Module nncombinator::layer

The various layers that make up a neural network and the traits they implement

Modules

• activation
  Implementation of Activation layers
• batchnormalization
  Batch normalization layer implementation
• bias
  Implementation of bias layer
• bridge
  Implementation of a layer for inverse transformation of the error type during back propagation
• input
  Implementation of Input layers
• linear
  Implementation of all full connected layers
• output
  Implementation of output layers

Enums

• DiffInput
  Differential input

Traits

• AddLayer
  Trait that defines the ability to add layers to a neural network.
• AddLayerTrain
  Trait that defines the ability to add a layer with learning capabilities to a neural network.
• AskDiffInput
  Trait that defines the function to query information to calculate the difference when applying the difference of neural networks.
• Backward
  Characteristics defining the internal implementation of the error back propagation method in neural networks
• BackwardAll
  Trait defining the implementation of error back propagation in neural networks
• BatchBackward
  Trait defining an implementation of error back propagation for neural networks with batch processing.
• BatchForward
  Trait defining the implementation of forward propagation of neural networks by batch processing.
• BatchForwardBase
  Trait defining the relevant type of implementation of forward propagation of neural networks by batch processing.
• BatchLoss
  Trait that defines the implementation of the process of calculating the loss during error back propagation of neural networks by batch processing.
• BatchPreTrain
  Trait that defines an implementation that calculates the results of forward propagation prior to the error back propagation process of a neural network through batch processing.
• BatchPreTrainBase
  Trait that defines the relevant type of implementation that calculates the results of forward propagation prior to processing the error back propagation of the neural network by batch processing.
• BatchTrain
  Trait that defines the implementation of neural network training by batch processing.
• Forward
  Trait defining the internal implementation of forward propagation of a neural network
• ForwardAll
  Trait defining the implementation of forward propagation of neural networks
• ForwardDiff
  Trait that defines the function of differential application of inputs in the process of forward propagation to neural networks.
• Loss
  Trait defining the calculation of the error during error back propagation.
• PreTrain
  Trait that defines the process of forward propagation performed prior to the process of error back propagation.
• Train
  Trait that defines the learning process of a neural network.
• TryAddLayer
  Trait defined functionality that attempts to add layers to a neural network.
• TryAddLayerTrain
  Trait that defines a function that seeks to add a learnable layer to a neural network