PatchEmbeddingLayer<T>
Layers · AiDotNet.NeuralNetworks.Layers
Implements a patch embedding layer for Vision Transformer (ViT) architecture.
Inheritance: LayerBase<T> → PatchEmbeddingLayer<T>
Implements: IShapeContract
For Beginners
This layer breaks an image into small square pieces (patches) and converts each patch into a numerical representation that can be processed by a transformer.
Think of it like cutting a photo into a grid of smaller squares, then describing each square with numbers. For example, a 224x224 image with 16x16 patches would be cut into 196 patches (14x14 grid), and each patch would be represented by a vector of numbers (the embedding).
This allows transformers, which were originally designed for text, to process images by treating the patches like “words” in a sentence.
How It Works
The patch embedding layer divides an input image into fixed-size patches and projects them into an embedding space. This is a key component of Vision Transformers, converting 2D spatial information into a sequence of embeddings that can be processed by transformer encoder blocks.
Constructors
| Constructor | Summary |
|---|---|
PatchEmbeddingLayer(int patchSize, int embeddingDim, IActivationFunction<T> activationFunction, IInitializationStrategy<T> initializationStrategy) |
|
PatchEmbeddingLayer(int patchSize, int embeddingDim, int expectedInputChannels, IActivationFunction<T> activationFunction, IInitializationStrategy<T> initializationStrategy) |
Eager-channel ctor: when the caller knows the input channel count at construction time (e.g. |
Properties
| Property | Summary |
|---|---|
bool SupportsTraining { get; } |
Indicates whether this layer supports training. |
Methods
| Method | Summary |
|---|---|
Dictionary<string, string> GetParameterRoles() |
Returns parameter roles for per-role optimizer configuration (e.g., weight decay exemption for biases). |
IReadOnlyList<OutputAxisContract> OutputAxesFor(int inputRank) |
|
IReadOnlyList<Tensor<T>> GetTrainableParameters() |
Returns all trainable parameter tensors marked with [TrainableParameter]. |
void ClearGradients() |
|
void ResetState() |
|
void SetTrainableParameters(IReadOnlyList<Tensor<T>> parameters) |
Replaces trainable parameter tensors (e.g., with ParameterBuffer views). |
void UpdateParameters(T learningRate) |
Updates the layer’s parameters using the calculated gradients. |
void UpdateParametersGpu(IGpuOptimizerConfig config) |
Updates layer parameters using GPU-resident optimizer. |
void ZeroGrad() |
Clears all gradient fields discovered by convention ({paramName}Gradient). |