View source on GitHub
|
Additional optimizers that conform to Keras API.
Modules
average_wrapper module
conditional_gradient module: Conditional Gradient optimizer.
cyclical_learning_rate module: Cyclical Learning Rate Schedule policies for TensorFlow.
lamb module: Layer-wise Adaptive Moments (LAMB) optimizer.
lazy_adam module: Variant of the Adam optimizer that handles sparse updates more efficiently.
lookahead module
moving_average module
novograd module: NovoGrad for TensorFlow.
rectified_adam module: Rectified Adam (RAdam) optimizer.
stochastic_weight_averaging module: An implementation of the Stochastic Weight Averaging optimizer.
weight_decay_optimizers module: Base class to make optimizers weight decay ready.
yogi module: Yogi: Extension of yogi adaptive nonconvex optimizer in Keras.
Classes
class AdamW: Optimizer that implements the Adam algorithm with weight decay.
class AveragedOptimizerWrapper: Updated base class for optimizers.
class ConditionalGradient: Optimizer that implements the Conditional Gradient optimization.
class CyclicalLearningRate: A LearningRateSchedule that uses cyclical schedule.
class ExponentialCyclicalLearningRate: A LearningRateSchedule that uses cyclical schedule.
class LAMB: Optimizer that implements the Layer-wise Adaptive Moments (LAMB).
class LazyAdam: Variant of the Adam optimizer that handles sparse updates more
class Lookahead: This class allows to extend optimizers with the lookahead mechanism.
class MovingAverage: Optimizer that computes a moving average of the variables.
class NovoGrad: The NovoGrad Optimizer was first proposed in [Stochastic Gradient
class RectifiedAdam: Variant of the Adam optimizer whose adaptive learning rate is rectified
class SGDW: Optimizer that implements the Momentum algorithm with weight_decay.
class SWA: This class extends optimizers with Stochastic Weight Averaging (SWA).
class Triangular2CyclicalLearningRate: A LearningRateSchedule that uses cyclical schedule.
class TriangularCyclicalLearningRate: A LearningRateSchedule that uses cyclical schedule.
class Yogi: Optimizer that implements the Yogi algorithm in Keras.
Functions
extend_with_decoupled_weight_decay(...): Factory function returning an optimizer class with decoupled weight