Module optimizers

Trait and implementations for optimizers.

Structs

• AdaDelta
  The AdaDelta optimizer with a learning rate
• AdaDeltaBuilder
  Builder for AdaDelta.
• AdaGrad
  The Adagrad optimizer.
• AdaGradBuilder
  Builder for AdaGrad.
• Adafactor
  The Adafactor optimizer.
• AdafactorBuilder
  Builder for Adafactor.
• AdafactorState
  State of the Adafactor optimizer.
• Adam
  The Adam optimizer.
• AdamBuilder
  Builder for Adam.
• AdamW
  The AdamW optimizer [1].
• AdamWBuilder
  Builder for AdamW.
• Adamax
  The Adamax optimizer, a variant of Adam based on the infinity norm [1].
• AdamaxBuilder
  Builder for Adamax.
• Lion
  The Lion optimizer [1].
• LionBuilder
  Builder for Lion.
• RmsProp
  The RMSprop optimizer [1].
• RmsPropBuilder
  Builder for RmsProp.
• Sgd
  Stochastic gradient descent optimizer.
• SgdBuilder
  Builder for Sgd.

Traits

• Optimizer
  Trait for optimizers.
• OptimizerState
  Trait for optimizer states.

Functions

• clip_grad_norm
  Clips the global norm of the gradients

Type Aliases

• AdafactorBeta1
  Type alias for the beta1 used in Adafactor
• AdafactorBuilderBeta1
  Option<f32> Type alias for the beta1 used in Adafactor builder due to limitation in the generate_builder macro
• AdafactorBuilderLr
  Option<Array>. Type alias for the learning rate used in Adafactor builder due to limitation in the generate_builder macro
• AdafactorEps
  Type alias for the epsilon values used in Adafactor builder
• AdafactorLr
  Type alias for the learning rate used in Adafactor
• Betas
  (f32, f32O). Type alias for betas in the Adam/AdamW/Adamax optimizer builders due to limitation in the generate_builder macro
• MaybeClippedGrads
  Type alias for clipped gradients that is returned by clip_grad_norm.
• State
  Type alias for common optimizer state.