Hyperparameter Optimization for Machine Learning

Learn grid and random search, Bayesian optimization, multi-fidelity models, Optuna, Hyperopt, Scikit-Optimize & more.

Created by: Soledad Galli
Data scientist | Instructor | Software developer

Created by: Train in Data Team
Data scientists | Instructors | Software engineers

What Will I Learn?

• Hyperparameter tunning and why it matters
• Cross-validation and nested cross-validation
• Hyperparameter tunning with Grid and Random search
• Bayesian Optimisation
• Tree-Structured Parzen Estimators, Population Based Training and SMAC
• Hyperparameter tunning tools, i.e., Hyperopt, Optuna, Scikit-optimize, Keras Turner and others

Requirements

• Python programming, including knowledge of NumPy, Pandas and Scikit-learn
• Familiarity with basic machine learning algorithms, i.e., regression, support vector machines and nearest neighbours
• Familiarity with decision tree algorithms and Random Forests
• Familiarity with gradient boosting machines, i.e., xgboost, lightGBMs
• Understanding of machine learning model evaluation metrics
• Familiarity with Neuronal Networks

Target audience

• Students who want to know more about hyperparameter optimization algorithms
• Students who want to understand advanced techniques for hyperparameter optimization
• Students who want to learn to use multiple open source libraries for hyperparameter tuning
• Students interested in building better performing machine learning models
• Students interested in participating in data science competitions
• Students seeking to expand their breadth of knowledge on machine learning