Hyperparameter Optimization

20 public types in this namespace, organized by kind.

Models & Types (13)

Type Summary
ASHAOptimizer<T, TInput, TOutput> Implements ASHA (Asynchronous Successive Halving Algorithm) for hyperparameter optimization.
BayesianOptimizer<T, TInput, TOutput> Implements Bayesian optimization for hyperparameter tuning using Gaussian Process regression.
BracketInfo Information about a Hyperband bracket.
EarlyStoppingState Represents the current state of early stopping.
EarlyStopping<T> Provides early stopping functionality for hyperparameter optimization and training.
GridSearchOptimizer<T, TInput, TOutput> Implements grid search hyperparameter optimization.
HyperbandOptimizer<T, TInput, TOutput> Implements Hyperband optimization for hyperparameter tuning with early stopping.
PopulationBasedTrainingOptimizer<T, TInput, TOutput> Implements Population-based Training (PBT) for hyperparameter optimization.
PopulationMemberInfo Information about a population member in PBT.
RandomSearchOptimizer<T, TInput, TOutput> Implements random search hyperparameter optimization.
RungStatistics Statistics for a single ASHA rung.
TrialPrunerStatistics Statistics about trial pruning.
TrialPruner<T> Provides trial pruning functionality for hyperparameter optimization.

Base Classes (1)

Type Summary
HyperparameterOptimizerBase<T, TInput, TOutput> Base class for hyperparameter optimization algorithms.

Enums (5)

Type Summary
AcquisitionFunctionType Types of acquisition functions for Bayesian optimization.
EarlyStoppingMode Mode for determining improvement in early stopping.
ExploitStrategy Strategy for exploiting better performers in PBT.
ExploreStrategy Strategy for exploring hyperparameters in PBT.
PruningStrategy Strategy for pruning trials.

Helpers & Utilities (1)

Type Summary
EarlyStoppingBuilder<T> Builder for configuring early stopping with fluent API.