optuna-integration

Optuna-Integration

This package is an integration module of Optuna, an automatic Hyperparameter optimization software framework. The modules in this package provide users with extended functionalities for Optuna in combination with third-party libraries such as PyTorch, sklearn, and TensorFlow.

[!NOTE] You can find more information in our official documentations and API reference.

Installation

Optuna-Integration is available via pip and on conda.

# PyPI
$ pip install optuna-integration

# Anaconda Cloud
$ conda install -c conda-forge optuna-integration

[!IMPORTANT] As dependencies of all the modules are large and complicated, the commands above install only the common dependencies. Dependencies for each module can be installed via pip. For example, if you would like to install the dependencies of optuna_integration.botorch and optuna_integration.lightgbm, you can install them via:

$ pip install optuna-integration[botorch,lightgbm]

[!NOTE] Optuna-Integration supports from Python 3.8 to Python 3.12. Optuna Docker image is also provided at DockerHub.

Integration Modules

Here is the table of optuna-integration modules:

Third Party Library	Example
BoTorch	Unavailable
CatBoost	CatBoostPruningCallback
Dask	DaskStorage
FastAI	FastAIPruningCallback
Keras	KerasPruningCallback
LightGBM	LightGBMPruningCallback / LightGBMTuner
MLflow	MLflowCallback
PyTorch Distributed	TorchDistributedTrial
PyTorch Ignite	PyTorchIgnitePruningHandler
PyTorch Lightning	PyTorchLightningPruningCallback
pycma	PyCma Sampler
SHAP	SHAP Importance Evaluator
scikit-learn	OptunaSearchCV
skorch	SkorchPruningCallback
TensorBoard	TensorBoardCallback
tf.keras	TFKerasPruningCallback
Weights & Biases	WeightsAndBiasesCallback
XGBoost	XGBoostPruningCallback
AllenNLP*	AllenNLPPruningCallback
Chainer*	ChainerPruningExtension
ChainerMN*	ChainerMNStudy
MXNet*	Unavailable

[!WARNING] * shows deprecated modules and they might be removed in the future.

Communication

• GitHub Discussions for questions.
• GitHub Issues for bug reports and feature requests.

Contribution

Any contributions to Optuna-Integration are more than welcome!

For general guidelines how to contribute to the project, take a look at CONTRIBUTING.md.

Reference

If you use Optuna in one of your research projects, please cite our KDD paper "Optuna: A Next-generation Hyperparameter Optimization Framework":

BibTeX

@inproceedings{akiba2019optuna,
  title={{O}ptuna: A Next-Generation Hyperparameter Optimization Framework},
  author={Akiba, Takuya and Sano, Shotaro and Yanase, Toshihiko and Ohta, Takeru and Koyama, Masanori},
  booktitle={The 25th ACM SIGKDD International Conference on Knowledge Discovery \& Data Mining},
  pages={2623--2631},
  year={2019}
}