Class Optimizer

Namespace

NeuralNetworks.Optimizers

Assembly

NeuralNetworks.dll

Base class for a neural network optimizer.

public abstract class Optimizer

Inheritance

object

Optimizer

Derived

AdamOptimizer

GradientDescentMomentumOptimizer

GradientDescentOptimizer

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Constructors

Optimizer(LearningRate)

Base class for a neural network optimizer.

protected Optimizer(LearningRate learningRate)

Parameters

learningRate LearningRate

Properties

LearningRate

public LearningRate LearningRate { get; }

Property Value

LearningRate

Methods

Update(object, Span<float>, ReadOnlySpan<float>)

protected abstract void Update(object paramsKey, Span<float> paramsToUpdate, ReadOnlySpan<float> paramGradients)

Parameters

paramsKey object

paramsToUpdate Span<float>

paramGradients ReadOnlySpan<float>

UpdateLearningRate(int, int, int)

public virtual void UpdateLearningRate(int steps, int epoch, int epochs)

Parameters

steps int

epoch int

epochs int

Update<T>(T, T)

public void Update<T>(T paramsToUpdate, T paramGradients) where T : notnull

Parameters

paramsToUpdate T

paramGradients T

Type Parameters

T