Applied Machine Learning: Ensemble Learning. Designed for those interested in strengthening their machine learning skills. This course covers the fundamental concepts of ensemble learning, with no formal scientific background required, and teaches techniques such as bagging, boosting, and stacking. Guided by instructor Matt Harrison, participants will learn how to implement these techniques with Python libraries such as scikit-learn and XGBoost, and by the end will be able to build efficient models for real-world problems. The course is integrated with GitHub Codespaces to provide a cloud-based development environment without the need for a local setup, allowing for hands-on practice anytime, anywhere. To learn more about this environment, read the “Using GitHub Codespaces” section.
What you will learn
- Collective Learning Concepts: Participants will be introduced to the definition of collective learning and the problem of overfitting, and understand its connection to the real world.
- Types of ensemble methods: This course examines types of ensemble methods such as bagging and random forests and teaches how to set parameters for random forests.
- Boosting and Gradient Boosting: The concepts of boosting, AdaBoost, and gradient boosting are covered along with hyperparameter tuning for them.
- XGBoost: Participants will learn the importance of XGBoost, learn how to code with it in practice, and tune hyperparameters for this model.
- Stacking: The concept of stacking, practical coding with StackingClassifier and evaluating it against individual models are taught.
- Implementation with Python: Throughout the course, participants will learn how to implement these methods using popular Python libraries such as scikit-learn and XGBoost.
- Solving Challenges: Practical challenges are provided for each section, helping participants put the concepts learned into practice.
This course is suitable for people who:
- Are eager to enhance their skills as a machine learning expert.
- Seek to understand and implement advanced collective learning methods.
- They want to enter the field without needing a formal background in data science.
- They plan to optimize their machine learning models for real-world tasks.
- Want to work with popular Python tools like scikit-learn and XGBoost.
Super Admin
