Hyperparameter tuning is a way to find the best machine learning model. We make it ridiculously easy to run hyperparameter sweeps using simple algorithms like grid search, to more modern approaches like bayesian optimization and early stopping.