optuna 4.0.0

A hyperparameter optimization framework

Stars: 10697, Watchers: 10697, Forks: 1016, Open Issues: 73

The optuna/optuna repo was created 6 years ago and the last code push was Yesterday.
The project is extremely popular with a mindblowing 10697 github stars!

How to Install optuna

You can install optuna using pip

pip install optuna

or add it to a project with poetry

poetry add optuna

Package Details

Author

Takuya Akiba

License

None

Homepage

None

PyPi:

https://pypi.org/project/optuna/

GitHub Repo:

https://github.com/optuna/optuna

Classifiers

• Scientific/Engineering
• Scientific/Engineering/Artificial Intelligence
• Scientific/Engineering/Mathematics
• Software Development
• Software Development/Libraries
• Software Development/Libraries/Python Modules

Related Packages

Errors

A list of common optuna errors.

Code Examples

Here are some optuna code examples and snippets.

GitHub Issues

The optuna package has 73 open issues on GitHub

• Make show_progress_bar tests simpler
• More control over how GridSampler prioritizes configurations after search space is exhausted
• Resuming Study
• integrate with Lightning ecosystem CI
• Update explanation for metrics of AllenNLPExecutor
• [RFC] Make trials editable after they are finished
• Investigate the necessary to use python 3.8 for documentation build
• Fix axis range issue in matplotlib contour plot
• When I retrain the model using the parameters of best_trial, I cannot achieve the F1 of optuna turning result
• Enable FloatDistribution and IntDistribution
• Introduce a rich benchmark environment in CI
• Fix type hints for Python 3.8.
• Handle -inf and inf values in RDBStorage
• Make tests of plot_optimization_history methods consistent
• [RFC] Extend the type annotation of catch argument of study.optimize

See more issues on GitHub