1. Google Vizier A Python-based research interface for blackbox and hyperparameter optimization — Google's scalable service for tuning ML models at enterprise scale.
2. Hyperopt Distributed Asynchronous Hyperparameter Optimization — a Python library supporting random search, Tree of Parzen Estimators (TPE), and adaptive TPE.
3. Optuna An open source hyperparameter optimization framework to automate hyperparameter search — works with any machine or deep learning framework through a simple API.
4. scikit-optimize Sequential model-based optimization in Python — minimizes expensive and noisy black-box functions using techniques like Gaussian processes and gradient boosted trees.
5. Talos Radically changes the Keras, TensorFlow (tf.keras), and PyTorch workflow by fully automating hyperparameter tuning and model evaluation with minimal code changes.
6. Bayesian Optimization Pure Python implementation of Bayesian global optimization with Gaussian processes — enabling efficient, probabilistic search over continuous hyperparameter spaces.
7. KerasTuner Easy-to-use, scalable hyperparameter optimization framework for Keras — solves the pain points of hyperparameter search with built-in tuners and callbacks.
8. NNI (Microsoft) Automates feature engineering, neural architecture search, hyperparameter tuning, and model compression for deep learning — an open source toolkit from Microsoft.
9. Ray / Ray Tune A unified framework for scaling AI and Python applications — Ray consists of a core distributed runtime and the Ray AIR toolkit for simplifying ML compute and HPO.
10. SHERPA A Python library for hyperparameter tuning of machine learning models — supporting sequential, parallel, and population-based optimization strategies.
11. Polyaxon Supports random search, grid search, and provides a simple interface for advanced approaches such as Hyperband and Bayesian Optimization for ML experiment management.
12. mlmachine A Python library that organizes and accelerates notebook-based machine learning experiments — including pipelines, feature engineering, and hyperparameter tuning workflows.
13. Dragonfly Open source Python library for scalable Bayesian optimisation — designed for optimizing expensive black-box functions with support for multi-fidelity and multi-objective settings.
14. FLAML Economical hyperparameter tuning module from Microsoft — frees users from manually tuning hyperparameters by finding accurate ML models with low compute cost.
15. HEBO (Huawei Noah) Heteroscedastic Evolutionary Bayesian Optimisation — a state-of-the-art HPO library from Huawei Noah's Ark Lab, winner of multiple NeurIPS optimization challenges.
16. Nevergrad (Meta) A gradient-free optimization platform from Meta (Facebook Research) — providing a suite of evolutionary and derivative-free algorithms for hyperparameter and black-box optimization.
17. SigOpt Model development platform for tracking runs, visualizing training, and scaling hyperparameter optimization for any model type, library, or infrastructure.
18. ZOOpt Zeroth-order optimization that does not rely on the gradient of the objective function — instead learns from samples of the search space for noisy or non-differentiable objectives.
19. GPyOpt Python open-source library for Bayesian Optimization developed by the Machine Learning group at the University of Sheffield — built on top of GPy for Gaussian process modeling.
20. Spearmint Software package for Bayesian optimization — automatically runs experiments and adjusts parameters to minimize an objective function in as few runs as possible.