How-to Guides

Task-oriented guides for common workflows. Each guide assumes you have already completed the Getting Started tutorial.

Optimization

• Configure Samplers: choose and configure Optuna samplers, set seeds for reproducibility
• Use Callbacks: stop optimization early with trial limits, timeouts, or custom conditions
• Score Multiple Metrics: evaluate configurations against multiple scoring metrics simultaneously
• Visualize Results: plot optimization history and parameter importance with Optuna's visualization tools

Integration

• Use in Pipelines: wrap or embed OptunaSearchCV in Scikit-Learn pipelines
• Route Metadata: pass sample weights and other metadata through to estimators and scorers
• Persist and Resume Studies: save optimization state to a database and resume from prior trials

Debugging

• Handle Errors: control what happens when a hyperparameter combination causes fitting to fail
• Troubleshooting: solutions to common problems with installation, search, pipelines, and storage

Project

• Contributing: set up a development environment, run tests, and submit changes