← All tutorials

Skill Zuno · skillzunos.com · Python Data Science

Optuna tutorial

Learn Optuna on Skill Zuno. This Optuna tutorial covers New, Easy, Medium, Hard, and Diff levels so you can study practical Optuna problems on skillzunos.com.

Open Optuna on Skill Zuno

Also see the Optuna interview questions.

Optuna is a hyperparameter optimisation framework for Python that uses efficient search algorithms and pruning to find good settings fast. You write an objective function that suggests parameters and returns a score, and Optuna's define-by-run API lets the search space depend on earlier choices. Interviews cover studies and trials, samplers such as TPE, pruning, storage and parallel search, and visualising results.

Optuna topics covered

  • Studies, trials and objectives — Writing an objective function and running study.optimize.
  • Defining the search space — suggest_float, suggest_int and suggest_categorical, with log scales and conditionals.
  • Samplers — TPE, CMA-ES, random and grid samplers, and when to use each.
  • Pruning — Stopping unpromising trials early with MedianPruner and Hyperband.
  • Storage and parallelism — Saving studies in a database and running trials across processes and machines.
  • Analysing results — Optimisation history, parameter importance and the Optuna dashboard.

What you can learn

  • Optuna tutorial for beginners and experienced developers
  • Level-based practice: New, Easy, Medium, Hard, and Diff
  • Python Data Science topics alongside Optuna on skillzunos.com

More Python Data Science tutorials

All tutorials