Base class for all layers. Do not instanciate. More...
Public Member Functions | |
| procedure, pass, public | get_num_parameters => layer_get_num_parameters |
| Returns the number of parameters. Implemented by layer_get_num_parameters. More... | |
| procedure, pass, public | set_parameters => layer_set_parameters |
| Setter for layer::parameters. Implemented by layer_set_parameters. More... | |
| procedure, pass, public | get_parameters => layer_get_parameters |
| Getter for layer::parameters. Implemented by layer_get_parameters. More... | |
| procedure, pass, public | tofile => layer_tofile |
| Saves the layer. Implemented by layer_tofile. More... | |
| procedure, pass, public | apply_forward => layer_apply_forward |
| Applies and linearises the layer. Implemented by layer_apply_forward. More... | |
| procedure, pass, public | apply_tangent_linear => layer_apply_tangent_linear |
| Applies the TL of the layer. Implemented by layer_apply_tangent_linear. More... | |
| procedure, pass, public | apply_adjoint => layer_apply_adjoint |
| Applies the adjoint of the layer. Implemented by layer_apply_adjoint. More... | |
Public Attributes | |
| integer(ik), public | input_size |
| The dimension of the input of the layer. More... | |
| integer(ik), public | output_size |
| The dimension of the output of the layer. More... | |
| integer(ik), public | batch_size |
| The batch size. More... | |
| integer(ik), public | num_parameters |
| The number of (trainable) parameters. More... | |
| class(linearactivation), allocatable, public | activation |
| The activation function. More... | |
| real(rk), dimension(:), allocatable, public | parameters |
| The vector of (trainable) parameters. More... | |
| real(rk), dimension(:, :), allocatable, public | forward_input |
| The storage for the layer::apply_forward method. More... | |
| real(rk), dimension(:, :), allocatable, public | tangent_linear_input |
| The storage for the layer::apply_tangent_linear method. More... | |
| real(rk), dimension(:, :), allocatable, public | adjoint_input |
| The storage for the layer::apply_adjoint method. More... | |
Detailed Description
Base class for all layers. Do not instanciate.
This class only exists to enable polymorphism.
Note
All attributes are public because they need to be accessed in subclasses and in networks. This could be avoided if all layers and networks would be part of the same module.
Member Function/Subroutine Documentation
◆ apply_adjoint()
| procedure, pass, public fnn_layer::layer::apply_adjoint |
Applies the adjoint of the layer. Implemented by layer_apply_adjoint.
◆ apply_forward()
| procedure, pass, public fnn_layer::layer::apply_forward |
Applies and linearises the layer. Implemented by layer_apply_forward.
◆ apply_tangent_linear()
| procedure, pass, public fnn_layer::layer::apply_tangent_linear |
Applies the TL of the layer. Implemented by layer_apply_tangent_linear.
◆ get_num_parameters()
| procedure, pass, public fnn_layer::layer::get_num_parameters |
Returns the number of parameters. Implemented by layer_get_num_parameters.
◆ get_parameters()
| procedure, pass, public fnn_layer::layer::get_parameters |
Getter for layer::parameters. Implemented by layer_get_parameters.
◆ set_parameters()
| procedure, pass, public fnn_layer::layer::set_parameters |
Setter for layer::parameters. Implemented by layer_set_parameters.
◆ tofile()
| procedure, pass, public fnn_layer::layer::tofile |
Saves the layer. Implemented by layer_tofile.
Member Data Documentation
◆ activation
| class(linearactivation), allocatable, public fnn_layer::layer::activation |
The activation function.
◆ adjoint_input
| real(rk), dimension(:, :), allocatable, public fnn_layer::layer::adjoint_input |
The storage for the layer::apply_adjoint method.
Array of reals with shape (layer::output_size, layer::batch_size).
Note
This array can be used by network classes to apply successive layers.
◆ batch_size
| integer(ik), public fnn_layer::layer::batch_size |
The batch size.
◆ forward_input
| real(rk), dimension(:, :), allocatable, public fnn_layer::layer::forward_input |
The storage for the layer::apply_forward method.
Array of reals with shape (layer::input_size, layer::batch_size).
Note
This array can be used by network classes to apply successive layers. In addition, this array is used to store the linearisation.
◆ input_size
| integer(ik), public fnn_layer::layer::input_size |
The dimension of the input of the layer.
◆ num_parameters
| integer(ik), public fnn_layer::layer::num_parameters |
The number of (trainable) parameters.
◆ output_size
| integer(ik), public fnn_layer::layer::output_size |
The dimension of the output of the layer.
◆ parameters
| real(rk), dimension(:), allocatable, public fnn_layer::layer::parameters |
The vector of (trainable) parameters.
All trainable parameters are stored into a single array of reals with shape (layer::num_parameters).
◆ tangent_linear_input
| real(rk), dimension(:, :), allocatable, public fnn_layer::layer::tangent_linear_input |
The storage for the layer::apply_tangent_linear method.
Array of reals with shape (layer::input_size, layer::batch_size).
Note
This array can be used by network classes to apply successive layers.
The documentation for this type was generated from the following file:
- src/fnn_layer.f90
