Class LearningRateSchedulerBase

Namespace

AiDotNet.LearningRateSchedulers

Assembly

AiDotNet.dll

Base class for learning rate schedulers providing common functionality.

public abstract class LearningRateSchedulerBase : ILearningRateScheduler

Inheritance

object

LearningRateSchedulerBase

Implements

ILearningRateScheduler

Derived

ConstantLRScheduler

CosineAnnealingLRScheduler

CosineAnnealingWarmRestartsScheduler

CyclicLRScheduler

ExponentialLRScheduler

LambdaLRScheduler

LinearWarmupScheduler

MultiStepLRScheduler

OneCycleLRScheduler

PolynomialLRScheduler

ReduceOnPlateauScheduler

SequentialLRScheduler

StepLRScheduler

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Remarks

This abstract base class implements the common behavior for all learning rate schedulers, including state management, step tracking, and serialization support.

For Beginners: This is the foundation that all learning rate schedulers build upon. It handles the common tasks like keeping track of what step we're on and saving/loading state so that training can be resumed from a checkpoint.

Constructors

LearningRateSchedulerBase(double, double)

Initializes a new instance of the LearningRateSchedulerBase class.

protected LearningRateSchedulerBase(double baseLearningRate, double minLearningRate = 0)

Parameters

baseLearningRate double

The initial learning rate.

minLearningRate double

The minimum learning rate (floor). Default is 0.

Fields

_baseLearningRate

The base (initial) learning rate.

protected double _baseLearningRate

Field Value

double

_currentLearningRate

The current learning rate.

protected double _currentLearningRate

Field Value

double

_currentStep

The current step count.

protected int _currentStep

Field Value

int

_minLearningRate

The minimum learning rate (floor).

protected double _minLearningRate

Field Value

double

Properties

BaseLearningRate

Gets the base (initial) learning rate.

public double BaseLearningRate { get; }

Property Value

double

CurrentLearningRate

Gets the current learning rate.

public double CurrentLearningRate { get; }

Property Value

double

CurrentStep

Gets the current step (iteration or epoch count depending on scheduler type).

public int CurrentStep { get; }

Property Value

int

Methods

ComputeLearningRate(int)

Computes the learning rate for a given step.

protected abstract double ComputeLearningRate(int step)

Parameters

step int

The step number.

Returns

double

The computed learning rate.

GetLearningRateAtStep(int)

Gets the learning rate for a specific step without advancing the scheduler.

public virtual double GetLearningRateAtStep(int step)

Parameters

step int

The step number to get the learning rate for.

Returns

double

The learning rate at the specified step.

GetState()

Gets the scheduler state for serialization/checkpointing.

public virtual Dictionary<string, object> GetState()

Returns

Dictionary<string, object>

A dictionary containing the scheduler state.

LoadState(Dictionary<string, object>)

Loads the scheduler state from a checkpoint.

public virtual void LoadState(Dictionary<string, object> state)

Parameters

state Dictionary<string, object>

The state dictionary to load from.

Reset()

Resets the scheduler to its initial state.

public virtual void Reset()

Step()

Advances the scheduler by one step and returns the new learning rate.

public virtual double Step()

Returns

double

The updated learning rate for the next step.