Class LogCoshLoss<T>

Namespace

AiDotNet.LossFunctions

Assembly

AiDotNet.dll

Implements the Log-Cosh loss function, a smooth approximation of Mean Absolute Error.

public class LogCoshLoss<T> : LossFunctionBase<T>, ILossFunction<T>

Type Parameters

T

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

Inheritance

object

LossFunctionBase<T>

LogCoshLoss<T>

Implements

ILossFunction<T>

Inherited Members

LossFunctionBase<T>.NumOps

LossFunctionBase<T>.Engine

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

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>)

Remarks

For Beginners: Log-Cosh loss is a smooth approximation of the Mean Absolute Error. It calculates the logarithm of the hyperbolic cosine of the difference between predictions and actual values.

This loss function has several desirable properties:

• It's smooth everywhere (unlike Huber loss which has a point where its derivative is not continuous)
• It's less affected by outliers than Mean Squared Error
• It behaves like Mean Squared Error for small errors and Mean Absolute Error for large errors
• Its derivatives are well-defined and bounded, which helps prevent gradient explosions during training

Log-Cosh loss is particularly useful for regression problems where:

• You want the smoothness of MSE but with better robustness to outliers
• You need stable gradients for model training
• You want a compromise between MSE and MAE

Methods

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

Calculates the derivative of the Log-Cosh loss function.

public override 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 Log-Cosh loss for each prediction.

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

Calculates the Log-Cosh loss between predicted and actual values.

public override 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 Log-Cosh loss value.

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

Calculates both Log-Cosh loss and gradient on GPU in a single efficient pass.

public override (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.