OptimizerGPU class abstract

Optimizers are used to update a parameters value to reduce the error of the system. For that the gradient of the requested value is used. Optimizers CAN be static so their execution tapes can be reused of they may require recompilation on every training step.

Implementers

• AdamGPU
• SGDGPU

Constructors

OptimizerGPU(List<GPUTensor> parameters, double learningRate)

Properties

hashCode → int

The hash code for this object.

no setter, inherited

learningRate ↔ double

getter/setter pair

parameters ↔ List<GPUTensor>

List of parameters the gradient is responsible for updating.

getter/setter pair

runtimeType → Type

A representation of the runtime type of the object.

no setter, inherited

Methods

noSuchMethod(Invocation invocation) → dynamic

Invoked when a nonexistent method or property is accessed.

inherited

refreshStep(CommandBuffer tape) → void

Called once before every replay of an already-compiled tape. Optimizers with no per-step state (SGD) can leave this as a no-op; Adam overrides it to patch its step count.

step(CommandBuffer tape) → void

Function to add an optimization step to a deferred execution tape.

toString() → String

A string representation of this object.

inherited

zeroGrad(CommandBuffer tape) → void

Function to reset the gradient of the parameter back to zero to make it ready for the next learning round.

Operators

operator ==(Object other) → bool

The equality operator.

inherited