►
Keras 3 API documentation /
KerasTuner API
KerasTuner API
The Hyperparameters class is used to specify a set of hyperparameters and their values, to be used in the model building function.
The Tuner subclasses corresponding to different tuning algorithms are called directly by the user to start the search or to get the best models.
The Oracle subclasses are the core search algorithms, receiving model evaluation results from the Tuner and providing new hyperparameter values.
The HyperModel subclasses are predefined search spaces for certain model families like ResNet and XceptionNet.
HyperParameters
- HyperParameters class
- Boolean method
- Choice method
- Fixed method
- Float method
- Int method
- conditional_scope method
- get method
Tuners
- The base Tuner class
- Objective class
- RandomSearch Tuner
- GridSearch Tuner
- BayesianOptimization Tuner
- Hyperband Tuner
- Sklearn Tuner
Oracles
- The base Oracle class
- @synchronized decorator
- RandomSearch Oracle
- GridSearch Oracle
- BayesianOptimization Oracle
- Hyperband Oracle