RepConvNBlock<T>

Models & Types · AiDotNet.ComputerVision.Detection.Backbones

YOLOv9’s RepConvN at training time: a 3x3 Conv and a 1x1 Conv (each with batch norm, no activation) summed, then SiLU. Re-parameterization into one 3x3 conv is an inference-time optimization that computes the same function.

Inheritance: LayerBase<T> → RepConvNBlock<T>

Implements: IShapeContract

Constructors

Constructor Summary
RepConvNBlock(int outChannels) Creates the block.

Properties

Property Summary
bool SupportsTraining { get; }

Methods

Method Summary
IReadOnlyList<ILayer<T>> GetSubLayers()
IReadOnlyList<OutputAxisContract> OutputAxesFor(int inputRank)
void ResetState()
void SetTrainingMode(bool isTraining)
void UpdateParameters(T learningRate)