Optuna

Optuna is an open source hyperparameter optimization framework designed for automating the search for optimal hyperparameters in machine learning workflows. It is suitable for data scientists, ML engineers, and researchers who want to efficiently tune ML and DL models using state-of-the-art algorithms across a variety of frameworks, including PyTorch, TensorFlow, Keras, Scikit-learn, XGBoost, and LightGBM. Optuna provides advanced features such as parallelization, LLM-powered dashboard interactions, and support for multi-objective optimization via Gaussian process-based Bayesian optimization.