Class LinearProjector<T>

Namespace

AiDotNet.SelfSupervisedLearning

Assembly

AiDotNet.dll

Linear projection head for self-supervised learning.

public class LinearProjector<T> : IProjectorHead<T>

Type Parameters

T

The numeric type used for computations (typically float or double).

Inheritance

object

LinearProjector<T>

Implements

IProjectorHead<T>

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Remarks

For Beginners: A linear projector is the simplest projection head - just a single linear transformation (matrix multiplication + bias). While simpler than MLP projectors, linear projectors can still be effective in some scenarios.

Architecture:

Input → Linear → Output
[d_in]   [d_out]

When to use Linear vs MLP:

• Use Linear for simplicity, lower compute, or when encoder is already powerful
• Use MLP for better downstream performance (recommended for most SSL methods)

Constructors

LinearProjector(int, int, bool, int?)

Initializes a new instance of the LinearProjector class.

public LinearProjector(int inputDim, int outputDim = 128, bool useBias = true, int? seed = null)

Parameters

inputDim int

Input dimension (encoder output size).

outputDim int

Output dimension (projection size).

useBias bool

Whether to include a bias term.

seed int?

Optional random seed for initialization.

Properties

HiddenDimension

Gets the hidden dimension (for MLP projectors).

public int? HiddenDimension { get; }

Property Value

int?

Remarks

Typical values: 2048-4096. Usually larger than output dimension.

InputDimension

Gets the input dimension expected by this projector.

public int InputDimension { get; }

Property Value

int

OutputDimension

Gets the output dimension produced by this projector.

public int OutputDimension { get; }

Property Value

int

Remarks

Typical values: 128-2048. SimCLR uses 128, MoCo uses 128, BYOL uses 256.

ParameterCount

Gets the total number of trainable parameters.

public int ParameterCount { get; }

Property Value

int

Methods

Backward(Tensor<T>)

Performs the backward pass through the projector.

public Tensor<T> Backward(Tensor<T> gradients)

Parameters

gradients Tensor<T>

The gradients from the loss with respect to projector output.

Returns

Tensor<T>

The gradients with respect to projector input (for encoder backprop).

ClearGradients()

Clears accumulated gradients.

public void ClearGradients()

GetParameterGradients()

Gets the gradients computed during the last backward pass.

public Vector<T> GetParameterGradients()

Returns

Vector<T>

A vector containing gradients for all parameters.

GetParameters()

Gets all trainable parameters of the projector.

public Vector<T> GetParameters()

Returns

Vector<T>

A vector containing all parameters.

Project(Tensor<T>)

Projects encoder output to the SSL embedding space.

public Tensor<T> Project(Tensor<T> input)

Parameters

input Tensor<T>

The encoder output tensor.

Returns

Tensor<T>

The projected embedding tensor.

Remarks

For Beginners: This transforms encoder features into a lower-dimensional space where the SSL loss is computed. The projection helps separate the pretraining objective from the learned representations.

Reset()

Resets the projector state (clears any internal buffers).

public void Reset()

SetParameters(Vector<T>)

Sets the parameters of the projector.

public void SetParameters(Vector<T> parameters)

Parameters

parameters Vector<T>

The parameter vector to load.

SetTrainingMode(bool)

Sets training or evaluation mode.

public void SetTrainingMode(bool isTraining)

Parameters

isTraining bool

True for training mode, false for evaluation.