Master hyperparameter tuning for linear models, random forests, and gradient-boosting trees with grid & random search, Bayesian optimization, and multi-fidelity methods.
If you're disappointed for whatever reason, you'll get a full refund.
Sole is a lead data scientist, instructor, developer advocate, author, and open-source software developer. She created and maintains Feature-engine, a popular Python library that simplifies feature engineering and selection through a comprehensive collection of scikit-learn-compatible transformers. Since its launch, Feature-engine has grown into a widely adopted project supported by an active community of users and contributors.
Sole is also the author of three books published by Packt: Python Feature Engineering Cookbook, Feature Selection in Machine Learning with Python, and Imbalanced Data: Myths, Mistakes and Modern Solutions. Through her courses, books, and open-source work, she helps data scientists build more effective machine learning models and apply best practices to real-world projects.
Build better machine learning models through systematic, efficient hyperparameter optimization.
This practical Python course teaches you how to tune models for tabular data using cross-validation, Grid Search, Random Search, Bayesian optimization, multi-fidelity methods, and Optuna.
You will learn how to run these techniques, and also how they work, when to use them, and how to avoid unreliable or unnecessarily expensive experiments.
Hyperparameters control how a machine learning model learns. Examples include tree depth, the number of estimators, regularization strength, and learning rate.
Hyperparameter optimization is the process of finding the combination of values that produces the best model for a particular dataset and business objective.
An effective optimization workflow requires:
This course teaches you how to bring these components together in a practical workflow.
By the end of the course, you will be able to:
Along the way, you will explore the rationale, advantages, limitations, and practical considerations behind each method.
The course combines clear explanations with practical Python demonstrations and reusable Jupyter notebooks.
You will implement optimization workflows with popular open-source machine learning tools, including scikit-learn and Optuna. The examples focus on real model-tuning decisions rather than abstract theory, helping you transfer what you learn to your own datasets and projects.
No previous experience with Bayesian optimization, multi-fidelity optimization, or Optuna is required.
