Applied Machine Learning: Ensemble Learning

Course details

Do you want to grow your skills as a machine learning practitioner, but don’t know where to begin? You don’t need any formal training in data science to start working toward your goal. In this course, instructor Matt Harrison guides you through the key concepts of ensemble learning. Explore different ensemble methods like bagging, boosting, and stacking and learn to implement them using popular Python libraries such as scikit-learn and XGBoost. By the end of this course, you’ll be equipped with the skills you need to implement and optimize ensemble models in real-world machine learning tasks.

Learning objectives
Understand the fundamental concepts of ensemble learning, including bagging, boosting, and stacking, and their practical applications.
Gain hands-on experience implementing ensemble models such as random forest, AdaBoost, gradient boosting, XGBoost, and stacking using Python libraries.
Learn how to tune hyperparameters for different ensemble models to optimize predictive performance.

Concepts

0. Introduction

01 - Ensemble learning - Boost your models' performance
02 - How to use Codespaces

1. Introduction to Ensemble Learning

03 - Definition of ensemble learning and the problem of overfitting
04 - Real-world relevance
05 - Types of ensembles

2. Bagging and Random Forests

06 - Concept of bagging
07 - Random forest example
08 - Parameter tuning for Random Forest
09 - Challenge - Tune Random Forest parameters
10 - Solution - Tune Random Forest parameters

3. Boosting and Gradient Boosting

11 - Concept of boosting
12 - AdaBoost and gradient boosting
13 - Hyperparameter tuning for boosting
14 - Challenge - tune AdaBoost model
15 - Solution - tune AdaBoost model

4. XGBoost

16 - Why XGBoost
17 - Hands-on coding with XGBoost
18 - Hyperparameter tuning for XGBoost
19 - Challenge - Tune XGBoost model
20 - Solution - Tune XGBoost model

5. Stacking

21 - Concept of stacking
22 - Hands-on coding with StackingClassifier
23 - Evaluation of stacking vs individual models
24 - Challenge - Create a stacked model
25 - Solution - Create a stacked model

Conclusion

26 - Next steps