Class LossFunctionBase<T>

Namespace

AiDotNet.LossFunctions

Assembly

AiDotNet.dll

Base class for loss function implementations.

public abstract class LossFunctionBase<T> : ILossFunction<T>

Type Parameters

T

The numeric type used for calculations (e.g., float, double).

Inheritance

object

LossFunctionBase<T>

Implements

ILossFunction<T>

Derived

CIoULoss<T>

DETRSetLoss<T>

DIoULoss<T>

GIoULoss<T>

BinaryCrossEntropyLoss<T>

CTCLossAdapter<T>

CategoricalCrossEntropyLoss<T>

CharbonnierLoss<T>

ContrastiveLoss<T>

CosineSimilarityLoss<T>

CrossEntropyLoss<T>

DiceLoss<T>

ElasticNetLoss<T>

ExponentialLoss<T>

FocalLoss<T>

HingeLoss<T>

HuberLoss<T>

JaccardLoss<T>

KullbackLeiblerDivergence<T>

LogCoshLoss<T>

MarginLoss<T>

MeanAbsoluteErrorLoss<T>

MeanBiasErrorLoss<T>

MeanSquaredErrorLoss<T>

ModifiedHuberLoss<T>

NoiseContrastiveEstimationLoss<T>

OrdinalRegressionLoss<T>

PerceptualLoss<T>

PoissonLoss<T>

QuantileLoss<T>

QuantumLoss<T>

RealESRGANLoss<T>

RootMeanSquaredErrorLoss<T>

ScaleInvariantDepthLoss<T>

SparseCategoricalCrossEntropyLoss<T>

SquaredHingeLoss<T>

TripletLoss<T>

WassersteinLoss<T>

WeightedCrossEntropyLoss<T>

PhysicsInformedLoss<T>

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Extension Methods

LossFunctionExtensions.ComputeGradient<T>(ILossFunction<T>, Tensor<T>, Tensor<T>)

LossFunctionExtensions.ComputeLoss<T>(ILossFunction<T>, Tensor<T>, Tensor<T>)

Constructors

LossFunctionBase()

Initializes a new instance of the LossFunctionBase class.

protected LossFunctionBase()

Fields

NumOps

Provides operations for the numeric type T.

protected readonly INumericOperations<T> NumOps

Field Value

INumericOperations<T>

Properties

Engine

Gets the global execution engine for vector operations.

protected IEngine Engine { get; }

Property Value

IEngine

Methods

CalculateDerivative(Vector<T>, Vector<T>)

Calculates the derivative (gradient) of the loss function.

public abstract Vector<T> CalculateDerivative(Vector<T> predicted, Vector<T> actual)

Parameters

predicted Vector<T>

The predicted values from the model.

actual Vector<T>

The actual (target) values.

Returns

Vector<T>

A vector containing the derivatives of the loss with respect to each prediction.

CalculateLoss(Vector<T>, Vector<T>)

Calculates the loss between predicted and actual values.

public abstract T CalculateLoss(Vector<T> predicted, Vector<T> actual)

Parameters

predicted Vector<T>

The predicted values from the model.

actual Vector<T>

The actual (target) values.

Returns

T

The loss value.

CalculateLossAndGradientGpu(IGpuTensor<T>, IGpuTensor<T>)

Calculates both loss and gradient on GPU in a single pass. Default implementation falls back to separate calls.

public virtual (T Loss, IGpuTensor<T> Gradient) CalculateLossAndGradientGpu(IGpuTensor<T> predicted, IGpuTensor<T> actual)

Parameters

predicted IGpuTensor<T>

The predicted GPU tensor from the model.

actual IGpuTensor<T>

The actual (target) GPU tensor.

Returns

(T Loss, IGpuTensor<T> Gradient)

A tuple containing the loss value and gradient tensor.

ValidateVectorLengths(Vector<T>, Vector<T>)

Validates that the predicted and actual vectors have the same length.

protected void ValidateVectorLengths(Vector<T> predicted, Vector<T> actual)

Parameters

predicted Vector<T>

The predicted values vector.

actual Vector<T>

The actual values vector.

Exceptions

ArgumentException

Thrown when vectors have different lengths.