Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
athena documentation
athena documentation

Contents:

  • About
  • Installation
  • Tutorials
    • Building a Basic Network
    • Training a Model
    • Accessing Network Outputs
    • Saving and Loading Models
    • Inverse Design
    • MNIST Classification Example
    • Function Approximation (Regression)
    • Residual Networks (ResNet)
    • Message Passing Neural Networks
    • Physics-Informed Neural Networks
    • LNO Rollout Example
    • Inverse Design Example
    • wandb Logging and Hyperparameter Optimisation
    • Creating Custom Layers
    • Creating Custom Activation Functions
    • Creating Custom Initialisers
    • Creating Custom Optimisers
    • Creating Custom Loss Functions
  • Layers
    • Core Layers
      • Activation Layer
      • Fully-Connected Layer
    • Convolutional Layers
      • 1D Convolutional Layer
      • 2D Convolutional Layer
      • 3D Convolutional Layer
    • Input Layers
      • Input Layer
    • Merge Layers
      • Add Layer
      • Concatenate Layer
    • Message Passing Layers
      • Duvenaud Message Passing Layer
      • Kipf Message Passing Layer
    • Neural Operator Layers
      • Neural Operator Layer
      • Fixed Laplace Neural Operator Layer
      • Dynamic Laplace Neural Operator Layer
      • Spectral Filter Layer
      • Graph Neural Operator Layer
      • Orthogonal Neural Operator Block
      • Orthogonal Attention Layer
    • Normalisation Layers
      • 1D Batch Normalisation Layer
      • 2D Batch Normalisation Layer
      • 3D Batch Normalisation Layer
    • Padding Layers
      • 1D Padding Layer
      • 2D Padding Layer
      • 3D Padding Layer
    • Pooling Layers
      • 1D Average Pooling Layer
      • 2D Average Pooling Layer
      • 3D Average Pooling Layer
      • 1D Max Pooling Layer
      • 2D Max Pooling Layer
      • 3D Max Pooling Layer
    • Recurrent Layers
      • Recurrent Layer
    • Regularisation Layers
      • Dropout Layer
      • 2D DropBlock Layer
      • 3D DropBlock Layer
    • Reshaping Layers
      • Flatten Layer
      • Reshape Layer
  • Optimisers
    • SGD Optimiser
    • Adam Optimiser
    • RMSprop Optimiser
    • Adagrad Optimiser
  • Activation Functions
    • Gaussian Activation
    • Linear Activation
    • ReLU Activation
    • Leaky ReLU Activation
    • Sigmoid Activation
    • Tanh Activation
    • Softmax Activation
    • Swish Activation
    • SELU Activation
    • No Activation
  • Initialisers
    • Glorot Uniform Initialiser
    • Glorot Normal Initialiser
    • He Uniform Initialiser
    • He Normal Initialiser
    • LeCun Uniform Initialiser
    • LeCun Normal Initialiser
    • Zeros Initialiser
    • Ones Initialiser
    • Identity Initialiser
    • Gaussian Initialiser
  • Loss Functions
    • Binary Cross Entropy Loss
    • Categorical Cross Entropy Loss
    • Mean Absolute Error Loss
    • Mean Squared Error Loss
    • Negative Log Likelihood Loss
    • Huber Loss
  • Training Configuration
    • train() Subroutine
    • Network Modes
    • Gradient Clipping
    • Learning Rate Decay
    • Regularisation
  • Fortran API
Back to top
GitHub repository
View this page
Edit this page

Merge LayersΒΆ

  • add_layer_type - Element-wise addition of multiple inputs

  • concat_layer_type - Concatenation of multiple inputs along a dimension

Next
Add Layer
Previous
Input Layer
Copyright © 2026, athena-developers
Made with Sphinx and @pradyunsg's Furo