Class CosineSimilarityLoss<T>

Namespace

AiDotNet.LossFunctions

Assembly

AiDotNet.dll

Implements the Cosine Similarity Loss between two vectors.

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

Type Parameters

T

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

Inheritance

object

LossFunctionBase<T>

CosineSimilarityLoss<T>

Implements

ILossFunction<T>

Inherited Members

LossFunctionBase<T>.NumOps

LossFunctionBase<T>.Engine

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

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: Cosine Similarity measures how similar two vectors are in terms of their orientation, regardless of their magnitude (size).

The formula for cosine similarity is: cos(θ) = (A·B)/(||A||×||B||) Where:

• A·B is the dot product of vectors A and B
• ||A|| and ||B|| are the magnitudes (lengths) of vectors A and B
• θ is the angle between vectors A and B

The loss is calculated as 1 - cosine similarity, so:

• A value of 0 means the vectors are perfectly aligned (very similar)
• A value of 1 means they are perpendicular (no similarity)
• A value of 2 means they point in exactly opposite directions

Cosine similarity loss is particularly useful for:

• Text similarity tasks (comparing document vectors)
• Recommendation systems
• Image retrieval
• Any task where the direction of vectors matters more than their magnitude

It's often preferred over Euclidean distance when working with high-dimensional sparse vectors.

Constructors

CosineSimilarityLoss()

Initializes a new instance of the CosineSimilarityLoss class.

public CosineSimilarityLoss()

Methods

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

Calculates the derivative of the Cosine Similarity Loss with respect to the predicted values.

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

Parameters

predicted Vector<T>

The predicted vector from the model.

actual Vector<T>

The actual (target) vector.

Returns

Vector<T>

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

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

Calculates the Cosine Similarity Loss between two vectors.

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

Parameters

predicted Vector<T>

The predicted vector from the model.

actual Vector<T>

The actual (target) vector.

Returns

T

A scalar value representing the cosine similarity loss.