This project was part of a competition aimed at optimizing 8 black box functions using Bayesian Optimization over a period of 12 weeks. The code was submitted at the end of the competition, after the results had been evaluated.
The code revolves around utilizing scikit-learn's GaussianProcessRegressor, coupled with Optuna for hyperparameter tuning, and Dask for parallel processing across a cluster.
- Adam de Zoete (@zootle)
- This project is licensed under the MIT License - please see the license file for details.