.ipynb
.pdf
repository
open issue
Binder
Colab
Live Code
Explainability of “Classical” Models
Explainability of “Classical” Models
¶
previous
7. Model Explanation
next
Explainability vs. Black Box Models
