Class LayerDimlp4

Inheritance Relationships

Base Type

Class Documentation

class LayerDimlp4 : public Layer

Represents a specialized neural network Dimlp layer using a staircase activation function, derived from the Layer class, with no weight or bias adaptation.

This class implements a layer in the neural network that uses a staircase activation function for its neurons. It overrides specific functions from the base Layer class to provide customized behavior for forward propagation without adapting weights or biases.

Public Functions

LayerDimlp4(float eta, float mu, float flat, int nbDown, int nbUp, int nbWeights, int nbWeightsForInit, int nbKnots)

Constructor for the LayerDimlp4 class.

Initializes the layer with the specified parameters and sets up the staircase activation function.

Parameters:
  • eta – Learning rate.

  • mu – Momentum.

  • flat – Flat spot elimination parameter.

  • nbDown – Number of neurons in the previous layer.

  • nbUp – Number of neurons in the current layer.

  • nbWeights – Total number of weights.

  • nbWeightsForInit – Number of weights for initialization.

  • nbKnots – Number of knots for the staircase activation function.