TOOLORA / Local reading edition
Optuna
Optimize model hyperparameters with objective functions and dynamic search spaces, pruning, parallel trials and multiple objectives.
Overview
Optuna treats each configuration and evaluation as a trial defined by an objective function. Search spaces follow program logic; samplers select parameters and pruners stop unpromising trials. Studies support parallel, multiobjective and constrained optimization. It optimizes existing training workflows rather than supplying a foundation model or replacing data-split and metric design.
Official sources and references
- Official websitehttps://optuna.org/
- Documentationhttps://optuna.readthedocs.io/en/stable/
Descriptions summarize public information; consult official websites for features, pricing and regional availability.